Home

Manual

1. 18 3 5 Grid caching GRIDSAVE NOGRIDSAVE 18 3 6 Grid symmetry GRIDSYM NOGRIDSYM 4 ever PULS ere rah Avie ea 185 Examples 19 1 Defining the input orbitals ORBITAL 19 2 Saving the localized orbitals SAVE 19 3 Choosing the localization method METHOD 19 4 Delocalization of orbitals DELOCAL 19 5 Localizing AOs LOCAO 19 6 Selecting the orbital space 19 6 1 Defining the occupied space OCC xi 93 93 93 93 93 94 94 95 96 96 96 97 97 97 97 97 97 98 98 98 98 99 99 100 100 100 100 100 101 101 101 101 102 103 103 103 103 104 104 105 106 106 CONTENTS xii 19 6 2 Defining the core orbitals CORE o 109 19 6 3 Defining groups of orbitals GROUP OFFDIAG 109 19 6 4 Localization between groups OFFDIAG enn 109 RO hee RR m Pub x dou m Seal eee E we 109 19 7 1 No reordering NOORDER 2l 110 19 7 2 Ordering using domains SORT leen 110 19 7 3 Defining reference orbitals REFORB o 110 19 7 4 Selecting the fock matrix FOCK llle 110 19 7 5 Selecting a density matrix DENSITY 111 19 8 Localization thresholds THRESH 111 19 9 Options for PM localization PIPEK 02 002 000 eee 111 19 10Printing options PRINT 4 4e 111 20 THE MCSCF PROGRAM MUL
2. Note that if the ci record is not found only an energy based optimization of the VB wavefunction can be carried out vb record name for the valence bond orbitals and structure coefficients as saved by a previous CASV B calculation If the VB wavefunction was previously saved in the AO basis the orbitals will be projected onto the present active space note that it is necessary to specify a record name for the molecular orbitals orb below for this to be possible orb record name for the molecular orbitals defining the CASSCF wavefunction This informa tion is necessary if one wants to output the valence bond orbitals in the atomic orbital basis trnint record name for the transformed CASSCF integrals These are required for the energy based criteria i e if CRIT ENERGY is specified and can be saved inside MULTI by the TRNINT sub command see 20 8 7 The default record name both here and in MULTI is 1900 1 36 6 Saving the VB wavefunction SAVE vb civb vbao vb record name for VB wavefunction default is first available record after 3200 2 i e orbitals and structure coefficients 36 THE VB PROGRAM CASVB 230 civb record name for valence bond full CI vector defined in terms of the CASSCF MOs default is 3300 2 Saving this vector is necessary for the calculation of further properties geometry optimization etc vbao record name for valence bond wavefunction in the AO basis Note that specifying orb in the S
3. 0 8papgq 34 In 1 2a z C DENSITY FUNCTIONAL DESCRIPTIONS 323 where q 1t x y 35 _ 95 Pa x 0 5 Xa 36 Pg 0 5 37 y 05 zp 37 online4 syntaxerror unexpected 38 z 2ur 39 Ps 0 5 4 r 0 5 x 40 online4 syntaxerror unexpected 41 dod 42 c 3 8 34 VTL 43 B 0 00375 44 and A 0 007 45 To avoid singularities in the limit p 0 G 0 01 p dz pU 22 Z 46 C 6 B88C Becke88 Correlation Functional Correlation functional depending on B86MGC exchange functional with empirical atomic pa rameters and u The exchange functional that is used in conjunction with B88C should replace B88MGC here See reference for more details In 1 K 0 8 Pappa _ alto 47 In 1 1 2 Y 0 01 p dz 2 int 2 where q 1 x y 48 E B Xo Pa x 0 5 coat e YE I 49 C DENSITY FUNCTIONAL DESCRIPTIONS sf ous Pavo ya di vn 1e Gg a t 0 63 z 2ur 059 elp UHA u 0 96 d u i455 c 3 8 W343 VT B 0 00375 and 0 007 To avoid singularities in the limit p 0 G otpa 1 gate C 7 B88 Becke88 Exchange Functional See reference for more details Borg K L p e eb am where c 3 8 34 P Vn and B 0 0042 To avoid singularities in the limit p 0 B px G p e Tio 6D x arcsinh a3 l 324 50 5
4. 119 O x y 2 3 p p os where o e Zp UP 11 8 C dZ Jp Jp 12 B 0 04918 13 A 0 132 14 c 0 2533 15 d 0 349 16 e 3 103 x 17 and F Z 5 a 18 C 2 B86MGC Xa y with Modified Gradient Correction B86 with modified gradient correction for large density gradients See reference for more details Bla o K Loya LFA 2 4 19 where c 3 8 34 Vx Q0 p 0 00375 21 and A 0 007 22 To avoid singularities in the limit p 0 4 3 B Xs G c p 23 C DENSITY FUNCTIONAL DESCRIPTIONS 322 C 3 B86R Xapy Re optimised Re optimised B of B86 used in part 3 of Becke s 1997 paper See reference 9 for more details o 1 B xs af K 24 L AQ C E where c 3 8 34 Y Q5 B 0 00787 26 and 0 004 27 To avoid singularities in the limit p 0 4 3 2 1 2 Xs CA B86 Xa y Divergence free semiempirical gradient corrected exchange energy functional A y in ref See reference for more details c ps ku 1 B Xs K 20 LTE RG id where c 3 8 W343 m 1 30 B 0 0076 31 and A 0 004 32 To avoid singularities in the limit p 0 E 4 3 1 2 g EP 1 BG T 1 1 xs C 5 B88CMASK Xq is the q component of an exchange functional with parameters and u to be used in conjunc tion with B88C See reference for more details Inf 0 q Y 0 01psdz 2 K
5. 39 2 17 Optimizing energy variables VARIABLE VARIABLE name 39 GEOMETRY OPTIMIZATION OPTG 264 Defines a variable name which holds the energy value to be optimized in using finite differences By default this is ENERGY 1 as set by the most recent program Other variables which can be used are ENERGY i holds last energy for state i ENERGR i holds last reference energy for state i ENERGD 1 holds last Davidson corrected energy for state i ENERGP i holds last Pople corrected energy for state i ENERGC holds CCSD QCI BCCD energy in CCSD T QCI T BCCD T calculations single state optimization ENERGT 1 holds CCSD T energy in CCSD T calculations single state ENERGT 2 holds CCSD T energy in CCSD T calculations single state ENERGT 3 holds CCSD T energy in CCSD T calculations single state These variables are set automatically by the CI and or CCSD programs It is the user s responsi bility to use the correct variable name an error exit occurs if the specified variable has not been defined by the last program or the user Note The use of the VARIABLE option triggers NUMERICAL so optimization can be very inefficient 39 2 18 Printing options PRINT PRINT code level Enables printing options Usually level should be omitted or 0 values of level gt 0 produce output useful only for debugging code can be HESSIAN prints the updated hessian matrix Note that its diagonal ele
6. Compute the gradient of the value of variable varname This implies numerical gradients The variable must be set in the corresponding energy calculation COORD ZMAT CART 3N coordinates with respect to which the gradient is evaluated See section 38 2 1 for more information DISPLACE ZMAT SYM UNIQUE CART Displacement coordinates to be used for numerical gradient The de fault is ZMAT if the geometry is given as a zmatrix which depends on variables and SYM symmetrical displacement coordinates other wise See section 38 2 1 for more information SYMMETRY AUTO NOSYM Symmetry to be used in wavefunction calculations of numerical AUTO ZMAT OPT3N RSTEP rstep DSTEP dstep ASTEP astep CENTRAL FORWARD FOURPOINT gradients This option is only relevant if DISPLACE UNIQUE CART If AUTO is given the maximum possible symmetry is used for each displacement This implies that the energy is independent of the sym metry used Note that this often not the case in MRCI or CASPT2 calculations The option can also not be used in local correlation cal culations logical Same as SYMMETRY AUTO logical Same as COORD ZMAT logical Same as COORD 3N Step length for distances in numerical gradient calculations in bohr The default is 0 01 Step length for symmetrical displacements in bohr The default is 0 01 Step length for angles in numerical gradient calculations in degree The defaul
7. DIAB orbref TYP E orbtype STATE state SP IN spin MS22ms2 SAVE orbsav ORB1 0rb1 ORB2 0rb2 PRINT pri METHOD method Here orbref is the record holding the orbitals of the reference geometry and orbsav is the record on which the new orbitals are stored If orbsav is not given recommended the new orbitals are stored in the default dump record 2140 2 or the one given on the ORBITAL directive see section 20 5 3 In contrast to earlier versions of MOLPRO it is possible that orbref and orbsav are the same The specifications TYPE STATE SPIN can be used to select specific sets of reference orbitals as described in section 4 11 orbl orb2 is a pair of orbitals for which the overlap is to be maximized These orbitals are specified in the form number sym e g 3 1 means the third orbital in symmetry 1 If orbl orb2 are not given the overlap of all active orbitals is maximized pri is a print parameter If this is set to 1 the transformation angles for each orbital are printed for each Jacobi iteration method determines the diabatization method method 1 default use Jacobi rotations method 2 use block diagonalization Both methods yield very similar results method 2 must only be used for CASSCF wavefunctions method 1 and method 2 as the positive values but AO overlap matrix of the current geometry is used This minimizes the change of the MO coefficients rather than maximizing the overlap to the neighbouring orb
8. N atom rhf nitrogen 14S state save orbitals to record 2110 on file 2 rhf for oxygen 3P state save orbitals to record 2120 on file 2 NO molecule read move move move move move orbitals of 291 c2v symmetry c2v symmetry atom ls orbital to output vector 1 1 2s orbital to output vector 3 1 2pz orbital 2px orbital 2py orbital examples sd no mergel com 2 3 to output vector to output vector to output vector move virtual orbitals of symmetry move virtual orbitals of symmetry move virtual orbitals of symmetry move virtual orbitals of symmetry read orbitals of O atom move all oxygen orbitals into place rotate 2s orbitals to make bonding and antibonding linear combinations rotate 2pz orbitals to make bonding and antibonding linear combinations set print option symmetrically orthonormalize the valence orbitals the resulting orbitals are printed save merged orbitals to record 2150 2 NO N M O A gI perform full valence casscf for NO 2Pix state 2Piy state start with merged orbitals One can also do the atomic calculations in the total basis set using dummy cards In this case the procedure is more complicated since the union of the two orbital spaces is over complete The calculation can be done as follows a SCF for the total molecule orbitals saved to 2100 2 b SCF for the N atom with dummy basis on the O atom orbi
9. 21 THE CI PROGRAM 137 GACPFE gacpfe Instead of diagonalizing the hamiltonian perform ACPF calculation or AQCC calculation Us ing the options GACPFI and GAPCPE The internal and external normalization factors gacpfi gacpfe may be reset from their default values of 1 2 nelec and 1 1 nelec 2 nelec 3 nelec nelec 1 respectively The ACPF and related methods are currently not robustly working for excited states Even though it sometimes works we do not currently recommend and support these methods for excited state calculations 21 3 3 Projected excited state calculations PROJECT record nprojc Initiate or continue a projected excited state calculation with information stored on record If nprojc gt 0 the internal CI vectors of nprojc previous calculations are used to make a projection operator If nprojc 1 this calculation is forced to be the first i e ground state with no projection If nprojc 0 then if record does not exist the effect is the same as nprojc 1 otherwise nprojc is recovered from the dump in record Thus for the start up calculation it is best to use project record 1 for the following excited calculations use project record At the end of the calculation the wavefunction is saved and the information in the dump record updated The project card also sets the tranh option so by default transition hamiltonian matrices are calculated For example to do successive calculations for three state
10. 28 8 2 The MULTP directive The MULTP directive turns on the multipole approximations for distant pairs as described in Ref 8 Further options can be given as described above for the LOCAL directive LOCAL MULTP options is equivalent to MULTP options The level of the multipole approximation can be chosen using option DSTMLT default 3 1 means dipole approximation 2 quadrupole approximation and so on 28 LOCAL CORRELATION TREATMENTS 188 The multipole approximation reduces the computational cost of LMP2 calculations for very large molecules but leads to some additional errors see Ref 8 It is normally not recom mended to be used in coupled cluster calculations and should never be used for computing in termolecular forces It can also not be used in geometry optimizations or gradient calculations 28 8 3 Saving the wavefunction SAVE The wavefunction can be saved for later restart using SAVE record where record has the usual form e g 4000 2 means record 4000 on file 2 If this directive is given the domain information as well as the amplitudes are saved for MPn the amplitudes are not saved If just the domain information should be stored the SAVE option on the LOCAL directive must be used cf section 28 3 28 8 4 Restarting a calculation START Local CCSD or QCISD calculations can be restarted using START record The record given must have been saved in a previous local calculation using the SAVE direct
11. 28 LOCAL CORRELATION TREATMENTS 184 28 6 1 Standard domains Standard domains are always determined first They are used to define strong close weak and distant pairs More accurate results can be obtained with extended domains as described in section 28 6 2 THRBP value CHGMIN value CHGMINH value CHGMAX value MAXBP maxbp MULLIKEN option MERGEDOM number Threshold for selecting the atoms contributing to orbital domains us ing the method of Boughton and Pulay BP As many atoms as needed to fulfill the BP criterion are included in a domain The order in which atoms are considered depends on the parameter MAXBP see below The default is THRBP 0 98 THRBP 1 0 includes all atoms into each orbital domain i e leads to full domains If no pairs are ne glected this should yield the canonical MP2 energy The criterion is somewhat basis dependent See section 28 9 4 for recommended values of this threshold determines the minimum allowed Mulliken or L wdin charge for an atom except H in a domain i e atoms with a smaller absolute charge are not included even if the THRBP criterion is not fulfilled default 0 01 as CHGMIN but used for H atoms default 0 03 If the atomic charge is larger than this value the atom is included independent of any ranking If maxbpz 1 the atoms are ranked according to their contribution to the Boughton Pulay overlap If maxbp 0 default the atoms are ranked accordi
12. H J Werner and W Meyer J Chem Phys 74 5794 1981 H J Werner Adv Chem Phys LXIX 1 1987 This program allows one to perform CASSCF as well as general MCSCF calculations For CASSCF calculations one can optionally use Slater determinants or CSFs as a N electron basis In most cases the use of Slater determinants is more efficient General MCSCF calculations must use CSFs as a basis A quite sophisticated optimization method is used The algorithm is second order in the orbital and CI coefficient changes and is therefore quadratically convergent Since important higher order terms in the independent orbital parameters are included almost cubic convergence is often observed For simple cases convergence is usually achieved in 2 3 iterations However convergence problems can still occur in certain applications and usually indicate that the active space is not adequately chosen For instance if two weakly occupied orbitals are of similar importance to the energy but only one of them is included in the active set the program might alternate between them In such cases either reduction or enlargement of the active orbital space can solve the problem In other cases difficulties can occur if two electronic states in the same symmetry are almost or exactly degenerate since then the program can switch from one state to the other This might happen near avoided crossings or near an asymptote Problems of this sort can be avoided by opt
13. e 127 20 9 1 Gradients for SA MCSCHF ee eee 128 20 9 2 Difference gradients for SA MCSCF ooo 128 20 9 3 Non adiabatic coupling matrix elements for SA MCSCH 128 RUE a P A Pd A Piae eee a 129 LETTURA DP PTT 129 bP ae a A a Bo ah a ae 129 131 iS E E GP chy BD Fe Gt take a ect NA UP D URN 131 TL C 132 p 132 21 2 2 Frozen coreorbital 22er 132 21 2 3 Closed shell orbitals 2l 132 NUR ad a Hee dow ADU A X WU A de wae es 132 21 2 5 Defining the statesymmetry o o 132 C he eae Bee ee ee 133 CT 133 CO 134 MONTES MM 135 olt dose eee ee dide n Dub Meat 135 ol pi pai o cani A 135 TUTUP 136 21 2 13 Restriction of classesofexcitatons 2l 136 Mer wm Pp ar a toe ew By ue Oa ae are dean oad p 136 da 136 21 3 2 Coupled Pair Functional ACPF AQCC 136 WR ITE ie atone ee 137 he O eae Gere ea 137 eae Wk ee eee E eee ae ee eee 137 21 36 Leyvelshits oo v uomo a x a Y e a 137 21 8 7 Maximum number of iterations llle 138 ae BE es a eeu 138 21 3 9 Selecting the primary configuration set rss 138 C e A N 138 pa de WD al Sh e ld Be 138 LL 139 A GR a a tee ani did Out Medi 139 21 3 14 Transition momentcalculations lll 139 La A a a an 4 139 21 3 16 Natural OrbitalsS ee 140 O A NN 140 E ae aio ao 141 21 4 Miscellane
14. y are evaluated for all pairs of u and v Default is NOSCORR The procedure is described by G Raos J Gerratt D L Cooper and M Raimondi Chem Phys 186 233 250 1994 ibid 251 273 1994 D L Cooper R Ponec T Thorsteinsson and G Raos Int J Quant Chem 57 501 518 1996 At present this analysis is only implemented for spin coupled wavefunctions 36 11 2 Printing weights of the valence bond structures For further details regarding the calculation of weights in CASVB see T Thorsteinsson and D L Cooper J Math Chem 23 105 26 1998 VBWEIGHTS keyl key2 Calculates and outputs weights of the structures in the valence bond wavefunction Pyg key specifies the definition of nonorthogonal weights to be used and can be one of CHIRGWIN Evaluates Chirgwin Coulson weights see B H Chirgwin and C A Coul son Proc Roy Soc Lond A201 196 1950 LOWDIN Performs a symmetric orthogonalization of the structures and outputs the corresponding weights INVERSE Outputs inverse overlap populations as in G A Gallup and J M Nor beck Chem Phys Lett 21 495 500 1973 ALL All of the above NONE Suspends calculation of structure weights The commands LOWDIN and INVERSE require the overlap matrix between valence bond struc tures and some computational overhead is thus involved 36 11 3 Printing weights of the CASSCF wavefunction in the VB basis For further details regarding the calcula
15. 42 6 Projecting orbitals PROJECT PROJECT namin file This command will read vectors from record namin file These vectors must have the same di mension as those of the current calculation All orbitals defined so far by the ORBI TAL MOVE and ADD directives are projected out of the input set The projected orbitals are then orthonor malized and moved to the undefined output vectors This should always yield a complete set of vectors 42 ORBITAL MERGING 289 42 7 Symmetric orthonormalization ORTH ORTH nj n ng Symmetrically orthonormalizes the first n vectors in each symmetry i These vectors must be supplied before by ORBITAL and MOVE or ADD directives 42 8 Schmidt orthonormalization SCHMIDT SCHMIDT nj no ng Schmidt orthonormalizes the first n vectors in each symmetry i These vectors must be supplied before by ORBI TAL and MOVE or ADD directives 42 9 Rotating orbitals ROTATE ROTATE iorbl sym iorb2 angle Will perform 2 x 2 rotation of orbitals iorb and iorb2 in symmetry sym by the specified angle in degree angle 0 means to swap the orbitals equivalent to angle 90 These vectors must be supplied before by ORBITAL and MOVE or ADD directives 42 10 Initialization of a new output set INIT INIT namout file Will initialize a new output set All previous vectors in the output set are lost unless they have been saved by a SAVE directive 42 11 Saving the merged orbitals SAVE
16. Print options Generally the value determines how much intermediate information is printed value 1 means no print default for all codes if value is omitted it is taken as zero which is usually appropriate Specification of higher values will generate more output The following codes are allowed ORBITAL Print molecular orbitals INTEGRAL Print integrals TIMING Print extra timing information DIAGONAL Print diagonal elements of Hamiltonian HAMILTONIAN Print much intermediate information 30 6 Interface to other programs DUMP causes the FCI diagonalization to be bypassed with input information and transformed inte grals being written to a formatted file FCIDUMP The format is as described in Comp Phys Commun 54 1989 75 31 SYMMETRY ADAPTED INTERMOLECULAR PERTURBATION THEORY 201 31 SYMMETRY ADAPTED INTERMOLECULAR PERTURBA TION THEORY 31 1 Introduction The SAPT symmetry adapted intermolecular perturbation theory program calculates the total interaction energy between closed shell molecules as a sum of individual first and second order pol 0 and dispersion B accompanied by their respective exchange counterparts EN ES a and Bc The latter ones arise due to electron exchange between the monomers when the molecules are close to each other and are sometimes denoted as Pauli repulsion Since all above terms are accessible through static and time dependent response density matrices of the monomers in princ
17. RS2 MIX nstates options STATE 1 istate for istate 2 nstates Further options can be given for instance a level shift At the end of each calculation the CASPT2 wavefunction is stored and at the end of the last CASPT2 calculation the Bloch Hamiltonian and the corresponding overlap matrix are automat ically assembled and printed The Hamiltonian is diagonalized after symmetrization following Brandow IJQC 15 207 1979 as well as with simple half sum averaging The MS CASPT2 energy and mixing coefficients printed in each case 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 149 The variable MSENERGY 1 with i 1 nstates is set to the multi state energies obtained with half sum diagonalization If a Level Shift is present MSENERGY 1 contains the multi state energies obtained with half sum diagonalization of the Bloch Hamiltonian whose diagonal ele ments CASPT2 energies have been corrected with the level shift Example SS SR CASPT 2 calculation for LiF SRevision 2006 1 r 3 4 5 6 7 8 9 10 ang i 1 geometry Li F l r i basis vtz F avtz hf Hartree Fock do i 1 4r loop over range of bond distances multi closed 3 0 0 0 occ 57 2270 state 2 ISA CASSCF for 2 states canonical ci examples rs2 MIX 2 INIT lif sr mscaspt2 com state 1 1 single state CASPT2 for reference state 1 el caspt2 i energy unmixed caspt2 energy for ground state rs2 MIX 2 state 1 2 single state
18. Some two electron transition properties for MCSCF wavefunctions e g L2 etc Population analysis Orbital localization Distributed Multipole Analysis A J Stone Automatic geometry optimization as described in J Comp Chem 18 1997 1473 Automatic calculation of vibrational frequencies intensities and thermodynamic proper ties Reaction path following as described in Theor Chem Acc 100 1998 21 Various utilities allowing other more general optimizations looping and branching e g for automatic generation of complete potential energy surfaces general housekeeping operations Geometry output in XYZ MOLDEN and Gaussian formats molecular orbital and frequency output in MOLDEN format Integral direct implementation of all Hartree Fock DFT and pair correlated methods MP CCSD MRCI etc as described in Mol Phys 96 1999 719 At present perturbative triple excitation methods are not implemented Local second order Mgller Plesset perturbation theory LMP2 and local coupled cluster methods as described in in J Chem Phys 104 6286 1996 Chem Phys Lett 290 143 1998 J Chem Phys 111 5691 1999 J Chem Phys 113 9443 2000 J Chem Phys 113 9986 2000 Chem Phys Letters 318 370 2000 J Chem Phys 114 661 2001 Phys Chem Chem Phys 4 3941 2002 Local density fitting methods as described in J Chem Phys 118 8149 2003 Phys Chem Chem Phys 5 3349 2003 Mol
19. These domains should then be reused in the subsequent calculations at all other intermolecular distances by using the START record option or the START directive see section 28 8 4 Only in this way the basis set superposition error is minimized and normally negligible of course this does not affect the BSSE for the SCF and therefore the basis set should be sufficiently large to make the SCF BSSE negligible 28 LOCAL CORRELATION TREATMENTS 195 Usually diffuse basis functions are important for obtaining accurate intermolecular interactions When diffuse basis sets are used it may happen that the Pipek Mezey localization does not yield well localized orbitals This problem can in most cases be overcome by using the directive PIPEK DELETE n as described in section 28 9 3 A final warning concerns local density fitting see sections 28 10 and 11 local fitting must not be used in counterpoise calculations since no fitting functions would be present on the dummy atoms and this can lead to large errors For examples and discussions of these aspects see Refs 21 23 28 10 Density fitted LMP2 DF LMP2 and coupled cluster DF LCCSD TO Density fitting LMP2 and LCCSD calculations can be performed by adding the prefix DF to the command name The input is as follows DF DF LMP 2 options LCCSD T options Options for density fitting can be mixed with any options for LOCAL Options for density fitting can also
20. erence 25 for more details K Y o RiS iXiYi 387 i 1 where n 19 388 Ri pa pg 389 e pm 558 oe 390 p vV 9aa Ogg x n Suu ud Bp l 391 pes Saa Ogg 2 Cua Op5 n D E 392 17 11 11 11 13 _ LU M ga L 393 t 7 6 413 3 2 5 3 7 312 513 7 513 7 2 53 2 7 6 4 3 3 2 5 3 75 393 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 394 v 0 0 0 0 1 1 1 1 2 2 2 0 0 0 0 0 0 0 0 395 w 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 396 and 0 142542 0 783603 0 188875 0 0426830 0 304953 0 430407 0 0997699 0 00355789 0 0344374 0 0192108 397 0 00230906 0 0235189 0 0331157 0 0121316 0 441190 2 27167 4 03051 2 28074 0 0360204 To avoid singularities in the limit p 0 Gea Y 1 2 0 ps ti Gss vi s wi ps im l 398 i 1 s C DENSITY FUNCTIONAL DESCRIPTIONS 348 C 37 TH4 Density an gradient dependent first and second row exchange correlation functional See refer ence 25 for more details K Y oiRiSiXiY 399 i l where n 19 400 po pg 401 S pm Pe 2 402 p vV 9aa Ogg x n Suu ud s 403 pes o Opp 2 Cua Opg y aa BB a OLOL pp n 404 p 13 t 7 6 4 3 3 2 5 5 31 s 1 ra l 53 1 T l 55 1 a La 11 6 4 3 3 2 5 3 pl 405 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 406 v 0 0 0 0 1 1 1 1 2 2 2 0 0 0 0 0 0 0 0 407 w
21. frozen MCSCF internal 16 CL 132 occupied 16 CL 132 FCI 199 MCSCE 113 ORBPERM ORBPRINT 98 125 ORBREL 233 ORTH BS 234 285 ORTHCON 234 O Q U 5 iwi PAIR 136 PAIRS 136 34 Parallel 2 PARAM 141 Plotting 73 POLARIZABILITY POP 209 population analysis 360 POTENTIAL 100 PRINT ITT I25 142 200 207 235 264 PROC 32 Procedures program structure 12 PROJECT 137 288 properties 206 CL 139 MCSCF PROPERTY 206 pseudopotential P SPACE 117 138 PUNCH 40 PUT QCI T61 oc1 131 QUAD 210 QUAD 10 quadrupole field RADIAL 102 RADIUS 209 reaction path 256 263 READ 230 READPUN 12 READVAR records 13 REF 133 References REF STATE 135 REL 212 Relativistic corrections 212 RELAX 164 RESTART 14 28 RESTRICT 115 134 RHF 90 RHF SCF 90 RI MP2 158 RKS 99 RKS SCF 99 ROOT 262 ROTATE 96 118 289 RS2 146 RS2 146 RS2C 146 RS3 146 RS3 146 Running MOLPRO SADDLE 231 SAMC 245 INDEX SAPT 201 save 93 08 TS 179 138 2291255 SCALE 244 SCF 90 SCHMIDT 289 SCORR 235 SCS MP2 159 SELECT 116 133 SERVICE 236 SET 41 HIFT D7 137 SHOW 53 sorted integrals SPECIAL 237 pecial Variables SPIN 16 SPINBASTS 229 START 94 117 139 229 23
22. needs not to be given CPP compute core polarization potential integrals HF RHF HF SCF or RHF SCF calls spin restricted Hartree Fock program open or closed shell UHF or UHF SCF calls spin unrestricted Hartree Fock program DFT calls the density functional program KS RKS call the Kohn Sham spin restricted density functional program UKS call the Kohn Sham spin unrestricted density functional program MULTI MCSCF or CASSCF calls MCSCF CASSCF program CASVB calls the CASVB valence bond program CI MRCI orCI PRO calls internally contracted MRCI program CIPT2 calls internally contracted CIPT2 program ACPF AQCC calls internally contracted MR ACPF program CEPA calls single reference CEPA program closed or open shell RS2 RS3 calls internally contracted multireference perturbation theory RS2C faster program for internally contracted multireference perturbation theory MP2 calls closed shell MP2 program MP3 calls closed shell MP3 program MP4 calls closed shell MP4 program CISD calls closed shell CISD program CCSD calls closed shell coupled cluster program BCCD calls closed shell Brueckner CCD program QCI QCSID calls closed shell quadratic configuration interaction program UCCSD calls spin unrestricted open shell coupled cluster program RCCSD calls spin restricted open shell coupled cluster program FCI or FULLCI calls determinant based full CI program Local correlation methods LMP2 calls closed shell local MP2 pro
23. value 1 means no print default for all codes In some of the cases listed below the speci fication of higher values will generate even more output than described The equal signs and zeros may be omitted All codes may be truncated to three characters The following codes are allowed max 7 per card ORBITALS print orbitals JOP 0 print operator list JOP 1 print coulomb operators in MO basis JOP 2 print coulomb operators in AO and MO basis KOP as JOP for internal exchange operators KCP 0 print paging information for CIKEXT 21 THE CI PROGRAM KCP 1 KCP 2 GSS 0 GSS 1 DPQ EPO HPO DPI DSS 143 print external exchange operators in MO basis print operators in AO and MO basis print paging information for CIDIMA print density matrix in MO basis print density matrix in AO and MO basis print energy denominators for pairs in addition print diagonal coupling coefficients in orthogonal basis print operators FPP print update information for pairs in each iteration print pair matrix updates MO basis in addition print pair matrices MO basis print CP in AO basis in CIKEXT print convergence information for internal CI print internal CI coefficients and external expansion coefficients as CP for singles print paging information for CICPS print matrices CPS in MO basis print paging information for CIGPQ print matrices GP at exit of CIGPQ print pagin
24. 0 6251173 0 6257931 0 6274755 0 ESCE 78283137 94445391 94512967 94681207 24 7 Saving the density matrix DM record ifil loop over requested methods perform calculation for given methods save energy in variable e lend loop over methods print a table with results title of table The effective first order density matrix is computed an stored in record record on file ifil This currently works for closed shell MP2 QCISD and CCSD See also NATORB 24 8 Natural orbitals NATORB RECORD record ifil PRINT nprint CORE 2natcor Calculate natural orbitals This currently only works for closed shell MP2 and QCISD The number of printed external orbitals in any given symmetry is nprint default 2 nprint 1 suppressed the printing The natural orbitals and the density matrix are saved in a dump record record on file ifil If record ifil is specified on a DM card see above this record is used If different records are specified on the DM and NATORB cards an error will result The record can also be given on the SAVE card Note that the effective density matrix of non variational methods like MP2 or QCISD does not strictly behave as a density matrix For instance it has non zero matrix elements between core and valence orbitals and therefore core orbitals are affected by the natural orbital transformation Also occupation numbers of core orbitals can be larger than 2 0 If CORE is given nat
25. 0 all orbitals of that symmetry are printed 43 20 Assigning matrix elements to a variable ELEM ELEM name matrix col row assigns elements col row of matrix to variable name col and row must be given in the form number isym where number is the row or column number in symmetry isym The product of the row and column symmetries must agree with the matrix symmetry 43 MATRIX OPERATIONS 299 43 21 Reading a matrix from the input file READ READ name TYPE type SUBTY PE subtype S YM symmetry FILE file values Reads a square matrix symmetry 1 from input or an ASCII file The values can be in free format but their total number must be correct Comment lines starting with or are skipped If the data are given in input the data block must be enclosed either by curley brackets or the first linbe must be BEGIN DATA and the last line END DATA If a filename is specified as option the data are read from this file In this case the BEGIN DATA END DATA lines in the file are optional and no data block must follow For compatibility with older versions the data can also be included in the input using the INCLUDE command see section 3 Ip In this case the include file must contain the BEGIN_DATA and END_DATA lines this is autopmatically the case if the file has been written using the MATROP WRITE directive type is a string which can be used to assign a matrix type If appropriate this
26. 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 408 and Q 0 0677353 1 06763 0 0419018 0 0226313 0 222478 0 283432 0 0165089 0 0167204 0 0332362 0 0162254 0 000984119 0 0376713 409 0 0653419 0 0222835 0 375782 1 90675 3 22494 1 68698 0 0235810 To avoid singularities in the limit p 0 G Paya yan oo 410 C DENSITY FUNCTIONAL DESCRIPTIONS 349 C 38 THGFCFO Density and gradient dependent first row exchange correlation functional FCFO FC open shell fitting See reference for more details n K OR S X Y 411 I where n 20 412 Ri Pa pg 413 2u 414 P Sua Opg x 2 See vei 415 py o Ogg 2 Cuan O n TW lem 416 p 11 11 11 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 312 513 72 302 513 2 716 413 3 2 5 3 417 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 418 v 0 0 0 0 1 1 1 1 2 2 2 2 0 0 0 0 0 0 0 0 419 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 420 and 0 864448 0 565130 1 27306 0 309681 0 287658 0 588767 0 252700 0 0223563 0 0140131 0 0826608 0 0556080 0 00936227 0 00677146 0 0515199 0 0874213 0 0423827 0 431940 0 691153 0 637866 1 07565 421 To avoid singularities in the limit p 0 G Y 1 20 Ps V6 7 m i Ps 499 TE 422 i l 5 C DENSITY FUNCTIONAL DESCRIPTIONS 350
27. 0 00000000 0 65349466 0 00000000 0 68871635 0 00000000 0 88758090 25 4 3 Calculate an EOM CCSD state most similar to a given CIS state This example shows how to force the convergence of the EOM CCSD program to a state which resembles at most a given CIS state EOM CCSD vector following procedure memory 2 m basis avdz geometry h f h r r 0 92 Ang hf save 2100 2 cis 4 4 exfile 6000 2 ccsd save 4000 2 om 4 4 checkovlp 1 exfile 6000 2 eompar inisingl 200 inidoubl 0 ccsd start 4000 2 eom 2 4 follow 2 exfile 6000 2 checkovlp 1 eompar inisingl 200 inidoubl 0 eomprint loce 1 define basis set z matrix define distance do SCF calculation save orbitals do CIS calculation save amplitudes do CCSD calculation save amplitudes do EOM CCSD calculation check overlap of singles with CIS vectors Stored in record given in exfile examples for first approximation take 2Bfeoenivgirecdl8F of approximate hamiltonian do CCSD calculation try to restart do EOM CCSD calculation for state closest to 2 4 CIS state check overlap of singles with CIS vectors stored in exfile print overlaps of sample and EOM vectors in each iteration In this example the CIS state 2 4 corresponds to the EOM CCSD state 1 4 25 5 Excited states with CIS Excitation energies can also be calculated using the Configuration Interaction Singles CIS method This method cannot be expected to give accurate results but can be u
28. 1 QI text mrci bee T2424 ci Ci wf 7 2 1 noexc save 5000 Ci wf 7 3 1 noexc save 5100 Ci wf 7 5 1 noexc save 5200 Ci wf 7 2 1 noexc save 5010 Ci wf 7 3 1 noexc save 5110 Ci wf 7 5 1 noexc save 5210 2 2 2 2 2 2 hlsmat ecp 5000 2 5100 2 5200 2 hlsmat ecp 5010 2 5110 2 5210 2 hlsmat ecp 6000 2 6100 2 6200 2 hlsmat ecp 6010 2 6110 2 6210 2 scf cass laver save save save mrci mrci mrci cass laver for 2Pz cf with minmal age 2P states casscf vector casscf vector casscf vector for 2Px state for 2Py state for 2Pz state cf with larger age 2P states active space for 2Px state for 2Py state for 2Pz state active space Ido spin orbit calculations 243 lokal term 0 s terme p terms with we d terms with we f terms with wei ECP SO for p ter ECP SO for d te ECP SO for f ter examples l ecp com 38 ENERGY GRADIENTS 244 38 ENERGY GRADIENTS 38 1 Analytical energy gradients MOLPRO uses two different gradient programs The CADPAC gradient program is based on the CADPAC integral routines by R D Amos Cur rently this program works for closed shell SCF high spin RHF and state averaged MCSCF In the MCSCF case the wavefunction must either be fully optimized or frozen core orbitals must be taken from a closed shell SCF calculation but this does not work in the case of state averaged MCSCP Note that CADPAC does not wo
29. 1 19802939 2 3 2 72079843 4 2 1 03813792 2 5 6 40131754 2 1 01158599 3 5 6 21328827 2 2 04193864 2 5 19 11604172 2 1 99631017 3 5 19 08465909 4 2 2 64971585 3 7 24 79106489 2 2 75335574 4 7 24 98147319 25 0 49970082 3 7 0 27936581 2 0 79638982 4 7 0 70184261 4 2 2 99860773 2 3 81 88444526 2 3 01690894 2 3 83 41280402 2 1 59415934 2 3 2 32392477 2 1 19802939 2 3 2 72079843 4 2 1 03813792 2 5 6 40131754 2 1 01158599 2 5 6 21328827 2 2 04193864 2 5 19 11604172 2 1 99631017 2 5 19 08465909 4 2 2 64971585 2 7 24 79106489 2 2 75335574 2 7 24 98147319 25 0 49970082 2 7 0 27936581 2 0 79638982 2 7 0 70184261 1 Iodine basis 1 s I 0 2027624 0 4080619 0 8212297 1 6527350 3 3261500 c 1 5 0 4782372 0 5811680 0 2617769 0 4444120 0 1596560 s 1 0 05 0 1007509 p 1 0 2027624 0 4080619 0 8212297 1 6527350 3 3261500 c 1 5 0 4251859 0 2995618 0 0303167 0 2064228 0 0450858 p 1 0 05 0 1007509 0 01 diffuse p Funktion wegen evt neg Part Ldg Gy L 0 22 044 f 1 0 3 1 htf ocGC Ll l Ll L wfyT 5 14 mn lti occ lyl 1 1 wf 7 2 1 wf 7 3 1 wf 7 5 1 Ci wf 7 2 1 save 6000 2 ci wf 7 3 1 save 6100 2 ci wf 7 5 1 save 6200 2 muluti occ lRZ 2 2 wf 7 2 1 wf 7 3 l wf 7 5 1 ci wf 7 2 1 save 6010 2 ci wf 7 3 1 save 6110 2 ci wf 7 5 1 save 6210 2 text casscf oce ply ty Lyt ci text casscf DOO TZ 2 42 el text mrci occ 1 1 1
30. 131729 030878 022829 1 009513 C 3 4 24645 792024 225006 12 34325 4 20192 1 379825 383453 1 044694 212106 453423 533465 99 2 2 2 2 2 2 2 99 wo ri bos 0 0O OQU O 0 errre Dos NO W end rhf el energy trhf occ 4 1 1 1 1 1 1 c10sed 4 1 1 1 1 1 wf 19 7 1 7 e2 energy de e2 el toev Delta E 0 075 eV 14 4 Example for ECP input from library 15 CORE POLARIZATION POTENTIALS 87 AuH CCSD T binding energy of the AuH molecule at r exp using the scalar relativistic 19 valence electron pseudopotential of the Stuttgart Koeln group gprint basis orbitals geometry au basis ecp au ECP 60MWB ECP input spd au ECP60MWB c 1 2 basis set f au 1 41 0 47 0 15 g au 1 2 0 4 spd h avtz c examples rhf auh ecp lib com treocsqd b s coreae l l l 1 3 el energy geometry h rhf e2 energy rAuH 1 524 ang molecular calculation geometry au h au rAuH hf ccsd t core 2 1 1 e3 energy de e3 e2 e1 toev binding energy 3 11 eV 15 CORE POLARIZATION POTENTIALS 15 1 Input options The calculation of core polarization matrix elements is invoked by the CPP card which can be called at an arbitrary position in the MOLPRO input provided the integrals have been calculated before The CPP card can have the following three formats e CPPINIT ncent
31. 245 215 1995 The shift can be specified on the RS2 or RS2C card RS2 Gn SHIF T shift RS2C Gn LSHIFT shift Typical choices for the shift is are 0 1 0 3 Only two figures after the decimal point are considered The shift affects the results the printed energies as well as the ENERGY variable include the energy correction for the shift as proposed by Roos and Andersson At convergence also the uncorrected energies are printed for comparison 22 6 Integral direct calculations RS2 RS2C and RS3 calculations with very large basis sets can be performed in integral direct mode The calculation will be direct if a global DIRECT or GDIRECT card appears earlier in the input Alternatively mainly for testing DIRECT can be specified as an option on the RSn C card RS2 Gn LSHIFT shift DIRECT RS2C Gn L SHIFT shift DIRECT 22 7 CASPT2 gradients P Celani and H J Werner J Chem Phys 119 5044 2003 CASPT analytic energy gradients are computed automatically if a FORCE or OPTG command follows see sections and 39 Analytical gradients are presently only available for RS2 calculations not RS2C and only for the standard 9 not G1 G2 etc Gradients can be computed for single state calculations as well as multi state MS MR CASPT2 see section 22 3 In single state calculations the gradient is automatically computed for the state computed in CASPT2 RSPT2 i e using STATE 1 2 the second state in the s
32. A 2 Installation of pre builtbinaries lens A 3 Installation from sourcefiles A 3 1 Overview 22r s IA CREW OTHO SN SEE dhe dnd dnp de aaa ake aaa hihi wiped FEET MM aia Pa a aaa balsa MM prono Up A 3 10 Getting and applying patches o o o A 3 11 Installation of documentation Recent Changes B 1 New features of MOLPRO2006 1 B 2 New features of MOLPRO2002 6 B 3 New features of MOLPRO2002 B 4 Features that were new in MOLPRO2000 B 5 Facilities that were new in MOLPRO98 C Density functional descriptions C l ALYP Lee Yang and Parr Correlation Functional C 3 B86R Xa y Re optimised A AA ARA Se cae Ua WES A ini NA A E AR T C 6_B388C Becke88 Correlation Functional C 7 B88 Becke88 Exchange Functional o o oo C 8 B95 Becke95 Correlation Functional o C 9 B97R Density functional part of B97 Re parameterized by Hamprecht et al C 10 B97 Density functional part of B97 22e C 11 BR Becke Roussel Exchange Functional o C 12 BRUEG Becke Roussel Exchange Functional Uniform Electron Gas Limit C 13 BW Becke Wigner Exchange Correlation Functional C 14 CS1 Colle Salvetti correlation function
33. BRUECKNER CANONICAL 120 CANORB 120 CASPROJ 31 CASSCF 112 27 CASVB 227 CCSD 160 CCSD 131 160 CCSD T CEPA 136 CHARGE 16 CHECK 160 164 CL 131 CI 131 CI PRO 131 CIS 168 CIS 168 CISD 162 CISD I31 CIWEIGHTS 235 CLEAR 53 CLEARALL 53 356 CLOSED 16 93 114 132 COEFF S 233 COMPRESS 69 CON 1 16 135 229 CONF IG 122 CONICAL 264 COORD 297 coordinates 257 B matrix 257 cartesian 257 natural internal Z Matrix COPT 126 CORE 16 109 132 199 COSMO 234 Cowan Griffin CPMCSCF 128 CPP CRD 73 CRIT 231 CUBE CUT 262 Darwin 212 DATA 14 40 DDR 218 DELETE 39 209 DELOCAL 108 DELSTRUC 234 DEMC 246 DENSITY 7 100 11 206 208 210 Density fitting 65 Density functionals ALYP 321 B86 322 B86MGC 321 B86R 322 B88 324 B88C 323 B88CMASK B22 B95 325 B97 327 B97R 326 BW 329 Cs1 B30 CS2 B30 DIRAC 330 G96 330 HCTH120 Ibal E Ei INDEX HCTH147 332 HCTH93 333 LTA 335 MKO0 336 MKOOB 336 P86 336 PBEC 338 PBEX 340 PBEXREV 340 PWS86 PW9IC PW91X 343 357 ECP effective core potential ELSEIF B0 ENDDO 29 ENDIF 30 ENDZ 1 EOM 164 EOM CCSD 164 EOMPAR 165 EOMPRINT 165 ERASE B9 Examples 22 PW92C STEST TH1 345 TH2 346 TH3 47 TH4 848
34. FROZEN MCSTART COREORB MCORB MCSAVE 51 Total charge of the molecule can be given instead of nelec number of electrons wavefunction symmetry This can be an array for state averaged cal culations as MCSYMM only used if MCSYMM is not present spin multiplicity minus one This can be an array for state averaged calculations but different spin multiplicities can only be used in de terminant CASSCE If only one value is specified this is used for all states as MCSPIN only used if MCSP IN is not present number of states for each symmetry in MCSCF as MCSTATE only used if MCSTATE is not present weight factors for all states defined by SYMMETRY and STATE Eigenvalues of TE for linear molecules for each state defined by SYM METRY and STATE records from which configurations can be selected and selection thre shold as MCSELECT only used if MCSELECT is not present can be used to define occupancy restrictions as MCRESTRCT only used if MCRESTRICT is not present if set to true or to one triggers use of CSFs number of occupied orbitals in each symmetry as MCOCC only used if MCOCC is not present number of optimized closed shell orbitals in each symmetry as MCCLOSED only used if MCCLOSED is not present number of frozen core orbitals in each symmetry as MCFROZEN only used if MCFROZEN is not present record of starting orbitals record of frozen core orbitals record for saving optimized o
35. IMP 3 2 with no closed shells inactive orbitals are used Note that this requires more CPU time than a standard CASPT2 calcu lation Moreover convergence of the CAS A method is often slow denominator shifts specified on a SHIFT card may be helpful in such cases In general we do not recommend the use of IHINT with nonzero values Default Interactions between reference configurations and singles are omitted Interactions between reference configurations and singles are included This causes a relaxation of the reference coefficients but may lead to intruder state problems After CASPT2 do variational CI using all internal configura tions and the first order wavefunctions of all states as a basis In this case the second order energy will correspond to the vari ational energy and the third order energy approximately to a Davidson corrected energy This is useful in excited state cal culations with near degeneracy situations 23 M LLER PLESSET PERTURBATION THEORY 158 23 M LLER PLESSET PERTURBATION THEORY Closed shell Mgller Plesset perturbation theory up to full fourth order MP4 SDTQ is part of the coupled cluster program The commands MP2 MP3 MP4 perform the MP calculations up to the specified order lower orders are included MP4 NOTRIPL performs MP4 SDQ calculations Normally no further input is needed if the MPn card directly follows the corresponding HF SCF Otherwise occupancies and orbitals can be
36. Local fitting can be restricted to certain programs using the following options LOCFIT If positive use local fitting in all programs in which it is avail able default 0 11 DENSITY FITTING 67 LOCFIT SCF If positive use local fitting in SCF default LOCFIT LOCFIT MP2 If positive use local fitting in DF LMP2 1 use orbital do mains 2 use pair domains default LOCF IT LOCFIT F12 If positive use local fitting in DF LMP2 F12 default LOCF IT LOCFIT_CCSD If positive use local fitting in DF LCCSD default LOCFIT LOCFIT 2EXT If positive use local fitting in LCCSD 2ext transformation de fault LOCFIT CCSD LOCFIT 3EXT If positive use local fitting in LCCSD 3ext transformation de fault LOCFIT CCSD LOCFIT_4EXT If positive use local fitting in LCCSD 4ext transformation de fault LOCFIT_CCSD LOCFIT_CPHF If positive use local fitting in CPHF default LOCF IT LOCFIT SCFGRD If positive use local fitting in gradient calculations default LOCFIT LOCORB If positive use localized orbitals in DF HF default 1 LOCTRA If positive use local screening in first half transformation de fault LOCFIT DSCREEN If positive enable density screening in LMP2 default 0 KSCREEN If positive enable fit basis Schwarz screening in LMP2 default depends on LOCTRA 11 1 4 Parameters for fitting domains The following options can be used to modify the domains used in loc
37. Phys 102 2311 2004 Analytical energy gradients for LMP2 and DF LMP2 as described in J Chem Phys 108 5185 1998 J Chem Phys 121 737 2004 111 e Explicit correlation methods as described in J Chem Phys 119 4607 2003 J Chem Phys 121 4479 2004 J Chem Phys 124 054114 2006 J Chem Phys 124 094103 2006 e Parallel execution on distributed memory machines as described in J Comp Chem 19 1998 1215 At present SCF DFT MRCL MP2 LMP2 CCSD T energies and SCF DFT gradients are parallelized when running with conventional integral evaluation integral direct and density fitted SCF DFT LMP2 and LCCSD T are also parallel The program is written mostly in standard Fortran 90 Those parts which are machine depen dent are maintained through the use of a supplied preprocessor which allows easy interconver sion between versions for different machines Each release of the program is ported and tested on a number of IBM RS 6000 Hewlett Packard Silicon Graphics Compaq and Linux systems A fuller description of the hardware and operating systems of these machines can be found at http www molpro net supported The program additionally runs on Cray Sun Convex Fujitsu and NEC SX4 platforms as well as older architectures and or operating systems from the primary list however testing is not carried out regularly on these systems and hand tuning of code may be necessary on porting A large library
38. THRINT default Product threshold for generation of 2 external integrals Defaults THR D2EXT THRPROD DTRAF THR DTRAF THRPROD default 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 59 THR D3EXT THREST D3EXT THRINT D3EXT THRPROD D3EXT THR_D4EXT THREST DAEXT THRINT DAEXT THRPROD_D4EXT THR_DCCSD THREST_DCCSD THRINT_DCCSD THRPROD_DCCSD THRMAX_DCCSD General threshold for generation of 3 external integrals If given this is used as a default for all D3EXT thresholds described be low Prescreening threshold for generation of 3 external integrals Defaults THR D3EXT THREST_DTRAF THR DTRAF THREST default Integral threshold for generation of 3 external integrals Defaults THR D3EXT THRINT_DTRAF THR DTRAF THRINT default Product threshold for generation of 3 external integrals Defaults THR D3EXT THRPROD_DTRAF THR DTRAF THRPROD default General threshold for generation of 4 external integrals If given this is used as a default for all D4EXT thresholds described be low Prescreening threshold for generation of 4 external integrals Defaults THR_D4EXT THREST_DTRAF THR_DTRAF THREST default Integral threshold for generation of 4 external integrals Defaults THR_D4EXT THRINT_DTRAF THR_DTRAF THRINT default Product threshold for generation of 4 external integrals Defaults THR_D4EXT THRPROD_DTRAF THR_DTRAF THRPROD default General threshold for ge
39. This also allows to calculate Brueckner orbitals for all three cases QCI and CCSD are identical in this case With no further input cards the wavefunction definition core closed and active orbital spaces symmetry corresponds to the one used in the most recently done SCF or MCSCF calcula tion By default a CASSCF reference space is generated Other choices can be made using the OCC CORE CLOSED WF SELECT CON and RESTRICT cards The orbitals are taken from the corresponding SCF or MCSCF calculation unless an ORBITAL directive is given The wavefunction may be saved using the SAVE directive and restarted using START The EXPEC directive allows to compute expectation values over one electron operators and the TRAN di rective can be used to compute transition matrix elements for one electron properties Natural orbitals can be printed and saved using the NATORB directive For excited state calculations see STATE REFSTATE and PROJECT 21 THE CI PROGRAM 132 21 2 Specifying the wavefunction 21 2 1 Occupied orbitals OCC nj n2 ng n specifies numbers of occupied orbitals including CORE and CLOSED in irreducible repre sentation number i If not given the information defaults to that from the most recent SCF MCSCF or CI calculation 21 2 2 Frozen core orbitals CORE ne 7n2 ng ni is the number of frozen core orbitals in irrep number i These orbitals are doubly occupied in all configurations i e not correlated
40. a B 1 CCo B e r a B T2 U2 V2 W2 X2 Y2 P5 e r o5 B Ti U1 Vi 1 X1 Y1 P1 o G a P ce 191 r a B 1 44 3428 CERE 192 a B 193 og AAA 194 1 e r t u v w x y p 2t 1 ur In 1 2 Ora L2 195 c 1 709921 196 T 0 031091 0 015545 0 016887 197 U 0 21370 0 20548 0 11125 198 V 7 5957 14 1189 10 357 199 W 3 5876 6 1977 3 6231 200 X 1 6382 3 3662 0 88026 201 Y 0 49294 0 62517 0 49671 202 P 1 1 1 203 A 0 72997 3 35287 11 543 8 08564 4 47857 204 B 0 222601 0 0338622 0 012517 0 802496 1 55396 205 C 1 0932 0 744056 5 5992 6 78549 4 49357 206 and A 0 006 0 2 0 004 207 C DENSITY FUNCTIONAL DESCRIPTIONS 335 C 21 LTA Local t Approximation LSDA exchange functional with density represented as a function of t See reference for more details K I 2E 23 208 where 0 3 E a 1 9 c5 54 9 Ss z i V and c 3 A 7300 1 a To avoid singularities in the limit p 0 G 1 2E 2 211 C 22 LYP Lee Yang and Parr Correlation Functional C Lee W Yang and R G Parr Phys Rev B 37 785 1988 B Miehlich A Savin H Stoll and H Preuss Chem Phys Letters 157 200 1989 ApappZ K 4 PPE ABa paps 47 78 18 2p 3 8 wes YABo oos s 22 ep 3 5 2 8 18 Oss 9p
41. allocates dynamic memory opens a punch file connects units to permanent files recovers file information includes other input files can be used to define default basis sets can be used to specify the geometry can be used to define the Z matrix can be used to control parallelization checks status of program steps controls global print levels controls global thresholds flags direct computation of integrals and for setting direct options controls computation of expectation values prints text stops execution controls do loops end of do loops controls conditional actions controls conditional actions end of IF block used to skip part of input and for loops over input no action data set management data set deletion performs matrix operations Define grid Dump data to grid Use cartesian basis functions Use spherical harmonic basis functions calls user supplied subroutine last line of input sets variables obsolete sets variables or numbers to their inverse obsolete sets variable arrays obsolete clears variables 4 GENERAL PROGRAM STRUCTURE 20 CLEARALL clears all variables GETVAR recovers variables from file SHOW displays the values of variables TABLE prints tables Wave function optimization INT calls the machine default integral program This is optional and needs not to be given LSINT calls the spin orbit integral program SORT calls two electron sorting program This is called automatically and
42. can be used swapping the coordinate axes if necessary This provides a mechanism for ensur ing that the same point group is used for example at all points in the complete generation of a potential energy surface allowing the safe re utilization of neighbouring geometry molecular orbitals as starting guesses etc 12 3 2 XYZ input Simple cartesian coordinates in Angstrom units can be read as an alternative to a Z matrix This facility is triggered by setting the MOLPRO variable GEOMTYP to the value XYZ before the ge ometry specification is given The geometry block should then contain the cartesian coordinates in Minnesota Computer Centre Inc XYZ format Variable names may be used as well as fixed numerical values The XYZ file format consists of two header lines the first of which contains the number of atoms and the second of which is a title The remaining lines each specify the coordinates of one atom with the chemical symbol in the first field and the x y z coordinates following A sequence number may be appended to the chemical symbol it is then interpreted as the atomic group number which can be used when different basis sets are wanted for different atoms of the same kind The basis set is then specified for this group number rather than the atomic symbol 12 GEOMETRY SPECIFICATION AND INTEGRATION 73 geomt yp xyz geomet ry 3 number of atoms This is an example of geometry input for water with an XYZ file O 0 0000
43. density 7000 2 7100 2 orbital 3140 2 2140 2 nergy el i1 e2 1 mixing 1 2 2 2 mixcildil mizanaci 1 223 noorient should always be used for diabatization This basis is too small for real application Reference geometry Displaced geometries Orbital dumprecord at reference geometry IMRCI record at reference geometry MRCI record at displaced geometries geometry C2v 11B1 and 1A2 states Save reference orbitals on reforb Dont use extra symmetries IMRCI at referenc 11B1 and 1A2 states Use orbitals from previous CASSCF Save MRCI wavefunction geometry Loop over different r values truncate dumpfile after referenc Set current r2 Wavefunction definition Starting orbitals Dump record for orbitals examples h2s diabl com Generate diabatic orbitals relative to referenc Dont use extra symmetries 11B1 and 1A2 states Use diabatic orbitals Save MRCI for displaced geometries Save adiabatic energies Compute transition densities at R2 Save transition densities on this record Compute transition densities between R2 and R1 Save transition densities on this record Densities for lt R2 R2 gt and R2 R1 Orbitals for lt R2 R2 gt and R2 R1 Adiabatic energies Compute mixing angle and diabatic energies Mixina anala obtained from ci wertore only geometry 35 QUASI DIABATIZATION 224 This calculation produces the following results Di
44. force samc 5103 1 compute NACME for states 2 1 3 1 See also test job 11 f nacme test 38 1 7 Non adiabatic coupling matrix elements NACM see Section 38 1 6 38 1 8 Difference gradients for SA MCSCF DEMC see Section 38 1 6 38 1 9 Example Calculate Gradients for Water alpha 104 degree set geometry parameters r 1 ang geometry 0 Idefine z matrix Hl o r H2 0 r H1l alpha basis vdz basis set examples hf Ido scf h20 forces com forces compute gradient for SCF mp2 mp2 calculation forces mp2 gradients multi casscf calculation forces casscf gradient 38 ENERGY GRADIENTS 247 38 2 Numerical gradients It is possible to compute gradients by finite differences using FORCE NUMERICAL options Numerical gradients are computed automatically if no analytical gradients are available for the last energy calculation By default no further input are needed and the gradient will be com puted for the last energy calculation The following options can be given on the FORCE com mand or on subsequent directives see subsequent sections STARTCMD command PROC procname VARIABLE varname The input between command and the current FORCE command de fines the energy calculation for which the gradient is computed This input section is executed for each displacement specifies a procedure to be executed for each displacement This must define a complete energy calculation and must not contain gradient or Hessian calculations
45. it will also if necessary revert and reapply patches In order to run patcher you should issue the command make patch This should be sufficient for most purposes pat cher will be built if has not yet been compiled and then it will contact the webserver apply any available patches and then return the patchlevel that you have reached If it is necessary to pass arguments to the patcher program then in the top level directory issue the command patcher apply revert list cache directory user password url 10cal verbose no action patchl patch2 A INSTALLATION OF MOLPRO 314 It can operate in one of three possible modes according to the options apply a default Apply i e install patches revert r Revert i e remove patches list 1 List available and installed patches The list of patches to remove or install can be given on the command line after all options as an explicit list of either patch names or in the case of application patch files Alternatively and usually for the case of application one can through options request either all patches that are in a local cache or all patches that are available The MOLPRO patches from the central web server default http www molpro net are cached by this program in a local directory default SHOME molpro cache Access to the web server typically has to be authenticated the first time you run this program you can s
46. lation 28 5 General Options LOCAL local Determines which method is used LOCAL 0 Conventional non local calculation LOCAL 1 Local method is simulated using canonical MOs The lo cal basis is used only at an intermediate stage to update the amplitudes in each iteration only for testing LOCAL 2 Calculation is done in local basis but without using lo cal blocking i e full matrices are used This is the most expensive method and only for testing LOCAL 3 Fully local calculation obsolete LOCAL 4 Fully local calculation default This is the fastest method for large molecules with many weak pairs and requires minimum memory PIPEK option If this option is given and option 0 the orbitals are localized using the Pipek Mezey technique If this option is not given or option 0 default the orbitals are localized unless localized orbitals are found in the orbital record cf ORBITAL directive and LOCALI command In the latter case the most recent localized orbitals are used Setting option 1 switches the localization off If option 1 the localized orbitals are printed Note Boys localization can only be performed using the LOCALI command The program will use the Boys orbitals if they are found in the orbital record and the PIPEK option is absent or option lt 0 SAVORB record Allows the localized and projected orbitals to be saved in record name ifil for later use e g plotting The two orbit
47. molpro h2 com h2 com contains ARE H2 file 2 h2 wf new punch h2 pun basis vdz ere geomet ry angstrom hl h2 h1 74 hf 3 As before but the file h2 wf is sent to the directory tmp wfu molpro W tmp wfu h2 com 5 2 Simple SCF calculations The first example does an SCF calculation for H5O using all possible defaults h2o A title r 1 85 theta 104 set geometry parameters Iz i i geometry ne z matrix geometry input examples po h20 scf com H2 0 r H1 theta hf closed shell scf In the above example the default basis set VDZ is used We can modify the default basis using BASIS directive 5 INTRODUCTORY EXAMPLES 23 h20 cc pVTZ basis A title r 1 85 theta 104 set geometry parameters geometry 0 z matrix geometry input H1 0 r examples H2 0 r H1 theta h20 scf vtz com basis VTZ luse VTZ basis hf closed shell scf 5 3 Geometry optimizations Now we can also do a geometry optimization simply by adding the card OPTG lexamples h20_scfopt_631g com Revision 2002 10 XU ENZO A title r 1 85 theta 104 set geometry parameters geometry 0 z matrix geometry input H1 0 r examples H2 0 r H1 theta h2o scfopt 631g com basis 6 31lg luse Pople basis set hf closed shell scf optg Ido scf geometry optimization 5 4 CCSD T The following job does a CCSD T calculation using a larger VTZ basis this includes an f function on oxygen and a d function on the hydrogens h2o A ti
48. onrootr 2 nrootr nstatr nstatr is the number of reference states for generating contracted pairs This may be larger or smaller than nstate If this card is not present nstatr nstate and nrootr i nroot i Roots for which no reference states are specified but which are specified on the STATE card or included by default if the nroot i are not specified explicitly on the STATE card will not be converged since the result will be bad anyway However it is often useful to include these states in the list nroot i since it helps to avoid root flipping problems Examples state 2 will calculate two states with two reference states state 2 refstate 1 2 will optimize second state with one reference state One external expansion vector will be generated for the ground state in order to avoid root flipping The results printed for state 1 are bad and should not be used unless the pair space is complete which might happen in very small calculations state 1 2 refstate 1 2 As the second example but no external expansion vectors will be generated for the ground state This should give exactly the same energy for state 2 as before if there is no root flipping which however frequently occurs state 2 accu 1 1 1 Will calculate second state with two reference states The ground state will not be converged only one iteration is done for state 1 This should give exactly the same energy for state 2 as the first example 21 T
49. r a B T3 Us Va Ws Xa Ys P3 01 B 1 E a B C 349 e r o B T Uz V2 W2 X5 Yo P2 I e r o B TU Vi Wi Xi Yi P o f a B C a B 1 a B 1 14 3428 350 r o B v3 xor B 350 a B a p 351 CB os G51 1 z ge a z Z a z 352 2 5 3 352 1 2t 1 Inj 1 1 2 e r t u v w x y p t 1 ur n TU rra 353 c 1 709921 354 T 0 031091 0 015545 0 016887 355 U 0 21370 0 20548 0 11125 356 V 7 5957 14 1189 10 357 357 W 3 5876 6 1977 3 6231 358 X 1 6382 3 3662 0 88026 359 Y 0 49294 0 62517 0 49671 360 and P511 361 C 33 STEST Test for number of electrons K ps 362 C DENSITY FUNCTIONAL DESCRIPTIONS 345 C 34 TH1 Tozer and Handy 1998 Density and gradient dependent first row exchange correlation functional See reference 23 for more details K Y o RiS XY 363 i 1 where n 21 364 Ri Pa pp 365 e 2u 366 P Oaa 088 x 119 V vei G67 p o Ogg 2 Gua O Y m aa PB a aa 3 wi 368 P 11 11 11 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 g 25h g 232 58 g 2 6 413 3 2 5 3 1 369 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 370 v 0 0 0 0 1 1 1 1 2 2 2 2 0 0 0 0 0 0 0 0 0 371 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 372 and 0 728255 0 331699 1 02946 0 235703 0 0876221 0 140
50. specifying accr and acca respectively 18 3 2 Radial integration grid RADIAL RADIAL method m scale no n n na Specify the details of the radial quadrature scheme Four different radial schemes are available specified by method 2 EM BECKE AHLRICHS or LOG with the latter being the default EM is the Euler Maclaurin scheme defined by C W Murray N C Handy and G J Laming Mol Phys 78 1993 997 m for which the default value is 2 is defined in equation 6 of the above as xn O _ 1 x a whilst scale default value 1 multiplied by the Bragg Slater radius of the atom gives the scaling parameter a LOG is the scheme described by M E Mura and P J Knowles J Chem Phys 104 1996 9848 It is based on the transformation r alog 1 x 2 with 0 lt x lt 1 and simple Gauss quadrature in x space The recommended value of m is 3 for molecular systems giving rise to the Log3 grid m 4 is more efficient for atoms is taken to be scale times the recommended value for amp given by Mura and Knowles and scale defaults to 1 BECKE is as defined by A D Becke J Chem Phys 88 1988 2547 It is based on the transformation UNS x a 3 r 1 3 3 using points in 1 lt x lt 1 and standard Gauss Chebyshev quadrature of the second kind for the x space quadrature Becke chose his scaling parameters to be half the Bragg Slater radius except for hydrogen for which
51. 1 1 1 1 2 2 2 2 440 and 0 864448 0 565130 1 27306 0 309681 0 287658 0 588767 0 252700 0 0223563 0 0140131 0 0826608 0 0556080 nn 0 00936227 To avoid singularities in the limit p 0 k O ps fi y Oss wi G 2 1 2 pio 442 C 41 THGFL Density dependent first row exchange correlation functional for closed shell systems See ref erence for more details K Y o 443 i l where n 4 444 Ri po Pp 445 t 7 6 4 3 3 2 5 3 446 and 1 06141 0 898203 1 34439 0 302369 447 C DENSITY FUNCTIONAL DESCRIPTIONS 352 C 42 VSXC See reference for more details K F x 2 P3 G3 F3 13 U3 V3 03 e Po pg Pa 0 e pg 0 Y ps FF Xs528 pi qi ri ti ui vi 04 ds 5 0 F Xs ZS po q2 2 12 u2 V2 02 448 where x Xa xg 449 Ts ws p 53 f 450 To Tp 2cf 451 Pa pg 53 451 x J KS 452 w 4zs Acf ind 2 4 2 2 p qx cz dx ex z fz 453 Parado arar Gea 9 A x 2 0 1 0 Es z 454 ge353 05 455 e a B a B stie m aun e l o B T3 U3 Vs Ws Xs Ys Ps 0 5 a B 1 E a B e 1 a B T2 U5 Vo W5 X5 Y2 P5 e 1 B Tj U1 Vi Wi Xs Y1 P1 o 5 a B Capt 456 3 3 5 I L o B 1 4347 am a By 457 opp G o B ENT 458 ditis 1 97r 0 27 2 459 24 2 1 e r t u v w x
52. 1 Using the molprocommand 000000000 0G 5 2 Simple SCF calculations o ooa ee 5 3 Geometry optimizations 2 lll ee 54 CCSD un Gace e fee a nee Bead ae ee ee he 5 5 CASSCF and MRCI 0 0000 00 000000848 Tales SA he ct deh he hh Gee ae ee ee ek Ae ht 5 8 Doleops x x93 cb ed ba ed Rue e Rohok e ei waew be awa ed das 6 PROGRAM CONTROL 61 Staringajob rs 62 Endigajob es 6 3 Restarting a job RESTART 44 4 els viii N EA 0 O N N DAA Rh pe Ree OO OC Ol 12 12 12 13 14 14 14 14 14 15 16 17 18 21 22 22 22 23 23 23 23 25 25 CONTENTS 6 4 Including secondary input files INCLUDE 6 5 Allocating dynamic memory MEMORY o o 6 6 DOloops DO ENDDO 4 4 e esr 6 66 1 Examplesfordoloops lens 6 7 Branching IF ELSEIF ENDIE eres 67 1 TF statements t ooo x e RRrc RO a ee ee RU Ed 6 7 2 GOTOcommands eh 6 7 3 Labels LABEL ee es 6 8 Procedures PROC ENDPROC 2 less 6 9 Text cards TEXT gt s 3 29 n em o EUR ee da Rs 6 10 Checking the program status STATUS lle 6 11 Global Thresholds GTHRESH 2s 6 12 Global Print Options GPRINT NOGPRINT o 6 13 One electron operators and expectation values GEXPEC 6 13 1 Example for computing expectation values 6 13 2 Example
53. 1000 dimer calculation at large separation text HF1 dummy 2 h2 second hf is now dummy hf accu 16 Iscf for first monomer mp2 mp2 for first monomer ehflinf energy save mp2 energy in variable forces compute mp2 gradient for first monomer text HF2 dummy f1 hl first hf is now dummy hf accu 16 scf for second monomer mp2 mp2 for second monomer ehf2inf energy save mp2 energy in variable forces compute mp2 gradient for second monomer add 1 ladd to previous gradient einf ehflinf ehf2inf total energy of unrelaxed momomers rff rff save reset HF HF distance to current value text CP calculation for HF1 MONOMER dummy 2 h2 second hf is now dummy hf accu 16 Iscf for first monomer mp2 mp2 for first monomer ehfl energy save mp2 energy in variable forces compute mp2 gradient for first monomer add 1 Isubtract from previous gradient text CP calculation for HF2 MONOMER dummy f1 h1 first hf is now dummy hf accu 16 scf for second monomer mp2 mp2 for second monomer ehf2 energy save mp2 energy in variable forces compute mp2 gradient for first monomer add 1 subtract from previous gradient text DIMER CALCULATION dummy reset dummies hf accu 16 Iscf for dimer mp2 mp2 for dimer edimer energy save mp2 energy in variable fForcas Losmnmita mn aradieaent for dimor 276 examples hfdimer cpcoptl com 39 GEOMETRY OPTIMIZATION OPTG 277 The next example
54. 1000087 C r 0 001667 4 z 0 11 o 0 023266 B 0 000007389 amp 8 723 5 0 472 k 0 0310907 0 01554535 1 61 7 1 0 10498 0 325 0 0047584 m 3 72744 7 06042 1 13107 and n 12 9352 18 0578 13 0045 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 C DENSITY FUNCTIONAL DESCRIPTIONS 338 C 26 PBEC PBE Correlation Functional See reference 3 for more details K p e po pp H d po pg 250 where d 1 12 ag fy 251 u a B 1 2 1 C o B 1 2 1 E a B 2 252 H d 0 B 1 2 u pa pg In a a v 253 A a B 2 2u au jJ c 254 1 0 0716 255 VK 256 v 16 V 257 T x 0 004235 258 Z 0 001667 259 r 8 r Z 260 a gt i000 i eae A a p E 23 266 262 d 0 007389 263 A 8 723 264 Y 0 472 265 e a B e r a B Ti U1 V1 W1 X Yi P1 e r a B T3 U3 Va Wa Xs Ys P3 o a B 1 E a B T e r a B Ta Us Va Wo X2 Y2 Ps gt e r o B T U1 Vi Wi X1 1 P1 o E a B E a B C DENSITY FUNCTIONAL DESCRIPTIONS 3 3 3 1 r a B 144 7347 Ces a B E B oc B 1 pis a gm 2 0 z 0 99 lt 9 y 1 e r t u v w x y p 2t 1 ur In 1 2 c 1 709921 C d a B K Q o B M Q a B 335 9789467 34
55. 1074 FREEZE thrfreez Freeze DFT grid and domains in local calculations if the step length is smaller than thrfreez default 0 01 Note The defaults for the convergence parameters can also be changed by using a global GTHRESH directive i e GTHRESH OPTSTEP step OPTGRAD grad ENERGY energy 39 1 4 Options to specify the optimization space If the geometry is given as Z matrix the default is to optimize the variables on which the Z matrix depends In case of xyz input always all 3N coordinates are optimized even if the xyz input depends on fewer variables If Cartesian z matrix input is used optimization in full space is only enforced if automatic orientation is requested using the ORIENT MASS or CHARGE options in the geometry block See opt space in section 39 2 2 for details SPACE ZMAT 3N Specifies the coordinates to be used in the optimization Z matrix optimization is only possible if the geometry is given as Z matrix OPT3N 3N logical Same as SPACE 3N ZMAT logical Same as SPACE ZMAT 39 1 5 Options to specify the optimization coordinates These options specify the coordinates in which the optimization takes place The default is to use local normal coordinates See opt coord in section 39 2 2 for details COORD NORMAL NONORMAL BMAT NORMAL logical Same as COORD NORMAL NONORMAL logical Same as COORD NONORMAL BMAT logical Same as COORD BMAT 39 1 6 Options for numerical gradients Nume
56. 2 5 Grid blocking factor DFTBLOCK DFTBLOCK nblock Respecify the number of spatial integration points treated together as a block in the DFT inte gration routines default 128 Increasing nblock may enhance efficiency on e g vector archi tectures but leads to increased memory usage 18 2 6 Dump integrand values DFTDUMP DFTDUMP file status Write out values of the integrand at grid points to the file file The first line of file contains the number of functional components there then follows a line for each functional giving the input key of the functional Subsequent lines give the functional number cartesian coordinates integrand value and integration weight with Fortran format 12 3F15 10 F23 15 18 3 Numerical integration grid control GRID Density functionals are evaluated through numerical quadrature on a grid in three dimensional space Although the sensible defaults will usually suffice the parameters that define the grid can be specified by using the GRID top level command which should be presented before the the DFT or KS commands that will use the grid Alternatively GRID and its subcommands can be presented as directives within the KS program GRID orb file status The integration grid is stored on record orb file default 1800 2 The information on disk con sists of two parts the parameters necessary to define the grid and a cache of the evaluated grid points and weights The latter is flagged as
57. 3 and 3 is usually fastest if large basis sets are used For compatibility with older versions if nonzero revert to old de faults Options set before this may be overwritten Threshold for Pipek Mezey localization The localization is assumed to be converged if all 2 x 2 rotation angles are smaller then thresh The default is 1 d 12 It can also be modified globally using GTHRESH LOCALI thresh Threshold for eliminating functions from pair domains whose norm is smaller then thresh after projecting out the occupied space The default is throrb 1 d 6 Threshold for eliminating redundant basis functions from pair do mains For each eigenvalue of Y lt thresh one function is deleted The default is 1 d 6 The method used for deleting functions depends on the parameters IDLEIG and IBASO Threshold for neglecting small fock matrix couplings in the LMP2 iterations default 1 d 8 Specifying a larger threshold speeds up the iterations but may lead to small errors in the energy In the initial iterations a larger threshold is chosen automatically It is gradually reduced to the specified final value during the iterations Threshold for deleting projected core orbitals The functions are only deleted if their norm is smaller than thresh default 0 1 The thresholds can also be specified on the THRESH directive 28 6 Options for selection of domains The following sections describe the most important options which affect the domains
58. 41 THE COSMO MODEL 285 The COSMO output file will be written after every converged SCF calculation The segment charges and potentials are corrected by the outlying charge correction For the total charges and energies corrected and uncorrected values are given The normal output file contains uncorrected values only It is recommended to use the corrected values from the output file Optimizations It is recommended to use optimizer that operates with gradients exclusively Line search tech niques that use energies tends to fail because of the energy discontinuities which may occur due to a reorganization of the segments after a geometry step For the same reasons numerical gradients are not recommended 41 BASIC THEORY COSMO is a continuum solvation model in which the solvent is represented as a dielectric continuum of permittivity e The solute molecule is placed in a cavity inside the continuum The response of the continuum due to the charge distribution of the solute is described by the generation of a screening charge distribution on the cavity surface This charge distribution can be calculated by solving the boundary equation of vanishing electrostatic potential on the surface of a conductor After a discretization of the cavity surface into sufficiently small segments the vector of the screening charges on the surface segments is q A lo where d is the vector of the potential due to the solute charge distribution on the seg
59. 6 0 Angstrom The same could be achieved as follows RVEC I 0 1 2 1 4 1 6 1 8 2 0 2 5 3 0 4 0 5 0 6 0 ANG DO I 1 RVEC R RVEC I ENDDO Up to 20 DO loops may be nested Each DO must end with its own ENDDO Jumps into DO loops are possible if the DO variables are known This can be useful in restarts since it allows to continue an interrupted calculation without changing the input all variables are recovered in a restart 6 PROGRAM CONTROL 6 6 1 Examples for do loops 30 The first example shows how to compute a potential energy surface for water SRevision 2006 H20 potential geometry x o hl o rl basis vdz angles 100 104 1 90 09 i h2 0 r2 i hl theta i 10 distances 1 6 1 7 1 8 1 9 2 0 i 0 do ith 1 tangles do irl 1 distances do ir2 1 irl i itl rl i distances irl r2 i distances ir2 theta 1 angles ith hf escf i energy ccsd t eccsd i energc eccsdt 1 energy enddo enddo enddo table r1 r2 theta escf head rl r2 theta scf ccsd ccsd t save h2o tab title Results for H20 sort 3 1 2 luse cs symmetry z matrix define basis set list of angles list of distances linitialize a counter loop over all angles H1 O H2 loop over distances for O H1 loop over O H2 distances rl ge r2 increment counter save rl for this geometry save r2 for this geometry save theta for this geometry do SCF calculati
60. Activates elimination of basis functions corresponding to core or bitals If nshell 1 only 1s functions are eliminated from projected space If nshell 2 default 1s functions on first row atoms and 1s 2s and 2p functions are eliminated on second row atoms Nothing is eliminated on H or He atoms If effective core potentials are used nothing is deleted at the corresponding atom Also functions are only deleted if the norm of the projected function is below THRCOR de fault 0 1 28 6 2 Extended domains There are two alternative modes for domain extensions either distance criteria REXT REXTS REXTC or REXTW can be used These are in Bohr and refer to the minimum distance between any atom in a standard orbital domain ij and another atom If an atom is found within the given distance all PAOs at this atom are added to the domain ij Alternatively connectivity criteria TEXT TEXTS IEXTC or IEXTW can be used These refer to the number of bonds between any atom contained in the standard domain ij and another atom The advantage of distance criteria is that they select also atoms within the given radius which are not connected to the present domain by bonds On the other hand the connectivity criteria are independent of different bond lengths e g for first and second row atoms Only one of the two possibilities can be used i e they are mutually exclusive REXT value Distance criterion for extension of all pair domains
61. CI vectors and orbital correction R El E2 H11 H22 H21 2 50 398 64296319 398 63384782 398 64296319 398 63384782 0 00000000 2 55 398 64572742 398 63666630 398 64509146 398 63730226 0 00231474 2 60 398 64911746 398 63771803 398 64648358 398 64035190 0 00480493 It is seen that the mixing obtained from the CI vectors only is now very different and mean ingless since the orbitals change significantly as function of geometry However the second calculations which accounts for this change approximately still gives results in quite good agreement with the calculation involving diabatic orbitals The final examples shows a more complicated input which also computes the non adiabatic coupling matrix elements In a two state model the NACME should equal the first derivative of the mixing angle In the example the NACME is computed using the 3 point DDR method NACMECD and also by finite difference of the mixing angle DCHI MIXCI 0 00 15427 27 87 MIXTOT 0 00 154226 27 88 MIXCI 0 00 19 11 35 83 MIXTOT 0 00 15 36 28 73 35 QUASI DIABATIZATION SRevision 2006 0 225 h2s Diabatization and NACME calculation memory 3 m gprint orbitals civector geometry x noorient S nb 9 fL h2 s r2 h1 theta basis avdz r1 2 5 theta 92 r 2 55 2 60 dr 0 0 01 0 01 reforb1 2140 2 refci 6000 2 savci 6100 2 text compute wavefunction at referenc r2 r1 noorient should alwa
62. CP is calculated this option taken when and only when no singles only K CP is calculated Implies that modified coupling co efficients are used K CP and K CP are calculated Default is IKCPS 2 except when single reference configuration when IKCPS 1 Option for density matrix routines all quantities in density matrix routines are recalculated for each intermediate symmetry max CPU min core quantities precalculated and stored on disk max I O min core quantities precalculated and kept in core min CPU max core If nonzero calculate intermediate orbitals for each pair Might improve convergence in some cases in particular if localized orbitals are used 21 3 18 Miscellaneous parameters PARAM codel value code2 value Redefine system parameters If no codes are specified the default values are displayed The following codes are allowed LSEG INTREL IVECT 0 IVECT 1 MINVEC IBANK LTRACK disc sector length number of integers per REAL 8 word scalar machine vector machine call MXMB in coupling coefficient routines if vector length larger than this value number of memory banks for vector machines If IBANK 1 vector strides which are multiples of IBANK are avoided where appropriate number of REAL 8 words per track or block for file alloca tion 21 THE CI PROGRAM 142 LTR determines how matrices are stored on disc If LTR LSEG all matrices start at sector
63. DKHn The desired DKH order DKHO and the chosen parametrization for the unitary transformations have to be specified by DKHO n n 2 14 DKHP m m 1 5 below the DKROLL 1 statement in the input file The possible parametrizations supported by MOLPRO are DKHP 1 Optimum parametrization OPT DKHP 2 Exponential parametrization EXP DKHP 3 Square root parametrization SQR DKHP 4 McWeeny parametrization MCW DKHP 5 Cayley parametrization CAY Example DKROLL 1 activate Douglas Kroll Hess one electron integrals DKHO 8 DKH order 8 DKHP 4 choose McWeeny parametrization for unitary transformations Note For DKHO gt 11 the values of some parameters in the file src common parameters h have to be suitably increased Only recommended for experts who do exactly know what they are doing For most cases DKHO 1 0 is sufficient Up to fourth order DKHO 4 the DKH Hamiltonian is independent of the chosen paramteri zation Higher order DKH Hamiltonians depend slightly on the chosen paramterization of the unitary transformations applied in order to decouple the Dirac Hamiltonian For details on the infinite order DKH Hamiltonians see M Reiher A Wolf JCP 121 2037 2047 2004 M Reiher A Wolf JCP 121 10945 10956 2004 For details on the different parametrizations of the unitary transformations see A Wolf M Reiher B A Hess JCP 117 9215 9226 2002 16 2 Example for computing
64. Dscf natorb C diff diffden write diffden denfil save C diff 2500 2 300 Z matrix geometry input bond length bond angle do scf calculation load mcscf density matrix load mcscf natural orbitals load mcscf canonical orbitals load scf density matrix load overlap matrix prints occupied casscf orbitals print element D 1 1 print element D 2 1 print element D 1 2 Itransform s into MO basis same as above print result should be unit matrix number of basis functions number of electrons Itrace of S MO form trace DS form SC S Cnat Itransform density to natural MO print diagonal elements transform D ao to canonical MO basis multiply d can by 1 diagonalizes density D can examples matrop com could also be done occupation numbers Same as above transforms canonical orbitals to natural orbitals prints new natural orbitals us sin make natural orbitals using MCSCF density D ao directly prints new natural orbitals form mcscf scf difference density make natural orbitals for difference density differenc writ density to ASCII file denfile should be the same as abo store natural orbitals for difference density in dump 1 This second example adds a quadrupole field to HO The result is exactly the same as using the QUAD command HO is overwritten by the modified one electron matrix and the nuclear energy is automatically changed
65. E eene Eus doo der i ge aerei t de e Ito EDD rmn 351 AA MMC 351 DR a eh da AAA a HE 352 C 43 VWN3 Vosko Wilk Nusair 1980 III local correlation energy 354 C 44 VWN5 Vosko Wilk Nusair 1980 V local correlation energy 355 356 1 HOW TO READ THIS MANUAL 1 1 HOW TO READ THIS MANUAL This manual is organized as follows The next chapter gives an overview of the general structure of MOLPRO It is essential for the new user to read this chapter in order to understand the conventions used to define the symmetry records and files orbital spaces and so on The later chapters which describe the input of the individual program modules in detail assume that you are familiar with these concepts The appendices describe details of running the program and the installation procedure Throughout this manual words in Typewriter Font denote keywords recognized by MOL PRO In the input these have to be typed as shown but may be in upper or lower case Numbers or options which must be supplied by the user are in italic In some cases various different forms of an input record are possible This is indicated as options and the possible options are described individually in subsequent subsections 2 RUNNING MOLPRO On Unix systems MOLPRO is accessed using the molpro unix command The syntax is molpro options datafile MOLPRO s execution is controlled by user prepared data if datafile is not given on the command lin
66. ENERGY ESCF define a macro HF do SCF calculation ESCF ENERGY store SCF energy in variable ESCF ULTI do CASSCF DEMC SSECORR store CASSCF correlation energy in variable DEMC RCI do MRCI DECI SSECORR store MRCI correlation energy in variable DECI Here is an example of advanced use of macros and string variables Revision 2006 0 test for parser text This fancy input demonstrates how string variables and macros can be used text basis vdz define basis set geometry 0 H O r define geometry z matrix text methods method rhf 2 casscf 2 mrci text active spaces spaces 3 1 1 3 4 2 2 1 3 5 2 2 text symmetries symset 1 2 1 2 3 1 2 text weight factors for state averaged casscf werghts p 1 4 T 13 1 2 4 DET1 0 554 0 5 2 7 text scf occupation set scfocc 3 2 1 examples text bond distance oh macros com r 1 85 hf do i 1 method loop over methods occ spaces i set active space for this run set symmetry symset i set symmetries for this run set weight weights i set weights for this run Smethod i now run method e 1 Senergy save energies in strings dipol i S dmz save dipole moments in strings enddo table method spaces symset weights e dipol title Results for OH r r basis Sbasis head method spaces symmetries weights energies dipole moments exit 8 6 Indexed Variables Vectors Var
67. For larger basis sets like cc pVTZ we recommend to use a slightly larger value of 0 985 to ensure that enough atoms are included in each domain For cc pVQZ recom mend THRBP 0 990 is recommended In cases of doubt compare the domains you get with a smaller basis e g cc pVDZ 28 LOCAL CORRELATION TREATMENTS 193 The choice of domains usually has only a weak effect on near equilibrium properties like equi librium geometries and harmonic vibrational frequencies More critical are energy differences like reaction energies or barrier heights In cases where the electronic structure strongly changes e g When the number of double bonds changes it is recommended to compare DF LMP2 and DF MP2 results before performing expensive LCCSD T calculations More balanced results and smooth potentials can be obtained using the MERGEDOM directive see section 28 8 6 28 9 5 Freezing domains In order to obtain smooth potential energy surfaces domains must be frozen The domain information can be stored using the SAVE option and recovered using the START option Alter natively the SAVE and START can be used see section 28 8 3 In the latter case also the CCSD amplitudes are saved restarted Freezing domains is particularly important in calculations of intermolecular interactions see section 28 9 8 Domains that are appropriate for larger ranges of geometries such as reaction pathways can be generated using the MERGEDOM directive sec tion The dom
68. G Hetzer M Schiitz H Stoll and H J Werner J Chem Phys 113 9443 2000 as well as LMP2 gradients as described in A El Azhary G Rauhut P Pulay and H J Werner J Chem Phys 108 5185 1998 are now available without special license The linear scaling LCCSD T methods as de scribed in M Sch tz and H J Werner J Chem Phys 114 661 2001 M Sch tz and H J Werner Chem Phys Lett 318 370 2000 M Sch tz J Chem Phys 113 9986 2000 will be made available at a later stage QCISD gradients as described in Phys Chem Chem Phys 3 4853 2001 are now available Additional and more flexible options for computing numerical gradients and performing geometry optimizations A large number of additional density functionals have been added together with support for the automated functional implementer described in Comp Phys Commun 136 310 318 2001 B RECENT CHANGES 318 5 6 10 B 4 Multipole moments of arbitrary order can be computed Further modules have been parallelized in particular the CCSD T and direct LMP2 codes The parallel running procedures have been improved The parallel version is available as an optional module The basis set library has been extended Some subtle changes in the basis set input it is not possible any more that several one line basis input cards with definitions for individual atoms follow each other Each new basis card super
69. If no CORE card is given the program uses the same core orbitals as the last CI calculation if there was none then the atomic inner shells are taken as core To avoid this behaviour and correlate all electrons specify CORE 21 2 3 Closed shell orbitals CLOSED n n2 ng nj is the number of closed shell orbitals in irrep number i inclusive of any core orbitals These orbitals do not form part of the active space i e they are doubly occupied in all reference CSFs however in contrast to the core orbitals see CORE these orbitals are correlated through single and double excitations If not given the information defaults to that from the most recent SCF MCSCF or CI calculation For calculations with closed shell reference function closed 0cc see CISD OCI and CCSD 21 2 4 Defining the orbitals ORBIT name file specifications name file specifies the record from which orbitals are read Optionally various specifications can be given to select specific orbitals if name file contains more than one orbital set For details see section Note that the IGNORE_ERROR option can be used to force MPn or triples calculations with non canonical orbitals The default is the set of orbitals from the last SCF MCSCF or CI calculation 21 2 5 Defining the state symmetry The number of electrons and the total symmetry of the wavefunction are specified on the WF card WF elec sym spin where elec 1s the number of electrons
70. LMP2 and two otherwise If given replaces value of SCREEN for DTRAF Maximum size of merged shells in the first quarter transforma tion step 0 not used Shells are only merged if their size is smaller than this value 0 not used Maximum size of merged shells in the second quarter transfor mation step 0 not used Shells are only merged if their size is smaller than this value 0 not used Maximum number of centres in merged shells 0 no limit Print parameter for DTRAF General thresholds for all direct integral transformations TL R_DTRAF REST_DTRAF RINT_DTRAF RPROD_DTRAF General threshold for DTRAF If given this is taken as default value for all thresholds described below AO prescreening threshold for DTRAF Defaults THR_DTRAF THREST default Integral threshold for DTRAF Defaults THR DTRAF THRINT default Product threshold for DTRAF Defaults THR DTRAF THRPROD default Thresholds specific to direct integral transformations THR D2EXT THREST D2EXT THRINT D2EXT THRPROD D2EXT General threshold for generation of 2 external integrals If given this is used as a default for all D2EXT thresholds described be low Prescreening threshold for generation of 2 external integrals Defaults THR D2EXT THREST DTRAF THR DTRAF THREST default Integral threshold for generation of 2 external integrals Defaults THR D2EXT THRINT_DTRAF THR DTRAF
71. LOAD command has slightly different options In all forms of LOAD name is an arbitrary string up to 16 characters long by which the loaded matrix is denoted in subsequent commands 43 2 1 Loading orbitals LOAD name ORB record specifications loads an orbital coefficient matrix from the given dump record If the record is not specified the last dump record is used Specific orbitals sets can be selected using the optional specifications as explained in section The keyword ORB needs not to be given if name ORB 43 2 2 Loading density matrices LOAD name DEN record specifications loads a density matrix from the given dump record If the record is not given the last dump record is used Specific orbitals sets can be selected using the optional specifications as ex plained in section The keyword DEN needs not to be given if name DEN 43 2 3 Loading the AO overlap matrix S LOAD name S loads the overlap matrix in the AO basis The keyword S needs not to be given if name S 43 2 4 Loading S LOAD name SMH loads S712 where S is the overlap matrix in the AO basis The keyword SMH needs not to be given if name SMH 43 2 5 Loading the one electron hamiltonian LOAD name HO LOAD name HO1 loads the one electron hamiltonian in the AO basis HO1 differs from HO by the addition of perturbations if present see sections 32 4 1 32 4 2 The keyword HO H01 needs not to be given if name H0 H01 The nuclear energy ass
72. MATROP Hints First generate a starting orbital guess by finding the eigenvectors of hO Store the orbitals in a record Basis and geometry are defined in the usual way before the first call to MATROP Then use a MOLPRO DO loop and call MATROP for each iteration Save the current energy in a variable note that the nuclear energy is stored in variable ENUC Also compute the dipole moment in each iteration At the end of the iteration perform a convergence test on the energy change using the IF command This must be done outside MATROP just before the ENDDO At this stage you can also store the iteration numbers energies and dipole moments in arrays and print these after reaching convergence using TABLE For the following geometry and basis set geometry 0o hl o r h2 0 r hl theta r 1 ang theta 104 basis vdz thresh 1 d 8 Z matrix geometry input bond length bond angle basis set convergence threshold the result could look as follows 43 MATRIX OPERATIONS scr has converged in 24 iterations m H 3 El 0000000 000000000000000000o0owz Io ol Ov R co ia i si Y Ae WwW Y JJ J aa a al 0214 0214 0214 0214 0214 0214 0214 0214 0214 0 73004 0NROowo0o 10 01 C0 Io ES N NM PO N L2 O0 o N Ww t eg dO UR VES Al s PEE Tee ve E Te e lle 4 oo o o o J J O00 000000000000 O N A 0214 It does not converge terribly fas
73. OPT command available in previous MOLPRO versions is no longer needed and not available any more OPTG key value key2 value The OP TG command can be used to perform automatic geometry optimizations for all kinds of wavefunctions For minimum searches it is usually sufficient to give just the OPTG command without further options or directives but many options are available which are described in the following sections Various optimization methods can be selected as described in section 39 2 1 MOLPRO allows minimization i e search for equilibrium geometries transition state optimization i e search for saddle points on energy surfaces and reaction path following The standard algorithms are based on the rational function approach and the geometry DIIS approach Also available is the quadratic steepest descent following method of Sun and Ruedenberg see J Sun and K Rueden berg J Chem Phys 99 5257 1993 This method is often advantageous in Transition State searches For a detailed discussion of the various minimization algorithms see F Eckert P Pu lay and H J Werner J Comp Chem 18 1473 1997 Reaction path following is described in F Eckert and H J Werner Theor Chem Acc 100 21 1998 Please refer to the references section for citations of the analytic gradient methods When analytical gradients are available for the optimized energy these will be used Otherwise the gradient will be computed numer
74. ORBITALS FOCK HO ORBEN OPER TRIANG SQUARE or VECTOR If type is not given but known from LOAD or another command this is assumed Orbitals density matrices fock matrices and orbital energies are saved to a dump record the same one should normally be used for all these quantities If type is HO the one electron hamiltonian is overwritten by the current matrix and the nuclear energy is modified according to the value associated to name The nuclear energy is also stored in the variable ENUC AII other matrices can be saved in triangular or square form to plain records using the TRI ANG and SQUARE options respectively for triangular storage the matrix is symmetrized before being stored Eigenvectors can be saved in plain records using the VECTOR option Only one matrix or vector can be stored in each plain record One electron operators can be stored in the operator record using SAVE name OPER PARITY np NUC opnuc CENTRE icen COORD x y z The user defined operator name can can then be used on subsequent EXPEC or GEXPEC cards np 1 0 1 for symmetric square antisymmetric operators respectively default 1 If CENTRE is specified the operator is assumed to have its origin at the given centre where icen refers to the row number of the z matrix input The coordinates can also be specified explicitly using COORD By default the coordinates of the last read operator are assumed or otherwise zero 43 MATRIX OPERATION
75. Other required properties can be specified using EXPEC card Excited state densities are saved if DM card is present For an example see section 25 4 2 If TRANS 2 transition moments among excited states are also calculated 25 EXCITED STATES WITH EQUATION OF MOTION CCSD EOM CCSD 165 It is possible to make the program to converge to a vector which resembles a specified singles vector This option is switched on by FOLLOW n card usually n 2 should be set FOLLOW card should be always accompanied with EXFILE record ifil card where record ifil contains singles vectors from a previous calculation see section 25 4 3 25 2 Options for EOMPAR card Normally no further input is needed However some defaults can be changed using the EOMPAR directive EOMPAR key1 valuel key2zvalue2 where the following keywords key are possible MAXDAV nv Maximum value of expansion vectors per state in Davidson procedure default 20 INISINGL ns Number of singly excited configurations to be included in initial Hamil tonian default 20 the configurations are ordered according to their energy Sometimes INISINGL should be put to zero in order to catch states dominated by double excitations INIDOUBL nd Number of doubly excited configurations to be included in initial Hamiltonian default 10 INIMAX nmax Maximum number of excited configurations to be included in initial Hamiltonian By default nmax ns nd MAXITER itmax Maxim
76. Phys 111 4523 1999 This diabatization can be done automatically and requires two steps first the active orbitals of a CASSCF calculation are rotated to maximize the overlap with the orbitals at the reference geometry This is achieved using the DIAB procedure described in section 20 5 8 Secondly the DDR procedure can be used to find the transformation among the CI vectors The following input is required DDR calls the DDR procedure ORBITAL orbl orb2 orbl and orb2 are the diabatic orbitals at the current and reference geometry respectively DENSITY trdml trdm2 trdml are the transition densities computed at the current geometry trdm2 are transition densities computed using the wavefunctions of the current bra and reference ket geometries MIXING statel state2 The given states are included in the diabatization ENERGY el e2 Adiabatic energies of the states If this input card is present the Hamiltonian in the basis of the diabatic states is computed and printed Alternatively the energies can be passed to DDR using the Molpro variable EADIA The results are printed and stored in the following Molpro variables provided the ENERGY directive or the EADIA variable is found Results including the first order orbital correction 35 QUASI DIABATIZATION 222 SMAT The first nstate x nstate elements contain the state overlap matrix bra index rans fastest UMAT The first nstate x nstate elements contain th
77. Phys 94 6708 1991 MP2 and LMP2 gradients A El Azhary G Rauhut P Pulay and H J Werner J Chem Phys 108 5185 1998 DF LMP2 gradients M Sch tz H J Werner R Lindh and F R Manby J Chem Phys 121 737 2004 QCISD and LQCISD gradients G Rauhut and H J Werner Phys Chem Chem Phys 3 4853 2001 CASPT gradients P Celani and H J Werner J Chem Phys 119 5044 2003 Geometry optimization F Eckert P Pulay and H J Werner J Comp Chemistry 18 1473 1997 Reaction path following F Eckert and H J Werner Theor Chem Acc 100 21 1998 Harmonic frequencies G Rauhut A El Azhary F Eckert U Schumann and H J Werner Spectrochimica Acta 55 651 1999 Moller Plesset Perturbation theory MP2 MP3 MP4 Closed shell Mgller Plesset Perturbation theory up to fourth order MP4 SDTQ is part of the coupled cluster code see CCSD Open shell Mgller Plesset Perturbation theory RMP2 R D Amos J S Andrews N C Handy and P J Knowles Chem Phys Lett 185 256 1991 Coupled Cluster treatments QCI CCSD BCCD C Hampel K Peterson and H J Werner Chem Phys Lett 190 1 1992 and references therein The program to compute the perturbative triples corrections has been developed by M J O Deegan and P J Knowles Chem Phys Lett 227 321 1994 Equation of Motion Coupled Cluster Singles and Doubles EOM CCSD T Korona and H J Werner J Chem Phys 118 30
78. Ps 0 E a B 1 0 B C e r B T2 U2 V2 W2 X2 Yo P3 e r o B Ty U1 Vi W1 X1 Y1 P1 o G a B ce 171 COT RE LODS 172 a p 2 P 173 a p C DENSITY FUNCTIONAL DESCRIPTIONS 333 1 t2 4 1 z en e z 13 3 174 1 e r t u v w x y p 2t 1 ur In TONES E mm 175 c 1 709921 176 T 0 031091 0 015545 0 016887 177 U 0 21370 0 20548 0 11125 178 V 7 5957 14 1189 10 357 179 W 3 5876 6 1977 3 6231 180 X 1 6382 3 3662 0 88026 181 Y 0 49294 0 62517 0 49671 182 P 1 1 1 183 A 0 54235 7 0146 28 382 35 033 20 428 184 B 0 56258 0 0171 1 3064 1 0575 0 8854 185 C 1 09025 0 7992 5 5721 5 8676 3 0454 186 and A 0 006 0 2 0 004 187 C 20 HCTH93 Handy least squares fitted functional See reference for more details K Pa Pg Pa 0 pg 0 Ao Ain d 24 A2 n 4 3 43 n d 4 A n d 34 Y e ps 0 Bo Bin s 7 A2 B n x s 7 A2 24 Bs n 5 2 A 188 Ba n Xs 7 A2 4 3 8 3425 Wx py 4 Co Cin Xs A3 n x 23 03 n 2 23 Ca n 05 2 33 where d V2 xa V2 xg 189 u 1 4 40 n 8 4 l 190 C DENSITY FUNCTIONAL DESCRIPTIONS 334 o B a B tres n nv aman e r a b T3 Us V3 Wa X5 Ya P o C
79. RE a 25 2 Options for EOMPARcardl 22e 25 3 Options for EOMPRINTcard len quM T ULP CDM MH 25 4 3 Calculate an EOM CCSD state most similar to a given CIS state 25 5 Excited states with CIS ooo ee or omo o omo m RR 26 OPEN SHELL COUPLED CLUSTER THEORIES 27 The MRCC program of M Kallay MRCC 27 1 Installing MRCC uce 203 0244 ee em xax eee de oe es as XR EROS 27 2 Running MRCC 4 ius 99 Paw a bare ho be Be he Xd are ad 28 LOCAL CORRELATION TREATMENTS 28 1 Introduction 28 2 Getting Started o s is e d owa e ie Ea e E e a e a EA e E E E EEEE E E ETO EE EEE E ve NE EE TOE E mad a ed E ET A E E RECS Seok ee xiv 146 147 148 148 150 152 152 152 152 155 156 158 158 158 159 160 160 161 161 161 162 162 162 162 162 163 163 164 164 165 165 166 166 167 168 168 169 170 170 170 CONTENTS XV Ph aq UR a NEAR aa ee ek 184 A 185 POSAO TETERE 185 nn 186 25 9 Directives se BR as a ck e ROO Rene Fue dun a ae ds 187 28 8 1 The LOCAL directive 222222 187 28 8 2 The MULTP directive s ss 222222 187 28 8 3 Saving the wavefunction SAVE llle 188 28 8 4 Restarting a calculation START 188 28 8 5 Correlating subsets of electrons REGION 188 28 8 6 Domain Merging MERGEDOM ll 189 189 25 0 Dome ITENI sane Ge xe po eor tee eee dr o ae AUR RU 190 28 91 Basissets ow ko koe ca ee ke eR Rao
80. REX 12 3 3 MOLPRO921nput less 12 4 Writing Gaussian XMol or MOLDEN input PUT MM TM a ra daa res se die oe A AN reer eee as ee ee E RI bee GS Bee eee EVENT aoe he bce jonas LE 13 BASIS INPUT 13 1 Overview sets and the basis library leen MEMOREM 13 3 Thebasissetlibrary Rn 13 4 Defaultbasissets les 13 5 Default basis sets forindividualatoms lll LA Pack RUSSE NUNC REN NINE E EE 13 7 Contracted setdefimtions 22e 15 8 Exaimples mox oe ake ek A Sj nomo moe o S0 14 EFFECTIVE CORE POTENTIALS 14 1 Input from ECP library Re 14 2 Explicit input for ECPs lt ee es ss es pagi agad taeka ad aea 14 3 Example for explicit ECPinput lens 14 4 Example for ECP input from library oaoa 15 CORE POLARIZATION POTENTIALS 15 1 Input options ece eR 15 2 ExampleforECP CPPB sh 16 RELATIVISTIC CORRECTIONS 16 1 Using the Douglas Kroll Hess Hamiltonian 16 2 Example for computing relativistic corrections llle 17 THE SCF PROGRAM 17 1 Optohs 2 g nap e edt ee mu E y Re RP Ron UR a a 17 1 1 Options to control HF convergence o o 17 1 2 Options for the diagonalization method 17 1 3 Options for convergence acceleration methods DIIS poca a aida LSU EI dats dira oe dr sa us uu6 0 coun eee ary I AMET 67 68 69 69 70 70 71 72 73 73 7
81. Roland Lindh for geometry optimizations This is done by prepending the optimization method with SL The following methods are supported SLRF Use the rational function approximation SLNR Use the Newton Raphson method SLC1 Use the C1 DIIS method SLC2 Use the C2 DIIS method When using DIIS methods SLC1 or SLC2 the DIIS parameters are specified in the same way as in standard molpro optimizer There are some differences when using the SLAPAF program 1 It is not possible to use Z matrix coordinates in the optimization 2 Instead one can explicitly define internal coordinates to be varied or fixed 3 Additional constraints can be imposed on the converged geometry in a flexible way 39 GEOMETRY OPTIMIZATION OPTG 268 39 3 1 Defining constraints Constraints and internal coordinates see below can be linear combinations of bonds angles etc The latter called here primitive internal coordinates can be specified before the constraints definition or directly inside The general definition of a primitive coordinate is PRIMITIVE NAME symbolic name explicit definition Or PRIM NAME symbolic name explicit definition Here symbolic name is the name given to the primitive coordinate if omitted it will be generated automatically This name is needed for further reference of this primitive coordinate explicit definition has the form type atoms type can be one of the following BOND Bond length d
82. SP If style is XYZ an XYZ file will be written see also section 12 3 2 If style is CRD the coordi nates will be written in CHARMm CRD format If style is MOLDEN an interface file for the MOLDEN visualization program is created further details and examples are given below If style is omitted the Z matrix current geometry and where applicable gradient are written file specifies a file name to which the data is written if blank the the data is written to the output stream If status is omitted or set to NEW any old contents of the file are destroyed otherwise the file is appended 12 4 1 Visualization of results using Molden Geometry molecular orbital and normal mode information when available is dumped by PUT MOLDEN in the format that is usable by MOLDEN examples h20 xyzinput com 12 GEOMETRY SPECIFICATION AND INTEGRATION 74 The interface to the gOpenMol program offers an alternative visualization possibility and is described in section 32 7 The example below generates all the information required to plot the molecular orbitals of water and to visualize the normal modes of vibration SRevision 2006 0 H20 geometry angstrom o h o roh h o roh h theta roh 1 0 theta 104 0 sh examples h20 put molden com optg frequencies print low img put molden h20 molden The example below does a difference density by presenting its natural orbitals to MOLDEN Note that it although
83. THGFC 351 THGFCFO 349 THGFCO 350 THGEL 351 VSXC 352 VWN3 354 VWNS B55 Density matrices DF LMP2 195 DF MP2 158 DFT 99 DFTBLOCK 101 DF TDUMP 101 DF TFACTOR 100 DFTTHRESH 100 Difference gradient s 128 Diabatization 221 DIIS DIP 210 D1P 210 dipole field DIRECT 56 98 distributed multipole analysis 208 DM 124 139 158 163 DMA 208 DO 29 DO loops 9 DONT 123 DUMMY 75 Dummy centres Q X 71 DUMP 200 ECP library allene_opt_bmat com 270 allene_optmp2 com 271 allene_optsc com 270 ar2_rel com 37 89 212 auh ecp lib com 86 bh mrci sigma delta com 145 caffeine opt diis comj7i cn sa casscf com 130 enaft com 106 cu ecp explicit com 86 field com 211 form freq com 282 h2 com 22 h2f merge com h20 c2v cs start com 95 h20_caspt2_opt com 153 h20_casscf com 129 h2o_ccsd com 162 h20 ccsdt vtz com 23 h2o_cepal com 144 h20 diffden molden com h2o_direct com 64 h20o dma com 209 h20 field com 211 h20_forces com 246 h20 gexpecl com 207 h h h h h h h h h h h h h h 20_gexpec2 com 36 20_manymethods com 26 B0 20_mrcc com 173 20 mrcc eom com 173 20 mrci vtz com 23 20 mscaspt2 opt com 154 20 pes ccsdt com 25 0 20 pop com 210 20 proce com 25 20_property com 207 20_put_mo
84. Use Newton Raphson method for specified iterations UNCOUPLE Do not optimize orbitals and CI coefficients simultaneously in the specified iterations This option will set DIAGCI for these iterations NULL No orbital optimization 20 6 3 Disabling the optimization In addition to the ITERATIONS directive described above some procedures can be be disabled more simply using the DONT directive DONT code code may be ORBITAL Do initial CI but don t optimize orbitals WAVEFUNC Do not optimize the orbitals and CI coefficients i e do only wavefunction analysis provided the orbitals and CI coefficients are supplied see START card WVEN Alias for WAVEFUNC ANAL Do no wavefunction analysis 20 6 4 Disabling the extra symmetry mechanism NOEXTRA This card disables the search for extra symmetries By default if extra symmetries are present each orbital is assigned to such an extra symmetry and rotations between orbitals of different extra symmetry are not performed 20 7 Calculating expectation values By default the program calculates the dipole expectation and transition moments Further expectation values or transition properties can be computed using the TRAN TRAN2 and EXPEC EXPEC2 directives 20 THE MCSCF PROGRAM MULTI 124 20 7 1 Matrix elements over one electron operators EXPEC oper oper2 0pern TRAN oper oper2 0Ppern Calculate expectation values and transition matrix elements for the given one elect
85. VTZ add the VTZ d function to the VDZ basis for H END BASIS SPD O VTZ use uncontracted s p d functions of basis VTZ for oxygen S H H07 use Huzinaga 7s for Hydrogen Ey 174 contract first four s functions P H 1 0 0 3 ladd two p functions for hydrogen END Several BASIS cards and or blocks can immediately follow each other Always the last spec ification for a given atom and type is used Defaults given using BASIS commands can be overwritten by specifications in the integral input If an individual basis function type is spec ified for an atom it is required that all other types are also defined For example in the above example no f functions are included for O even if the global default would include f functions Also defining the s functions for hydrogen switches off the default basis set for hydrogen and so the p functions must be defined Instead of the atomic symbol the atom group number can also be used The same input forms are also possible as direct input to the integral program In contrast to MOLPRO92 now the atomic symbol can be used in field 2 of a basis specification instead of the atom group number 13 BASIS INPUT 82 SPD O VTZ luse VTZ basis for all oxygen atoms SPD 1 VTZ luse VTZ basis for atom group 1 Instead of the BASIS END block one can also use the structure BASIS If a basis is not specified at all for any unique atom group then the program assumes a default For further detai
86. Werner and W Meyer J Chem Phys 73 2342 1980 H J Werner and W Meyer J Chem Phys 74 5794 1981 H J Werner Adv Chem Phys LXIX 1 1987 Internally contracted MRCI H J Werner and P J Knowles J Chem Phys 89 5803 1988 P J Knowles and H J Werner Chem Phys Lett 145 514 1988 See also H J Werner and E A Reinsch J Chem Phys 76 3144 1982 H J Werner Adv Chem Phys LXIX 1 1987 Excited states with internally contracted MRCI P J Knowles and H J Werner Theor Chim Acta 84 95 1992 Internally contracted MR ACPF QDVPT etc H J Werner and P J Knowles Theor Chim Acta 78 175 1990 The original reference to uncontracted MR ACPF QDVPT MR ACQO are R J Gdanitz and R Ahlrichs Chem Phys Lett 143 413 1988 R J Cave and E R Davidson J Chem Phys 89 6798 1988 P G Szalay and R J Bartlett Chem Phys Lett 214 481 1993 Multireference perturbation theory CASPT2 CASPT3 H J Werner Mol Phys 89 645 1996 P Celani and H J Werner J Chem Phys 112 5546 2000 Coupling of multi reference configuration interaction and multi reference perturbation theory P Celani H Stoll and H J Werner Mol Phys 102 2369 2004 vi Analytical energy gradients and geometry optimization Gradient integral evaluation ALASKA R Lindh Theor Chim Acta 85 423 1993 MCSCF gradients T Busch A Degli Esposti and H J Werner J Chem
87. all electron calculations and ECPLSX ECPLSY or ECPLSZ in ECP calculations Since the spin orbit program is part of the MRCI program the TRANLS card must be preceded by a MR CI card For the case that the matrix elements are computed for MCSCF wave functions one has to recompute and save the CI vectors using the MRCI program see chapter 21 using the NOEXC directive to avoid inclusion of any further excitations out of the MCSCF reference function If in the MRCI step several states of the same symmetry are computed simul taneously using the STATE directive the matrix elements are computed for all these states Note that the OCC and CLOSED cards must be the same for all states used in a TRANLS calculation The selection rules for the Ms values are AM 1 for the LSX and LSY operators and AM 0 for the LSZ operator Note that 2Ms has to be specified and so the selection rules applying to the difference of the input values are 0 or 2 In all electron SO calculations the value of the calculated spin orbit matrix element is saved in atomic units in the MOLPRO variables TRLSX TRLSY and TRLSZ for the x y and z components respectively For ECP LS calculations the variables TRECPLSX TRECPLSY and TRECPLSZ are used Note that for imaginary matrix elements i e for the x and z components of the SO Hamiltonian the matrix elements are imaginary and the stored real values have to be multiplied by i If matrix elements for several states a
88. all stored in the file CONFIG generated by configure Subsequently make ARCH procname will link the desired version where procname can be p3 p4 or athlon This will generate the executable molpros 2006 0 i14 procname exe If the ARCH option is not given the last one configured will be generated In addition a file molpros_2006_0_i14_procname rc will be generated for each case which defines the running environment and may also contain system dependent tuning parameters see section A 3 7 A specific executable can then be requested using molpro rcfile molpros 2006 0 i4 procname rc input More conveniently one can set the Unix environment variable MOLPRO RCFILE to molpros 2006 0 i4 procname rc and then simply use molpro without an option The recommended mechanism is to set the environment variable MOLPRO_RCF ILE in the default environment cshrc profile as appropriate on a given machine Similarly different MPP version can also be installed in one MOLPRO tree but the tree for parallel and serial versions must be distinct In this case one can run configure for tcgmsg mpi and or myrinet and in addition with 3 p4 and or ath1lon and then link using make MPPLIB libname where libname can be tcgmsg mpi or mnyrinet The ARCH and MPPLIB options can be combined e g make MPPLIB ibname ARCH procname and this will generate the executable molprop_2006_0_i4_procname_libname exe and the default file
89. all trigonometric functions use or produce angles in degrees 3 9 Variables 3 9 1 Setting variables Data and results can be stored in MOLPRO variables Variables can be of type string floating or logical and may be used anywhere in the input The syntax for setting variables is VARNAME 1 expression unit VARNAME 2 expression unit where unit is optional If a variable is undefined zero is assumed Variables are useful for running the same input with different actual parameters e g geometries or basis function exponents and to store and manipulate the results Arrays are variables with an index in parenthesis e g var 1 The number of elements in an array var is var The array length can be reset to zero by the CLEAR directive or simply by modifying var Variables and variable arrays may be displayed at any place in the output by the SHOW command and whole tables of variables can be generated using the TABLE command For more details about variables see section 8 3 9 2 String variables Special care is necessary when using strings In order to avoid unexpected results either a has to be prefixed whenever a string variable is set or the string has to be given in quotes Possible forms are namecstring name string name string variable Sname string variable Examples stringl This is a string define a string variable Text in quotes is not converted to upper case string2 stringl lassign string variable str
90. and may be specified using one or more CON cards note that the RESTRICT and SELECT keywords are not used in CASVB CON nj n2 3 n4 The configurations can be specified by occupation numbers exactly as in MULTI see sec tion 20 4 3 so that n is the occupation of the ith valence bond orbital Alternatively a list of Nact orbital numbers in any order may be provided the program determines which definition applies The two cards CON 1 0 1 2 and CON 1 3 4 4 are thus equivalent If no configurations are specified the single covalent configuration 16 ONact is assumed 36 4 2 Selecting the spin basis SPINBASIS key key may be chosen from KOTANI default RUMER PROJECT or LTRUMER specifying the basis of spin eigenfunctions used in the definition of valence bond structures PROJECT refers to spin functions generated using a spin projection operator LTRUMER to Rumer functions with the so called leading term phase convention 36 5 Recovering CASSCF CI vector and VB wavefunction The appropriate MOLPRO records may be specified explicitly using the START directive an alternative is the vbdump mechanism described in section 36 2 1 START ci vb orb trnint ci record name for the CASSCF CI vector The CI vector must have been dumped previ ously using either of the SAVE NATORB CANONICAL or LOCALI directives see sec tion 20 5 4 A default value for ci is determined from the most recent vbdump record s
91. are printed in one column so all variables used must be defined for the same range and corresponding elements should belong together For example if in a calculation one has stored R 1 THETA i ECI i foreach geometry i one can print these data simply using TABLE R THETA ECI By default the number of rows equals the number of elements of the first variable This can be changed however using the RANGE subcommand The first ten columns of a table may contain string variables For instance hf etot 1 energy method 1 program cpu 1 cpustep ccsd etot 2 energy method 2 program cpu 2 cpustep qci etot 3 energy method 3 program cpu 3 cpustep table method etot cpu prints a table with the SCF CCSD and OCT results in the first second and third row respec tively For other use of string variables and tables see e g the examples h20_tab com and oh macros com The apparence of the table may be modified using the following commands which may be given in any order directly after the the TABLE card HEADING headl head2 Specify a heading for each column By default the names of the variables are used as headings FORMAT format Specify a format for each row in fortran style format must be enclosed by quotes Normally the program determines au tomatically an appropriate format which depends on the type and size of the printed data FTYP typl typ2 typ3 Simplified form to modify t
92. atom2 respectively For instance DEFAULT VTZ O AVTZ H VDZ uses VTZ as the default basis sets but sets the basis for oxygen to AVTZ and for hydrogen to VDZ This name conventions for the atom specific basis sets work exactly as described above for default basis sets The keyword DEFAULT can be abbreviated by DEF Any DEFAULT basis set defined in a basis set block supercedes a previous one given outside the basis block The specifications SET DEFAULT at om name are all optional If DEFAULT is not given the previous default as specified on the last previous BASIS card is used 13 BASIS INPUT 81 If no further primitive basis set specifications follow one can also use the one line form BASIS DEFAULT VTZ O AVTZ H VDZ or BASIS VTZ O AVTZ H VDZ Both of these are equivalent to BASIS DEFAULT VTZ O AVTZ H VDZ END Note that any new BASIS card supercedes all previous basis input except for the default basis unless this is given The optional additional primitive basis set specifications see next section are appended to the given atom specific basis sets 1 e the union of atom specific and primitive basis set definitions is used for the atom Examples BASIS DEFAULT VTZ use cc pVTZ basis as default H VDZ use cc pVDZ for H atoms END This could also be written as BASIS DEF VTZ H VDZ BASIS DEFAULT VTZ use cc pVTZ basis as default H VDZ use cc pVDZ for H atoms D H
93. boundaries which optimizes I O but unused space is between matrices both on disc and in core With LTR 1 all matrices are stored dense This might increase I O if much paging is necessary but reduce I O if everything fits in core NCPUS Maximum number of CPUs to be used in multitasking 21 4 Miscellaneous thresholds THRESH codel value code2 value If value 0 the corresponding threshold is set to zero otherwise 10 value The equal signs may be omitted If no codes are specified the default values are printed The following codes are allowed max 7 per card ZERO numerical zero THRDLP delete pairs if eigenvalue of overlap matrix is smaller than this threshold PNORM delete pair if its norm is smaller than this threshold all pairs are normalized to one for a closed shell case PRINTCI print CI coefficients which are larger than this value INTEG omit two electron integrals which are smaller than this value ENERGY convergence threshold for energy see also ACCU card COEFF convergence threshold for coefficients see also ACCU card SPARSE omit coefficient changes which are smaller than this value EQUAL set values in the internal vector and the diagonal elements equal if they differ by less than this value Useful for keeping track of symmetry 21 5 Print options PRINT codel value code2 value Print options Generally the value determines how much intermediate information is printed
94. computation If however a new basis block is presented in the input then the program marks as outdated any quantities such as integrals that have been calculated with the old basis set subsequent job steps will then use the new basis c Even tempered basis sets type atom E VEN nprim ratio centre dratio or type atom EVEN NPRIM nprim RAT I O ratio CENTRE centre DRAT I O dratio Generates a generalized even tempered set of functions The number of functions n is specified by nprim their geometric mean c by centre the mean ratio of successive exponents r by ratio and the variation of this ratio d by dratio If centre is not given the previous basis of the same type is extended by diffuse functions If in this case ratio is not given r is determined from the exponents of the last two function of the previous basis If this is not possible the default r 2 5 is adopted d 1 the default specifies a true even tempered set but otherwise the ratio between successive exponents changes linearly the exponents are given explicitly by 1 loge loge n 1 2 i logr 5 n 1 2 i logd i 1 2 n Example 1 SP 1 VT2 C SP l1 EVEN l generates the generally contracted s and p triple zeta basis sets for atom 1 and extends these by one diffuse function Example 2 SPD 1 VTZ DELETE 1 C SP 1 EVEN NPRIM 2 RATIO 2 5 generates the generally contracted s p triple zeta basis sets for atom 1 Two energy optimized d fun
95. coupling matrix element CONICAL 6100 1 save information for optimization of conical intersection Force SAMC 5101 1 compute gradient for state 1 CONICAL 6100 1 save information for optimization of conical intersection Force SAMC 5102 1 compute gradient for state 2 CONICAL 6100 1 save information for optimization of conical intersection optg startcmd multi find conical intersection This second example optimizes the singlet triplet intersection in LiH ground state is Sin glet excited state is Triplet 39 GEOMETRY OPTIMIZATION OPTG 267 examples 1ih2 _SOTO com Revision 2002 10 ERA ilz basis sto 3g geometry nosym Li Hl LiT H2 Li r H1 theta r 3 7 theta 160 hf wf 4 1 0 multi occ 7 examples wf 4 1 0 singlet state 1ih2 SOTO com wf 4 1 2 triplet state CPMCSCF GRAD 1 1 spin 0 accu 1 0d 7 record 5101 1 cpmcscf for gradient of singlet state CPMCSCF GRAD 1 1 spin 1 accu 1 0d 7 record 5100 1 cpmcscf for gradient of triplet state Force SAMC 5101 1 state averaged gradient for singlet state CONICAL 6100 1 NODC save information for OPTCONICAL Force SAMC 5100 1 state averaged gradient for triplet state CONICAL 6100 1 NODC save information for OPTCONICAL optg startcmd multi gradient 1 d 6 find singlet triplet crossing point 39 3 Using the SLAPAF program for geometry optimization It is optionally possible to use the SLAPAF program written by
96. dr ASTEP da where dr is the displacement for distances or Cartesian coordinates in bohr and da is the displacement for angles in degree The value of RSTEP is used for symmetrical displacements The step sizes for individual variables can be modified using VARSTEP varname value where the value must be in atomic units for distances and in degree for angles 38 2 4 Active and inactive coordinates By default numerical gradients are computed with respect to all variables on which the Z matrix depends or for all 3N coordinates if there are no variables or XYZ inputstyle is used One can define subsets of active variables using ACTIVE variables If this card is present all variables which are not specified are inactive Alternatively INACTIVE variables In this case all variables that are not given are active 38 3 Saving the gradient in a variables Tf the directive VARSAV 38 ENERGY GRADIENTS 250 is given the gradient is saved in variables GRADX GRADY GRADZ GRADX n is the derivative with respect to x for the n th atom The atoms are in the order as printed This order can be different from the order in the input z matrix since the centres are reordered so that all atoms of the same type follow each other 39 GEOMETRY OPTIMIZATION OPTG 251 optgeo tex Revision 2006 1Patch 2006 1 optgairectives 39 GEOMETRY OPTIMIZATION OPTG Automatic geometry optimization is invoked using the OPTG command The
97. e a B a B tres n v mana e r a B T3 Us Va Wa Xs Ys P3 6 o B 1 Ela B Hel BI i Vo iiA er T V W XPO CA gt 151 1 a B 1 4 43428 152 r o B v3 eB 152 a 153 C a B TS 153 1 1 2 43 a ge 2 z 154 e 24 2 2 nt AT 155 A ST U VW X y p ur In t vr wr xr yrP gt c 1 709921 156 T 0 031091 0 015545 0 016887 157 0 21370 0 20548 0 11125 158 V 7 5957 14 1189 10 357 159 W 3 5876 6 1977 3 6231 160 C DENSITY FUNCTIONAL DESCRIPTIONS 332 X 1 6382 3 3662 0 88026 161 Y 0 49294 0 62517 0 49671 162 P111 163 A 0 51473 6 9298 24 707 23 110 11 323 164 B 0 48951 0 2607 0 4329 1 9925 2 4853 165 C 1 09163 0 7472 5 0783 4 1075 1 1717 166 and A 0 006 0 2 0 004 167 C 19 HCTH147 Handy least squares fitted functional See reference for more details K e pa pg Pa 0 pg O Ao Ain d 24 A2 m 4 24 A3 m 4 24 A4 n d 24 Y e ps 0 Bo Bin 05 7 32 Ba n 5 32 Bs n 5 7 32 168 Ba n x 22 3 8 3425 VT ps 49 Co Cin xs 2 34 C n s 7 A3 7 C n 05 5 33 Ca n xs A3 4 where d V2 xa V2 xg 169 uU 170 n 0 1 Dru 170 e a B a B stre ntm aan e r o B T3 Us Va Ws Xs Ys
98. eccsd eccsdt produce a table with results head rl1 r2 theta scf ccsd ccsd t modify column headers for table save h20 tab save the table in file h20 tab title Results for H20 basis basis title for table SorU 3 15 2 sort table This produces the following table Results for H20 basis VDZ R1 R2 THETA SCF CCSD CCSD T 1 6 1 6 100 0 1949 901 3 38 76 20140563 76 20403920 Die 7 1 6 100 0 76 00908379 76 21474489 76 21747582 17 1 7 100 0 76 02060127 76 22812261 76 23095473 2 0 1 9 110 0 76 01128923 76 22745359 76 23081968 2 0 2 0 110 0 76 00369171 76 22185092 76 22537212 You can use also use DO loops to repeat your input for different methods SRevision 2006 0 h20 benchmark method hf fci ci cepa 0 cepa 1 cepa 2 cepa 3 mp2 mp3 mp4 N qci ccsd bccd qci t ccsd t bccd t casscf mrci acpf basis dz Double zeta basis set geometry 0o hl o r h2 0 r hl theta Z matrix for geometry r 1 ang theta 104 Geometry parameters do i 1 method Loop over all requested methods examples Smethod i call program h2o manymethods co e i energy save energy for this method enddo escf e 1 scf energy efci e 2 fci energy table method e scf fci print a table with results Title for table title Results for H20 basis Sbasis R r Ang Theta Stheta degr This calculation produces the following table 5 INTRODUCTORY EXAMPLES Results for H20 basis DZ METHOD HF 75 FCI 76
99. either by supplying the ENEPART direc tive ENEPART epart iepart or by giving the parameters as options on the command line The epart parameter determines the cutoff distance for intramolecular bond lengths in a u default 3 a u and is used to automatically determine the individual monomer subunits of the cluster The iepart parameter enables the energy partitioning if set to a value larger than zero default 1 Additionally if iepart is set to 2 a list of all intermolecular pair energies and their components is printed The output section produced by the energy partitioning algorithm will look similar to the fol lowing example energy partitioning enabled centre groups formed for cutoff au 3 00 1 01 H11 H12 28 LOCAL CORRELATION TREATMENTS 190 2 02 H21 H22 energy partitioning relative to centre groups intramolecular correlation 43752663 xchange dispersion 00000037 dispersion energy 00022425 ionic contributions 00007637 The centre groups correspond to the individual monomers determined for epart 3 In the present example two water monomers were found The correlation energy is partitioned into the four components shown above The exchange dispersion dispersion and ionic components reflect directly the related intermolecular components of the complex while the intramolecu lar correlation contribution to the interaction energy has to be determined by a super molecular calculation i e by
100. else has been specified on an ORBITAL card if present By default the orbitals are not printed and the hamiltonian is not diagonalized for the new orbitals The following options can be specified in any order CI Diagonalize the hamiltonian in the basis of the computed natu ral orbitals and print the configurations and their associated co efficients This has the same effect as the GPRINT CIVECTOR directive see section 6 12 By default only configurations with coefficients larger than 0 05 are printed This threshold can be modified using the THRESH see section 20 8 2 or GTHRESH see section 6 11 options STATE state Compute natural orbitals for the specified state state has the form istate isym e g 3 2 for the third state in symmetry 2 In contrast to earlier versions isym refers to the number of the irreducible representation and not the sequence number of the state symmetry It is therefore independent of the order in which WF cards are given The specified state must have been optimized If STATE is not given and two or more states are averaged the natural orbitals are calculated with the state averaged density matrix default SPIN ms2 Compute natural orbitals for states with the specified spin ms2 equals 2 x S i e O for singlet 1 for doublet etc This can be used to together with STATE to select a specific state in case that states of different spin are averaged If STATE is not spec ified the state averaged de
101. energy for excited state enddo table r el caspt2 e2 caspt2 el mscaspt2 e2 mscaspt2 title MS MR CASPT2 for LiF plot file lif mr mscaspt2 plot This produces the plot MR MR CASPT for LiF 107 T T T T T T T 107 05 107 1 e e El CASPT2 m nm E2 CASPT2 El MSCASPT2 4 4 A E2 MSCASPT2 107 2 L L L L L L L One can clearly see that this gives smoother potentials than the SS SR CASPT 2 calculation in the previous section Also the avoided crossing is shifted to longer distances which is due to the improvement of the electron affinity of F examples lif mr mscaspt2 com 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 152 22 4 Modified Fock operators in the zeroth order Hamiltonian The g1 g2 and g3 operators proposed by Andersson Theor Chim Acta 91 31 1995 as well as a further g4 operator may be used g4 makes CASPT2 calculations size extensive for cases in which a molecule dissociates to high spin open shell RHF atoms The index n of the operator to be used is specified on the RS2 RS2C or RS3 card RS2 option RS2C option RS3 option where option can be G1 G2 G3 or G4 This option can be followed or preceded by other valid options 22 5 Level shifts Level shifts are often useful to avoid intruder state problems in excited state calculations MOL PRO allows the use of shifts as described by Roos and Andersson Chem Phys Lett
102. for SA MCSCF For computing difference gradients use CPMCSCF DGRAD statel state2 ACCU thresh RECORD record where state and state2 specify the two states considered e g 2 1 3 1 for the second and third states in symmetry 1 The gradient of the energy difference will be computed Both states must have the same symmetry record specifies a record on which the gradient information is stored the default is 5101 1 thresh is a threshold for the accuracy of the CP MCSCF solution The default is 1 d 7 The gradients are computed by a subsequent call to FORCES or OPTG 20 9 3 Non adiabatic coupling matrix elements for SA MCSCF For computing non adiabatic coupling matrix elements analytically use CPMCSCF NACM statel state2 AC CU thresh RECORD record where state and state2 specify the two states considered e g 2 1 3 1 for the second and third states in symmetry 1 Both states must have the same symmetry record specifies a record on which the gradient information is stored the default is 5101 1 This will be read in the subse quent gradient calculation thresh is a threshold for the accuracy of the CP MCSCF solution The default is 1 d 7 NADC and NADK are an aliases for NADC and SAVE is an alias for RECORD The matrix elements for each atom are computed by a subsequent call to FORCES Note this program is not yet extensively tested and should be used with care 20 THE MCSCF PROGRAM MULTI 129 20 10 Optimizing valenc
103. generating the density can be introduced through a p Ta Livo gt Tp Ivo T To 5 8 All of the available functionals are of the general form F Ps Ps Oss Oss Oss Ts Ts Us Us d rK psps Os Oss Oss Ts Ts Ds Us 9 where is the conjugate spin to s Below is a list of keywords for the functionals supported by MOLPRO Additionally there are a list of alias keywords deatailed in the next section for various combinations of the primary functionals listed below Xa with Modified Gradient Correction XaPy Re optimised Becke88 Correlation Functional Becke88 Exchange Functional Becke95 Correlation Functional Density functional part of B97 Re parameterized by Hamprecht et al Density functional part of B97 Becke Roussel Exchange Functional Becke Roussel Exchange Functional Uniform Electron Gas Limit Becke Wigner Exchange Correlation Functional Colle Salvetti correlation functional Colle Salvetti correlation functional 18 THE DENSITY FUNCTIONAL PROGRAM 105 Slater Dirac Exchange Energy Gill s 1996 Gradient Corrected Exchange Functional Handy least squares fitted functional Handy least squares fitted functional Handy least squares fitted functional Local t Approximation Lee Yang and Parr Correlation Functional Exchange Functional for Accurate Virtual Orbital Energies Exchange Functional for Accurate Virtual Orbital Energies P86 PBE Correlation Functional Revised PBE Exchange Functiona
104. geometry and must be specified separately on the FROZEN card Otherwise the program is likely to stop with error non orthogonal core orbitals The program remembers where to take the core orbitals from if these have been specified on a FROZEN card in a previous MCSCF calculation 20 5 2 Rotating pairs of initial orbitals ROTATE orbl sym orb2 sym angle Performs a 2 x 2 rotation of the initial orbitals orb and orb2 in symmetry sym by angle degrees With angle 0 the orbitals are exchanged ROTATE is meaningful only after the START card See MERGE for other possibilities to manipulate orbitals 20 5 3 Saving the final orbitals ORBITAL record file The orbitals are dumped to record record file Default for record is 2140 and file 2 This default record number is incremented by one for each subsequent MCSCF calculation in the same job see section 4 11 Therefore if several different MCSCF calculations at several geometries are performed in one job each MCSCF will normally start with appropriate orbitals even if no ORBITAL or START card is present The ORBITAL card can be omitted if a NATORB CANORB or LOCORB card is present since orb can also be specified on these cards the same defaults for orb as above apply in these cases 20 5 4 Saving the CI vectors and information for a gradient calculation Old form obsolete SAVE cidump refsav grdsav New form 20 THE MCSCF PROGRAM MULTI 119 SAVE CI cidump REF refsav GRD
105. given with newline forced after n elements Note that the total length of the format should not exceed 100 characters a left margin of 30 characters is always needed A wild card format can be used to show several variables more easily SHOW qm dm shows all variables whose names begin with OM and DM Note that no letters must appear after the i e the wild card format is less general than in UNIX commands See the TABLE command for another possibility to tabulate results 8 10 Clearing variables Variables can be deleted using CLEAR namel name2 Wild cards can be used as in SHOW e g CLEAR ENERG clears all variables whose names begin with ENERG All variables can be cleared using CLEARALL The length of vectors can be truncated simply by redefining the length specifier R 2 truncates the array R to length 2 Higher elements are no longer available but could be redefined Setting R 0 is equivalent to the command CLEAR R 8 11 Reading variables from an external file Variables can be read from an external file using READVAR filename Such files can be save for instance by the geometry optimization program and reused later to recover a certain optimized geometry The format of the input in filename is the same as for ordinary input 9 TABLES AND PLOTTING 54 9 TABLES AND PLOTTING 9 1 Tables Variables can be printed in Table form using the command TABLE varl var2 The values of each variable
106. grdsav This directive must be placed before any WF or STATE cards The options can be given in any order cidump record name for saving the CI vectors By default the vectors are only written to a scratch file If NATORB CANORB or LOCORB cards are present cidump should be specified on these cards At present there is hardly any use of saved CI vectors and therefore this option is rarely needed refsav record name for saving the orbital configurations and their weights for use in subsequent MULTI or CI calculations using the SELECT directive If wavefunctions for more than one state symmetry are optimized in a state averaged calculation the weights for each state symmetry are saved separately on records refsav istsym 1 x 100 where istsym is the sequence number of the WF card in the input If several NATORB CANORB or LOCORB cards are present the record number is increased by 1000 for each subsequent orbital set Note that this option implies the use of CSFs even of no CONF IG card see section 0 6 1 is present grdsav record name for saving the information which is needed in a subsequent gradient calcu lation This save is done automatically to record 5000 1 if the input contains a FORCE or OPTG card and therefore the GRD option is normally not required 20 5 5 Natural orbitals NATORB record options Request to calculate final natural orbitals and write them to record record The default for record is 2140 2 or what
107. if the variable DEFAULT is set during make install ie make DEFAULT 1 install then a symbolic link is made to INS TBIN name Furthermore If the file INSTBIN molpro does not already exist or if the variable DEFAULT is set to molpro A INSTALLATION OF MOLPRO 313 duringmake installthenasymbolic link is made from INSTBIN nameto INSTBIN molpro The overall effect of this cascade of links is to provide in the normal case the commands molpro and one or both of molpros serial and molprop parallel for normal use with the long names remaining available for explicit selection of particular variants As with the uninstalled program the environment variable MOLPRO RCFILE can be used to override the choice of configuration file For normal single variant installations none of the above has to be worried about and the molpro command will be available from directory INSTLIB During the install process the key from HOME molpro token is copied to INSTLIB token so that the key will work for all users of the installed version When the program has been verified and or installed the command make clean can be used to remove compilation logs nake veryclean will remove all binary and object files retain ing only those files included in the original distribution it is usually recommended that this is not done as it implies that to apply future updates and bug fixes the whole program will have to be recompiled A 3 10 Getting and applying
108. in curley bracketq The general format is either COMMAND options Or COMMAND options directives data Examples for commands are HF MP2 CCSD T MCSCF MRCI Examples for direc tives are OCC CLOSED WF PRINT Directives can be in any order unless otherwise noted Data can follow certain directives For the format of options directives and data see subsections and 3 6 respectively In the following such a sequence of input will be denoted a command block Special command blocks are the geometry and basis blocks The options given on the command line may include any options relevant to the current program For instance in DF LMP2 R12 this could be options for density fitting local explicit and or thresholds Alternatively options can be specified on individual directives like DFIT options LOCAL options EXPLICIT options THRESH options In these cases only the options belong to the corresponding directive are valid thus if an option for EXPLICIT would be specified e g on the DF IT directive an error would result This error would be detected already in the input prechecking stage Depending on the parameter STRICTCHECK in file Lib variable registry the program may tolerate directives given after commands without curley brackets The program checks for ambiguities in the input A directive is considered ambiguous if a command or procedure with the same name is known and the directive is not in a comma
109. input and output dimensions must be identical If orb is nonzero but orb2 is zero orb2 is set to the last orbital in symmetry sym2 If sym2 0 sym2 is set to sym1 off is an offset in the output vector relative to the global offset set by OFFSET directive fac has no effect for move The elements istart to iend of the input vector are moved If istart 0 and iend 0 the whole input vector is moved The usage of the MOVE directive is most easily understood by looking at the examples given below See also ADD and EXTRA commands 42 3 Adding orbitals to the output set ADD ADD orbl syml orb2 sym2 orb3 sym3 ioff fac istart iend This adds orbitals orb1 syml to orb2 sym2 to the output vectors starting at orb3 sym3 The input vectors are scaled by the factor fac If fac 0 fac is set to 1 0 For other details see 42 ORBITAL MERGING 288 MOVE command Note however that the output vectors which have already been defined are not skipped as for MOVE See also MOVE and EXTRA commands 42 4 Defining extra symmetries EXTRA EXTRA exsym orb1 syml orb2 sym2 orb3 sym3 ioff fac istart iend Works exactly as MOVE but only input vectors with extra symmetry exsym are considered If orbl symI and orb2 sym2 are zero all input vectors are moved to the output set ordered according to increasing extra symmetries Examples EXTRA 1 4 1 will move the next 4 orbitals in symmetry 1 which have extra symme try 1 Orbitals which have been move
110. is private or public 6 configure prompts for the destination directory INSTLIB for installation of ancil lary files which are required for program execution 7 configure will attempt to contact the molpro webserver and download an appropriate licence key if it does not a token in the file SHOME molpro token This token will be copied to INSTLIB token during installation 8 configure prompts for the destination directory for documentation This should nor mally be a directory that is mounted on a worldwide web server 9 configure prompts for the destination directory for the CGI scripts that control the delivery of documentation This might be the same directory as h but some web servers require a particular special directory to be used The latter two parameters are relevant only if the documentation is also going to be installed from this directory see below The following command line options are recognized by configure batch disables the prompting described above i8 i4 forces the use of 8 or 4 byte integers respectively L lib specifies any additional directories containing system libraries to be scanned at link time blas 0 1 2 3 4 specifies system BLAS level as described above mpp nompp controls whether compilation is to be for MPP parallelism see above ifort pgf path controls whether the Intel ifort Portland pgf or Pathscale path compiler is to be used on Linux systems
111. it can also be built with a GA library made with the LAPI target configure prompts for the type default tcgmsg and then for the directory holding the associated libraries Normally t cgmsg is recommended which is most efficient on most systems and also most easily installed If a myrinet network is available myrinet should be chosen This requires in addi tion to the usual MPI libraries the gm library and mpirun gm rather than mpirun At present the myrinet option has been tested only on Linux systems The name of the MOLPRO executable is generated from the program version number the library type and the machine architecture It is then possible to install different versions simultaneously in the same MOLPRO tree see section A 3 4 When building Global Arrays on Linux the default is tcgmsg You should build with something similar to make TARGET where TARGET is LINUX on a 32 bit Linux system LINUX64 on a 64 bit system On other platforms consult the README for a list of valid targets The parallel job launcher needed to start molpro can be found at tcgmsg ipcv4 0 parallel and should be copied into your PATH or it s location specified in configure In some cases you will need to specify the compiler you use when building molpro make TARGET FC where for example FC i fort for the Intel compiler FC pgf 90 for Portland or FC pathf90 for Pathscale When building with MPICH you should use something similar t
112. keyl key2 Specify print options TABLE HEADING TITLE WARNING FORMAT SORT The default is print for the first three and noprint for the last three NOPRINT keyl key2 Disable print for given keys NOPUNCH Don t write data to the punch file data are written by default RANGE start end Specify start and end indices of the variables to be printed STATISTICS Print also linear regression and quadratic fits of the data columns 9 2 Plotting P LOT CMD 7 unix plot command F I LE plotfile NOP LOT Execute a plotting program using the table as data PLOT is a subcommand of TABLE and must follow TABLE or any of its valid subcommands given in the previous section unix plot command consists of the unix command needed to start the plotting program followed by any required options The whole thing should normally be enclosed in quotation marks to preserve lower case letters The default is xmgrace At present only the xmgrace grace gracebat and xmgr programs with all numerical data are supported although use of xmgr is deprecated and may not be possible in future versions By default the input file for the plotting program is saved in molpro_plot dat The name of the plotfile can be modified using the FILE or PLOTFILE option FILE implies that the plot is not shown on the screen but all plot data are saved in the given file The plot on the screen can also be suppressed with the NOPLOT option The following addit
113. list must be closed under the orbital permutation induced by label for this to be possible The operator is defined in terms of its action on the active MOs as specified by one or more of the keywords IRREPS COEFFS or TRANS any other keyword will terminate the definition of this symmetry operator If no further keyword is supplied the identity is assumed for abel The alternative format S YMELM abel sign key 1 key 2 may also be used 36 10 2 The IRREPS keyword IRREPS ji i2 3 36 THE VB PROGRAM CASVB 233 The list ij i2 specifies which irreducible representations as defined in the CASSCF wave function are antisymmetric with respect to the abel operation If an irreducible representation is not otherwise specified it is assumed to be symmetric under the symmetry operation 36 10 3 The COEFFS keyword COEFF S i Iy 3 The list ij i2 specifies which individual CASSCF MOs are antisymmetric with respect to the abel operation If an MO is not otherwise specified it is assumed to be symmetric under the symmetry operation This specification may be useful if for example the molecule possesses symmetry higher than that exploited in the CASSCF calculation 36 10 4 The TRANS keyword TRANS nqim Ur Cit CDI Cans Specifies a general ngim X Naim transformation involving the MOs ij i4 specified by the c coefficients This may be useful for systems with a two or three dimensional irreducible r
114. maximum angular momentum in the basis set can be reduced using syntax such as BASIS VOZ D which would omit the f and g functions that would normally be present in the VQZ basis set BASIS VOZ D P would specify additionally a maximum angular momentum of 1 on hydrogen i e would omit d orbitals on hydrogen For generally contracted basis sets an extended syntax can be used to explicitly give the number of contracted functions of each angular momentum For example BASIS ROOS 3s2p1d 2s generates a 6 31G sized basis set from the Roos ANO compilation 13 5 Default basis sets for individual atoms More specific basis set definitions for individual atoms can be given BASIS input blocks which have the following general form BASIS SET type type can be ORBITAL DENSITY or any other name as used in basis specifications for density fitting optional default ORBITAL sets the default basis to name Use basis namel for atoml Use basis namel for atom2 DEFAULT name atoml namel 1 1 1 1 atom2 name2 primitive basis set specifications ladditional basis functions SET type specify basis of another type in following lines END Any number of basis sets can be be given in a basis block The default and atom specifications can also be merged to one line separated by commas DEFAULT name atoml name 1 atom2 name2 Here the basis sets namel name2 overwrite the default basis set name for specific atoms atoml
115. more with large blocksize file sys tems and where binary files are large during compilation Typically 50Mb is needed for the finally installed program Large calculations will require larger amounts of disk space 5 One or more large scratch file systems each containing a directory that users may write on There are parts of the program in which demanding I O is performed simultane ously on two different files and it is therefore helpful to provide at least two filesystems on different physical disks if other solutions such as striping are not available The directory names should be stored in the environment variables TMPDIR STMPDIR2 STMPDIR3 These variables should be set before the program is installed preferably in profile or cshro since at some stages the installation procedures will check for them cf section A 3 6 A INSTALLATION OF MOLPRO 306 6 If the program is to be built for parallel execution then the Global Arrays toolkit is needed We recommend version 4 0 1 although earlier versions should also work This is avail able from http www emsl pnl gov docs global and should be installed prior to com piling MOLPRO In some installations GA uses the tcgmsg parallel harness on others it sits on an existing MPI subsystem and on others it makes use of the native parallel subsystem e g LAPI MOLPRO can be built to use any of these although it is not nor mally recommended to use MPI where other possibil
116. no copying will take place On some main frames the scratch directory is erased automatically after a job has terminated and in such cases a different I directory e g SHOME int can be specified environment variables will be expanded at run time In view of the large integral file sizes this should be used with care however Note that in parallel runs with more than 1 processor the integral file will never be copied and cannot be restarted This determines the destination of permanent wavefunction dump files used for storing information like orbitals or CI vectors etc These files are essential for restarting a job As explained for the integral files above permanent wavefunction files will be copied to directory after completion of the job The default for directory is SHOME wfu where key is the licence key obtainable as described in section A 1 The default local memory and GA memory should be checked to be appropriate for the hardware environment The number of processors or their identity can be specified explicitly in the configuration file but very often it is neither desirable nor nec essary to do so Where possible the molpro program extracts a rea sonable default for the node specification from the controlling batch system e g LoadLeveler PBS Usually the user will want to either specify n explicitly on the command line or rely on molpro s at tempts to get it from the batch system A INSTALLATION OF MOLPRO 3
117. occupied orbitals NACTV 0 Number of active virtual orbitals SACC 0 Spin adapted coupled cluster DBOC 0 Diagonal BO correction MEMORY 1 Memory TOL ENERGY 1 0 Energy convergence threshold FREO 0 0 Frequency for dynamic polarizabilities FILE fort Name for MRCC fortran files CONVER ICONV 0 See mrcc manual CS 1 See mrcc manual DIAG 0 See mrcc manual MAXEX 0 See mrcc manual SPATIAL 1 See mrcc manual a 1 means default value taken from MOLPRO 27 THE MRCC PROGRAM OF M KALLAY MRCC Table 10 Methods available in the MRCC program MRCC parameters Key METHOD LEVEL Notes CI n configuration interaction methods CISD 0 2 CISDT 0 3 CISDTO 0 4 CI N 0 N Specify excitation level N using LEVEL CC N coupled cluster methods CCSD 1 2 CCSDT 1 3 CCSDTO 1 4 CC N 1 N Specify excitation level N using LEVEL CC N 1 N coupled cluster methods CCSD T 2 3 CCSDT O 2 4 CC N 1 N 2 N Specify excitation level N using LEVEL CC N 1 N coupled cluster methods Also computes n corrections CCSD T 3 3 CCSDT Q 3 4 CC N 1 N 3 N Specify excitation level N using LEVEL CC n 1 n L methods also computes n and n corrections CCSD T L 4 3 CCSDT 0 L 4 4 CC N 1 N L 4 N Specify excitation level N using LEVEL CC n 1a methods CCSDT 1A 5 3 CCSDTQ 1A 5 4 CC N 1A 5 N Specify excitation level N using LEVEL CC n 1b methods CCSDT 1B 6 3 CCSDTQ 1B 6 4 CC N 1B 6 N Specify
118. of options for Brueckner calculations Normally none of the options has to be specified and the BCCD command can be used to perform a Brueckner CCD calculation orbbrk if nonzero the Brueckner orbitals are saved on this record ibrstr First iteration in which orbitals are modified default 3 ibrueck Iteration increment between orbital updates default 1 brsfak Scaling factor for singles in orbital updates default 1 24 THE CLOSED SHELL CCSD PROGRAM 162 24 4 Singles doubles configuration interaction CISD Performs closed shell configuration interaction CISD The same results as with the CI pro gram are obtained but this code is somewhat faster Normally no further input is needed For specifying DIIS directives see section 24 5 The DIIS directive DIIS itedis incdis maxdis itydis This directive allows to modify the DIIS parameters for CCSD QCISD or BCCD calculations itedis First iteration in which DIIS extrapolation may be performed de fault 4 incdis Increment between DIIS iterations default 1 maxdis Maximum number of expansion vectors to be used default 6 itydis DIIS extrapolation type itedis 1 default residual is minimized itedis 2 AT is minimized In addition there is a threshold THRDIS which may be modified with the THRESH directive DIIS extrapolation is only done if the variance is smaller than THRDIS 24 6 Examples 24 6 1 Single reference correlation treatment
119. of strong and weak pairs default 3 The primary domains are extended according to RDOMAUX_MP2 or IDOMAUX_MP 2 11 DENSITY FITTING FITDOM_CCSD RDOMAUX_SCF IDOMAUX_SCF RDOMAUX_CORE IDOMAUX_CORE RDOMSCF_START IDOMSCF_START RDOMSCF_FINAL IDOMSCF_FINAL RDOMAUX_MP 2 IDOMAUX MP2 RDOMAUX CCSD IDOMAUX CCSD RDOMAUX_CPHF RDOMAUX SCFGRD SCSGRD 68 Similar to FITDOM_MP2 but used for LCCSD 2 ext transfor mation Distance criterion for fitting domain extension in SCF default 5 0 Connectivity criterion for fitting domain extension in SCF de fault 0 Distance criterion for core orbital fitting domain extension in SCF default RDOMAUX SCF Connectivity criterion for core orbital fitting domain extension in SCF default IDOMAUX_SCF Distance criterion for fitting domain extension in the initial SCF iterations default 3 0 Connectivity criterion for fitting domain extension in the initial SCF iterations default 1 Distance criterion for fitting domain extension in the final SCF iterations default RDOMAUX SCF Connectivity criterion for fitting domain extension in the final SCF iterations default IDOMAUX SCF Distance criterion for fitting domain extension in LMP2 The default value depends on FITDOM MP2 Connectivity criterion for fitting domain extension in LMP2 The default value depends on FI TDOM MP2 Distance criterion for fitting domain
120. of this program must acknowledge the above See also H J Werner and E A Reinsch J Chem Phys 76 3144 1982 H J Werner Adv Chem Phys 59 1 1987 The command CI or CI PRO calls the program The command CISD calls fast closed shell CISD program The command QCI calls closed shell quadratic CI program The command CCSD calls closed shell coupled cluster program The following options may be specified on the command line NOCHECK Do not stop if no convergence DIRECT Do calculation integral direct NOSING Do not include singly external configurations NOPAIR Do not include doubly external configurations not valid for single reference methods MAXIT value Maximum number of iterations MAXITI value Maximum number of microiterations for internals SHIFTI value Denominator shift for update of internal configurations SHIFTS value Denominator shift for update of singles SHIFTP value Denominator shift for update of doubles THRDEN value Convergence threshold for the energy THRVAR value Convergence threshold for the CI vector This applies to the square sum of the changes of the CI coefficients 21 1 Introduction The internally contracted MRCI program is called by the CI command This includes as special cases single reference CI CEPA ACPF MR ACPF and MR AQCC For closed shell reference functions a special faster code exists which can be called using the CISD QCI or CCSD commands
121. on the ORBITAL directive see section 20 5 3 In contrast to earlier versions of MOLPRO it is possible that orbref and orbsav are the same The specifications TYPE STATE SPIN can be used to select specific sets of reference orbitals as described in section 4 11 orbl orb2 is a pair of orbitals for which the overlap is to be maximized These orbitals are specified in the form number sym e g 3 1 means the third orbital in symmetry 1 If orbl orb2 are not given the overlap of all active orbitals is maximized pri is a print parameter If this is set to 1 the transformation angles for each orbital are printed for each jacobi iteration Using the defaults described above the following input is sufficient in most cases DIAB orbref Using Molpro98 is is not necessary any more to give any GEOM and DISPL cards The displacements and overlap matrices are computed automatically the geometries are stored in the dump records along with the orbitals The diabatic orbitals have the property that the sum of orbital and overlap contributions in the non adiabatic coupling matrix elements become approximately zero such that the adiabatic mixing occurs only through changes of the CI coefficients This allows to determine the mix ing angle directly from the CI coefficients either in a simple way as described for instance in J Chem Phys 89 3139 1988 or in a more advanced manner as described by Pacher Cederbaum and K ppel in J Chem Phys 89 7
122. orbitals which are occupied in any of the reference CSFs In the MCSCF FROZEN orbitals are doubly occupied in all CSFs and frozen not optimized while closed denotes all doubly occupied orbitals frozen plus optimized In the CI and CCSD programs core orbitals are those which are not correlated and closed orbitals are those which are doubly occupied in all reference CSFs OCC CORE and CLOSED commands are generally required in each program module where they are relevant however the program remembers the most recently used values and so the com mands may be omitted if the orbital spaces are not to be changed from their previous values Note that this information is also preserved across restarts Note also as with the WF informa tion sensible defaults are assumed for these orbital spaces For full details see the appropriate program description 4 11 Selecting orbitals and density matrices ORBI TAL DENSITY As outlined in section 4 3 the information for each SCF or MCSCF calculation is stored in a dump record Dump records contain orbitals density matrices orbital energies occupa tion numbers fock matrices and other information as wavefunction symmetries etc Subse quent calculation can access the orbitals and density matrices from a particular record using the ORBITAL and DENSITY directives respectively These input cards have the same structure in all programs The general format of the ORBI TAL and DENSITY direct
123. pair energies plus ba sis information ipri 2 Debugging output THRBINV Threshold below which non physical eigenvalues are projected from approximate B matrices THRINT Threshold for integral screening Local variants of the DF MP2 F12 methods are available invoked by the commands DF LMP2 F12 or DF LMP2 R12 Special options for these local variants are PAIRS Specifies which pairs to be treated by R12 or F12 STRONG CLOSE WEAK ALL pairs up to the given level are in cluded The default is STRONG DEBUG Parameter for debug print LOCFIT F12 If set to one use local fitting Default is no local fitting LOCF1T_F12 0 LOCFIT R12 Alias for LOCFIT F12 Local fitting is not recommended in R12 calculations 29 EXPLICITLY CORRELATED METHODS 198 FITDOM Determine how the base fitting domains are determined 0 Fitdomains based on united operator domains 1 Fitdomains based in orbital domains 2 Fitdomains based on united pair domains using strong pairs 3 Fitdomains based on united pair domains using strong close and weak pairs default RDOMAUX Distance criterion for density fitting domain extensions in case of lo cal fitting The default depends on FITDOM IDOMAUX Connectivity criterion for density fitting domain extensions in case of local fitting RAODOM Distance criterion for RI domain extensions Zero means full RI basis default If this parameter is chosen to be nonzero it must be rather large
124. performing counterpoise corrected geometry optimizations see section 39 4 7 13 BASIS INPUT 13 1 Overview sets and the basis library Basis functions are used in Molpro not just for representing orbitals but also for providing aux iliary sets for density fitting see 11 and for simplifying integrals through approximate identity resolution in explicitly correlated methods see 29 In order to accommodate this the program maintains internally a number of different sets The first of these always has the name ORBI TAL and is the primary basis set for representing orbitals and others can be defined as necessary as described below or else are constructed automatically by the program when required In the 13 BASIS INPUT 78 latter case the density fitting and other modules attempt to guess a reasonable libary fitting ba sis that should be appropriate for the orbital basis set it is advisable to check the choice when using anything other than a standard orbital basis set The basis sets may either be taken from the program library or may be specified explicitly or any combination Optionally the basis function type can be chosen using the CARTESIAN or SPHERICAL commands 13 2 Cartesian and spherical harmonic basis functions MOLPRO uses spherical harmonics 5d 7f etc by default even for Pople basis sets like 6 31G This behaviour may be different to that of other programs However cartesian functions can be requested using the CAR
125. programs with identical functionality the preferred code is SEWARD R Lindh which is the best on most machines ARGOS R M Pitzer is avail able as an alternative and in some cases is optimum for small memory scalar machines Also two different gradient integral codes namely CADPAC R Amos and ALASKA R Lindh are available Only the latter allows the use of generally contracted symmetry adapted gaussian basis functions Effective Core Potentials contributions from H Stoll e Many one electron properties e Some two electron properties e g fs L L2 L L etc Closed shell and open shell spin restricted and unrestricted self consistent field Density functional theory in the Kohn Sham framework with various gradient corrected exchange and correlation potentials Multiconfiguration self consistent field This is the quadratically convergent MCSCF procedure described in J Chem Phys 82 1985 5053 The program can optimize a weighted energy average of several states and is capable of treating both completely gen eral configuration expansions and also long CASSCF expansions as described in Chem Phys Letters 115 1985 259 Multireference CI As well as the usual single reference function approaches MP2 SDCI CEPA this module implements the internally contracted multireference CI method as described in J Chem Phys 89 1988 5803 and Chem Phys Lett 145 1988 514 Non variational variants e g MR ACPP as describ
126. running procedures for MPP machines Parallel direct scf and scf gradients are working These features are only available with the MPP module which is not yet being distributed 13 Important bugfixes for DFT grids CCSD with paging finite field calculations without core orbitals spin orbit coupling 14 Many other internal changes As an additional service to the MOLPRO community an electronic mailing list has been set up to provide a forum for open discussion on all aspects of installing and using MOLPRO The mailing list is intended as the primary means of disseminating hints and tips on how to use Molpro effec tively It is not a means of raising queries directly with the authors of the program For clearly demonstrable program errors reports should continue to be sent to molpro support molpro net however how to questions sent there will merely be redirected to this mailing list In order to subscribe to the list send mail to molpro user request molpro net containing the text subscribe for help send mail containing the text help Messages can be sent to the list molpro user molpro net but this can be done only by subscribers Previous postings can be viewed in the archive at http www molpro net molpro user archive irrespective of whether or not you subscribe to the list Experienced Molpro users are encouraged to post responses to queries raised Please do contribute to make this resource mutually useful B 5 Facil
127. saddle point optimization In the present implementation a saddle point search is possible with the rational function method METHOD RF the geometry DIIS method METHOD DIIS and the quadratic steepest descent method of Sun and Ruedenberg METHOD SRTRANS Note that convergence is usually much more difficult to achieve than for minimizations In par ticular a good starting geometry and a good approximation to the hessian is needed The latter is achieved by evaluating the hessian numerically see section 39 2 7 or using a precomputed hessian see section 39 2 6 39 2 12 Setting a maximum step size STEP STEP steplength drmax damax drmaxl damaxl steplength is the initial step length in the scaled parameter space default 0 3 In the AH method this is dynamically adjusted and can have a maximum value ahmax see TRUST drmax is the initial max change of distances in bohr default 0 3 In the AH method this is dynamically adjusted up to a maximum value of drmax1 default 0 5 bohr damax is the initial max change of angles in degree default 2 In the AH method this is dynamically adjusted up to a maximum value of damax1 default 10 degrees 39 2 13 Redefining the trust ratio TRUST TRUST ratio ahmax ratio determines the radius around the current minimum in which points are used to update the Hessian with the conjugate gradient method default 0 5 see also UPDATE ahmax is the maximum step size allowed in the Augme
128. should be any of the ones used in the LOAD command In addition SUBTYPE can be specified if necessary This describes e g the type of orbitals or density matrices e g for natural orbitals TYPE ORB and SUBTYPE NATURAL The matrix symmetry needs to be given only if it is not equal to 1 43 22 Writing a matrix to an ASCII file WRITE WRITE name filename status Writes a matrix to an ASCII file If filename is not given the matrix is written to the output file otherwise to the specified file filename is converted to lower case If filename PUNCH it is written to the current punch file If status NEW ERASE or em REWIND a new file is written otherwise as existing file is ap pended 43 23 Examples The following example shows various uses of the MATROP commands 43 MATRIX OPERATIONS SRevision 2006 0 h20 matrop examples geometry 0o hl o r h2 0 r hl theta r 1 ang theta 104 hf multi natorb canonical matrop load D ao DEN 2140 2 load Cnat ORB 2140 2 natural load Ccan ORB 2140 2 canonical load Dscf DEN 2100 2 load S prio Cnat 4 1 2 elem d11 Dscf 1 1 1 elem d21 Dscf 2 1 1 elem d12 Dscf 1 1 2 PRR tran S_mo s Cnat print S_mo trace Nao S_mo trace Nel D_ao S mult SC S Cnat tran D_nat D_ao SC prid D_nat dmo D can D ao Ccan add D neg 1 D can diag U EIG D neg mult Cnat1 Ccan U prio Cnat1 4 1 2 natorb Cnat2 D ao prio Cnat2 4 1 2 add diffden D ao 1
129. sigma state save orbitals to record 2100 2 default examples Define active and inactive orbitals ne cn sa casscf com Start with RHF orbitals from above Save configuration weights for CI in record 4000 2 Define the four states Print natural orbitals and associated ci coefficients Compute matrix elements over LZ compute expectation values for LZZ Example for an RASSCF restricted active space calculation for N2 including SCF determi nant plus all double excitations into valence orbitals The single excitations are excluded D symmetry CSF method used SRevision 2006 0 N2 geometry N1 N2 N1 r r 2 2 hf occ 3 1 1 2 wf 14 1 save 2100 2 tmultiroeccr3 dr Li Ll freeze 1 1 2100 2 config wf 14 1 restrict 0 2 93 5 1 rtestrrtet l 1 349 print refl natorb ci print Cydia AC l5 lef geometry input bond length scf calculation Define occupied orbitals Define frozen core scf orbitals Use CSF method Define state symmetry Restriction to singles and doubles Take out singles Print configurations Print natural orbitals and CI coeffs examples n2 rasscf com 21 THE CI PROGRAM 131 21 THE CI PROGRAM Multiconfiguration reference internally contracted configuration interaction Bibliography H J Werner and P J Knowles J Chem Phys 89 5803 1988 P J Knowles and H J Werner Chem Phys Lett 145 514 1988 All publications resulting from use
130. space can be one of the following ZMAT Optimize all variables on which the Z matrix depends default if the geometry is given as Z matrix 3N Optimize all 3N cartesian coordinates default if the Z matrix depends on no variables or if xyz input is used Z Matrix input coordinates will be destroyed if 3N is used opt coord determines the coordinates in which the optimization takes place By default local normal coordinates are used Optionally cartesian coordinates or natural internal coordinates can be used opt coord can be one of the following NORMAL Optimization in local normal coordinates This is default if the Model Hessian is used to approximate the Hessian NONORM Don t use local normal coordinates BMAT filename Use Pulay s natural internal coordinates see G Fogarasi X Zhou P W Taylor and P Pulay J Am Chem Soc 114 8191 1992 P Pulay G Fogarasi F Pang J E Boggs J Am Chem Soc 101 2550 1979 Optionally the created coordinates as well as additional in formations about this optimization are written to the specified file These coordinates resemble in part the valence coordinates used by vibrational spectroscopist and have the advantage of decreasing cou pling between different modes This often increases the speed of con vergence The use of this option is highly recommended especially in minimization of large organic molecules with rings Nevertheless you should keep in mind that these
131. specified as in the CI program The resulting energies are stored in variables as explained in section 8 8 23 1 Expectation values for MP2 One electron properties can be computed as analytical energy derivatives for MP2 This cal culation is much more expensive than a simple MP2 and therefore only done if an EXPEC card follows the MP 2 card the GEXPEC directive has no effect in this case The syntax of the EXPEC card is explained in section 6 13 For an example see section 24 6 1 The density matrix can be saved using DM record ifil See also sections and 23 2 Density fitting MP2 DF MP2 RI MP2 DF MP 2 options invokes the density fitted MP2 program The present implementation works only without sym metry RI MP2 is an alias for the command DF MP2 The following options can be specified BASIS MP 2 basis Fitting basis set basis can either refer to a basis set defined in a BASIS block or to a default fitting basis set only available for correlation consistent basis sets If a correlation consistent orbital basis set is used the corresponding MP2 fitting basis is generated by default In all other cases the fitting basis must be defined THRAO value Screening threshold for 3 index integrals in the AO basis THRMO value Screening threshold for 3 index integrals in the MO basis THROV value Screening threshold for 2 index integrals of fitting basis THRPROD value Screening product threshold for first half transforma
132. specified on the command line However if a DF IT or GDFIT directive is given outside of a command block the specified options are used globally in all subsequent density fitting calculations in the same run The options specified on a global DF IT directive are also passed down to procedures However if a DFIT is given within a procedure the corresponding options are used only in the same procedure and procedures called from it When the procedure terminates the options from the previous level are recovered 11 1 Options for density fitting The options described in this section have sensible default values and usually do not have to be given Many options described below have alias names These can be obtained using HELP CFIT ALIASES 11 1 1 Options to select the fitting basis sets BASIS Basis set for fitting Default set corresponding to the orbital basis BASIS COUL Basis set for Coulomb fitting default BASIS BASIS EXCH Basis set for exchange fitting default BASIS 11 DENSITY FITTING 66 BASIS_MP2 Fitting basis set for DF MP2 default BASIS BASIS CCSD Fitting basis set for DF LCCSD default BASIS 11 1 2 Screening thresholds THRAO Threshold for neglecting contracted 3 index integrals in the AO basis default 1 d 8 THRMO Threshold for neglecting half transformed 3 index integrals de fault 1 d 8 THRSW Threshold for Schwarz screening default 1 d 5 THROV Threshold for neglecting 2 i
133. stored in the same record In that case and also when there are just two states 34 NON ADIABATIC COUPLING MATRIX ELEMENTS 219 whose spatial symmetry is not 1 it is necessary to specify for which states the coupling is to be computed using the STATE directive STATE state state where state is of the form istate isym the symmetries of both states must be the same and it is therefore sufficient to specify the symmetry of the first state As an example the input for first order and second order calculations is given below The cal culation is repeated for a range of geometries and at the end of the calculation the results are printed using the TABLE command In the calculation shown the diabatic CASSCF orbitals are generated in the two CASSCF calculations at the displaced geometries by maximizing the overlap with the orbitals at the ref erence geometry This is optional and within the numerical accuacy does not influence the final results However the relative contributions of the orbital overlap and CI contributions to the NACME are modified If diabatic orbitals are used which change as little as possible as function of geometry the sum of overlap and orbital contribution is minimized and to a very good approximation the NACME could be obtained from the CI vectors alone 34 NON ADIABATIC COUPLING MATRIX ELEMENTS SRevision 2006 0 lif non adiabatic coupling memory 1 m basis f avdz li vdz TIO 10i 0 145
134. stored on disk This behaviour may be overridden by using the input command gdirect see section TO to force evaluation of integrals on the fly If the integrals are stored on disk immediately after evaluation they are sorted into complete symmetry packed matrices so that later program modules that use them can do so as efficiently as possible The options for the integral sort can be specified using the AOINT parameter set using the input form AOINT keyl valuel key2 value2 The following summarizes the possible keys together with their meaning and default values c final Integer specifying the compression algorithm to be used for the final sorted integrals Possible values are 0 no compression 1 compression using 1 2 4 or 8 byte values 2 2 4 or 8 bytes 4 4 8 bytes and 8 Default 0 c_sortl Integer specifying the compression algorithm for the interme diate file during the sort Default O c_seward Integer specifying the format of label tagging and compression written by the integral program and read by the sort program Default 0 compress Overall compression c final c seward and c_sortl are forced internally to be not less than this parameter Default 1 thresh Real giving the truncation threshold for compression Default 0 0 which means use the integral evaluation threshold GTHRESH TWOINT io String specifying how the sorted integrals are written Possi ble values are molpro standard MOLPRO re
135. that gradients should be computed for this lattice point 0 means no gradient outfile specifies a file name to which the lattice gradient is written if blank it will be written to the output stream VARGRAD logical Stores the lattice gradient in variable VARGRAD NUCONLY logical Disables gradient evaluation with respect to the lattice independent of flag in the lattice file REMOVE logical Removes the lattice Symmetry is not supported for lattice gradients 12 7 Redefining and printing atomic masses The current masses of all atoms can be printed using MASS PRINT The atomic masses can be redefined using MASS type symbol mass The optional keyword type can take either the value AVER AGE for using average isotope masses or ISO TOPE for using the masses of the most abundant isotopes This affects only the rotational constants and vibrational frequencies As in most quantum chemistry packages the default for type is AVERAGE If INIT is given all previous mass definitions are deleted and the defaults are reset Individual masses can be changed by the following entries where symbol is the chemical symbol of the atom and mass is the associated mass Several entries can be given on one MASS card and or several MASS cards can follow each other The last given mass is used Note that specifying different isotope masses for symmetry related atoms lowers the symmetry of the system if the molecular centre of mass i
136. the code as built will run correctly Installation The program can be run directly from the source tree in which it is built but it is usually recommended to run the procedure that installs the essential components in standard system directories A 3 2 Prerequisites The following are required or strongly recommended for installation from source code 1 A Fortran 90 compiler Fortran77 only compilers will not suffice On most systems the latest vendor supplied compiler should be used For IA32 Linux for example Intel Pentium or AMD athlon the recommended compilers are the Intel Compiler ifort version 9 0 or higher or the Portland pg 90 compiler version 6 0 or higher For Opteron and EM64T systems the recommended compilers are Portland version 6 0 or higher Pathscale compiler pathf90 version 2 3 or higher or the Intel Compiler version 9 0 or higher The full list of supported compilers can be found at http www molpro net supported 2 GNU make freely available from http www fsf org and mirrors GNU make must be used most system standard makes do not work In order to avoid the use of a wrong make and to suppress extensive output of GNU make it may be useful to set an alias e g alias make gmake s 3 The GNU wget utility for batch mode http transfers although not needed for installation is essential for any subsequent application of patches that implement bug fixes 4 About 10GB disk space strongly system dependent
137. the d working directory to avoid copying of large integral files W wavefunction file repository issimilarto wavefunction file repository except that it refers to the directory for the wavefunction files 2 3 and 4 X xml output specifies that the output file will be a well formed XML file suit able for automatic post processing Important data such as input geometries and results are tagged and the bulk of the normal de scriptive output is wrapped as XML comments no xml output switches off this behaviour and forces a plain text output file to be produced L 1library directory specifies the directory where the basis set library files LIBMOL are found 1 file 1 directory directory directory specifies the directory where the run time file 1 will be placed overriding directory for this file only 2 3 4 5 6 7 8 and 9 may be used similarly Normally these options should not be given since the program tries to use what is given in d to optimally distribute the I O There are a number of other options for tuning and system parameters but these do not usually concern the general user It is not usually necessary to specify any of these options the defaults are installation dependent and can be found in the system configuration file molpro rc in the same directory as the molpro command itself 2 0 2 Running MOLPRO on parallel computers MOLPRO will run on distributed memory multip
138. unused SCREEN lt 2 THREST is unused MAXRED Maximum number of iterations after which thresholds are re duced to their final values in CI and CCSD calculations If MAXRED 0 the final thresholds will be used in CI and CCSD from the beginning same as THRMAX 0 but MAXRED has no effect on DSCF In the latter case a fixed value of 10 is used VARRED Thresholds are reduced to their final values if the sum of squared amplitude changes is smaller than this value SWAP Enables or disables label swapping in SEWARD Test purpose only Specific options for direct SCF DFOCK THREST_DSCF Final prescreening threshold in direct SCF If given it replaces the value of THREST THRMAX_DSCF Initial prescreening threshold in direct SCF This is used for the first 7 10 iterations Once a certain accuracy is reached the threshold is reduced to THREST_DSCF 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 58 SWAP _DFOCK Enables or disables label swapping in fock matrix calculation test purpose only General options for direct integral transformation DTRAF PAGE_DTRAF SCREEN_DTRAF MAXSHLO MINSHLO1 MAXSHLQ2 DTRAF MINSHLQ2 DTRAF MAXCEN_DTRAF DTRAF DTRAF PRINT_DTRAF Selects the transformation method PAGE_DTRAF 0 use minimum memory algorithm requiring four integral evaluations PAGE DTRAF 1 use paging algorithm leading to the mini mum CPU time one integral evaluation for DMP2
139. vector 1 2 no merge2 com vector 1 3 to output vector 1 2 to output vector 1 3 Gh Ch CR CT rotate 2pz orbitals to make bonding and antibonding linear combinations set print option symmetrically orthonormalize the valence orbitals the resulting orbitals are printed Project valence orbitals out of scf orbitals of the molecule and add virtual orbital set save merged orbitals to record 2150 on file 2 remove dummies perform full valence casscf for NO 2Pi state 2Pi state start with merged orbitals 43 MATRIX OPERATIONS 293 43 MATRIX OPERATIONS MATROP MATROP performs simple matrix manipulations for matrices whose dimensions are those of the one particle basis set To do so first required matrices are loaded into memory using the LOAD command To each matrix an internal name an arbitrary user defined string is assigned by which it is referenced in further commands After performing operations the resulting matrices can be saved to a dump record using the SAVE directive Numbers e g traces or individual matrix elements can be saved in variables code may be one of the following LOAD Loads a matrix from a file SAVE Saves a matrix to a file ADD Adds matrices TRACE Forms the trace of a matrix or of the product of two matrices MULT Multiplies two matrices TRAN Transforms a matrix DMO Transforms density into MO basis NATORB Computes natural orbitals DIAG Diago
140. we describe how to define these orbitals and how to save the optimized orbitals In a CASSCF calculation one has the choice of transforming the final orbitals to natural orbitals the first order density matrix is diagonalized to pseudo canonical orbitals an effective Fock operator is diagonalized or of localizing the orbitals 20 5 1 Defining the starting guess START record options 20 THE MCSCF PROGRAM MULTI 118 record dump record containing starting orbitals As usual record has the form irec ifil where irec is the record number e g 2140 and fil the file number usually 2 The options can be used to select orbitals of a specific type for details see section If this card is missing the program tries to find suitable starting orbitals as follows First Try to read orbitals from the record specified on the ORBITAL card or the corresponding default see ORBITAL All files are searched Second Try to find orbitals from the most recent MCSCF calculation All files are searched Third Try to find orbitals from the most recent SCF calculation All files are searched If no orbitals are found a starting orbital guess is generated It is often useful to employ MCSCF orbitals from a neighbouring geometry as starting guess this will happen automatically if orbitals are found see the above defaults Note however that frozen core orbitals should always be taken from an SCF or MCSCF calculation at the present
141. will act as default definitions for all subsequent OPT IM keywords If only one optimization step is required the OPT IM keyword need not be specified When only a machine generated guess is available CASVB will attempt to define a sequence of optimization steps chosen such as to maximize the likelihood of successful convergence and to minimize CPU usage To override this behaviour simply specify one or more OPTIM cards 36 9 6 Multi step optimization A loop over two or more optimization steps may be specified using ALTERN Niter FINALTER With this specification the program will repeat the enclosed optimization steps until either all optimizations have converged or the maximum iteration count Niter has been reached 36 10 Point group symmetry and constraints The problems associated with symmetry adapting valence bond wavefunctions are considered for example in T Thorsteinsson D L Cooper J Gerratt and M Raimondi Theor Chim Acta 95 131 1997 36 10 1 Symmetry operations SYMELM label sign Initiates the definition of a symmetry operation referred to by label any three characters sign can be or it specifies whether the total wavefunction is symmetric or antisymmetric under this operation respectively A value for sign is not always necessary but if provided constraints will be put on the structure coefficients to ensure that the wavefunction has the correct overall symmetry note that the configuration
142. 0 7246512 2 00 dr 0 01 geometry li f 1li rlif rlif 3 hf occ 4 1 1 multi closed 3 wf 12 1 state 2 orbital 2140 2 do i 1 r rlif r i multi closed 3 wf 12 1 state 2 orbital 2140 2 ci state 2 noexc save 6000 2 dm 8000 2 rlif r i dr multi closed 3 wf 12 1 state 2 start 2140 2 orbital 2141 2 diab 2140 2 ci state 2 noexc save 6001 2 ci trans 6000 2 6001 2 dm 8100 2 rlif r i dr multi closed 3 wf 12 1 state 2 start 2140 2 orbital 2142 2 diab 2140 2 ci state 2 noexc save 6002 2 ci trans 6000 2 6002 2 dm 8200 2 ddr dr 2140 2 2141 2 8100 2 nacmelp i nacme ddr dr 2140 2 2142 2 8200 2 nacmeim i nacme ddr 2 dr orbital 2140 2 2141 2 2142 2 density 8000 2 8100 2 8200 2 nacme2 1 nacme end do nacmeav nacmelp nacmelm 0 5 table r nacmelp nacmelm nacmeav nacme2 title Non adiabatic couplings for LiF 220 define basis define bond distances define increment define geometry first calculation at R 3 SCF ICASSCF 3 inactive orbitals Two 1A1 states dump orbitals to record 2140 2 loop over geometries set bond distance CASSCF 3 inactive orbitals Two 1A1 states lOverwrite previous orbitals by present ones ICI for 2 states no excitations save wavefunction to record 6000 2 save transition densities to record 8000 2 lincrement bond distance by dr same CASSCF as above Two 1A1 stat
143. 0 9 Coupled perturbed MCSCF The coupled perturbed MCSCF is required for computing gradients with state averaged orbitals non adiabatic couplings difference gradients or polarizabilities We note that the present im plementation is somewhat preliminary and not very efficient 20 THE MCSCF PROGRAM MULTI 128 20 9 1 Gradients for SA MCSCF For computing state averaged gradients use CPMCSCF GRAD state SP IN spin MS2 ms2 ACCU thresh RECORD record where state specifies the state e g 2 1 for the second state in symmetry 1 for which the gra dients will computed spin specifies the spin of the state this is half the value used in the corresponding WF card e g O Singlet 0 5 Doublet 1 Triplet Alternatively MS2 can be used where ms2 2 spin i e the same as specified on WF cards The specification of SPIN or MS2 is only necessary if states with different spin are state averaged record specifies a record on which the gradient information is stored the default is 5101 1 thresh is a threshold for the accuracy of the CP MCSCF solution The default is 1 d 7 The gradients are computed by a subsequent call to FORCES or OPTG Note if for some reason the gradients are to be computed numerically from finite energy differ ences it is in state averaged calculations necessary to give instead of the CPMCSCF input the following SAVE GRAD 1 Otherwise the program will stop with an error message 20 9 2 Difference gradients
144. 000000 0 0000000000 0 1302052882 H 1 4891244004 0 0000000000 1 0332262019 H 1 4891244004 0 0000000000 1 0332262019 hf The XYZ format is specified within the documentation distributed with MSCI s XMol package Note that MOLPRO has the facility to write XYZ files with the PUT command see section 12 4 12 3 3 MOLPRO92 input A subset of the MOLPRO92 atom specification commands are retained for compatibility These may be interspersed with Z matrix lines and are of the form A group atom x y z A group atom POL r 0 0 giving respectively cartesian or polar coordinates of the atom to be added Note that the in ternal coordinate specifications NPCC CCPA TCT LC RCP RCF are no longer avail able and Z matrix input should be used instead If any MOLPRO92 style atom specifications appear in the input the NOORIENT option is en forced and the handling of symmetry is slightly different No automatic search for symmetry takes place and all symmetry required should be specified Furthermore only symmetry unique atoms need be given the others being generated automatically 12 4 Writing Gaussian XMol or MOLDEN input PUT The PUT command may be used at any point in the input to print or write to a file the current geometry The syntax is PUT style file status info If style is GAUSSIAN a complete Gaussian input file will be written in that case info will be used for the first route data line and defaults to
145. 055 0 51777034 0 50728914 0 51252974 0 51252974 11 0 0 76672943 0 76125391 0 76399167 0 76399167 Tiles 0 42565202 0 42750263 0 42657733 0 42657733 12 0 0 19199878 0 19246799 0 19223338 0 192233 38 Note that the sign changes because of a phase change of one of the wavefunctions In order to keep track of the sign one has to inspect both the orbitals and the ci vectors 35 QUASI DIABATIZATION The DDR procedure can also be used to generate quasi diabatic states and energies for MRCI wavefucntions CASSCF case can be treated as special case using the NOEXC directive in the MRCI The quasi diabatic states have the propery that they change as little as possible relative to a reference geometry with other words the overlap between the states at the current geometry with those at a reference geometry is maximized by performing a unitary transformation among the given states Preferably the adiabatic and diabatic states should be identical at the reference geometry e g due to symmetry For instance in the examples given below for the B1 and 1A states of H2S Ca geomtries are used as reference and at these geometries the states are unmixed due to their different symmetry At the displaced geometries the molecular symmetry is reduced to Cs Both states now belong to the 14 irreducible representation and are strongly mixed For a description and application of the procedure described below see D Simah B Hartke and H J Werner J Chem
146. 06 2003 Open shell coupled cluster RCCSD UCCSD P J Knowles C Hampel and H J Werner J Chem Phys 99 5219 1993 Erratum J Chem Phys 112 3106 2000 Local MP2 LMP2 G Hetzer P Pulay and H J Werner Chem Phys Lett 290 143 1998 M Sch tz G Hetzer and H J Werner J Chem Phys 111 5691 1999 G Hetzer M Sch tz H Stoll and H J Werner J Chem Phys 113 9443 2000 See also references on energy gradients and density fitting Local Coupled Cluster methods LCCSD LQCISD LMP4 C Hampel and H J Werner J Chem Phys 104 6286 1996 M Sch tz and H J Werner J Chem Phys 114 661 2001 M Sch tz Phys Chem Chem Phys 4 3941 2002 See also references on energy gradients and density fitting Local triple excitations M Sch tz and H J Werner Chem Phys Lett 318 370 2000 M Sch tz J Chem Phys 113 9986 2000 M Sch tz J Chem Phys 116 8772 2002 Density fitting methods vil DFT Poisson fitting F R Manby P J Knowles and A W Lloyd J Chem Phys 115 9144 2001 DF MP2 DF LMP2 H J Werner F R Manby and P J Knowles J Chem Phys 118 8149 2003 DF LCCSD M Sch tz and F R Manby Phys Chem Chem Phys 5 3349 2003 DF HF R Polly H J Werner F R Manby and Peter J Knowles Mol Phys 102 2311 2004 DF LMP2 gradients M Sch tz H J Werner R Lindh and F R Manby J Chem Phys 121 737 2004
147. 1 52 53 54 55 56 57 58 59 60 61 62 63 C DENSITY FUNCTIONAL DESCRIPTIONS 325 C 8 B95 Becke95 Correlation Functional tau dependent Dynamical correlation functional See reference for more details E K 1 1 a xg 64 Fe ps 0 y H 1 v Xs 7 where E Pa Pp Pa 0 e pg 0 65 I 0 0031 66 Ped 67 H 3 5 623 x 2 3 ps I 68 v 0 038 69 e a B a B tres n t aa e r o B T3 Us Va Wa Xs Ys P o E a B 1 Ea C e r B T2 U2 V2 W2 X2 Ya P3 e r o5 B Ty U1 Vi W1 X1 Y1 P1 o 5 a B cen 70 1 a B 114434208 e 71 r o p x a4 B 71 a p a 5 72 a B a4 p 72 1 4 3 1 43 _2 qe EU ieee 73 o nt 2 ur in 1 412 74 u e r t U V W X Y p ur t v T wr xr yrP l c 1 709921 75 T 0 031091 0 015545 0 016887 76 U 0 21370 0 20548 0 11125 77 C DENSITY FUNCTIONAL DESCRIPTIONS 326 V 7 5957 14 1189 10 357 78 W 3 5876 6 1977 3 6231 79 X 1 6382 3 3662 0 88026 80 Y 0 49294 0 62517 0 49671 81 and P 1 1 1 82 To avoid singularities in the limit p 0 _ Fe ps 0 MIES G 83 C 9 B97R Density functional part of B97 Re parameterized by Hamprecht et al Re parameterization of the B97 functional in a s
148. 1 geomet ry O H1 O R H2 0 R H1 THETA hf accu 12 multi state averaged casscf for various triplet states closed 2 wf 10 1 2 state 3 wf 10 2 2 S w S h Cf tate 2 t 105 352 canonical 2140 2 rs2 mix 3 root 2 shift 0 2 loptimized second 3B2 state wf 10 3 2 13B2 wavefunction symmetry state 3 linclude 3 states optg gradient 1 d 5 geometry optimization using analytical gradients e opt 1 msenergy 2 loptimized ms caspt2 energy r_opt 1 r optimized bond distance theta_opt 1 theta loptimized bond angle method 1 rs2 analytical rs2 mix 3 shift 0 2 WESL0 372 13B2 wavefunction symmetry state 3 tinclude 3 states optg variable msenergy 2 gradient 1 d 5 fourpoint Igeometry optimization using numerical gradients e_opt 2 msenergy 2 loptimized ms caspt2 energy r opt 2 2r loptimized bond distance theta opt 2 theta loptimized bond angle method 2 2 rs2 numerical table method r opt theta opt e opt digits 4 4 8 This produces the table METHOD R OPT THETA OPT E OPT rs2 analytical 2 4259 96 7213 75 81630628 rs2 numerical 2 4259 96 7213 75 81630628 22 8 Coupling MRCI and MRPT2 The CIPT2 method P Celani H Stoll and H J Werner Mol Phys 102 2369 2004 For particularly difficult cases with strong intruder problems or in which second order pertur bation theory fails to predict reliable results a new method that couples MRCI and CASPT2 has been develop
149. 1 32 ang rch 1 08 ang acc 120 degree Geometry Cl Z matrix input C2Z cl Pee Ol cl recy G2 45 C3 c2 rcocG cl r180 g1 20 hl cl rch c2 acoc q1 0 h2 Cl f2ch c2 a00 h1 190 h3 G3 rch c2 acc hl1 90 h4 c3 rch c2 acc h2 90 examples allene optmp2 com optg procedure runmp2 use procedure optmp2 runmp2 hf mp2 procedure definition 39 4 4 Optimization using geometry DIIS kk CAFE EINE memory l m basis sto 3g geomtyp xyz geomet ry 24 nano0ozananazzzaaN hf 0 zs Ze 0 T 39 GEOMETRY OPTIMIZATION OPTG 272 lexamples caffeine opt diis com Revision 2002 10 cartesian coordinates XYZ format CAFFEINE CARTESIAN COORDINATES 8423320060 0 3654865620 0 0000000000 2841017540 1 1961236000 0 0000000000 0294818880 1 1042264700 0 0000000000 0774743850 2 5357317920 0 0000000000 6472646000 0 6177952290 0 0000000000 4531962870 2 3678913120 0 0000000000 6373131870 1 1735112670 0 0000000000 7812691930 0 7688916330 0 0000000000 6771444680 1 6306355000 0 0000000000 6106752160 1 9349693060 0 0000000000 9202890400 1 2510058880 0 0000000000 9202462430 3 1094501020 0 0000000000 8623938560 1 4824503660 0 0000000000 4552156930 0 6811094280 0 0000000000 0878150460 3 2451913360 0 0000000000 4989252090 3 4222116470 0 8897886280 4989252090 3 4222116470 0 8897886280 0071905670 3 7148499490 0 0000000000 4903070930 1 2888938190 0 89077
150. 11 Summary of local mul tp options and their default values Parameter Alias Default value Meaning General Parameters LOCAL 4 determines which program to use MULTP 0 turns on multipole approximations for distant pairs SAVEDOM SAVE 0 specifies record for saving domain info RESTDOM START 0 specifies record for reading domain info PIPEK LOCORB 0 activates or deactivates PM localization CANONICAL 0 allows to use canonical virtual orbitals for testing PMDEL CPLDEL 0 discards contributions of diffuse functions in PM localization SAVORB SAVLOC 0 specifies record for saving local orbitals DOMONLY 0 if 1 only domains are made if 2 only orbital domains are made Parameters to define domains THRBP DOMSEL 0 98 Boughton Pulay selection criterion for orbital domains CHGMIN 0 01 determines the minimum allowed atomic charge in domains CHGMINH 0 03 as CHGMIN but used for H atoms default 0 03 CHGMAX 0 40 If the atomic charge is larger than this value the atom is always included in the domain MAXANG MAXL 99 angular momentum restriction for BP domain selection MAXBP 0 determines how atoms are ranked in BP procedure MULLIKEN LOCMUL 0 determines the method to determine atomic charges MERGEDOM 0 merges overlapping domains DELCOR IDLCOR 2 delete projected core AOs up to certain shell DELBAS IBASO 0 determines how to remove redundancies Distance criteria for domain extensions REXT REXTS REXTC REXTW Conn
151. 12 A 3 7 Tuning MOLPRO can be tuned for a particular system by running in the root directory the command molpro tuning com This job automatically determines a number of tuning parameters and appends these to the file bin molpro rc Using these parameters MOLPRO will select the best BLAS routines depending on the problem size This job should run on an empty system It may typically take 10 minutes depending on the processor speed and you should wait for completion of this run before doing the next steps A 3 8 Testing At this stage it is essential to check that the program has compiled correctly The makefile target test 1 e command make test will do this using the full suite of test jobs and although this takes a significantly long time it should always be done when porting for the first time A much faster test which checks the main routes through the program can be done using make quicktest For parallel installation it is highly desirable to perform this validation with more than one running process This can be done conveniently through the make command line as for example make MOLPRO OPTIONS n2 test If any test jobs fail the cause must be investigated It may be helpful in such circumstances to compare the target platform with the lists of platforms on which MOLPRO is thought to function at http www molpro net supported If after due efforts to fix problems of a lo cal origin the problem cannot be resolved the de
152. 2 M d a B 0 5v e r a B x 3 7Z de Ves d N o B a K d a B 0 2500000000A7 In 2X N a B N a B 245 _ 4 op N a B 2 7 e ao 1 7 VOss V236 Yn ipro Q 1 2 T 0 031091 0 015545 0 016887 U 0 21370 0 20548 0 11125 V 7 5957 14 1189 10 357 W 3 5876 6 1977 3 6231 X 1 6382 3 3662 0 88026 Y 0 49294 0 62517 0 49671 and P At To avoid singularities in the limit p 0 G p ps 0 T C 0 Ps 0 vr wr xr m n 339 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 C DENSITY FUNCTIONAL DESCRIPTIONS 340 C 27 PBEXREV Revised PBE Exchange Functional Changes the value of the constant R from the original PBEX functional See reference for more details K 2 1 2E 2ps 285 where Vani 3 3 2 F roas 2 5 286 2 3 S 1 12 s 287 2 F s 14 R R 1 22 288 R 1 245 289 u 1387 290 and 5 0 066725 291 To avoid singularities in the limit p 0 G 1 2E 2p 292 C 28 PBEX PBE Exchange Functional See reference for more details K 2 1 2E 2ps 293 where f 3 3 2 F E n 3 4 ee S 294 62 1 12 Wa 295 2 F s 14 R R 1 57 296 R 0 804 297 u IBST 298 and 0 066725 299 To avoid singularities in the limit p 0 G 1 2E 2
153. 2 7 32 PROPERTIES AND EXPECTATION VALUES 214 32 6 4 AXIS direction of grid axes AXIS X y Z x y z specify the unnormalised direction cosines of one of the three axes defining the grid Up to three AXIS commands can be given but none is required Axes need not be orthogonal By default the first axis is the cartesian x the second is orthogonal to the first and to the cartesian z and the third is orthogonal to the first two 32 6 5 BRAGG spatial extent of grid Based on the direction of the coordinate axes a parallelopiped in the usual case of orthogonal axes a cuboid is constructed to contain the molecule completely The atoms are assumed to be spherical with an extent proportional to their Bragg radii and the constant of proportionality can be changed from the default value using BRAGG scale After the parallelopiped has been constructed the grid is laid out with equal spacing to cover it using the number of points specified on the CUBE command 32 6 6 ORIGIN centroid of grid ORIGIN x y z x y z specify the centroid of the grid It is usually not necessary to use this option since the default should suffice for most purposes 32 6 7 TITLE user defined title TITLE title Set a user defined title in the cube file 32 6 8 DESCRIPTION user defined description DESCRIPTION description Set a user defined description in the cube file 32 6 9 Format of cube file The formatted cube file contains the
154. 21 THE CI PROGRAM 133 sym is the number of the irreducible representation spin defines the spin symmetry spin 25 singlet 0 doublet 1 trip let 2 etc The WF card must be placed after any cards defining the orbital spaces OCC CORE CLOSED The REF card can be used to define further reference symmetries used for generating the con figuration space see REF 21 2 6 Additional reference symmetries REF synm This card which must come after the WF directive defines an additional reference symmetry used for generating the uncontracted internal and singly external configuration spaces This is sometimes useful in order to obtain the same configuration spaces when different point group symmetries are used For instance if a calculation is done in C symmetry it may happen that the two components of a II state one of which appears in A and the other in A come out not exactly degenerate This problem can be avoided as in the following example for a doublet A state WES 1 Sy Ly define wavefunction symmetry 1 REF 2 define additional reference symmetry 2 and for the doublet A state Wey Lb yey ee define wavefunction symmetry 2 REF 1 define additional reference symmetry 1 For linear geometries the same results can be obtained more cheaply using C2 symmetry WF 15 2 1 define wavefunction symmetry 2 REF 1 define additional reference symmetry 1 REF 3 define additional
155. 2p7 3 ps es 212 where o e 9579 113 213 c dZ gt 214 B 0 04918 215 A 0 132 216 c 0 2533 217 d 0 349 218 3 252 3 e ig 67 Q19 and d 1 C DENSITY FUNCTIONAL DESCRIPTIONS 336 C 23 MKOOB Exchange Functional for Accurate Virtual Orbital Energies MKOO0 with gradient correction of the form of B88X but with different empirical parameter See reference for more details p ps idi Xs a Dc NM 1 6B x arcsinh x el where B 0 0016 222 To avoid singularities in the limit p 0 T ps B p x5 223 G Ts 1 40 1 4 6ppy arcsinh x5 C 24 MK00 Exchange Functional for Accurate Virtual Orbital Energies See reference for more details ps K E 20 ee DB 224 C 25 P86 Gradient correction to VWN See reference for more details e C r o K pe dp 225 where r 1 4434 a 226 V np le 227 pre 228 p e A oy 1 nC 229 2 4n y 9 rya 230 y g 6 50 C 9 4 230 cr 231 A q k1 11 m3 n1 232 A q ko l2 m2 n2 233 C DENSITY FUNCTIONAL DESCRIPTIONS 0 q k3 13 m3 n3 q A p c d A in zi 2carctan 82 Q c d 7 X x c d 2x c 337 234 cp m LL 2 e 2p aretan EEA iem toe X x c d 2x c Q c d V4d e X i c d ci d 2 0 d 0 00739007 0 007390075 5 7 d V2 112 1 26 1 2 1 296 0 002568 ar Br 1 6r 6872
156. 3 1074 a u and the maximum energy change to be less than 1 1076 H or the maximum component of the gradient to be less then 3 107 a u and the maximum component of the step to be less then 3 1074 a u It is also possible to use the convergency criterion of the Gaussian program package It is somewhat weaker than the MOLPRO criterion and requires the maximum component of the gradient to be less then 4 5 1074 a u and the root mean square RMS of the gradient to be less then 3 1074 a u as well as the maximum component of the optimization step to be less then 0 0018 a u and the RMS of the optimization step to be less then 0 0012 a u MAXI T maxit maximum number of optimization cycles The default is 50 GRADIENT thrgrad required accuracy of the optimized gradient The default is 3 1074 ENERG Y threnerg required accuracy of the optimized energy The default is 1 1076 STEP thrstep convergence threshold for the geometry optimization step The de fault is 3 1074 BAKER logical Use Baker s convergency criteria see J Baker J Comp Chem 14 1085 1993 GAUSSIAN logical Use Gaussian convergency criteria 39 GEOMETRY OPTIMIZATION OPTG 253 SRMS thrsrms sets for Gaussian convergency criterion the required accuracy of the RMS of the optimization step The default is 0 0012 GRMS thrgrms sets for Gaussian convergency criterion the required accuracy of the RMS of the gradient The default is 3
157. 3 74 74 75 75 76 76 77 TI 78 78 79 80 82 84 84 84 85 85 86 86 87 87 88 88 88 89 CONTENTS 17 2 Defining the wavefunction 17 2 1 Defining the number of occupied orbitals in each symmetry 1722 Specifying closed shell orbitals oe c MEE TM 17 4 3 Starting with a previous density matrix ror 17 6 Using additional point group symmetry a a a 17 8 Polarizabilities 17 9 Miscellaneous directives 17 9 1 Level shifts 17 9 2 Maximum number of iterations 17 9 3 Convergence threshold 179 4 Printoptions 179 5 Interpolation 17 9 6 Reorthonormalization of the orbitals 179 7 Direct SCH 18 THE DENSITY FUNCTIONAL PROGRAM 18 1 Option o o 182 Directives 18 2 1 Density source DENSITY ODENSITY 18 2 2 Thresholds DFTTHRESH 18 2 3 Exact exchange computation EXCHANGE o 18 2 4 Exchange correlation potential POTENTIAL 18 2 5 Grid blocking factor DF TBLOCK 18 2 6 Dump integrand values DF TDUMP 18 3 Numerical integration grid control GRID 18 3 1 Target quadrature accuracy GRIDTHRESH 18 3 2 Radial integration grid RADIAL 18 3 3 Angular integration grid ANGULAR 18 3 4 Atom partitioning of integration grid VORONOI
158. 367 1988 Below we present an example for the first two excited states of H2S which have B and A symmetry in Cz and A symmetry in Cs We first perform a reference calculation in C sym metry and then determine the diabatic orbitals for displaced geometries in Cs symmetry Each subsequent calculation uses the previous orbitals as reference One could also use the orbitals of the Ca calculation as reference for all other calculations In this case one would have to take out the second last input card which sets re orb 2141 2 33 DIABATIC ORBITALS SRevision 2006 0 H2S diabatic A states basis VDZ geometry x planeyz noorient s hl s rl h2 s r2 hl theta gprint orbitals civector text reference calculation for C2V theta 92 12 r1 2 3 r2 2 3 ihipoce 1 22 wt is Ly multi occ 9 2 closed 4 1 wf 18 2 state 2 orbital 2140 2 reforb 2140 2 217 use cc pVDZ basis set use Cs symmetry fix orientation of the molecule dont allow automatic reorientation Z matrix geometry input global print options reference geometry scf calculation for ground state define active and inactive spaces two A states 1B1 and 1A2 in C2v examples save orbitals to 2140 2 P h2s diab com text calculations at displaced geometries rd 2 4 2 5 2 6 do i 1 rd r2 rd i multi occ 9 2 closed 4 1 wf 18 2 state 2 orbital 2141 2 diab reforb reforb 2141 2 enddo define a
159. 500 2 vbdump casvb Overlap based VB using save 3200 2 the spin coupled wavefunction casvb Energy based VB calculation start 3200 2 save 3220 2 crit energy multi occ 4 1 2 closed 1 Fully variational VB calculation vb start 3220 2 save 3240 2 print 2 memory 4 m n2s2 model a Variational calculation for N2S2 geometry X VY Z al n 2 210137753 0 0 NOTE other choices of active space a2 n 2 2101377535 0 07 give alternative competing models d3 58 0 22 2T0137453 05 a4 s 0 2 210137753 0 basis VTZ cartesian hf wf 46 1 imulti occ T7T 4 5 2 4 2 2 0 closed 7 4 5 2 1 0 1 0 natorb cl save 3500 2 tmultisocec 7 42 5 2 0 2 2 0 closed 7 4 5 2 17 07 1 03 vot goce LLY Fully variational VB calculation r 2 8 bohr and geometry optimization basis s 1 921 300000 138 700000 31 940000 9 353000 3 158000 1 157000 k 1 6 0 001367 0 010425 0 049859 0 160701 0 344604 0 425197 s 1 0 444600 0 076660 0 028640 p 1 1 488000 0 266700 0 072010 0 023700 k 1 2 0 038770 0 236257 5 2 13 36 2 013 50 4538 1233 k 1 2 0 032828 0 231204 geometry 1i h 1i r int hf wf 4 1 multi occ 4 0 0 0 closed 0 0 0 0 natorb ci save 3500 2 multi maxiter 20 vb 36 THE VB PROGRAM CASVB 238 optg 37 SPIN ORBIT COUPLING 239 37 SPIN ORBIT COUPLING 37 1 Introduction Spin orbit matrix elements and eigenstates can be computed using either the Breit Pauli BP o
160. 6 STATE 115 135 STATUS STEP 126 262 String variables 43 STRONG 234 STRUC 230 Summary of keywords 18 SYM 96 SYMELM 232 symmetry 70 WF card 16 additional MCSCE 123 SCF B6 Integral program SYMPROJ 233 System variables 44 02 UN TABLE 54 Tables 54 TEST 126 THERMO 281 THRESH 111 127 142 160 TRAN 124 TRAN2 124 TRANH 137 TRANS 139 164 233 TRNINT 127 TRUST 262 UHF 90 UHF SCE 90 UKS 99 UKS SCF 99 UNCOMPRESS 69 ys D UPDATE 260 VARIABLE variables Indexed Introduction Setting Special String System VB 227 VB VBDUMP VBWEIGHTS 235 Vector operations vibrational frequencies 280 VORONOT wavefunction definition WEIGHT 115 wF 15 93 1 14 132 199 WRITE 236 XYZ 72 73 Z matrix ZMAT nj
161. 63360 4903070930 1 2888938190 0 8907763360 6289534570 2 5638654230 0 0000000000 1360211370 1 5529079440 0 0000000000 6817059520 0 0685850980 0 8931597470 6817059520 0 0685850980 0 8931597470 XYZ save optimized geometry in file caffeine xyz optg savexyz caffeine coord bmat method diis optg coord bmat method diis savexyz caffeine xyz 39 4 5 Transition state of the HCN HNC isomerization Optimization in natural internal coordinates Optimization method Geometry DIIS same as above The first example shows how to do a MP2 transition state optimization The initial Hessian is taken from a previous HF frequency calculation examples caffeine opt diis com 39 GEOMETRY OPTIMIZATION OPTG 273 lexamples hcn mp2 ts com Revision 2002 10 HCN NHC Isomerization Transition State Optimization and Frequencies 11 1 18268242 ang 1221 40745082 ang al 55 05153416 degree basis 3 21G geometry nosymm C NT dS les H 2 12 1 al examples hen mp2 ts com hf HF SCF frequencies analytical Vibrational frequencies for HF SCF analytical Hessian mp2 MP2 optg root 2 method rf readhess Transition State Search using Rational Function Optimize frequencies Vibrational frequencies for MP2 numerical Hessian The second example shows how to do a CCSD T optimization with an MP2 hessian Note that currently the CCSD T gradient is computed numerically using finite energy differences and t
162. 73043 27 THE MRCC PROGRAM OF M KALLAY MRCC 174 h20 excitation energies memory 8 m gthresh energy 1 d 8 geometry o hl o r h2 o r hl theta theta 104 r l ang basis vdz hf 1i 0 s 2 number of states in each symmetry do sym 1 4 loop over irreps ccsd eom s 0 1 sym S p molpro save energy mrcc method ccsd symm sym nstates 2 pemrcc save energy mrcc method ccsdt symm sym nstates 2 p mrcc save_energy s 1 enddo examples h20 mrcc eom com table method prog states e exc sort 3 save_energy procedure to save results in variables nogprint variable el energy 1 do i 1 energy ii ii 1 e ii energy i method ii program prog ii p states ii i 0 1 sym xc ii e ii 1 toev end do This yields METHOD PROG STATES E EXC CCSD OLPRO 1 1 76 23580212 0 000 CCSD RCC Eod 76 23580212 0 000 CCSDT RCC Es il 76 23922746 0 000 CCSD OLPRO 1 2 76 23580212 0 000 CCSD RCC L2 76 23580212 0 000 CCSDT RCC I2 76 23922746 0 000 CCSD OLPRO gees 76 23580212 0 000 CCSD RCC L3 76 23580212 0 000 CCSDT RCC 1 3 76 23922746 0 000 CCSD OLPRO 1 4 76 23580212 0 000 CCSD RCC 4 76 23580212 0 000 CCSDT RCC 1 4 76 23922746 0 000 CCSD OLPRO 271 75 85033256 10 489 CCSD RCC 251 75 85033257 10 489 CCSDT RCC 2 1 75 85316687 10 505 CCSD OLPRO 2 2 75 95093334 7 152 CCSD RCC 252 75 95093335 7 1752 CCSDT RCC 22 75 95299013 7 789 CCSD OLPRO BD 75 77630664 12 504
163. 854 0 0336982 0 0353615 0 00497930 0 0645900 0 0461795 0 00757191 373 0 00242717 0 0428140 0 0744891 0 0386577 0 352519 2 19805 3 72927 1 94441 0 128877 To avoid singularities in the limit p 0 y Oss 1 20 Ps Gs 8 3 Ms p Wi ps d l 374 i C DENSITY FUNCTIONAL DESCRIPTIONS 346 C 35 TH2 Density and gradient dependent first row exchange correlation functional See reference 24 for more details K Y o RiS XY 375 i l where n 19 376 Ri Pa pg 377 e 2u 378 P Ooa Opp x 2 San 1 50 379 p o Ogg 2 aa O Y m aa PB a Qo 3 M 380 p 13 17 11 11 11 E pt lI pint pe 381 t p57 7 6 4 3 3 2 5 3 1272 55 6 5 3 6 2 5 3 6 2 7 6 4 3 3 2 5 3 381 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 382 v 0 0 0 0 0 1 1 1 1 2 2 2 0 0 0 0 0 0 0 383 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 384 and 0 678831 1 75821 1 27676 1 60789 0 365610 0 181327 0 146973 0 147141 0 0716917 0 0407167 0 0214625 0 000768156 0 0310377 385 0 0720326 0 0446562 0 266802 1 50822 1 94515 0 679078 To avoid singularities in the limit p 0 n v Oss G L 1 29 ps V0ss 8 3 Ps h p 386 C DENSITY FUNCTIONAL DESCRIPTIONS 347 C 36 TH3 Density and gradient dependent first and second row exchange correlation functional See ref
164. 9 3 3 Additional options for SLAPAF Some options can be passed to the SLAPAF program Options are specified with SLOPT sub directive opt method slnr slopt opt1 opt2 parl par2 opt 3 The available options are CART Use eigenvectors of the approximate Hessian expressed in cartesian coordinates as the definition of internal coordinates NOMA Don t impose any restrictions on the step size UORD Order the gradients and displacement vectors according to Schlegel prior to the update of the Hessian Default is no reordering HWRS Use force field weighted internal coordinates default 39 GEOMETRY OPTIMIZATION OPTG 270 RS P NOHW PRBM RTHR Thra Thrb Thrt MODE index FIND GNRM thr MEP NMEP npoints Activate RS P RFO as default for transition state search default is RS I RFO Use unweighted internal coordinates Print B matrix Thresholds for redundant coordinate selection for bonds bends and torsions respectively Default 0 2 0 2 0 2 Hessian vector index for mode following when calculating transition states Enable unconstrained optimization for constrained cases when look ing for transition states see MOLCAS manual Threshold for FIND default 0 2 see MOLCAS manual Perform minimum energy path MEP search Number of MEP points to find in MEP calculation For more information please consult the MOLCAS manual 39 4 Examples 39 4 1 Simple HF optimization
165. ALI is a separate command which calls the localization program and not recognized by MULTI In order to avoid confusion it is recom mended to use LOCORB rather then LOCAL as subcommand within MULTI 20 5 8 Diabatic orbitals In order to construct diabatic states it is necessary to determine the mixing of the diabatic states in the adiabatic wavefunctions In principle this mixing can be obtained by integration of the non adiabatic coupling matrix elements Often it is much easier to use an approximate method in which the mixing is determined by inspection of the CI coefficients of the MCSCF or CI wavefunctions This method is applicable only if the orbital mixing is negligible For CASSCF wavefunctions this can be achieved by maximizing the overlap of the active orbitals with those of a reference geometry at which the wavefunctions are assumed to be diabatic e g for symmetry reasons The orbital overlap is maximized using using the new DIAB command in the MCSCF program Only the active orbitals are transformed 20 THE MCSCF PROGRAM MULTI 121 This procedure works as follows first the orbitals are determined at the reference geometry Then the calculations are performed at displaced geometries and the diabatic active orbitals which have maximum overlap with the active orbitals at the reference geometry are obtained by adding a DIAB directive to the input Old form Molpro96 obsolete DIAB orbref orbsav orbl orb2 pri New form
166. Becke J Chem Phys 104 1040 1995 F A Hamprecht A J Cohen D J Tozer and N C Handy J Chem Phys 109 6264 1998 P A Stewart and P M W Gill J Chem Faraday Trans 91 4337 1995 J C Slater Phys Rev 81 385 1951 P M W Gill Mol Phys 89 433 1996 A D Boese N L Doltsinis N C Handy and M Sprick J Chem Phys 112 1670 2000 M Ernzerhof and G Scuseria J Chem Phys 111 911 1999 F R Manby and P J Knowles J Chem Phys 112 7002 2000 J P Perdew Phys Rev B 33 8822 1986 Y Zhang and W Yang Phys Rev Lett 80 890 1998 J P Perdew and Y Wang Phys Rev B 33 8800 1986 D J Tozer and N C Handy J Chem Phys 108 2545 1998 D J Tozer and N C Handy D J Tozer and N C Handy Mol Phys 94 707 1998 D J Tozer N C Handy and W H Green Chem Phys Lett 273 183 1997 T V Voorhis and G E Scusseria J Chem Phys 109 400 1998 S H Vosko L Wilk and M Nusair Can J Phys 58 1200 1980 A INSTALLATION OF MOLPRO 304 A Installation of MOLPRO A 1 Obtaining the distribution materials MOLPRO is distributed to licensees on a self service basis using the world wide web Those enti tled to the code should obtain it from http www molpro net download supplying the username and password given to them The web pages contain both source code and binaries although not everyone is entitled to source code and binaries ar
167. C 39 THGFCO Density and gradient dependent first row exchange correlation functional See reference 26 for more details K Y o RiS XY 423 i 1 where n 20 424 Ri Pa pp 425 e 2u 426 P VOaa Opg X 1 2 aa na BB 427 o Ogg 2 aa O Y aa BB a Qo 428 P 11 11 11 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 SA g 232 58 G12 116 413 3 2 5 3 429 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 430 v 0 0 0 0 1 1 1 1 2 2 2 2 0 0 0 0 0 0 0 0 431 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 432 and 0 962998 0 860233 1 54092 0 381602 0 210208 0 391496 0 107660 0 0105324 0 00837384 0 0617859 0 0383072 0 00526905 433 0 00381514 0 0321541 0 0568280 0 0288585 0 368326 0 328799 1 22595 1 36412 To avoid singularities in the limit p 0 y Oss 1 20 Ps Gs 8 3 Ms p Wi ps d l 434 i C DENSITY FUNCTIONAL DESCRIPTIONS 351 C 40 THGFC Density and gradient dependent first row exchange correlation functional for closed shell sys tems Total energies are improved by adding DN where N is the number of electrons and D 0 1863 See reference for more details K Y oiRyX 435 i l where n 12 436 Ri Pa pp 437 VOaa 5pg x i2 ax usi p 438 ps 11 11 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 6 2 58 FL 439 v 0 0 0 0
168. CASPT2 for reference state 2 e2 caspt2 i energy lunmixed caspt2 energy for excited state el_mscaspt2 i msenergy 1 ms caspt2 energy for ground state e2_mscaspt2 i msenergy 2 ms caspt2 energy for excited state enddo table r el caspt2 e2 caspt2 el mscaspt2 e2 mscaspt2 title SS SR CASPT2 for LiF plot file lif sr mscaspt2 plot This produces the plot 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 150 SS SR CASPT2 for LiF 107 T T T T T T T 107 05 107 1 e e El CASPT2 m mu E2 CASPT2 El MSCASPT2 4 4 4 E2 MSCASPT2 22 3 2 Performing MS MR CASPT2 calculations In the case of multi state multi reference CASPT2 calculations only a single run is needed RS2 MIX nstates options STATE nstates Example MS MR CASPT2 calculation for LiF 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 151 SRevision 2006 1 r 3 4 5 6 7 8 9 10 ang i 1 geometry Li F l r i basis vtz F avtz hf Hartree Fock do i 1 r loop over range of bond distances multi closed 3 0 0 0 occ 5 2 270 state 2 ISA CASSCF for 2 states canonical ci rs2 MIX 2 state 2 2 state CASPT2 with 2 reference states el_caspt2 i energy 1 unmixed caspt2 energy for ground state e2_caspt2 i energy 2 unmixed caspt2 energy for ground state el_mscaspt2 i msenergy 1 ms caspt2 energy for ground state e2_mscaspt2 i msenergy 2 ms caspt2
169. CCSD RCC 2 9 75 77630665 12 504 CCSDT RCC 2295 75 71972816 12 504 CCSD OLPRO 2 4 75 87776149 9 743 CCSD RCC 2 4 75 87776150 9 743 CCSDT RCC 2 4 75 88051189 9 761 27 THE MRCC PROGRAM OF M KALLAY MRCC Open shell ground state calculations for O2 02 tests memory 8 m gthresh energy 1 d 8 geometry 01 02 01 r1 r1 2 2 set state 1 symmetry 4 spin 2 Triplet sigma state basis vdz rhf uccsd t method 1 2 UCCSD T MOLPRO e 1 energy rccsd t method 2 2 RCCSD T MOLPRO e 2 energy mrcc method ccsdt dir mrecdir method 3 CCSDT MRCC e 3 energy mrcc method ccsdtq restart 1 dir mrccdir method 4 CCSDT MRCC e 4 energy table method e This yields METHOD E UCCSD T MOLPRO 149 9815472 RCCSD T MOLPRO 149 9812566 CCSDT MRCC 149 9816705 CCSDT MRCC 149 9832255 175 examples 02 mrcc com 28 LOCAL CORRELATION TREATMENTS 176 28 LOCAL CORRELATION TREATMENTS 28 1 Introduction The local correlation program of MOLPRO can currently perform closed shell LMP2 LMP3 LMP4 SDTQ LCISD LQCISD T and LCCSD T calculations For large molecules all methods scale linearly with molecular size provided very distant pairs are neglected and the integral direct algorithms are used Much higher efficiency is achieved by using density fitting DF approximations to compute the integrals Density fitting is available for all local methods up to LCC
170. CP P maxcpp PRINT_CPP maxcpp PROJECT_CPP maxcpp POLAR I value 17 1 8 Printing options PRINT ORBPRINT value DEBUG value to be described to be described to be described to be described If nonzero compute analytical dipole polarizabilities Number of virtual orbitals to be printed If value 0 the occu pied orbitals are printed Option for debug print 17 THE SCF PROGRAM 93 17 2 Defining the wavefunction The number of electrons and the total symmetry of the wavefunction are specified on the WF card WF elec sym spin where elec is the number of electrons sym is the number of the irreducible representation spin defines the spin symmetry spin 2 x S singlet 0 doublet 1 triplet 2 etc Note that these values take sensible defaults if any or all are not specified see section 4 8 17 2 1 Defining the number of occupied orbitals in each symmetry OCC n1 N2 Ng To avoid convergence problems in cases with high symmetry this card should be included when ever the occupation pattern is known in advance n is the number of occupied orbitals in the irreducible representation i The total number of orbitals must be equal to elec spin 2 see WE card 17 2 2 Specifying closed shell orbitals CLOSED nj n2 ng This optional card can be used in open shell calculations to specify the number of closed shell orbitals in each symmetry This makes possible to force specific states in the absence of a
171. Chem S23 1989 199 is printed and stored in the variable T1DIAG for later analysis 24 Coupled cluster CCSD The command CCSD performs a closed shell coupled cluster calculation Using the CCSD T command the perturbative contributions of connected triple excitations are also computed If the CCSD is not converged an error exit will occur if triples are requested This can be avoided using the NOCHECK option CCSD T NOCHECK In this case the T correction will be computed even if the CCSD did not converge Note NOCHECK has no effect in geometry optimizations or frequency calculations For further information on triples corrections see under RCCSD 24 THE CLOSED SHELL CCSD PROGRAM 161 24 2 Quadratic configuration interaction QCI OCT or QCISD performs quadratic configuration interaction QCISD Using the OCI T or QCISD T commands the contributions of connected triples are also computed by perturbation theory Normally no further input is needed if the OCT card follows the corresponding HF SCF Otherwise occupancies and orbitals can be specified as in the CI program For modifying DIIS directives see section 24 5 For avoiding error exits in case of no convergence see CCSD T 24 3 Brueckner coupled cluster calculations BCCD BCCD SAVE record PRINT TYPE type BCCD performs a Brueckner coupled cluster calculation and computes Brueckner orbitals With these orbitals the amplitudes of the singles vanis
172. D T HF KS MP2 and all local correlation methods can be prepended by DF to invoke density fitting 4 13 Mor Pno help The help command can be used to obtain a short description of commands input parameters and variables The syntax is HELP set name keys where set is either COMMAND VARIABLE or the name of the input set e g THRESH PRINT LOCAL EOM CFIT and name is the name of the parameter If name is blank all parameters of the set are shown Optionally keys can be specified to request specific information e g 5 INTRODUCTORY EXAMPLES 22 short description long description default_value type program If keys are not given short description is assumed Currently help is only available for a limited number of parameters and commands However the database will be extended in the near future 5 INTRODUCTORY EXAMPLES This section explains some very simple calculations in order to help the new user to understand how easy things can be 5 1 Using the molpro command 1 Perform a simple SCF calculation for molecular hydrogen The input is typed in directly and the output is sent to the terminal molpro basis vdz geometry angstrom hl h2 h1 74 hf 1 2 The same calculation with the data taken from the file h2 com The output is sent to h2 out On completion the file h2 pun is returned to the current directory and the file h2 wf to the directory SHOME wfu this is the default
173. DF LCCSD T H J Werner and M Sch tz in prepation Explicitly correlated methods with density fitting DF MP2 R12 F R Manby J Chem Phys 119 4807 2003 DF MP2 F12 A J May and F R Manby J Chem Phys 121 4479 2004 DF LMP2 R12 loc H J and F R Manby J Chem Phys 124 054114 2006 DF LMP2 F12 loc F R Manby H J Werner T B Adler and A J May J Chem Phys 124 094103 2006 Full CI FCI P J Knowles and N C Handy Chem Phys Letters 111 315 1984 P J Knowles and N C Handy Comp Phys Commun 54 75 1989 Distributed Multipole Analysis DMA A J Stone Chem Phys Letters 83 233 1981 Valence bond D L Cooper T Thorsteinsson and J Gerratt Int J Quant Chem 65 439 1997 D L Cooper T Thorsteinsson and J Gerratt Adv Quant Chem 32 51 67 1998 See also An overview of the CASVB approach to modern valence bond calculations T Thorsteinsson and D L Cooper in Quantum Systems in Chemistry and Physics Volume 1 Basic problems and models systems eds A Hernndez Laguna J Maruani R McWeeny and S Wilson Kluwer Dordrecht 2000 pp 303 26 Spin orbit coupling A Berning M Schweizer H J Werner P J Knowles and P Palmieri Mol Phys 98 1823 2000 Diabatization procedures H J Werner and W Meyer J Chem Phys 74 5802 1981 H J Werner B Follmeg and M H Alexander J Chem Phys 89 3139 1988 D Simah B Hart
174. DMP 2 one integral evaluation This is only useful in local DMP2 cal culations with many distant pairs General threshold for generation of 2 external integrals in DMP2 If given this is used as a default for all DMP2 thresholds de scribed below Prescreening threshold for generation of 2 external integrals Defaults THR DMP2 THREST_DTRAF THR DTRAF THREST default Integral threshold for generation of 2 external integrals Defaults THR DMP2 THRINT DTRAF THR DTRAF THRINT default Product threshold for generation of 2 external integrals Defaults THR DMP2 THRPROD DTRAF THR DTRAF THRPROD default Specific options for direct local MP2 LMP 2 DTRAF THR LMP2 THREST LMP2 THRO1_LMP2 Selects the transformation method for direct LMP 2 DTRAF gt 0 generates the 2 external integrals exchange op erators first in AO basis and transforms these thereafter in a second step to the projected local basis The disk storage re quirements hence scale cubically with molecular size DTRAF 1 generates the 2 external integrals exchange op erators directly in projected basis The disk storage require ments hence scale linearly with molecular size This together with PAGE DTRAF 0 is the recommended algorithm for very large molecules cf linear scaling LMP2 chapter 28 DTRAF 2 alternative algorithm to generate the exchange operators directly in projected basis Usually this alg
175. EPART 0 If nonzero do energy partitioning EPART 3 0 cutoff parameter for determining individual monomers Parameters for redundancy check using DELBAS 1 not recommended TYPECHECK TYPECHK 1 activates basis function type restrictions DELSHL IDLSHL 1 determines if whole shells are to be deleted DELEIG IDLEIG 1 determines how to select redundant functions DELCMIN CDELMIN 0 1 parameter for use with DELEIG 1 Parameters for choosing operator domains in LCCSD OPDOM IOPDOM 5 determines how operator domains are determined for LCCSD RMAXJ 8 distance criterion for J operator list RMAXK 8 distance criterion for K operator list RMAXL 15 distance criterion for L operator list RMAX3X 5 distance criterion for 3 ext integral list RDOMJ 0 distance criterion for K operator domains RDOMK 8 distance criterion for J operator domains IMAXJ 5 connectivity criterion for J operator list IMAXK 5 connectivity criterion for K operator list IMAXL 8 connectivity criterion for L operator list IMAX3X 3 connectivity criterion for 3 ext integral list IDOMJ 0 connectivity criterion for K operator domains IDOMK 5 connectivity criterion for J operator domains Miscellaneous options SKIPDIST SKIPD 3 determines at which stage weak and distant pairs are eliminated ASYDOM JITERM 0 parameter for use of asymmetric domains LOCSING LOCSNG 0 determines virtual space used for singles PIPEKAO LOCAO 0 activates AO localizati
176. F THRDTRAF THREST_DTRAF THRAO_DTRAF THR_DTRAF THREST THRINT_DTRAF THRAO_DTRAF THR_DTRAF THRINT THRPROD DTRAF THRP DTRAF THR DTRAF THRPROD THR D2EXT THR2EXT THR DTRAF THREST D2EXT THRAO D2EXT THR D2EXT THREST DTRAF THRINT D2EXT THRSO D2EXT THR D2EXT THRINT DTRAF THRPROD D2EXT THRP D2EXT THR D2EXT THRPROD DTRAF THR_D3EXT THR3EXT THR_DTRAF THREST_D3EXT THRAO_D3EXT THR_D3EXT THREST DTRAF THRINT_D3EXT THRSO_D3EXT THR D3EXT THRINT DTRAF THRPROD D3EXT THRP D3EXT THR D3EXT THRPROD DTRAF THR_D4EXT THR4EXT THR_DTRAF THREST_D4EXT THRAO_D4EXT THR_D4EXT THREST DTRAF THRINT_D4EXT THRSO_D4EXT THR_D4EXT THRINT DTRAF THRPROD_D4EXT THRP_D4EXT THR_D4EXT THRPROD DTRAF THR_DCCSD THRCCSD THR_DTRAF THREST DCCSD THRAO DCCSD THR DCCSD THREST DTRAF THRINT DCCSD THRSO DCCSD THR_DCCSD THRINT DTRAF THRPROD DCCSD THRP DCCSD THR DCCSD THRPROD DTRAF THRMAX DCCSD THRMAX DTRAF THRMAX THR DMP2 THRDMP2 THR DTRAF THREST DMP2 THRAO DMP2 THR DMP2 THREST DTRAF default THRINT_DMP2 THRSO_DMP2 THR_DMP 2 THRINT DTRAF default THRPROD DMP2 THRP_DMP2 THR DMP2 THRPROD_DTRAF default THR_LMP 2 THRLMP2 THR DTRAF THREST LMP2 THRAO LMP2 THR_LMP 2 THREST DTRAF default THRO1_LMP2 THRO1 THR_LMP2 THRPROD_DTRAF default THRO2_LMP2 THRO2 THR_LMP 2 THRINT DTRAF default THRAO ATTEN THRATTEN THREST LMP2 THR DKEXT THRKEXT THREST DKEXT THRAO DKEXT THR DKEXT THREST THRINT_DKEXT THRSO_DKEXT THR_DKEXT THRINT TH
177. HE CI PROGRAM 136 21 2 12 Specifying correlation of orbital pairs PAIR iorbl isyl iorb2 isy2 np is a request to correlate a given orbital pair np l singlet pair np 1 triplet pair np 0 singlet and triplet pair if possible Default is to correlate all electron pairs in active and closed orbitals See also PAIRS card PAIRS iorbl isy iorb2 isy np Correlate all pairs which can be formed from orbitals iorbl isy through iorb2 isy2 Core orbitals are excluded Either iorb2 must be larger than iorb or isy2 larger than isy If iorbl isyl iorb2 isy2 the PAIRS card has the same effect as a PAIR card PAIR and PAIRS cards may be combined If no PAIR and no PAIRS card is specified all valence orbitals are correlated The created pair list restricts not only the doubly external configurations but also the all internal and semi internals 21 2 13 Restriction of classes of excitations NOPAIR No doubly external configurations are included NOSINGLE No singly external configurations are included NOEXC Perform CI with the reference configurations only 21 3 Options 21 3 1 Coupled Electron Pair Approximation CEPA ncepa 0 lt ncepa lt 3 Instead of diagonalizing the hamiltonian perform CEPA calculation CEPA type ncepa This is currently available only for single configuration reference functions 21 3 2 Coupled Pair Functional ACPF AQCC ACPF options AQCC options where options can be GACPF I gacpfi
178. IMIT directive is used No account is taken of symmetry every site in a symmetry equivalent set must be specified explicitly The radius of the site may also be specified default 1 0 DELETE name Delete all atoms with the name given from consideration as a multipole site Note that original atoms from the integral program have names 1 2 3 as printed in integral output DELETE ALL deletes all atoms and gives the multipoles with respect to the origin only 32 2 7 Defining the radius of multipole sites RADIUS RADIUS name r Assign radius r to all sites with the name given The program moves multipoles at an overlap centre P to the site S for which the value of P S r S is smallest In the absence of a RADIUS directive all sites are given radius 1 32 2 8 Notes and references The multipoles produced by this analysis are given in their spherical harmonic definitions Explicit formulae for translating between the cartesian and spherical harmonic definitions of the multipole moments are given in Explicit formulae for the electrostatic energy forces and torques between a pair of molecules of arbitrary symmetry S L Price A J Stone and M Alderton Molec Phys 52 987 1984 For examples of the use of DMA analysis see Price and Stone Chem Phys Lett 98 419 1983 Buckingham and Fowler J Chem Phys 79 6426 1983 32 2 9 Examples The following input calculates SCF multipole moments for water SRevisio
179. LOCAL 120 Local correlation 176 LOCALT 108 Localization space 109 359 LOCAO 108 LOCORB 120 loops 14 LQUANT 117 Macros in string variables 44 MASS Mass velocity 212 Matrix operations 293 MATROP 293 MAXDAV 138 MAXITER 97 126 38 231 MCSCR MCSCF MEMORY Memory allocation MERGE 287 METHOD 255 MOLDEN 73 molpro 1 Molpro help 21 Molpro2000 Molpro2002 Molpro2006 1 Molpro98 319 molpro basis MOVE 287 MP2 158 MP2 F12 196 MP2 R12 196 MP3 158 MP4 158 MPP D MPP systems 2 MPPX B Mulliken analysis 209 MULTI 112 MULTI NACM 128 246 NACME 128 218 NATORB 1 19 140 163 NELEC 16 NOCASPROJ 23 NOCHECK 160 164 NOEXC 136 NOEXTRA 123 NOGPRINT NOGRIDSAVE 103 NOGRIDSYM 103 Non adiabatic coupling NONLINEAR 127 NONUCLEAR 208 INDEX NOORDER 110 NOPAIR 136 NOS INGLE 136 NOSYMPROJ 233 NUMERICAL 247 261 Numerical gradients 247 NUMHES 259 c 16 93 109 113 132 199 FFDIAG 109 FFSET D88 PEN 93 PTG 51 PTIM D32 PTION 140 156 263 B 230 RB1T 199 RBITAL 7 03 08 10 118 1132208 orbital localization 108 orbital manipulation 287 orbital spaces Orbitals 17 orbitals closed CL 132 MCSCE 114 closed shell 16 core CL 132 FCI 199
180. MOLDEN has internal features for difference density plots the approach show here is more general in that it bypasses the restriction to STO 3G 3 21G 4 31G and 6 31G basis sets Revision 2006 0 gprint orbitals geometry y planexz 0 H1 0 r h2 0 r h1 alpha r 1 8 alpha 104 int hf wf 10 1 orbital 2100 2 multi wf 10 1 orbital 2140 2 ened examples load dscf density 2100 2 load scf density h2o diffden molden c load dmcscf density 2140 2 load mcscf density add ddiff gdgmcscf l dsct compute dmcscf dscf natorb neworb1 dscf natorb neworb2 dmcscf natorb neworbs ddiff save neworbs 2110 2 save ddiff 2110 2 put molden h2o_ddens molden orb 2110 2 12 5 Geometry Files Using the format GEOMETRY file the geometry definitions are read from file instead of inline This file must contain all informa tion of the symmetry block i e symmetry specifications optional z matrix or xyz input 12 6 Lattice of point charges LATTICE INFILE input file OUTF I LE output file VARGRAD NUCONLY REMOVE A lattice of point charges is included in the calculation through the use of this card An external file input file should be given as input with the following format 12 GEOMETRY SPECIFICATION AND INTEGRATION 75 Comment line number of point charges N x1 y1 z1 q1 flagl xN yN zN qN flagN The x y and z fields stand for the point charge coordinates in A q for its charge and flag 1 indicates
181. MOLPRO User s Manual Version 2006 1 H J Werner Institut f r Theoretische Chemie Universit t Stuttgart Pfaffenwaldring 55 D 70569 Stuttgart Federal Republic of Germany P J Knowles School of Chemistry Cardiff University Main Building Park Place Cardiff CF10 3AT United Kingdom May 2006 Copyright 2006 University College Cardiff Consultants Limited Introduction to MOLPRO MOLPRO is a complete system of ab initio programs for molecular electronic structure calcula tions designed and maintained by H J Werner and P J Knowles and containing contributions from a number of other authors As distinct from other commonly used quantum chemistry packages the emphasis is on highly accurate computations with extensive treatment of the electron correlation problem through the multiconfiguration reference CI coupled cluster and associated methods Using recently developed integral direct local electron correlation methods which significantly reduce the increase of the computational cost with molecular size accurate ab initio calculations can be performed for much larger molecules than with most other pro grams The heart of the program consists of the multiconfiguration SCF multireference CI and coupled cluster routines and these are accompanied by a full set of supporting features The package comprises Integral generation for generally contracted symmetry adapted gaussian basis functions spdfghi There are two
182. Multi State CASPT2 Multi state CASPT2 is implemented as described by Finley et al CPL 288 299 1998 Cur rently this can only be used with the RS2 program i e not with RS2C There are two different modes in which MS CASPT2 calculations can be performed i Each of the states to be mixed is computed independently and finally all states are mixed In the following such calculations will be denoted SS SR CASPT2 single state single reference CASPT2 There is one contracted reference state for each CASPT2 calculation that is spe cific for the state under consideration This is the cheapest method but there are no gradients available in this case It is the users responsibility to make sure that no state is computed twice ii All nstates states are treated together with nstates contracted reference states This is more expensive but should give a more balanced description since the different reference states can mix in the CASPT2 It is required that nstates equals the number of states specified on the STATE directive For this case denoted MS MR CASPT 2 multi state multi reference CASPT2 analytical energy gradients are available see section 22 7 22 3 Performing SS SR CASPT2 calculations If one wants to mix together nstates CASPT2 wavefunctions a nstates single state single reference CASPT2 calculations must be run The first calculation must use RS2 MIX nstates INIT options STATE 1 1 and the subsequent ones
183. NT DFOCK value 92 If nonzero use density screening default Max disk size in Byte for semi direct calculations currently disabled Max memory buffer size for semi direct calculations currently disabled Threshold for writing integrals to disk currently disabled Print option for direct Fock matrix calculation 17 1 5 Special options for UHF calculations NATORB record UNOMIN unomin UNOMA X unomax Save natural charge orbitals in given record Minimum occpation number for UNO CAS default 0 02 Maximum occupation number for UNO CAS default 1 98 17 1 6 Options for local density fitting calculations Please refer section 1 for more options regarding density fitting The following options affect local density fitting as described in H J Werner F R Manby and P J Knowles J Chem Phys 118 8149 2003 and R Polly H J Werner F R Manby and Peter J Knowles Mol Phys 102 2311 2004 Note that local fitting affects the accuracy LOCFIT locfit RDOM locfit RDOMC locfit DOMSEL domesel If nonzero use local fitting for exchange If gt 1 also use local fitting for Coulomb not recommended Radius for fitting domain selection in local fitting default 5 bohr Radius for fitting domain selection for core orbitas in local fit ting default RDOM Criterion for selecting orbital domains in local fitting default 0 1 17 1 7 Options for CPP and polarizabilities CPP cpp MAX
184. Number of geminal functions default 6 GEM CENTRE Centre of even tempered geminal exponents if GEM BASIS EVEN default 1 0 29 EXPLICITLY CORRELATED METHODS 197 GEM RATIO Ratio of even tempered geminal exponents if GEM BASIS EVEN default 3 0 GEM BETA Exponent for Slater type frozen geminal or parameter for weight func tion in other frozen geminal models default 1 4 GEM OMEGA Exponent for weighting function default 1 which means a value derived from GEM BETA GEM MOM Exponent for r in omega fitting default 0 GEM M Exponent for r in weighting function default 0 GEM MAXIT Max iteration in geminal optimization default 200 GEM PRINT Print parameter for geminal optimization default 0 GEM DEBUG Debug option for geminal optimization default 0 GEM_ACC Convergence threshold for geminal line search default 0 001 GEM_FAC Scaling factor for exponents in geminal optimization default 1 0 GEM_METHOD Geminal optimization method augmented Hessian AH or Newton Raphson NR default AH GEM_TRUST Trust ratio in AH geminal optimization default 0 4 GEM_SHIFT Hessian shift in AH geminal optimization default 0 GEM_NUMERICAL Flags numerical integration in geminal optimization default 0 GEM_PLOT Geminal plot file default blank Options only available for the canonical version PRINT ipri Select output level for canonical methods 1pri 0 Standard output ipri 1 Standard output plus
185. OCALIZATION 109 19 6 Selecting the orbital space By default only the valence orbitals are localized in order to ensure invariance of subsequent electron correlation treatments This behaviour can be modified using the OCC and CORE direc tives 19 6 1 Defining the occupied space OCC OCC 01 02 defines the highest orbital o in each symmetry i to be localized 19 6 2 Defining the core orbitals CORE CORE C1 C2 The first c orbitals in each symmetry are treated as core orbitals and not localized Thus orbitals cj 1 to oj are localized in symmetry i 19 6 3 Defining groups of orbitals GROUP OFFDIAG GROUP orbl orb2 orb3 This card defines groups of orbitals to be localized as follows GROUP 1 1 2 1 3 1 a group of orbitals 1 3 in symmetry 1 GROUP 1 1 3 1 equivalent to previous example GROUP 3 1 5 1 8 1 this group includes orbitals 3 5 6 7 8 in symmetry 1 Orbitals in different groups are localized independently Orbitals not included in any group are unchanged 19 6 4 Localization between groups OFFDIAG OFFDIAG If this card is present localize between groups instead of within groups 19 7 Ordering of localized orbitals ORDER type If type CHARGE the orbitals are ordered according to their charge centroids default If type F OCK the orbitals are ordered according to increasing diagonal elements of the fock operator PIPEK or increasing Coulson additive orbital energies B
186. OTAL MRCI ENERGY 75 79831209 ci occ 4 1 2 closed 2 core 1l wf 9 1 1 save 7100 1 TOTAL MRCI ENERGY 75 71309879 ci trans 7300 1 7100 1 1y Transition moment lt 1 3 X 1 1 gt 0 14659810 a u ITransition moment lt 1 3 LY 1 1 gt 0 96200488i a u BH singlet Sigma and Delta states r 2 1 geometry b h b r HEs occ 3 mE br Ls multi occ 3 1 1 frozen 1 wf 6 1 state 3 lquant 0 2 0 wf 6 4 lquant 2 tran lz examples SG rica RUP bh mrci sigma delta Sigma states nergies 25 20509620 24 94085861 ci occ 3 1 1 core 1 wf 6 1 state 2 1 3 Delta states nergies 24 98625171 ci occ 3 1 1 core 1 wf 6 1 state 1 2 Delta state xy component IclroOCc 3 1 1 00fe6 l1 Wf 06 4 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 146 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTUR BATION THEORY Bibliography Original RS2 RS3 H J Werner Mol Phys 89 645 661 1996 New internally contracted RS2C P Celani and H J Werner J Chem Phys 112 5546 2000 All publications resulting from use of this program must acknowledge the above The commands RS2 options RS2C options RS3 options are used to perform second or third order perturbation calculations RS3 always includes RS2 as a first step For closed shell single reference cases this is equivalent to MP2 or MP3 but a different program is used RS2C calls a new more efficient second order program see below which should normal
187. OYS This requires a Fock operator from the preceding energy calculation For localization of Hartree Fock orbitals this operator is stored in the dump record and automatically found For localization of MCSCF orbitals an effective fock operator is computed from the MCSCF density matrix see DENSITY option Alternatively a dump record of a previous SCF calculation can be specified on the FOCK card and then the fock operator is read from this record For degenerate orbitals further ordering according to the the coordinates of charge centres is attempted first according to largest z coordinates then according to x then y 19 ORBITAL LOCALIZATION 110 19 7 1 No reordering NOORDER NOORDER If this card is present the localized orbitals are not reordered This is useful if localized orbitals are used as starting guess and it is intended that their order remains unchanged 19 7 2 Ordering using domains SORT SORT THRCHCHG charge THREIG eps GROUP igrp REVERT centrelist This directive only works for Pipek Mezey localization The orbitals are ordered according to domains and the given centrelist The contributions of the centres to domains are determined by L wdin charges Only centres with charges greater than THRCHCHG default 0 4 are included in these domains The orbitals are reordered according to the following criteria 1 The primary centre in a domain is the one with largest charge the secondary centre the one with t
188. PTIMIZATION OPTG 252 gradient of Hessian calculations numerical gradients will be com puted automatically for the optimized energy or variable However the procedure can include the calculation of analytical gradients for instance for counter poise corrected optimizations in which a linear combination of several gradient calculations is needed VARIABLE varname Optimize the value of variable varname This implies numerical gra dients 39 1 2 Options for optimization methods METHODZRF AH DIIS OSD OSTPATH SRMIN SRTRANS STSTEEP Optimization method to be used See section 39 2 1 for details ROOT 1 2 Minimum search 1 default or transition state search 2 DIRECTION idir Determines step length and direction in reaction path following see section 39 2 16 STEPMAX value Max step length in one optimization step For more detailed specifi cations see section 39 2 12 TRUST value Trust ratio for Augmented Hessian method default 0 5 AHMAX value Maximum step size allowed in the Augmented Hessian procedure This refers to the scaled parameter space default 0 5 CUT value Threshold for ortho normalization used in conjugate gradient update of Hessian default 1 d 3 ROTATE logical If true the Cartesian coordinates are transformed to mini mize rotations default true 39 1 3 Options to modify convergence criteria The standard MOLPRO convergency criterion requires the maximum component of the gradient to be less then
189. QT UE CEPA 0 16 CEPA 1 16 CEPA 2 16 CEPA 3 16 MP2 76 MP3 10 MP4 E QCI TD CCSD 16 BCCD 76 QCI T 16 CCSD T 16 BCCD T 16 CASSCF 16 MRCI 16 ACPF 16 m 99897339 13609416 12844693 13490643 13304720 13431548 13179688 12767140 12839400 13487266 13461684 13431854 13410586 13555640 13546225 13546100 05876129 13311835 13463018 R 1 Ang E ESCE 00000000 13712077 12947355 13593304 13407381 13534209 13282349 12869801 12942062 13589927 13564345 13534515 13513247 13658301 13648886 13648762 05978790 13414496 13565679 Theta 104 degr E EFCI 13412017 00000000 00764722 00118773 00304696 00177868 00429728 00842276 00770015 00122149 00147732 00177561 00198830 00053776 00063191 00063315 07733286 00297580 00146398 27 One can do even more fancy things like for instance using macros stored as string variables See example oh_macros com for a demonstration 6 PROGRAM CONTROL 28 6 PROGRAM CONTROL 6 1 Starting a job The first card of each input should be fext where fext is arbitrary If file 1 is restarted text must always be the same The effect of this card is to reset all program counters etc If the card is omitted text assumes its default value which is all blank 6 2 Ending a job The end of the input is signaled by either an end of fil
190. R Roussel Phys Rev A 39 3761 1989 1 K Us 12 7 LP U 126 where Us 1 e xe 12 1b 127 xe b 128 np 128 and x is defined by the nonlinear equation xe 21 Ap 12 TE 30 129 where Qs Vs 2yD 6 130 Oss Ds Ts 1 131 and y 1 132 C 12 BRUEG Becke Roussel Exchange Functional Uniform Electron Gas Limit A D Becke and M R Roussel Phys Rev A 39 3761 1989 As for BR but with y 0 8 C 13 Bw Becke Wigner Exchange Correlation Functional Hybrid exchange correlation functional comprimising Becke s 1998 exchange and Wigner s spin polarised correlation functionals See reference for more details K 4cpoppp 1 D Vp de e a d 3 u ip Ps Xs 1 6BXsarcsinh x5 where a 3 8 V343 Y 1 134 B 0 0042 135 c 0 04918 136 and d 0 349 137 To avoid singularities in the limit p 0 G a ps B p s 138 1 6Bxsarcsinh Xs C DENSITY FUNCTIONAL DESCRIPTIONS 330 C 14 Cs1 Colle Salvetti correlation functional R Colle and O Salvetti Theor Chim Acta 37 329 1974 C Lee W Yang and R G Parr Phys Rev B 37 785 1988 CS1 is formally identical to CS2 except for a reformulation in which the terms involving v are eliminated by integration by parts This makes the functional more economical to evaluate In the limit of exact quadrature CS 1 and CS2 are identical but small numerical difference
191. REXTS value Distance criterion for extension of strong pair domains REXTC value Distance criterion for extension of strong and close pair domains REXTW value Distance criterion for extension of strong close and weak pair do mains IEXT value Connectivity criterion for extension of all pair domains IEXTS value Connectivity criterion for extension of strong pair domains IEXTC value Connectivity criterion for extension of strong and close pair domains IEXTW value Connectivity criterion for extension of strong close and weak pair domains By default domains are not extended i e the default values of all parameters listed above are zero Note that the pair classes are determined on the basis of the standard domains and therefore domain extensions have no effect on the pair lists Also note that the computational effort increases with the fourth power of the domain sizes and can therefore increase quite dramatically when extending domains This does not affect the linear scaling behaviour in the asymptotic limit 28 6 3 Manually Defining orbital domains DOMAIN It is possible to define the domains by hand using the DOMAIN directive DOMAIN orbital atoml atom2 28 LOCAL CORRELATION TREATMENTS 186 where orbital has the form iorb isym e g 3 1 for the third orbital in symmetry 1 and atomi are the atomic labels as given in the Z matrix geometry input or alternatively the Z matrix row numbers All basis functions
192. RPROD DKEXT THRP DKEXT THR DKEXT THRPROD THRMAX DKEXT THRMAX a AE is the requested accuracy in the energy default 1 d 6 b The thresholds are reduced if the overlap matrix has small eigenvalues c The default thresholds for DMP2 and LMP2 are 0 1 AE 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 64 10 1 Example for integral direct calculations SRevision 2006 0 memory 2 m method hf mp2 ccsd qci bccd multi mrci acpf rs3 some methods basis vdz basis geometry 0o hl o r h2 0 r hl theta geometry gdirect direct option r 1 ang theta 104 bond length and ang xamples do i 1 method loop over methodo direct com Smethod i run method i e i energy save results in variables dip i dmz enddo table method e dip print table of results This jobs produces the following table METHOD E DIP HF 76 02145798 0 82747348 MP2 76 22620591 0 00000000 CCSD 76 23580191 0 00000000 QCI 76 23596211 0 00000000 BCCD 76 23565813 0 00000000 MULTI 76 07843443 0 76283026 MRCI 16 23369819 0 76875001 ACPF 76 23820180 0 76872802 RS3 76 23549448 0 75869972 11 DENSITY FITTING 65 11 DENSITY FITTING Density fitting can be used to approximate the integrals in spin restricted Hartree Fock HF density functional theory KS second order Mgller Plesset perturbation theory MP 2 and all levels of closed shell local correlation methods LMP 2 LMP 4 LQCISD T LCCSD T Den sity fitting is in
193. Rum m RE E RUP E 190 usce eta us bee un VENE UR CIS TUB et 191 PP 191 28 9 4 Orbital domains 222A 192 iia a ae ok do a E Y o ia 193 28 9 6 PairClasses 1 odo e or a a ad 193 Bo ow Wd we XO qe 9 m SLE 193 28 9 8 Intermolecularinterac ons 00 0000 eee eee 194 sl 195 196 199 ge wie ee Hare cal eee te de ce ao da a et 199 aa ee De ate hae aa ee E Bee eo oe 199 303 Frozen core orbitals len 199 30 4 Defining thestatesymmetry o e e 199 E px eM E EE 200 A a 200 201 A A 201 e Bk Ae Bk a at de e d wl eoe o d A 201 51 5 DEI SAPT bucal Rok onc Romo Eo e De Re e we ges De es d es x 203 Vr dee PM ee age ty Ae RA 203 31 5 Density ht ng ecu ooo a ew Ro REOR os y iow E R PR 204 daa a Ad aoe ao a dad wx 204 32 PROPERTIES AND EXPECTATION VALUES 206 A e Regu mos Sow oe e e eos e GO 206 32 1 1 Calling the property program PROPERTY 206 32 1 2 Expectation values DENSITY lens 206 32 1 3 Orbital analysis ORBITAL lle 206 prado URS sea ae 206 ea bale A at aaa e e a 207 o ira a a ee ee aa ds 207 LLL 208 32 2 1 Calling the DMA program DMA o 208 32 2 2 Specifying the density matrix DENSITY een 208 322 3 Linear molecules LINEAR GENERAL ees 208 CONTENTS 32 2 4 Maximum rank of multipoles LIMIT ls 32 2 5 Omitting nuclear contributions NONUCLEAR p
194. S 296 If NATURAL orbitals are generated and saved in a dump record the occupation numbers are automatically stored as well This is convenient for later use e g in MOLDEN 43 4 Adding matrices ADD ADD result fac1 mat1 fac2 mat2 calculates result fac mat fac2 mat2 The strings result matl mat2 are internal names specifying the matrices matl mat2 must exist otherwise an error occurs If result does not exist it is created The factors fac1 fac2 are optional may be variables If not given one is assumed The nuclear values associated to the individual matrices are added accordingly and the result is associated to result 43 5 Trace of a matrix or the product of two matrices TRACE TRACE variable matl factor Computes variable factor tr matl TRACE variable matl mat2 factor Computes variable factor trace matl mat2 The result of the trace operation is stored in the MOLPRO variable variable which can be used in subsequent operations If factor is not given one is assumed 43 6 Setting variables SET SET variable value Assigns value to MOLPRO variable variable where value can be an expression involving any number of variables or numbers Indexing of variable is not possible however 43 7 Multiplying matrices MULT MULT result matl mat2 facl fac2 calculates result fac2 result facl matl mat2 The strings result matl mat2 are the in
195. SD T as well as for analytical LMP2 gradients Only iterative triples methods like LCCSDT 1b can currently not be done with density fitting The errors introduced by DF are negligible and the use of the DF methods is highly recom mended Linear scaling can be obtained in DF LMP2 using the LOCFIT option see Ref 11 in DF LCCSD T the most important parts also scale linearly but some transformation steps scale quadratically Energy gradients are available for LMP2 DF LMP2 DF SCS LMP2 and LQCISD in the latter case only for LOCAL 1 i e the local calculation is simulated using the canonical program and savings only result from the reduced number of pairs Local explicitly correlated methods DF LMP2 R 12 and DF LMP2 F12 are described in section Before using these methods it is strongly recommended to read the literature in order to under stand the basic concepts and approximations A recent review 1 and Ref 2 may be suitable for an introduction References Review 1 H J Werner and K Pfl ger On the selection of domains and orbital pairs in local correla tion treatments Ann Rev Comp Chem in press preprint available under ht tp www theochem uni stu General local Coupled Cluster 2 C Hampel and H J Werner Local Treatment of electron correlation in coupled cluster CCSD theory J Chem Phys 104 6286 1996 3 M Sch tz and H J Werner Local perturbative triples correction T with linear cos
196. TART directive is a precondition for this keyword It may be useful for plotting of orbitals or for providing a guess to be used in the interpretation of a CASSCF solution employing a different active space It is normally advisable to use records on file 2 for vb civb and vbao 36 7 Specifying a guess GUESS key 1 key 2 The GUESS keyword initiates the input of a guess for the valence bond orbitals and structure co efficients key i can be either ORB STRUC or READ These keywords modify the guess provided by the program or specified by the START directive It is thus possible to modify individual orbitals in a previous solution to construct the starting guess 36 7 1 Orbital guess ORB C1 C2 Cmact Specifies a starting guess for valence bond orbital number i The guess is specified in terms of the mact active MOs defining the CASSCF wavefunction Note that the definition of these MOs will depend on how the CI vector was dumped i e which of the SAVE NATORB CANONICAL or LOCALI directives was used see section 20 5 4 Use of one of the three latter keywords is recommended 36 7 2 Guess for structure coefficients STRUC C C5 CNVB Specifies a starting guess for the NV B structure coefficients If this card is not provided and no guess specified by START the perfect pairing mode of spin coupling is assumed for the spatial configuration having the least number of doubly occupied orbitals Note
197. TESIAN command CARTESIAN If this command is encountered the logical MOLPRO variable CARTES IAN is set to true 1 0 and all subsequent calculations use cartesian basis functions This is remembered across restarts One can switch back to spherical harmonics using the command SPHERICAL 13 3 The basis set library The basis set library consists of a set of plain text files together with an associated index that constitute a database of commonly used basis sets primitive gaussians and associated contractions and effective core potentials These files can be found in the source tree as lib libmol and lib libmol index but it is usually more convenient to query the database using one of the provided tools Many of the basis sets are taken directly from the Pacific Northwest National Laboratory basis set database but there are others notably the Stuttgart effective core potentials and bases A simple command line interface to the database is provided through the 1ibmol program It requires the environment variable LIBMOL to point to the 1ib directory but this will default to the location of the source tree at compile time so it is often not necessary to specify it The command line syntax 1s libmol pprint eelement k key t type format where the parameters are print Output level 0 means list matching keys 1 means print also the entry element Specify chemical element If omitted all elements are searc
198. TEXFAC factor multiplying exact exchange in KS Example for the use of these variables for a state averaged MCSCF note that system variables can only be modified using the SET command see section 8 4 SET NELEC 9 defines number of electrons SET SPIN 1 defines wavefunction to be a doublet SET SYMMETRY 1 2 3 defines wavefunction symmetries for state averaged calculation SET STATE 2 1 1 defines number of states to be averaged in each symmetry WEIGHT 2 2 1 1 defines weights for the above four states OCC 5 2 2 number of occupied orbitals in each symmetry CLOSED 2 number of closed shell orbitals in symmetry 1 MCORB 3100 2 record for optimized orbitals MULTI do mescf with above parameters 8 9 Displaying variables Variables or the results of expressions can be displayed in the output using SHOW and TABLE 8 VARIABLES 53 8 9 1 The SHOW command The general form of the SHOW command is as follows SHOW ncol format expression where expression can be an expression or variable ncol is the number of values printed per line default 6 and format is a format default 6F15 8 This can be used to print vectors in matrix form The specification of ncol and format is optional Assume that E is a vector SHOW E prints E using defaults SHOW n E prints E with n elements per line if n 6 more than one line is needed but in any case a new line is started after n elements SHOW n 10f10 4 E prints E in the format
199. TI correspond to those used in MCSCF 3 DEFINITION OF MOLPRO INPUT LANGUAGE 8 3 5 Options Options have the general form NAME value where value can be a number and expression or a string Several options are separated by comma or blank NAME must begin with a character A Z If options are given on a COMMAND line or on directives within a command block they are valid only for the corresponding program see Sec B 3 If options are given in a procedure they are valid only in the procedure and procedures called from the current procedure whenever a procedure is terminated the options of the previous level are restored Options can also be single keywords like SYM or NOSYM In this case the option is switched on or off depending whether or not the key begins with NO Alternatively such logical options can also be set or unset using NAME ON or NAME OFF For instance SYM OFF is equivalent to NOSYM Furthermore YES and NO are aliases for ON and OFF respectively 3 6 Data Data are defined as a sequence of numbers expressions or strings separated by commas or blanks Generally the order of data is essential Empty fields are interpreted as zeros Strings and variables must begin with a character A Z If or follows blank and directly precedes a number or variable it is interpreted as sign and not a binary operator If there are no blanks before and after such operators or blanks follow them they are interpreted as binary
200. TI 112 20 1 Structureoftheinput se 112 bid eh eh eS GE ADE S RS ee ees 113 vdd opi ho bh Be ARR A e a Re ee 113 20 2 2 Frozen core orbitals o ee ee 113 20 2 3 Closed shell orbitals 2 0 20 20 02 0020 00000 114 rcm 114 Torr 114 20 3 14 Defining the statesymmetry ens 114 20 3 2 Defining the number of states in the present symmetry 115 cope 115 e aia RUE S PLS ee 115 i Rok uo od ao x ike dd 115 OTTENUTO 116 D ues uode x dus o ef al elec nos 116 20 4 4 Selecting the primary configuration set 117 Dn 117 rop e 117 Au C 117 Sue 2 BOE GA OR Ee NIE Reg eee aes 118 ok mig n RO WIR XO A pce 118 20 5 5 Natural orb tals a 119 20 5 06 Pseudo canonical orbitals ee ee 120 205 7 Localized orbital 22e 120 205 8 Diabatic orbitals 120 a O a o ge 122 20 6 1 Selecting the CI method o o o 122 20 6 2 Selecting the orbital optimization method 122 Tw 123 oats soa adc 123 baa hae ee hoe Ae of bale ek ws 123 bea sete a re Ses 124 ste Gee aie et 124 DEC a Negro sete eee oats eae 124 ret ee re 124 C ghia weak Diy ee E TTD EC 125 Tu 125 20 83 Maximum number of iterations 126 CONTENTS xiii rk He dee Bape Ge oh AN 126 ar he Sky ee et eee ceca a 126 MENS 127 20 8 7 Saving transformed IntegralS o o 127 209 Coupled perturbed MOSCE oo o
201. TION TREATMENTS 178 T1 TO plus one perturbative update of the triples amplitudes If the accuracy of TO is insufficient very rarely the case this can be used to improve the accuracy The computational cost is at least twice as large as for TO In contrast to TO the triples amplitudes must be stored on disk which can be a bottleneck in calculations for large molecules Also the memory requirements are substantially larger than for TO T1C As T1 but a caching algorithm is used which avoids the simultaneous storage of all triples amplitudes on disk as is the case for T1 or TF Hence T1C requires less disk space but more CPU time than T1 The more disk space is made available for the caching algorithm using the TIDISK option on the local card see below the less CPU time is used TF Full iterative triples calculation With full domains and without weak pair approx imations this gives the same result as a canonical T calculation Typically 3 5 iterations are needed and therefore the computational effort is 2 3 times larger than for T1 The disk and memory requirements are the same as for T1 The TO energy is also computed and printed TFULL and FULL are aliases for TF TA As TF but the T1 energy is also computed Since the first iteration is different for T1 the convergence of the triples iterations is slightly different with TF and TA TF being somewhat faster in most cases TALL and ALL are aliases for TA Density
202. TRDMZ for transition dipole moments If more than two states are computed the index is i 1 i 2 2 j where i gt j gt 1 are state numbers In a state averaged calculation states are counted sequentially for all state symmetries For instance in the following state averaged MCSCF MULTI WF 14 1 0 STATE 3 WF 14 2 0 STATE 2 WF 3 0 the states are counted as i 12 3 4 Symmetry 1 1 1 2 RootinSyn 1 2 3 1 NNU WO 8 8 2 Variables recognized by the program All variables described below are checked by the program but not set except NELEC and SPIN If these are not defined by the user the program uses its internal defaults The variables have no effect if the corresponding input cards are present Variables recognized by the SCF program CHARGE Total charge of the molecule can be given instead of nelec NELEC number of electrons SPIN spin multiplicity minus one SCFSYM METRY wavefunction symmetry SYMMETRY as SCFSYMM only used if SCFSYMM is not present SCFOC C number of occupied orbitals in each symmetry for SCF SCFCL OSED number of closed shell orbitals in each symmetry for SCF SCFORB record of saved orbitals in SCF SCFSTART record of starting orbitals used in SCF Variables recognized by the MCSCF program 8 VARIABLES CHARGE NELEC MCSYM METRY SYMMETRY MCSPIN SPIN MCSTATE STATE WEIGHT LQUANT MCSELECT SELECT MCRESTRICT RESTRICT CONFIG MCOC C OCC MCCL OSED CLOSED MCFROZEN
203. This functional needs to be mixed with 0 1943 exact exchange See reference 9 for more details K e pa pg Pu 0 Pg 0 Ao Ain 4 21 A2 m 4 34 T Y e ps 0 Bo Bum s gt a Bo n x s 7 A2 2 105 3 8 3428 Y ps 9 Co Cin xs 3 Ca n xs 3 where d 1 2 Xa 1 2 xg 106 C DENSITY FUNCTIONAL DESCRIPTIONS 328 ta 107 e a B a B tres n nv mana e r a B T3 U3 V3 Ws X5 Ys P3 o E a B 1 E a B e r a B 75 U5 V5 W5 X5 Yo P3 e r o5 B Ti U1 Vi W1 X1 Y1 P1 o G a P ce 108 r a B 1 44343 e warp 109 tap LL 110 AS ain e r t u v w x y p 2t 1 ur ln 1 2 l 112 t vr wr xr yrPt c 1 709921 113 T 0 031091 0 015545 0 016887 114 U 0 21370 0 20548 0 11125 115 V 7 5957 14 1189 10 357 116 W 3 5876 6 1977 3 6231 117 X 1 6382 3 3662 0 88026 118 Y 0 49294 0 62517 0 49671 119 P 1 1 1 120 A 0 9454 0 7471 4 5961 121 B 0 1737 2 3487 2 4868 122 C 0 8094 0 5073 0 7481 123 and A 0 006 0 2 0 004 124 To avoid singularities in the limit ps 0 G Ps 0 Bo Bin 05 5 2 Ba n Xs A2 dom 3 8 34 Y p Co Cin xs A3 n 05 2 33 C DENSITY FUNCTIONAL DESCRIPTIONS 329 C 11 BR Becke Roussel Exchange Functional A D Becke and M
204. URE 16 Table 3 Numbering of the irreducible representations in the four dimensional groups Cay Can D No Name Function Name Function Name Function 1 A1 S Z Ag S XY A S 2 Bi X XZ Au z B3 xX yz 3 B y YZ B x y By y xz 4 A2 xy B XZ yZ B xy Table 4 Numbering of the irreducible representations in the two dimensional groups C C5 C No Name Function Name Function Name Function 1 A S X Y Xy A Z Xy Ag S Xy XZ YZ 2 A 2 X2 YZ B X Y XZ YZ u X Y Z WF NELEC nelec SYM METRY irrep spin spin CHARGE charge where nelec is the total number of electrons irrep is the number of the irreducible representation and spin equals 2 x S where S is the total spin quantum number Instead of nelec also charge can be given which specifies the total charge of the molecule For instance for a calculation in Cp symmetry with 10 electrons WF 10 3 0 denotes a 1B state and WF 10 1 2 a A state The charge can also be defined by setting the variable CHARGE SET CHARGE charge This charge will be used in all energy calculations following this input Not that SET is required since CHARGE is a system variable cf section 8 4 Although in principle each program unit requires a WF command in practice it is seldom nec essary to give it The program remembers the information on the WF card and so one might typically specify the information in an SCF calculation but then not in subseque
205. VEL Set to 1 for first order terms E 1 and E 1 pl exch to 2 for additional sec ond order exchange induction terms E and ES ud and 3 for all first and second order terms including then also EN and ES _ disp default 3 SAPT FITLEVEL Level of density fitting approximations in SAPT which can have val ues 0 to 3 default 0 SAPT ICPKS Switch between iterative 21 and non iterative 0 solution of coupled perturbed Kohn Sham equations default 0 SAPT CPKSTHR Threshold for density matrix convergency in the coupled perturbed Kohn Sham program default 1 d 6 SAPT CPKMAXIT Maximum number of iterations in the coupled perturbed Kohn Sham program default 50 SAPT FROZENA Number of frozen electrons in the response calculations for monomer A default 0 C6 Calculate dispersion coefficients for the two monomers The following parameters are of importance if SAPT_FITLEVEL gt O SAPT NFRO DISP Number of frequencies for the Casimir Polder integration default 12 SAPT NORM DISP Norm for the density fitting which can be either COULOMB or NATURAL default COULOMB SAPT_DISP_N4 Can speedup the calculation of the dispersion energy by N scaling default 1 31 SYMMETRY ADAPTED INTERMOLECULAR PERTURBATION THEORY 205 THR_XCKERN FIT_XCKERN SAPT_DISK COMPRESS_THR UNCOUPLED THRAO THRMO THROV THRPROD THRSW Density threshold for the xc kernel matrix elements default 1 d 8 Fit both side
206. Z R ANG EGRST E EXC 2 1 E EXC 1 2 E_EXC 1 4 0 80 100 23687380 0 56664 0 41204 0 56934 0 81 100 24094256 0 56543 0 40952 0 56812 0 82 100 24451598 0 56422 0 40695 0 56690 etc 25 4 EOM CCSD transition moments for hydrogen fluoride This example shows how to calculate and store CCSD and EOM CCSD density matrices cal culate dipole and quadrupole moments transition moments from the ground to excited states are calculated and how to use the EOM CCSD excited state density for Mulliken population analysis SRevision 2006 0 Properties and transition moments for several lowest states of hydrogen fluoride memory 2 m basis avdz define basis set geometry h f h r z matrix r 0 92 Ang define distance hf do SCF calculation NICA UM ccsd do CCSD calculation dm 5600 2 density matrices will be stored here expec qm require quadrupole moments om 3 1 2 2 2 3 2 4 trans 1 do EOM CCSD calculation properties pop density 5600 2 state 2 4 make population analysis for state 2 4 25 EXCITED STATES WITH EQUATION OF MOTION CCSD EOM CCSD 168 This calculation produces the following table Final Results for EOM CCSD moments in a u State Exc Energy eV X 2l 14 436 Right transition moment 0 00000000 Left transition moment 0 00000000 Dipole strength 0 45007246 Oscillator strength 0 15917669 Dipole moment 0 00000000 etc Y Z
207. abatic energies for H2S obtained from CI vectors R El E2 H11CI H22CI H21CI 2 50 398 64296319 398 63384782 398 64296319 398 63384782 0 00000000 2 55 398 64572746 398 63666636 398 64509901 398 63729481 0 00230207 2 60 398 64911752 398 63771802 398 64662578 398 64020976 0 00471125 Diabatic energies for H2S obtained from CI vectors and orbital correction R El E2 HAT H22 H21 2 50 398 64296319 398 63384782 398 64296319 398 63384782 0 00000000 2 55 398 64572746 398 63666636 398 64509941 398 63729441 0 00230139 2 60 398 64911752 398 63771802 398 64662526 398 64021027 0 00471160 The results in the first table are obtained from the CI contribution to the state overlap matrix only while the ones in the second table include a first order correction for the orbitals In this case both results are almost identical since the DI AB procedure has been used to minimize the change of the active orbitals This is the recommended procedure If simply natural orbitals are used without orbital diabatization the following results are obtained from the otherwise unchanged calculation Diabatic energies for H2S obtained from CI vectors R El E2 H11CI H22CI H21CI 2 50 398 64296319 398 63384782 398 64296319 398 63384782 0 00000000 2 55 398 64572742 398 63666630 398 64475612 398 63763760 0 00280315 2 60 398 64911746 398 63771803 398 64521031 398 64162518 0 00541050 Diabatic energies for H2S obtained from
208. acements not rec ommended Note that the displacement type for gradient and hessian must be the same icalc 0 Recalculate the complete Hessian matrix numerically after each hstep optimization steps default icalc 1 Recalculate selected Hessian matrix elements if the relative deviation of this element before and after update see UPDATE sec tion 39 2 9 is larger than thresh If thresh is not specified a default value of thresh 0 05 i e a maximum deviation of 596 is used icalc 2 Recalculate complete Hessian matrix if the RMS deviation of the Hessian matrix before and after update is larger than thresh If thresh is not specified a default value of Threshold for partial or dynamical update of hessian see above 39 2 8 Hessian elements HESSELEM HESSELEM value activel active2 sets the starting value for hessian matrix element between active variables activel active2 to value If active2 is omitted it defaults to active diagonal element As many HESSELEM directives as needed may be given 39 2 0 Hessian update UPDATE UPDATE T YPE type MAX maxupd This directive chooses the update type and limits the number of points used for the hessian update to maxupd The default number of steps used in hessian update procedures is 5 If there are symmetry constraint in the coordinates of the optimization the default number may be lower than five In minimizations type may be BFGS IBFGS CGRD NONE Use BFGS
209. ains are automatically frozen in geometry optimizations and frequency calculations see section 28 9 7 28 9 6 Pair Classes The strong close weak and distant pairs are selected using distance or connectivity criteria as described in more detail in section 28 7 Strong pairs are treated by CCSD all other pairs by LMP2 In triples calculations all orbital triples ijk are included for which ij ik and jk are close pairs In addition one of these pairs is restricted to be strong The triples energy depends on the strong and close pair amplitudes The close pair amplitudes are taken from the LMP2 calculation Thus increasing the distance or connectivity criteria for close and weak pairs will lead to more accurate triples energies While for near equilibrium properties like geometries and harmonic vibrational frequencies the default values are normally appropriate larger distance criteria are sometimes needed when computing energy differences in particular barrier heights In cases of doubt RWEAK should first be increased until convergence is reached and then RCLOSE can be varied as well Such tests can be performed with small basis sets like cc pVDZ and the optimized values then be used in the final calculations with large basis sets 28 9 7 Gradients and frequency calculations Geometry optimizations 15 17 and numerical frequency calculations 18 20 can be performed using analytical energy gradients 15 17 for local MP2 LMP2 geom
210. al fitting These pa rameters only have an effect if LOCFIT 1 The local fitting domains are determined in two steps first primary orbital domains are deterimined In the LMP2 and LCCSD programs the primary orbital domains are the same as used for excitation domains and determined by the Boughton Pulay procedure as described in Sect Depending on the value of FI TDOM MP2 or FITDOM CCSD for LMP2 and LCCSD respectively either the orbital domains are used di rectly or united pair domains are generated In DF HF the primary orbital domains include all basis functions at atoms which have L wdin charges greater or equal to THRCHG_SCF In the second step the primary fitting domains are extended using either distance criteria RDOMAUX in bohr or bond connectivity criteria IDOMAUX IDOMAUX 1 means to include all functions at atoms wich are at most one bond distant from the primary domains By default distance criteria are used However if IDOMAUX ge 0 the distance criteria are ignored and connectivity is used THRCHG SCF Parameter to select the primary orbital domains in local ex change fitting default 0 1 All atoms are include which have L owdin charges greater than this value The primary domains are extended according to RDOMAUX SCF or IDOMAUX_SCF FITDOM_MP2 Parameter to select primary fitting domains in LMP2 transfor mation default 3 1 use orbital domains 2 use united orbital domains of strong pairs 3 use united orbital domains
211. al sets are stored in the same dump record and can be restored at later stages using ORBITAL record T YPE LOCAL or ORBITAL record T YPE P ROJECTED respectively 28 LOCAL CORRELATION TREATMENTS 182 DOMONLY value DSTMLT level INTERACT If value gt 0 only domains are made but no energy is computed This can be used to check and save the domains for later use Determines the expansion level of the multipole expansion of distant pairs e g 1 means dipole approximation 2 quadrupole approxima tion and so on The default for MULTP is 3 Automatically determine individual molecules in a calculation and set appropriate pair lists for computing interaction energies See section 28 9 8 for more details Parameters for energy partitioning IEPART value EPART cutoff Miscellaneous options SKIPDIST skipdist ASYDOM jiterm LOCSING locsing MAXANG max enables disables energy partitioning iepart 0 Energy partitioning is disabled iepart 1 Energy partitioning is enabled iepart 2 Energy partitioning is enabled Additionally a list of all pair energies and their components is printed Cutoff parameter to determine individual monomers in a cluster i e centre groups Should be somewhat larger than the largest intramolec ular bond length given in a u Test parameter Its value should only affect the efficiency but not in fluence the results skipdist 1 Weak and distant pairs are set to ze
212. all o C 15 CS2 Colle Salvetti correlation functional o C 16 DIRAC Slater Dirac Exchange Energy o C 17 G96 Gill s 1996 Gradient Corrected Exchange Functional C 18 HCTH120 Handy least squares fitted functional C 19 HCTH147 Handy least squares fitted functional C 20 HCTH93 Handy least squares fitted functional joe eis wed dcus HERE IEEE TN EE C 22 LYP Lee Yang and Parr Correlation Functional AM C 24 MK00 Exchange Functional for Accurate Virtual Orbital Energies C25 PS Cie L5 xxm E xem box ue Ws ded fes us RE ee HEUS Fono e E Rede XX 301 303 304 304 304 304 304 305 306 309 310 311 312 312 312 313 315 316 316 317 317 318 319 CONTENTS xxi v atan ee a 338 ed tak 340 E eee ee ee Be eee 340 Puoi gw ee ade ere E Sd dE eee eee aca 341 C 30 PW91C Perdew Wang 1991 GGA Correlation Functional 341 C 3 PW91X Perdew Wang 1991 GGA Exchange Functional 343 MTM 344 C 33 STEST Test for number of electrons oo o 344 C 34 TH1 Tozer and Handy 1998 o o o o o o 345 C35 THAI hon a a a a Ra Rb 346 SEDE EP wt CL 347 C37 THAN nie dae uo ra RON ose RG Aem M doe NR d P DER 348 C38 LHGECROJ 24 a A ws a Nr PES RUE 349 Meek pP 350 CAO THGEG ra ut
213. appropriately The subsequent SCF calculations use the modified one electron operator Note that it is usually recommended to add fields with the DIP QUAD or FIELD commands 43 MATRIX OPERATIONS define field strength SRevision 2006 0 memory 2 m R 0 96488518 ANG THETA 101 90140469 geometry H1 O H1 R H2 0 R H1 THETA hf wf 10 1 field 0 05 matrop load h0 h0 load xx oper xx load yy oper yy load ZZ Oper zz 301 load one electron hamiltonian load second moments add h01 h0 field zz 0 5 field xx 0 5 field yy ladd second moments to h0 and store in h01 save h01 1210 1 h0 save ho hf Ido scf with modified h0 examples matropfield com matrop load h0 h0 load ho load qmzz oper qmzz load quadrupole moment qmzz add h01 h0 field qmzz add quadrupole moment to h0 same result as above with second moment save h01 1210 1 h0 save ho hf Ido scf with modified h0 quad field ladd quadrupole field to h0 hf do scf with modified h0 same result as above with matrop field zz field xx 0 5 field yy 0 5 field add general field same result as above hf Ido scf with modified h0 same result as above with matrop field zz field same as before with separate field commands field xx 0 5 field field yy 0 5 field hf Ido scf with modified h0 same result as above with matrop 43 24 Exercise SCF program Write a closed shell SCF program for H20 using
214. arches invoked with the ROOT option see section 39 2 11 key can be RF Rational Function method default DIIS Pulay s Geometry DIIS method see above QSD Quadratic Steepest Descent Transition State search using the image Hessian method see J Sun and K Ruedenberg J Chem Phys 101 2157 1994 The use of this option is recommended for transition state searches especially in complicated cases The optimization step is checked and the Hessian is recalculated when approaching a troublesome region of the PES Thus this method is somewhat safer and often faster in reaching convergence than the RF or DIIS method The Hessian recalculation safeguard may be turned off using the METHOD QSD NOHESS input card SRTRANS Old version of QSD For reaction path following the input key is OSDPATH Quadratic Steepest Descent reaction path following This methods determines reaction paths intrinsic reaction coordinates IRCs by following the exact steepest descent lines of subsequent quadratic ap proximations to the potential energy surface The Hessian matrix is calculated numerically at the first optimization step and subsequently updated by Powell or BFGS update If a given arc length of the steep est descent lines is exceeded the Hessian is recalculated numerically see OPTION section 39 2 16 For details see J Sun and K Rueden berg J Chem Phys 99 5269 1993 It is also possible to recalculate the Hessian after each m steps usin
215. ard matrix operations can be performed with MATROP e g printing records linearly combining or multiplying matrices or forming the trace of a product of two matrices 4 6 Memory allocation MOLPRO allocates memory dynamically as required by the user on the MEMORY card Thus it is not necessary to maintain different versions of the program with different memory sizes If the MEMORY command is omitted the program will use a default memory size which depends on the hardware used and how the program was installed Note that on Unix machines the default memory can be set on the molpro command line using the flag m 4 7 Multiple passes through the input Itis possible to perform loops over parts of the input using DO loops very much as in FORTRAN programs DO loops may be nested to any reasonable depth This can be conveniently used for instance to compute automatically whole potential energy surfaces 4 8 Symmetry MOLPRO can use Abelian point group symmetry only For molecules with degenerate symme try an Abelian subgroup must be used e g Cay or D for linear molecules The symmetry group which is used is defined in the integral input by combinations of the symmetry elements x y and z which specify which coordinate axes change sign under the corresponding generat ing symmetry operation It is usually wise to choose z to be the unique axis where appropriate essential for C and C27 The possibilities in this case are shown
216. are merged Selection threshold for configurations read from disc records recl rec2 This applies to the norm of all CSFs for each or bital configuration Specifies from which state vector the configurations are se lected This only applies to the case that the configurations were saved in a state averaged calculation If refstat is zero or not specified the configurations are selected from all states If refstat is greater than zero then the specified reference state is used If refstat is less than zero then all appropriate reference states are used Lastly if refstat is of the form istatl istat2 states istat through istat2 are used maximum number of open shells in the selected or generated configurations 21 2 8 Occupation restrictions RESTRICT nmin nmax orb orb2 orby This card can be used to restrict the occupation patterns in the reference configurations Only configurations containing between nmin and nmax electrons in the specified orbitals orb orb Orb are included in the reference function If nmin and nmax are negative configurations with exactly abs nmin and abs nmax electrons in the specified orbitals are deleted This can be used for instance to omit singly excited configurations The orbitals are specified in the form number sym where number is the number of the orbital in irrep sym Several RESTRICT cards may follow each other The RESTRICT cards must follow the WF or REF cards to which th
217. arizabilities using finite fields 32 PROPERTIES AND EXPECTATION VALUES 212 SRevision 2006 0 H20 finite field calculations r 1 85 theta 104 set geometry parameters geometry 0 z matrix input HO H2 0 r H1 theta basis avtz define default basis field 0 0 005 0 005 define finite field strengths method hf mp4 ccsd t casscf mrci k 0 do i 1 field loop over fields dip field i ladd finite field to H do m 1 method loop over methods k k 1 method m calculate energy examples e k energy save energy h2o field com enddo enddo k 0 n method do m 1 method k k 1 energ m e k dipmz m e k n k 2 n field 2 field 3 dipole moment as first energy derivative dpolz m e k n e k 2 n 2 e k field 2 field 1 field 3 field 1 polarizability enddo table method energ dipmz dpolz title results for H20 r R theta theta basis Sbasis 32 5 Relativistic corrections Relativistic corrections may be calculated within the Cowan Griffin approach by computing ex pectation values of the mass velocity and 1 electron Darwin integrals these should be generated using the property integral program with keyword REL The expectation values can be computed within the SCF MCSCF and CI programs in the usual way using the EXPECT command again with the keyword REL The mass velocity and Darwin terms and their sum are subsequently available through the MOLPRO variable
218. as SYMMETRY AUTO logical Same as SYMMETRY NOSYM Step length for distances in numerical gradient calculations in bohr The default is 0 01 Step length for symmetrical displacements in bohr The default is 0 01 Step length for angles in numerical gradient calculations in degree The default is 1 logical Use 4 point formula for accurate numerical gradient logical Force the use of numerical gradients even if gradients are available 39 1 7 Options for computing Hessians By default an approximate Hessian model Hessian is used Optionally a Hessian can be computed in the optimization or read from a previous Hessian or frequency calculation NUMHE SS hstep HESSREC record READHESS HESSPROC procname HESSVAR varname HESSCENT If given a numerical Hessian is computed in each hstep th iteration If hstep 0 or not given only an initial Hessian is computed Read initial Hessian from the given record If record is not given or zero the last computed Hessian is used logical Same as HESSREC 0 specifies a procedure to be used for computing the Hessian This pro cedure must be define a complete energy calculation orbital optimiz ation and correlation treatment A different method can be used than for the optimized energy For instance an MP2 Hessian can be used for CCSD T optimizations or a CASPT2 Hessian for MRCI opti mizations By default the same procedure is used for the Hessian as
219. ation coordinates COORD o o 257 39 2 3 Displacement coordinates DISPLACE o 258 39 24 Defining active geometry parameters ACTIVE 258 39 25 Defining inactive geometry parameters INACTIVE 258 39 2 6 Hessian approximations HESSIAN rens 258 39 2 7 Numerical Hessian NUMHESS ee 259 39 2 8 Hessian elements HESSELEM len 260 39 29 Hessian update UPDATE 2l 260 39 2 10 Numerical gradients NUMERICAL 261 39 2 11 Transition state saddle point optimization ROOT 262 39 2 12 Setting a maximum step size STEP o 262 39 2 13 Redefining the trust ratio TRUST o o 262 39 2 14 Setting a cut parameter CUT 2 2 ooo o 262 39 2 15 Line searching LINESEARCH o 00002 eee 263 39 2 16 Reaction path following options OPTION 263 39 2 17 Optimizing energy variables VARIABLE 263 39 2 18 Printing options PRINT ens 264 39 2 19 Conical Intersection optimization CONICAL 264 m 267 ee pra dodo a a oe 268 AR NN 269 39 3 3 Additional options for SLAPAF o o o o o 269 A ee Beek wp NS ID gee 270 eH 270 39 4 2 Optimization using natural internal coordinates BMAT 270 a ee ee aa Sk ed 271 39 44 Optimization using geometry DI S 271 39 4 5 Transition state of th
220. ation sets CSFs only PSPACE print list of configurations making up the primary space ORBITALS print orbitals see also ORBPRINT NATORB print natural orbitals see also ORBPRINT VIRTUALS print virtual orbitals see also ORBPRINT CIVECTOR print CI vector better use CANORB or NATORB INTEGRAL print transformed integrals for testing only DENSITY print density matrices HESSIAN print hessian DIAGONAL print diagonal elements of hessian GRADIENT print gradient LAGRANGI print Lagrangian STEP print update vector ADDRESS print addressing information for testing only DEBUG print debugging information CI2 print debugging information in routine ci2 Warning may be long IO print debugging information in I O routines 20 8 2 Convergence thresholds Convergence thresholds can be modified using ACCURACY GRADIENT conv STEP sconv NERGY econv where 20 THE MCSCF PROGRAM MULTI 126 conv Threshold for orbital gradient default 10 econv Threshold for change of total energy default 1076 sconv Threshold for size of step default 1073 The default values can be modified using the global GTHRESH command see section 6 1 1 Normally the above default values are appropriate 20 8 5 Maximum number of iterations MAXITER maxit maxit is maximum number of iterations default 6 If the calculation does not converge in the default number of iterations you should first think about the reason before increas
221. be available within any particular time MOLPRO on the WWW The latest information on MOLPRO including program updates can be found on the worldwide web at location http www molpro net References All publications resulting from use of this program must acknowledge the following MOLPRO version 2006 1 a package of ab initio programs H J Werner P J Knowles R Lindh F R Manby M Sch tz P Celani T Korona G Rauhut R D Amos A Bernhardsson A Berning D L Cooper M J O Deegan A J Dobbyn F Eckert C Hampel and G Hetzer A W Lloyd S J McNicholas W Meyer and M E Mura A Nicklass P Palmieri R Pitzer U Schumann H Stoll A J Stone R Tarroni and T Thorsteinsson see http www molpro net Some journals insist on a shorter list of authors in such a case the following should be used instead MOLPRO version 2006 1 a package of ab initio programs H J Werner P J Knowles R Lindh F R Manby M Sch tz and others see http www molpro net Depending on which programs are used the following references should be cited Integral evaluation SEWARD R Lindh U Ryu and B Liu J Chem Phys 95 5889 1991 Integral direct Implementation M Sch tz R Lindh and H J Werner Mol Phys 96 719 1999 MCSCF CASSCF H J Werner and P J Knowles J Chem Phys 82 5053 1985 P J Knowles and H J Werner Chem Phys Lett 115 259 1985 See also H J
222. be given on a DF IT directive see section Tip The most important options for density fitting in local methods are BASIS MP 2 string Fitting basis set used in LMP2 and in LCCSD for integrals with up to 2 external orbitals If a correlation consistent basis set is used e g cc pVTZ the corresponding fitting basis for MP2 us used by default cc pVTZ MP2FIT Otherwise the fitting basis set must be defined in a preceding basis block see section 13 BASIS_CCSD string Fitting basis set used in LCCSD for integrals over 3 and 4 external orbitals The default is BASIS MP2 and this is usually sufficient However the accurate approximation of 4 external integrals in LCCSD requires larger fitting basis sets than LMP2 Therefore in order to minimize fitting errors it is recommended to use the next larger fit ting basis e g BASIS CCSD VOZ for orbital basis VTZ LOCFIT value If LOCFIT 1 local fitting is enabled This is necessary to achieve linear scaling in DF LMP2 see Refs 11 14 The errors introduced by local fitting are usually very small but there are some exceptions For instance LOCFIT 1 must not be used in counterpoise calcula tions see section 28 9 8 Note that for small molecules LOCFIT 1 can be more expensive than LOCFIT 0 For further details and options for density fitting see section 11 29 EXPLICITLY CORRELATED METHODS 196 29 EXPLICITLY CORRELATED METHODS Explicitly correlated MP2 R12 and MP2 F12 calculations ca
223. between the atoms are filled with additional segments Now the A matrix can be set up The matrix elements will be calculated from the basis grid points of the segments for close and medium segment distances governed by the di sex value 41 THE COSMO MODEL 286 or using the segment centres for large segment distances Outlying charge correction The non vanishing electron density outside the cavity causes an error that can be corrected by the outlying charge correction This correction uses the potential on the so called outer surface defined by the radii R rsolv x routf to estimate a correction term for the screening charges and the energies A Klamt and V Jonas J Chem Phys 105 9972 9981 1996 The correction will be performed once at the end of a converged SCF calculation All corrected values can be found in the COSMO output file 42 ORBITAL MERGING 287 42 ORBITAL MERGING Orbitals can be manipulated using the MERGE facility For instance this allows the construction of molecular orbitals from atomic orbitals to merge and orthogonalize different orbital sets or to perform 2 x 2 rotations between individual orbitals Other orbital manipulations can be performed using the LOCALI program see section 19 or the MATROP program section 43 The merge program is called using MERGE namout file All subcommands described in the following sections may be abbreviated by three characters namout file specifies the output da
224. c d X p c d 486 Q c d V 4d c 487 X i c d ci 4 d 488 k 0 0310907 0 01554535 1 61 489 1 0 409286 0 743294 0 0047584 490 m 13 0720 20 1231 1 13107 491 and n 42 7198 101 578 13 0045 492 INDEX 355 C 44 VWN5 Vosko Wilk Nusair 1980 V local correlation energy VWN 1980 V functional The fitting parameters for Ag r y appear in the caption of table 7 in the reference See reference for more details K pe 493 where x 1 4 vee E l 494 ca Pe Pe 495 p e 2 ay 1 2 nC 496 y 0 0 O 9n 497 h 4 9 1 498 E A q k l m n 499 q k2 lh2 m m 500 a q k3 13 m3 n3 501 q A p c d A in zi 2carctan 222 Q c d cp in E25 2 c 2p arctan 222 Q c d X p c d 502 Q c d V 4d c 503 X i c d P ci d 504 k 0 0310907 0 01554535 1 61 505 l 0 10498 0 325 0 0047584 506 m 3 72744 7 06042 1 13107 507 and n 12 9352 18 0578 13 0045 508 INDEX Index comments in input 5 ck comma 5 a end of input record 5 ACCURACY 97 125 ACPF 136 ACTIVE 258 ADD 209 244 275 287 ALTERN 232 ANGULAR 103 AOINT 69 AQCC 136 arrays Atomic mass 75 BASIS basis cartesian spherical harmonic basis set 77 contraction even tempered primitive BCCD BMAT 257
225. card contains a number of input fields Input fields may be up to 256 characters wide and contain either expressions or strings The fields can be separated by commas or blanks We recommend the general use of commas in order to avoid unexpected results Each line may start with a label A label is separated from the body of the line by a colon The colon is part of the label The length of the label must not exceed 6 characters including the colon and the labels must be unique Labels may be useful with GOTO commands Example GOTO START START CCSD T 3 DEFINITION OF MOLPRO INPUT LANGUAGE 6 Here START is a label and CCSD T is a command Strings containing blanks can be entered using quotes For instance This is a string is interpreted as one string but This is a string is a sequence of four strings in four subsequent fields Strings in quotes are not converted to upper case Input lines may be concatenated using at the end of the line s to be continued Any number of lines may be concatenated up to a total length of 1024 characters only 500 characters are possible on older IBM systems Filenames may be up to 31 characters long provided that long filenames are supported by the Unix system used An exception are older CRAY systems which allow only 8 characters for the names of binary MOLPRO files 3 0 Commands A command invokes a particular program It may be followed by local input for this program enclosed
226. ce first non zero x coordinate to be positive Similarly ZSIGNY ZSIGNZ can be set for the y and z coordinates respectively If is used instead of as last char acter the corresponding coordinate is forced to be negative This can be useful to fix the orientation of the molecule across different calculations and geometries Alternatively the system variables Z8 IGNX ZSIGNZ ZSIGNZ can be set to positive or negative values to achieve the same effect For the C5 and D2 point groups force the primary plane to be xz instead of the default yz The geometry builder attempts by swapping coordinate axes to place as many atoms as possi ble in the primary plane so for the particular case of a planar molecule this means that all the atoms will lie in the primary plane The default implements recommendation 5a and the first part of recommendation 5b specified in J Chem Phys 55 1997 1955 PLANEYZ and PLANEXY may also be specified but note that the latter presently generates an error for Czy The general form of an atom specification line is group atom pi P P2 Q P3 p J or alternatively group atom p1 X y z where group atom P atomic group number optional Can be used if different basis sets are used for different atoms of the same kind The basis set is then referred to by this group number and not by the atomic symbol chemical symbol of the new atom placed at position po T
227. ce for each system Concerning the more technical parameters in the DFT monomer calculations it is recommended to use lower convergence thresholds and larger intergration grids compared to standard Kohn Sham calculations 31 4 High order terms It has been found that third and higher order terms become quite important if one or both monomers are polar As no higher than second order terms are currently implemented in SAPT one may use a non correlated estimation of those terms by using supermolecular Hartree Fock see e g 7 This can be done by adapting the following template dimer hf edm energy monomer A dummy monomer2 hf save ca ema energy sapt monomerA monomer B 31 SYMMETRY ADAPTED INTERMOLECULAR PERTURBATION THEORY 204 dummy lt monomer1 gt hf start atdens save cb emb energy sapt monomerB linteraction contributions sapt sapt level 2 intermol ca ca cb cb sup edm ema emb 1000 mH dHF esup elpol elex e2ind e2exind which stores the resulting 6 HF term in dHF 31 5 Density fitting In order to be able to study interactions between extended monomers one can use density fitting to approximate the integrals in SAPT 7 For this one may use the input sapt intermol ca ca cb cb fitlevel 3 dfit basis coul jkfit basis exch jkfit basis mp2 2mp2fit cfit scf 3 with in the basis section defined jkfit and mp2fit fitting basis sets see section 11 31 6 Options SAPT LE
228. cedes previous ones Either all specifications must be given on one BASIS card or a basis input block must be used BASIS NAME is now entirely equivalent to BASIS NAME i e a global default basis set is defined and the variable BASIS is set in both cases Pseudopotential energy calculations can now be performed with up to i functions gradi ents with up to functions Many internal changes have been made to make MOLPRO more modular and stable Sup port has been added for recent operating systems on Compaq HP SGI SUN and Linux The patching system has been improved Features that were new in MOLPRO2000 Relative to version 98 1 there are the following principal changes and additions There was a fundamental error in the derivation of the spin restricted open shell coupled cluster equations in J Chem Phys 99 5129 1993 that is also reflected in the RCCSD code in MOLPRO version 98 1 and earlier This error has now been corrected and an er ratum has been published in J Chem Phys 112 3106 2000 Fortunately the numerical implications of the error were small and it is not anticipated that any computed properties will have been significantly in error There was a programming error in the transformation of gradients from Cartesian to in ternal coordinates which in some cases resulted in slow convergence of geometry opti mizations The error is now fixed Vibrational frequencies formerly by default used averag
229. centred at the given atoms are included into the domain For instance DOMAIN 3 1 C1 C2 defines a domain for a bicentric bond between the carbon atoms C1 and C2 The DOMAIN di rective must be given after any OCC CLOSED or CORE directives Note that the order of the localized orbitals depends on the localization procedure and could even change as function of geometry and therefore manual DOMAIN input should be used with great care The domains of all orbitals which are not explicitly defined using DOMAIN directive are determined automati cally as usual 28 7 Options for selection of pair classes There are two alternative modes for defining the pair classes either distance criteria RCLOSE RWEAK RDIST RVDIST can be used These are in Bohr and refer for a given orbital pair ij to the minimum distance R between any atom in the standard orbital domains i and any atom in the standard orbital domains j Alternatively the connectivity criteria ICLOSE IWEAK IDIST IVDIST can be used These refer to the minimum number of bonds between any atom contained in the standard domain i and any atom contained in the standard domain j The advantage of using connectivity criteria is the independence of the bond lengths while the advantage of distance criteria default is that they are also effective in non bonding situations Only one of the two possibilities can be used i e they are mutually exclusive The use of distance criteria is the
230. cient way is to use static and time dependent DFT theory This variant of SAPT termed as DFT SAPT 2 6 has in contrast to Hartree Fock SAPT the appeal ing feature that the polarisation terms E EQ E a are potentially exact i e they come out exactly if the exact exchange correlation xc potential and the exact frequency dependent xc response kernel of the monomers were known On the other hand this does not hold for the exchange terms since Kohn Sham theory can at best give a good approximation to the exact density matrix of a many body system It has been shown 6 that this is indeed the the case and therefore DFT SAPT has the potential to produce highly accurate interaction energies compa rable to high level supermolecular many body perturbation or coupled cluster theory However in order to achieve this accuracy it is of crucial importance to correct the wrong asymptotic behaviour of the xc potential in current DFT functionals 3 5 This can be done by using e g ks 1da asymp lt shift gt which activates the gradient regulated asymptotic correction approach of Gr ning et al J Chem Phys 114 652 2001 for the respective monomer calculation The user has to supply a shift parameter for the bulk potential which should approximate the difference between the exact ionisation potential of the monomer and the negative HOMO energy obtained from the respective standard Kohn Sham calculation Note that this needs to be done only on
231. cluding secondary input files INCLUDE INCLUDE file echo Insert the contents of the specified file in the input stream In most implementations the file name given is used directly in a Fortran open statement If the parameter echo is nonzero the included file is echoed to the output in the normal way but by default its contents are not printed The included file may itself contain INCLUDE commands up to a maximum nesting depth of 10 6 5 Allocating dynamic memory MEMORY MEMORY n scale Sets the limit on dynamic memory to floating point words If scale is given as K n is multiplied by 1000 if scale is M n is multiplied by 1 000 000 Note The MEMORY card must precede all FILE cards Examples MEMORY 90000 allocates 90 000 words of memory MEMORY 500 K allocates 500 000 words of memory MEMORY 2 M allocates 2 000 000 words of memory 6 6 DO loops DO ENDDO DO loops can be constructed using the DO and ENDDO commands The general format of the DO command is similar to Fortran DO variable start end increment L unit where start end increment may be expressions or variables The default for increment is 1 In contrast to Fortran these variables can be modified within the loop to be used with care For instance DR 0 2 DO R 1 0 6 0 DR ANG IF R EQ 2 DR 0 5 IF R EQ 3 DR 1 0 ENDDO performs the loop for the following values of R 1 0 1 2 1 4 1 6 1 8 2 0 24 54 3 305 4 04 50
232. commands of the present and lower level procedures will be printed If ECHO is specified in the main input file all subsequent procedures are printed Certain important input data can be passed to the program using variables For instance occu pancy patterns symmetries number of electrons and multiplicity can be defined in this way see section B 8 for more details This allows the quite general use of procedures For example assume the following procedure has been defined in molproi rc PROC MRCI IF INTDONE EQ 0 INT IF SCFDONE EQ 0 THEN SCF 6 PROGRAM CONTROL 33 ENDIF ULTI CI ENDPROC This procedure can be used for a calculation of a vertical ionization potential of H2O as follows R 1 ANG Set bond distance THETA 104 DEGREE Set bond angle BASIS VTZ Define basis set GEOMETRY Geometry input block O Z matrix H1 0 R H2 0 R H1 THETA ENDG End of geometry input HF RCI Compute mrci energy of water using defaults EH20 ENERGY save mrci energy in variable EH20 SET NELEC 9 Set number of electrons to 9 SET SYMMETRY 2 Set wavefunction symmetry to 2 HF MRCI Compute mrci energy of H20 2B2 state IPCI ENERGY EH20 TOEV Compute MRCI ionization potential in eV Note At present all variables are global i e variables are commonly known to all procedures and all variables defined in procedures will be
233. coordinates are constructed au tomatically and there exist exotic bond structures which might not be treated properly e g weakly bonded species as in transition state optimizations In such a case if the BMAT optimization converges slowly or leads to symmetry breaking errors you should try another optimization method and or cartesian or Z Matrix coordinates If the option NOROT is given the cartesian coordinates are not transformed to minimize rota tions 39 GEOMETRY OPTIMIZATION OPTG 258 39 2 3 Displacement coordinates DISPLACE DISPLACE displacement type see section 38 2 L for details 39 2 4 Defining active geometry parameters ACTIVE ACTIVE param Declares variable name param to be active in the optimization By default initially all variables on which the geometry depends are active inclusion of an ACTIVE card makes all parameters inactive unless explicitly declared active see also INACTIVE 39 2 5 Defining inactive geometry parameters INACTIVE INACTIVE param Declares variable name param to be inactive in the optimization If any ACTIVE card appears in the input this card is ignored see also ACTIVE 39 2 6 Hessian approximations HESSIAN By default the MOLPRO geometry optimization utilizes a force field approximation to the hes sian Model Hessian see R Lindh A Bernhardsson G Karlstr m and P Malmqvist Chem Phys Lett 241 423 1995 which speeds up convergence significantly Th
234. cord holding start orbitals SAVE ORBITAL record Dump record for orbitals MAXTT maxit Maximum number of iterations default 60 SHIFTA SHIFTC shifta Level shift for closed shell orbitals in RHF default 0 3 and amp spin orbitals in UHF default 0 SHIFTB SHIFTO shiftb Level shift for open shell orbitals in RHF and B spin orbitals in UHF default 0 17 THE SCF PROGRAM 9 NITORD nitord In open shell calculations the orbitals are reordered after each iteration to obtain maximum overlap with the orbitals from the previous iteration This takes only effect after nitord iterations The default is nitord maxit 4 if no start card is present and nitord 1 if a START card is found NITOCC nitocc Starting with iteration nitocc the occupation pattern is kept fixed The default depends on the quality of the starting guess NITCL nitcl If the iteration count is smaller than nitcl only the closed shell part of the Fock matrix is used default nitcl 0 NITORT nitort The orbitals are reorthonormalized after every nitort iterations The default is nitort 10 POTFAC potfac Scale factor for potential energy in first iteration default 1 0 17 1 2 Options for the diagonalization method In calculations with very large basis sets the diagonalization time becomes a significant fraction of the total CPU time This can be reduced using the orbital rotation method as described in R Polly H J Werner F R Manby an
235. cord on file 1 and ea f Exclusive access file ea f is permissible only if the program has been configured for MPP usage and at present molpro is implemented only for serial execution molpro is required if the integrals are to be used in a restart job For max imum efficiency on a parallel machine eaf should be used since in that case the integrals are distributed on separate processor local files For backward compatibility purposes two convenience commands are also defined COMPRESS is equivalent to AOINT COMPRESS 1 and UNCOMPRESS is equivalent to AOINT COMPRESS 0 12 GEOMETRY SPECIFICATION AND INTEGRATION 70 12 2 Symmetry specification If standard Z matrix input is used MOLPRO determines the symmetry automatically by de fault However sometimes it is necessary to use a lower symmetry or a different orientation than obtained by the default and this can be achieved by explicit specification of the symmetry elements to be used as described below On the first card of the integral input directly after the INT card or as first card in a geometry block generating symmetry elements can be given which uniquely specify the point group The dimension of the point group is 2 number of fields given Each field consists of one or more of X Y or Z with no intervening spaces which specify which coordinate axes change sign under the corresponding generating symmetry operation It is usually wise to choose z to be the unique ax
236. corz 1 the core orbitals are frozen by excluding them from the natural orbital transformation 25 EXCITED STATES WITH EQUATION OF MOTION CCSD EOM CCSD 164 25 EXCITED STATES WITH EQUATION OF MOTION CCSD EOM CCSD Excitation energies for singlet states can be computed using equation of motion EOM ap proach For the excitation energies the EOM CCSD method gives the same results as linear response CCSD LR CCSD theory Accurate results can only be expected for singly excited states The states to be computed are specified on an EOM input card which is a subcommand of CCSD The following input forms are possible EOM statel state2 state3 Computes the given states Each state is specified in the form number sym e g 5 3 means the fifth state in symmetry 3 Note that state 1 1 corresponds to the ground state CCSD wavefunction and is ignored if given EOM n1 syml n2 sym2 computes the first n1 states in symmetry sym1 n2 in sym2 etc EOM n1 syml n2 syml computes states n1 through n2 in symmetry sym The different forms can be combined e g EOM 3 1 2 2 2 3 5 3 computes states 1 3 in symmetry 1 the second excited state in symmetry 2 and the second through fifth excited states in symmetry 3 Note that state 1 1 is the ground state CCSD wave function By default an error exit will result if the CCSD did not converge and a subsequent EOM cal culation is attempted The error exit can be av
237. ction file optional gprint options global print options optional gthresh options global thresholds optional gdirect options global direct optional gexpec opnames global definition of one electron operators optional basis basisname basis specification If not present cc pVDZ is used geometry geometry specification varl value var2 value setting variables for geometry and or wavefunction definitions command options program or procedure name directive data option directives for command optional end of command block n end of input optional If the memory card is given it should be the first card after the optional title card If any file cards are given they should follow immediately The order of basis geometry gprint gdirect gthresh gexpec and variable definitions is arbitrary It is possible to call sev eral programs one after each other It is also possible to redefine basis set and or geometry between the call to programs the program will recognize automatically if the integrals have to be recomputed 4 2 Files MOLPRO uses three sequential text files namely the input file the output file and the punch file The punch file is a short form of the output which contains the most important data and results such as geometries energies dipole moments etc The punch file can be processed by the separate program READPUN which selects specific results by keywords and is able to produce o
238. ctions of Dunning are in cluded The last s and p functions are deleted and replaced by two even tempered functions with ratio 2 5 d 3 term tempered basis sets type atom EVEN3 nprim Q D y Generates a 3 parameter set of nprim functions with exponents given by 2 500 pese dejo 2 mn nprim x e Regular even tempered basis sets type atom EVENR nprim aa ap bb bp Generates an even tempered set of nprim functions according to the regular prescription de scribed in M W Schmidt and K Ruedenberg J Chem Phys 71 1970 3951 If any of the parameters aa ap bb bp is zero or omitted the values are taken from table III of the above 14 EFFECTIVE CORE POTENTIALS 84 13 7 Contracted set definitions a C first last c1 c2 cn cn L General specification of a contracted function first last defines the range of primitives to be contracted The order corresponds to the primitives as specified on the previous input card c c2 are the last first 1 contraction coefficients Continuation onto a subsequent card is permitted as shown b C Use default contractions from the library This applies to both the number of contracted primitives and also to the number of different contraction sets c nC first last n contracted functions taken from library first last defines the range of primitives to be con tracted If n is omitted and first last is specified n 1 If first last is omitted the library defau
239. d Peter J Knowles Mol Phys 102 2311 2004 MINROT minrot If minrot Q the orbital rotation method is employed Explicit diagonalization of the full Fock matrix is performed in the first minrot iterations and in the last iteration If minrot 0 a default is used which depends on the starting guess NEXPR nexpr Number of terms used in the exponential expansion of the uni tary orbital transformation matrix default 4 DEROT nexpr Energy gap used in the orbital rotation method For orbitals within derot hartree of the HOMO orbital energy the Fock matrix is constructed and diagonalized default 1 0 JACOB I jacobi If nonzero use Jacobi diagonalization 17 1 3 Options for convergence acceleration methods DIIS For more details see IPOL directive IPTYP iptyp Interpolation type default DI IS see IPOL directive IPNIT DIIS START ipnit First iteration for DIIS interpolation IPSTEP DIIS STEP ipstep Iteration increment for DIIS interpolation MAXDIS MAXDIIS maxdis Max number of Fock matrices used in DIIS interpolation de fault 10 17 1 4 Options for integral direct calculations DIRECT logical If given do integral direct HF THRMIN THRDSCF_MIN value Final integral screening threshold for DSCF THRMAX THRDSCF MAX value Initial integral screening threshold for DSCF THRINT THRDSCF value Same as THRDSCF_MIN 17 THE SCF PROGRAM PRESCREEN value DISKSIZE value BUFSIZE value THRDISK value PRI
240. d before are skipped EXTRA 2 1 1 will move all orbitals of symmetry 1 which have extra symmetry 2 Orbitals which have been moved before are skipped EXTRA will move all orbitals all symmetries and order them according to extra symmetries EXTRA 3 1 1 0 8 Will move all orbitals which have extra symmetry 3 in all symmetries Orbitals which have been moved before are skipped See also ADD and MOVE commands 42 5 Defining offsets in the output set OFFSET OFFSET iof iofo iofs Sets offsets in the output vector for symmetries 1 to 8 In subsequent MOVE or ADD commands the input vectors are moved to the locations iof in the output vectors The offset for individual ADD or MOVE commands can be modified by the parameter ioff on these cards This card should immediately follow the orbital directive to which it applies Generally this card is only needed if the dimensions of input and output vectors are not identical If the dimensions of the input orbital sets are smaller than the current basis dimension the offsets are determined automatically in the following way each time an orbital set is read in the previous input orbital dimensions are added to the offsets Hence this works correctly if the orbital sets are given in the correct order and if the individual dimensions add up to the current total dimension If this is not the case the offsets should be specified on an OFFSET card which must follow the orbital directive
241. d calculation all irreps are retained for which a non zero weight has been specified in the wavefunction definition The SYMPROJ keyword may also be used in combination with constraints 36 THE VB PROGRAM CASVB 234 36 10 7 Freezing orbitals in the optimization FIXORB j in 3 This command freezes the orbitals specified in the list ij z to that of the starting guess Alternatively the special keywords ALL or NONE may be used These orbitals are eliminated from the optimization procedure but will still be normalized and symmetry adapted according to any ORBREL keywords given 36 10 8 Freezing structure coefficients in the optimization FIXSTRUC in 3 Freezes the coefficients for structures ij i5 Alternatively the special keywords ALL or NONE may be used The structures are eliminated from the optimization procedure but may still be affected by normalization or any symmetry keywords present 36 10 9 Deleting structures from the optimization DELSTRUC ij i2 ALL NONE Deletes the specified structures from the wavefunction The special keywords ALL or NONE may be used A structure coefficient may already be zero by symmetry as defined by SYMELM and ORBREL in which case deleting it has no effect 36 10 10 Orthogonality constraints ORTHCON kKey 1 key 2 The ORTHCON keyword initiates the input of orthogonality constraints between pairs of valence bond orbitals The sub keywords key i can b
242. d can be either BOYS or PIPEK By default the valence orbitals from the last energy calculation are localized using the Boys criterion Only orbital subsets which leave the energy invariant are transformed These defaults can be modified using the optional commands described in the following sections 19 1 Defining the input orbitals ORBITAL ORBITAL record file specifications The orbitals to be localized are read from dump record record file A state specific orbital set can be selected using specifications as explained in section Default are the orbitals calculated last 19 2 Saving the localized orbitals SAVE SAVE record file This specifies the dump record where the localized orbitals are stored If the dump record already exists the localized orbitals are added to it Default is the input record cf ORBITAL 19 3 Choosing the localization method METHOD METHOD method The localization method method can be either BOYS or PIPEK This can also be specified as argument on the LOCALI card see above 19 4 Delocalization of orbitals DELOCAL DELOCAL If this card is present the orbitals are delocalized 19 5 Localizing AOs LOCAO LOCAO If this card is present the number of AOs contributing to each MO is minimized This can be useful to rotate degenerate orbitals e g px py pz in an atom so that pure orbitals in this case px py pz result This implies Pipek Mezey localization 19 ORBITAL L
243. d defined on a PSPACE card before the first WF or STATE WEIGHT SELECT PUNCSF if WF is not given card is global i e valid for all state symmetries State specific thresholds can be defined by placing a PSPACE card after the corresponding WF card In the latter case the PSPACE card can be followed by CON cards which define state specific P space configurations 20 4 5 Projection to specific A states in linear molecules Since MOLPRO can only use Abelian point groups e g C instead of C for linear molecules A states as well as X states occur in the irreducible representation number 1 for example Sometimes it is not possible to predict in advance to which state s the program will converge In such cases the LOUANT option can be used to specify which states are desired LOUANT lam 1 lam 2 lam nstate lam i is the A quantum number of state i i e O for X states 1 for II states 2 for A states etc The matrix over A will be constructed and diagonalized in the P space configuration basis The eigenvectors are used to transform the P space hamiltonian into a symmetry adapted basis and the program then selects the eigenvectors of the correct symmetry The states will be ordered by symmetry as specified on the LOUANT card within each symmetry the states will be ordered according to increasing energy 20 5 Restoring and saving the orbitals and CI vectors MULTI normally requires a starting orbital guess In this section
244. d to lower case on unix machines 4 3 Records Record names are positive integers and are usually referred to in the format record file e g 2100 2 means the record called 2100 on file 2 Note that these names are quite arbitrary and their numerical values have nothing to do with the order of the records in the file Record names lt 2000 are reserved for standard quantities e g integrals properties etc and you should never use these in an input unless you know exactly what you are doing Some important default records to remember are 2100 RHF dump record closed and open shell 2200 UHF dump record 2140 MCSCF dump record 4100 CPHF restart information 5000 MCSCF gradient information 5100 CP MCSCF gradient information 5200 MP2 gradient information 5300 Hessian restart information 5400 Frequencies restart information 6300 Domain restart information If an input contains several wavefunction calculations of the same type e g several MCSCF calculations with different active spaces the record number will be increased by 1 for each calculation of the same type Thus the results of the first SCF calculation in an input are stored in dump record 2100 2 the second SCF in record 2101 2 the first MCSCF in 2140 2 the second MCSCF in 2141 2 and so on Note that these numbers refer to the occurrence in the input and not on the order in which the calculations are performed in the actual run If an input or part of it is repeated u
245. default Using connectivity criteria for pair selection requires to set the option USE_DIST 0 USE DIST default 1 If nonzero use distance criteria otherwise connectivity criteria CHGMIN PAIRS Only atoms in the primary domains are considered for the pair classi fication if the atomic L wdin charge is larger than CHGMIN PAIRS default value 0 2 This criterion was introduced in order to reduce the dependence of the pair selection on localization tails RCLOSE default 1 Strong pairs are defined by 0 lt RUP lt RCLOSE Close pairs are defined by RCLOSE lt RD lt RWEAK RWEAK default 3 Weak pairs are defined by RWEAK lt RJ lt RDIST RDIST default 8 Distant pairs are defined by RDIST lt RJ lt RVDIST RVDIST default 15 Very distant pairs for which R gt RVDIST are neglected ICLOSE default 1 Strong pairs are separated by less that ICLOSE bonds Close orbital pairs are separated by at least ICLOSE bonds but less than IWEAK bonds IWEAK default 2 Weak orbital pairs are separated by at least IWEAK bonds but less than IDIST bonds IDIST default 5 Distant orbital pairs are separated by at least IDI ST bonds but less than IVDIST bonds IVDIST default 8 Very distant orbital pairs neglected are separated by at least IVDIST bonds 28 LOCAL CORRELATION TREATMENTS 187 KEEPCL default 0 If KEEPCL 1 the LMP2 amplitudes of close pairs are included in the computation of the strong pair LCCSD
246. details RI BASIS basis Select the basis for the resolution of the identity RI In case of R12 methods this should be chosen to be a large uncontracted AO basis at least AVQZ For F12 methods we have found that the Hartree Fock JKFIT basis sets perform well for the RI despite having been optimized for other purposes ANSATZ ansatz Select the explicitly correlated ansatz ansatz for the canonical meth ods The ansatz takes the form 2A 2 A or 2A The optional invokes additional approximations based on the extended Brillouin approximation that result in increased efficiency The optional back ward quote standing in for prime results in the inclusion of some small terms required for full orbital invariance Since the terms are cheap to compute the flexibility not to include them is provided for historical reasons Whatever ansatz is chosen all levels of theory are computed that do not entail the evaluation of additional integrals Currently only ansatz 2 A is implemented in the local version with the additional approximation that only diagonal ijij pairs are in cluded in the correlation factor GEM BASIS Basis set name for geminal expansion atom labels are ignored This can either be OPTFULL full nonlinear fit of the geminal expansion EVEN even tempered fit or refer to a set name defined in a previous BASIS block Default is OPTFULL GEM TYPE Frozen geminal type LINEAR or SLATER default is SLATER GEM NUMBER
247. dients to the given file APPEND logical If given existing SAVEACT and or SAVEGRD files are ap pended 39 2 Directives for OPTG An alternative way to specify options is to use directives as described in this section In some cases this allows more detailed specifications than with the options on the OPTG command In particular directives ACTIVE or INACTICE can be used to define the optimization space in more detail 39 2 1 Selecting the optimization method METHOD METHOD Key key defines the optimization method For minimization the following options are valid for key RF Rational Function method default AH Augmented Hessian method This is similar to RF algorithm but uses a more sophisticated step restriction algorithm DIIS Pulay s Geometry DIIS method As an an additional option you may add the number of geometries to be used in GDIIS interpolation de fault 5 and the interpolation type i e the subspace in which the GDIIS interpolation is made 39 GEOMETRY OPTIMIZATION OPTG 256 METHOD DIIS number type type may be GRAD interpolation using the gradients default work ing good for rigid molecules STEP interpolation using Quasi Newton steps which could be advantageous in dealing with very floppy molecules ENER interpolation using energies which is an intermediate between the above two OSD Quadratic steepest descent method of Sun and Ruedenberg SRMIN Old version of OSD For transition state se
248. dirty whenever any parameters are changed and whenever the geometry changes if the cache is dirty then when an attempt is made to use the grid it will be recalculated otherwise the cached values are used If status is OLD an attempt to restore the grid from a previous calculation is performed effec tively the old grid provides a template of parameters which can be adjusted using the parameter commands described below If status is NEW the grid is always created with default parameters If status is UNKNOWN the default a new grid is created either if record orb file does not exist otherwise the old grid is used The GRID command may be followed by a number of parameter modifying subcommands The currently implemented default parameters are equivalent to the following input commands GRIDTHRESH 1e 5 0 0 RADIAL LOG 3 1 0 20 25 25 30 ANGULAR LEBEDEV 0 0 0 0 LMIN 3 5 5 7 LMAX 53 53 53 53 VORONOI 10 GRIDSAVE GRIDSYM 18 3 1 Target quadrature accuracy GRIDTHRESH GRIDTHRESH acc accr acca 18 THE DENSITY FUNCTIONAL PROGRAM 102 Specify the target accuracy of integration Radial and angular grids are generated adaptively with the aim of integrating the Slater Dirac functional to the specified accuracy acc is an overall target accuracy and is the one that should normally be used radial and angular grid target accuracies are generated algorithmically from it However they can be adjusted individually by
249. ds DFTTHRESH DFTTHRESH Keyl valuel key2 value2 Sets various truncation thresholds key can be one of the following TOTAL Overall target accuracy per atom of density functional De faults to the value of the global threshold GRID or the value specified by option GRID For proper use of this threshold other thresholds should be left at their default value of zero ORBITAL Orbital truncation threshold DENSITY Density truncation threshold FOCK Fock matrix truncation threshold 18 2 3 Exact exchange computation EXCHANGE EXCHANGE factor For Kohn Sham calculations compute exchange energy according to Hartree Fock formalism and add the contribution scaled by factor to the fock matrix and the energy functional Other wise the default is factor 0 1 e the exchange is assumed to be contained in the functional and only the Coulomb interaction is calculated explicitly DFTFACTOR facl fac2 Provide a factor for each functional specified The functionals will be combined accordingly By default all factors are one 18 2 4 Exchange correlation potential POTENTIAL POTENTIAL rec fil For stand alone DFT calculations compute exchange correlation potential pseudo matrix ele ments defined formally as the differential of the sum of all specified functionals with respect to elements of the atomic orbital density matrix The matrix is written to record rec on file fil 18 THE DENSITY FUNCTIONAL PROGRAM 101 18
250. ds not to be specified Specifies a particular state in the form istate isym For instance 2 1 refers to the second state in symmetry 1 This can be used if density matrices or natural orbitals have been computed for different states in a state averaged CASSCF calculation If not given the state averaged orbitals are used The specification of isym is optional it can also be defined using the SYMMETRY key Density type This can be one of CHARGE charge density SPIN UHF spin density TRANSITION transition density matrix The default is CHARGE Specifies a particular state symmetry Alternatively the state symme try can be specified using STATE see above Spin quantum number i e 0 for singlet 1 2 for doublet 1 for triplet etc Alternatively MS2 can be used 2Ms i e O for singlet 1 for doublet 2 for triplet etc Alternatively SPIN can be used Number of electrons Set number of orbitals The orbital sets are numbered in the order they are stored If OVL is specified the starting orbitals are obtained by maximizing the overlap with previous orbitals By default this is used if the basis dimension of the previous orbitals is different then the current one If OVL is specified this procedure is used even if the basis dimensions are the same which is occasionally useful if the contraction scheme changed If NOCHECK is specified some consistency checks for finding correct orbitals are skipped and error messages l
251. dz R 2 0 RO R Theta 100 geomet ry O Hl O R H2 0 R H1 THETA hf accu 12 multi closed 2 rs2 shift 0 3 ignoreshift lignore shift in computing gradient i e no cp caspt2 optg gradient 1 d 5 e_opt 1 energy r_opt 1 r theta_opt 1 theta method 1 rs2 analytical ignore examples rs2 shift 0 3 lexact gradient with shift h20 caspt2 opt com optg gradient 1 d 5 e opt 2 energy r opt 2 2r theta opt 2 theta method 2 2 rs2 analytical exact rs2 shift 0 3 numerical gradient with shift optg gradient 1 d 5 numerical fourpoint use four point numerical gradient e opt 3 energy r_opt 3 r theta_opt 3 theta method 3 rs2 numerical rs2c shift 0 3 numerical gradient of rs2c with shift optg gradient 1 d 5 fourpoint use four point numerical gradient e_opt 4 energy r_opt 4 r theta_opt 4 theta method 4 rs2c numerical table method r_opt theta_opt e_opt digits 4 4 8 This produces the Table METHOD R OPT THETA OPT E OPT rs2 analytical ignore 1 8250 102 1069 76 22789382 rs2 analytical exact 1 8261 102 1168 76 22789441 rs2 numerical 1 8261 102 1168 76 22789441 rs2c numerical 1 8260 102 1187 76 22787681 MS CASPT2 geometry optimization for the second excited B state if HO 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 155 Revision 2006 1 Xx kk memory 8 m gthresh energy 1 d 12 basis vdz R 2 0 RO R Theta 100 step 0 00
252. e or a card All input following the card is ignored Alternatively a job can be stopped at at some place by inserting an EXIT card This could also be in the middle of a DO loop or an IF block If in such a case the card would be used an error would result since the ENDDO or ENDIF cards would not be found 6 3 Restarting a job RESTART In contrast to MOLPRO92 and older versions the current version of MOLPRO attempts to recover all information from all permanent files by default If a restart is unwanted the NEW option can be used on the FILE directive The RESTART directive as described below can still be used as in MOLPRO92 but is usually not needed RESTART 71 72 13 T4 5 The r specify which files are restarted These files must have been allocated before using FILE cards There are two possible formats for the r a 0 rj 10 Restart file r and restore all information b r name nr Restart file nr but truncate before record name If all r 0 then all permanent files are restarted However if at least one r is not equal to zero only the specified files are restarted Examples RESTART will restart all permanent files allocated with FILE cards default RESTART 1 will restart file 1 only RESTART 2 will restart file 2 only RESTART 1 2 3 will restart files 1 3 RESTART 2000 1 will restart file 1 and truncate before record 2000 6 PROGRAM CONTROL 29 6 4 In
253. e the data is read from standard input and program results go to standard output Otherwise data is taken from datafile and the output is written to a file whose name is generated from datafile by removing any trailing suffix and appending out If the output file already exists then the old file is appended to the same name with suffix out 1 and then deleted This pro vides a mechanism for saving old output files from overwriting Note that the above behaviour can be modified with the o or s options Unless disabled by options the user data file is prepended by one or more default procedure files if these files exist These are in order of execution the file molproi rc in the system direc tory containing the molpro command itself SHOME molproirc and molproi rc 2 0 1 Options Most options are not required since sensible system defaults are usually set Options as detailed below may be given in order of decreasing priority on the command line in the environment variable MOLPRO or in the files molpro rc SHOME molprorc andmolpro rcin the system directory o output outfile specifies a different output file x executable executable specifies an alternative MOLPRO executable file d directory directory directory2 specifies a list of directories in which the pro gram will place scratch files For detailed discussion of optimal specification see the installation guide s nobackup disables the mechanism
254. e CASSCF wavefunction not generally required c Definition of the valence bond wavefunction d Recovery and or storage of orbitals and vectors e Manual input of starting guess optional g Optimization control f Definition of molecular symmetry and possible constraints on the VB wavefunction h Wavefunction analysis 1 Further general options Items a and b should precede everything else in the input apart from this commands may come in any order 36 THE VB PROGRAM CASVB 228 36 2 Defining the CASSCF wavefunction CASVB is interfaced with the determinant part of MULTI i e CONFIG CSF must not be specified When this program is run prior to CASVB the CI vector must dumped using one of the directives SAVE NATORB CANONICAL or LOCALI see section 20 5 4 The three latter are recommended 36 2 1 The VBDUMP directive VBDUMP vbdump If present the VBDUMP card must occur first in the CASVB input It is not required for variational calculations Note that in the majority of cases e g if a CASVB run occurs immediately after MULTI or for variational calculations explicit specification of dump records with vbdump is not required Wavefunction definitions may be restored here using VBDUMP cards see also Section 20 8 6 The default record name vbdump is 4299 2 If a VBDUMP card is not present and record 4299 2 does not exist then CASVB will attempt to generate the wavefunction information automatical
255. e HCN HNC isomerization 272 39 4 6 Reaction path of the HCN HNC isomerization 274 z NN 275 40 VIBRATIONAL FREQUENCIES FREQUENCIES 280 40 1 Numerical hessian using energy variables VARIABLE 281 40 2 Thermodynamical properties THERMO o 281 40 3 Examples 5 moro omo 4a ea Ba bE dade e a E 282 CONTENTS xix 284 II 285 287 42 1 Defining the input orbitals ORBITAL o 287 42 2 Moving orbitals to the output set MOVE o o 287 42 3 Adding orbitals to the output set ADD o o 287 42 4 Defining extra symmetries EXTRA es 288 42 5 Defining offsets in the output set OFFSET llle 288 42 6 Projecting orbitals PROJECT es 288 42 7 Symmetric orthonormalization ORTH llle 289 42 8 Schmidt orthonormalization SCHMIDT 22e 289 42 9 Rotating orbitals ROTATE 2r 289 42 10Initialization of a new output set INIT ens 289 EA AAN ee ae ee ee ee 289 42 12Printing options PRINT aoaaa aaa 289 UMS Ee Re 6 Rowe Gd Ue XO UR x A e ee d 290 42 T9 E 290 ook uude ee ee ark ee uH EXE 290 43 MATRIX OPERATIONS 293 43 1 Calling the matrix facility MATROP ee 293 43 2 Loading matrices LOAD o o e e 294 Bond a eee BSS a RUE o 294 votes a dut a rl 294 A Bee 294 ds te helada e este 294 a a det 294 apra 295 gea bes cab renos 295
256. e Model Hessian is parameterized for the elements up to the third row Alternatively the model Hessian of Schlegel can be used or the Hessian can be computed numerically see also section 39 2 7 HESSIAN options where options can be MODEL Use Lindh s Model Hessian in optimization default MODEL SCHLEGEL Use Schlegel s Model Hessian MODEL VDW Add vdW terms to Lindh s Model Hessian SCHLEGEL same as MODEL SCHLEGEL VDW same as MODEL VDW NOMODEL Don t use Model Hessian approximation to the hessian NUMERICAL hstep Recompute Hessian after hstep iterations This disables the use of a model hessian If hstep 0 the Hessian is only computed in the first iteration Default parameters are used for computing the numerical Hessian unless modified using options as described for the NUMHESS directive see Sect Any option valid for the NUMHESS direc tive may also follow the NUMERICAL option on the HESSIAN direc tive READ RECORD HESSREC record Read Hessian from given record If record is not given or zero the last computed hessian will be read See section 39 2 7 for more details about numerical Hessians 39 GEOMETRY OPTIMIZATION OPTG 259 UPDATE type Method used for hessian update See section 39 2 9 for possibilities and details MAXUPD maxupd Max number of hessian updates The count is reset to zero each time a hessian is computed If the Model Hessian is disabled NOMODEL and no Hessian is read or compu
257. e allowed on procedure calls However specific options may be set using directives within the procedure and these are then valid for all programs within the procedure which follow the directive When execution of the procedure is finished the previous global options are restored The hierarchy in which options are processed is as follows Global options Options in procedures Command line options Options given on directives within a command block 4 GENERAL PROGRAM STRUCTURE 12 The last option set is then actually used Thus options specified on command lines or within command blocks have preference over procedure options and procedure options have prefer ence over global options 4 GENERAL PROGRAM STRUCTURE This chapter gives an overview of the most important features of MOLPRO For the new user it is essential to understand the strategies and conventions described in this section in particular the meaning of files and records and the use of symmetry This chapter will focus on general aspects detailed information about each command will be given in later chapters Information about commands and parameters can also be obtained using the MOLPRO help facility see section 4 13 4 1 Input structure A typical MOLPRO input has the following structure title title optional memory 4 m memory specification optional file 1 name int permanent named integral file optional file 2 name wfu permanent named wavefun
258. e at a ee ae Se eae 239 37 3 Calculation of individual SO matrix elements 239 pM 240 ada a eae ld EC E d 240 ee rr ee 241 RR EUR OR XMEUR E ACE ede fene s 241 c r 241 oir 241 ob EEE 242 38 ENERGY GRADIENTS 244 PEUT AA 244 38 1 1 Adding gradients ADD o o e e 244 38 1 2 Scaling gradients SCALE o o e e 244 38 1 3 Defining the orbitals for SCF gradients ORBITAL 245 38 1 4 MCSCF gradients MCSCF o e ee 245 SOTTO 245 38 1 6 State averaged MCSCF gradients with CADPAC 245 38 1 7 Non adiabatic coupling matrix elements NACM 246 38 1 8 Difference gradients for SA MCSCF DEMC 246 amp owoksu ede ket dr dedii lop ake BEE deb d ARE ede 246 DAMA ROD A 247 38 2 1 Choice of coordinates COORD 248 38 2 2 Numerical derivatives of a variable 249 CONTENTS xviii T 249 38 2 4 Active and inactive coordinates o 249 e a e BM 249 39 GEOMETRY OPTIMIZATION OPTG 251 dad dla a a ot eed gh a ld a EA oh es Bees 251 39 1 1 Options to select the wavefunction and energy to be optimized 251 CMT 252 prieta 252 Sor er 253 sso easel aes ee 253 LEM done at ae DE 253 a Fone ahs ane a ate ai aoe os 254 39 1 8 Miscellaneous options lees 255 pco AA dee T 255 39 2 1 Selecting the optimization method METHOD 255 39 2 2 Optimiz
259. e atomic masses rather than those of the most common isotopes which is now the default behaviour MCSCF second derivatives author Riccardo Tarroni added preliminary version only without symmetry Frequency and geometry optimization programs are modified so that they can use the analytic Hessian New internally contracted multi reference second order perturbation theory code author Paolo Celani through command RS2C as described in P Celani and H J Werner J Chem Phys 112 5546 2000 EOM CCSD for excited states author Tatiana Korona QCISD dipole moments as true analytical energy derivatives author Guntram Rauhut Linear scaling CPU and memory LMP2 as described by G Hetzer P Pulay and H J Werner Chem Phys Lett 290 143 1998 M Sch tz G Hetzer and H J Werner J Chem Phys 111 5691 1999 B RECENT CHANGES 319 9 Improved handling of basis and geometry records 98 1 and 99 1 dump files can be restarted but in case of problems with restarting old files add RESTART NOGEOM im mediately after the ile card Also if there are unjustified messages coming up in very large cases about ORBITALS CORRESPOND TO DIFFERENT GEOMETRY try ORBITAL record NOCHECK This can happen for cases with more than 100 atoms since the old version was limited to 100 10 Reorganization and generalization of basis input Increased basis library 11 Counterpoise geometry optimizations 12 Improved
260. e bond calculations written by T Thorsteinsson and D L Cooper 1996 2005 This program can be used in two basic modes a variational optimization of quite general types of nonorthogonal MCSCF or modern va lence bond wavefunctions b representation of CASSCF wavefunctions in modern valence form using overlap rela tively inexpensive or energy based criteria Bibliography T Thorsteinsson D L Cooper J Gerratt P B Karadakov and M Raimondi Theor Chim Acta 93 343 66 1996 D L Cooper T Thorsteinsson and J Gerratt Int J Quant Chem 65 439 51 1997 D L Cooper T Thorsteinsson and J Gerratt Adv Quant Chem 32 51 67 1998 T Thorsteinsson and D L Cooper in Quantum Systems in Chemistry and Physics Volume 1 Basic problems and models systems eds A Hern ndez Laguna J Maruani R McWeeny and S Wilson Kluwer Dordrecht 2000 pp 303 26 All publications resulting from use of this program should acknowledge relevant publications There is a more complete bibliography at http www liv ac uk dlc CASV B html 36 1 Structure of the input All CASVB sub commands may be abbreviated by four letters The general input structure can be summarized as follows a For generating representations of CASSCF wavefunctions the program is invoked by the command CASVB For variational optimization of wavefunctions it is normally invoked inside MULTI by the sub command VB see 20 10 b Definition of th
261. e bond wavefunctions VB Using this keyword the optimization of the CI coefficients is carried out by CASVB The VB keyword can be followed by any of the directives described in section Energy based opti mization of the VB parameters is the default and the output level for the main CASVB iterations is reduced to 1 20 11 Hints and strategies MCSCF is not a black box procedure like SCF For simple cases for example a simple CASSCF with no CLOSED orbitals this program will converge in two or three iterations For more complicated cases you may have more trouble In that case consider the following e Always start from neighbouring geometry orbitals when available this is the default e The convergence algorithm is more stable when there are no CLOSED orbitals 1 e or bitals doubly occupied in all configurations but fully optimized Thus a reasonable ap proach is to make an initial calculation with CLOSED replaced by FROZEN all doubly occ frozen e If still no success you can switch off the coupling between CI coefficients and orbital rotations for a few iterations e g ITERATIONS UNCOUPLE 1 T0 2 and or disable the simultaneous optimization of internal orbitals amp CI e g ITERATIONS DONT INTERNAL 1 TO 2 You can often get a clue about where the program starts to diverge if you include IPRINT MICRO in the data Also consider the general remarks at the beginning of
262. e input is exactly the same as for closed shell CCSD except that RCCSD or UCCSD are used as keywords By default the open shell orbitals are the same as used in the RHF reference function but this can be modified using OCC CLOSED and WF cards Perturbative triples corrections are computed as follows RCCSD T UCCSD T triples corrections are computed as defined by J D Watts J Gauss and R J Bartlett J Chem Phys 98 8718 1993 RCCSD T UCCSD T corrections are computed without contributions of single excita tions sometimes called CCSD T CCSD RCCSD T UCCSD T triples corrections are computed as defined by M J O Deegan and P J Knowles Chem Phys Letters 227 1994 321 In fact all three contributions are always computed and printed The following variables are used to store the results here CCSD stands for either UCCSD or RCCSD ENERGY total energy for method specified in the input ENERGC total CCSD energy without triples ENERGT 1 total CCSD T energy ENERGT 2 total CCSD T energy ENERGT 3 total CCSD T energy It should be noted that in open shell cases the triples energy slightly depends on the treatment of core orbitals In MOLPRO pseudo canonical alpha and beta spin orbitals http dx doi org 10 1016 S0009 2614 91 85118 G are generated by block diagonalizing the corresponding Fock matrices in the space of valence orbitals leaving frozen core orbitals untouched Some other progra
263. e not available for every platform Execution of MOLPRO whether a supplied binary or built from source requires a valid licence key Note that the key consists of two components namely a list of comma separated key value pairs and a password string and these are separated by amp In most cases the licence key will be automatically downloaded from the website when building or installing the software A 2 Installation of pre built binaries Binaries are given as RPM see http www rpm org packages which are installed in the stan dard way There are RPMs tuned for the Pentium III Pentium 4 and Athlon architectures These also support parallel execution There is a generic serial rpm which should run on all IA32 ar chitectures You can install using the command rpm Uhv molpro mpp 2006 1 0 p4 rpm where the filename of the rpm has the format molpro mpp serial 2006 1 PATCHLEVEL ARCH rpm where PATCHLEVEL is a number denoting the patchlevel of the rpm and ARCH denotes the architecture At present these RPMs are not relocatable and will install under usr local If a licence key is set in the MOLPRO_KEY environment variable or the rpm finds a licence key which has been cached in HOME molpro token from a previous install then that key will be installed with the software If the rpm cannot find a key or automatically down load it from the molpro website then the user will be prompted to run the post install script usr local bin molp
264. e one of ORTH PAIRS GROUP STRONG or FULL as described below Orthogonality constraints should be used with discretion Note that or thogonality constraints for an orbital generated from another by symmetry operations using the ORBREL keyword cannot in general be satisfied ORTH ij i2 Specifies a list of orbitals to be orthogonalized All overlaps between pairs of orbitals in the list are set to zero PAIRS i i Specifies a simple list of orthogonalization pairs Orbital i is made orthogonal to i 3 to i4 etc GROUP abel i i Defines an orbital group to be used with the ORTH or PAIRS keyword The group is referred to by label which can be any three characters beginning with a letter a z Labels defining dif ferent groups can be used together or in combination with orbital numbers in ORTH or PAIRS ij i specifies the list of orbitals in the group Thus the combination GROUP AZZ 1 2 GROUP BZZ 3 4 ORTH AZZ BZZ will orthogonalize the pairs of orbitals 1 3 1 4 2 3 and 2 4 STRONG 36 THE VB PROGRAM CASVB 235 This keyword is short hand for strong orthogonality The only allowed non zero overlaps are between pairs of orbitals 2n 1 2n FULL This keyword is short hand for full orthogonality This is mainly likely to be useful for testing purposes 36 11 Wavefunction analysis 36 11 1 Spin correlation analysis NO SCORR With this option expectation values of the spin operators
265. e precomputed and stored on disk using the command LSINT LX LY1L Z X Y and Z specify the components to be computed If none of these is given all three are evaluated The advantage of precomputing the integrals is that they can then be used in any number of subsequent SO calculations but this may require a large amount of disk space note that there are 6 times as many integrals as in an energy calculation If the LSINT card is not given the integrals are recomputed for one component at a time whenever needed and destroyed at the end of the SO calculation This reduces the disk space by a factor of 3 but may be expensive in terms of CPU if several SO calculations e g for MCSCF and MRCI wavefunctions are carried out The input for spin orbit ECPs is described in section 14 Of course in ECP LS calculations the LSINT card is not needed 37 3 Calculation of individual SO matrix elements Individual spin orbit matrix elements can be computed within the MRCI program using TRANLS recordl file record2 file bra2ms ket2ms lsop where recordl file Record holding the bra wavefunction record2 file Record holding the ket wavefunction Both records must have been generated using the SAVE directive of the MRCI program 37 SPIN ORBIT COUPLING 240 bra2ms 2 x Ms value of the bra wavefunction ket2ms 2 x Ms value of the ket wavefunction Isop Cartesian component of the Spin orbit Hamiltonian This can be one of LSX LSY or LSZ in
266. e sum of the density matrix element changes The default is accu 10 17 THE SCF PROGRAM 98 17 9 4 Print options ORBPRINT print test This determines the number of virtual orbitals printed at the end of the calculation By default print 0 1 e only the occupied orbitals are printed print 1 suppresses printing of orbitals entirely test 1 has the additional effect of printing the orbitals after each iteration 17 9 5 Interpolation IPOL ptyp ipnit ipstep maxdis This command controls DIIS interpolation iptyp can be DIIS direct inversion of the iterative subspace This is the default and yields mostly fastest convergence DM obsolete No effect in MOLPRO98 HFM obsolete No effect in MOLPRO98 NONE No interpolation ipnit is the number of the iteration in which the interpolation starts ipstep is the iteration in crement between interpolations maxdis is the maximum dimension of the DIIS matrix default 10 17 9 6 Reorthonormalization of the orbitals ORTH nitort The orbitals are reorthonormalized after every nitort iterations The default is nitort 10 17 9 7 Direct SCF DIRECT options If this card is present the calculation is done in direct mode See section 10 for options Nor mally it is recommended to use the global GDI RECT command to request the direct mode See section IO for details 18 THE DENSITY FUNCTIONAL PROGRAM 99 18 THE DENSITY FUNCTIONAL PROGRAM Density functional theory calc
267. e the energy is differ entiated twice Using numerical differentiation the dipole derivatives and the IR intensities are also calculated Note that numerical hessians cannot be computed when dummy atoms holding basis functions are present The accuracy of the hessian is determined by method which can be one of the following ANALYTICAL use analytical second derivatives of the energy At present analyti cal second derivatives are only possible for closed shell Hartree Fock HF and MCSCF wavefunctions without symmetry It is not yet pos sible to calculate IR intensities analytically Note that due to techni cal reasons the analytical MCSCF second derivatives have to be com puted in the MCSCF program using e g multi cpmcscf hess see MULTI before they can be used in FREQUENCIES If analyt ical MCSCF second derivatives are available FREQUENCIES will use them by default CENTRAL use central differences high quality force constants default NUMERICAL differentiate the energy twice using central differences FORWARD use forward differences low quality force constants During the numerical calculation of the hessian the symmetry of the molecule may be lowered Giving SYMM AUTO the program uses the maximum possible symmetry of the molecular wave function in each energy gradient calculation and this option therefore minimizes the computa tional effort With SYMM NO no symmetry is used during the frequency calculation default For si
268. e transformation matrix HDIA The first nstate nstate 1 2 elements contain the lower triangle of the diabatic hamiltonian MIXANG Non adiabatic mixing angle in degree This is available only in the two state case The corresponding results obtained from the CI vectors only without orbital correction are stored in the variables SMATCI UMATCI HDIACI and MIXANGCI The way it works is most easily demonstrated for some examples In the following input the wavefunction is first computed at the C2 reference geometry and then at displaced geometries 35 QUASI DIABATIZATION SRevision 2006 0 h2s Diabatization memory 3 m gprint orbitals civector geometry x noorient S nb 9 fL h2 s r2 h1 theta basis avdz r1 2 5 theta 92 r 2 50 2 55 2 60 reforb 2140 2 refci 6000 2 savci 6100 2 text compute wavefunction at referenc r2 r1 ihfiroco 90 2 wf l8 2 44 orbital 2100 2 imulti o Gc 9 2 closed 4 1 wf 18 2 state 2 natorb reforb noextra LciroOGGC 9 2sc6Llosed 4 15 wf 18 2 0 state 2 orbital reforb save refci Text Displaced geometries do i 1 r data truncate savcitl r2 r i multi occ 9 2 closed 4 1 wf 18 2 0 state 2 start reforb orbital 3140 2 diab reforb noextra citocc9 2 closed 4 1 wf 18 2 0 state 2 orbital diabatic save savci el i energy 1 e2 i energy 2 ci trans savci savci dm 7000 2 ci trans savci refci dm 7100 2 ddr
269. e valid only locally for the current program However if they are given outside a command block they act globally and are used for all programs executed after the input has been encountered Local options have preference over global options The following directives can be either local or global Wavefunction definition OCC CORE CLOSED FROZEN WF Thresholds and options LOCAL DFIT DIRECT EXPLICIT THRESH PRINT GRID If such options are given outside a command block a context can be specifified DIRECTIVE data CONTEXT context e g OCC 3 1 1 CONTEXT HF OCC 4 1 2 CONTEXT MCSCF CONTEXT can be any valid command name or any valid alias for this but internally these are converted to one of the following HF Hartree Fock and DFT MC MCSCF and CASSCF CC single reference correlation methods as implemented in the CCSD program CI multiref erence correlation methods as implemented in the MRCI program The directive will then be applied to one of the four cases Several contexts can be specified separated by colon e g CONTEXT HF CCSD If only a single context is given no colon shortcuts for the specifying the CONTEXT option are obtained by postfixing context to the command name e g OCC HF 3 1 1 OCC_MCSCF 4 1 2 If no context is given the default is HF The default occupations for single reference methods e g MP2 CCSD are the ones used in HF the defaults for multireference methods e g RS2 MRC
270. e wavefunction If nmin and nmax are negative configurations with exactly abs nmin and abs nmax electrons in the specified orbitals are deleted This can be used for instance to omit singly excited configurations The orbitals are specified in the form number sym where number is the number of the orbital in irrep sym Several RESTRICT cards may follow each other RESTRICT only works if a CONFIG card is specified before the first WF card RESTRICT cards given before the first WF cards are global i e are active for all state symme tries If such a global restrict card is given variable MC RESTRICT is not used Additional state specific RESTRICT cards may be given after a WF card These are used in addition to the global orbital restrictions If neither state specific nor global RESTRICT cards are found the values from the variable MC RESTRICT are used 20 THE MCSCF PROGRAM MULTI 116 20 4 2 Selecting configurations SELECT ref1 ref2 refthr refstat mxshrf This card is used to specify a configuration set other than a CAS which is the default This option automatically triggers the CONFIG option which selects CSFs rather than determinants Configurations can be defined using CON cards which must follow immediately the SELECT card Alternatively if ref is an existing MOLPRO record name the configurations are read in from that record and may be selected according to a given threshold ref recl file rec17 2000 The configu
271. ectivity criteria for domain extensions IEXT IEXTS IEXIC IEXTW Parameters to select pair classes USE DIST RCLOSE RWEAK RDIST RVDIST ICLOSE IWEAK IDIST IVDIST CHGMIN PAIRS KEEPCL CLOSEP WEAKP DISTP VERYD CHGMINP oooo 0 0 0 0 0 UY ha criterion for all pair domains criterion for strong pair domains criterion for strong and close pair domains criterion for strong close and weak pair domains criterion for all pair domains criterion for strong pair domains criterion for strong and close pair domains criterion for strong close and weak pair domains determines if distance of connectivity criteria are used distance criterion for selection of weak pairs distance criterion for selection of weak pairs distance criterion for selection of distant pairs distance criterion for selection of very distant pairs connectivity criterion for selection of weak pairs connectivity criterion for selection of weak pairs connectivity criterion for selection of distant pairs connectivity criterion for selection of very distant pairs determines minimum charge of atoms used for pair classification determines if close pairs are included in LCCSD 28 LOCAL CORRELATION TREATMENTS 180 Parameter Alias Default value Meaning Parameter for multipole treatment of exchange operators DSTMLT 3 multipole expansion level for distant pairs Parameters for energy partitioning I
272. ectric field gradients DMS 1 DMSXX DMSYX DMSZX DMSXY DMSYY DMSZY DMSXZ DMSYZ DMSZZ diamagnetic shielding tensor LOP 1 LX LY LZ Angular momentum operators Le Lj La LOP2 1 LXLX LYLY LZLZ one electron parts of products of LXLY LXLZ LYLZ angular momentum operators The symmetric combinations HL HE etc are computed VELO 1 D DX D DY D DZ velocity LS 1 LSX LSY LSZ spin orbit operators ECPLS 1 ECPLSX ECPLSY ECPLSZ ECP spin orbit operators 7 FILE HANDLING 39 7 FILE HANDLING 7 1 FILE The FILE directive is used to open permanent files which can be used for later restarts The syntax in MOLPRO94 and later versions is FILE file name status file is the logical MOLPRO file number 1 9 name is the file name will be converted to lower case status can be one of the following UNKNOWN A permanent file is opened If it exists it is automatically restarted This is the default OLD Same effect as UNKNOWN No error occurs if the file does not exist NEW A permanent file is opened If it already exists it is erased and not restarted ERASE Same effect as NEW SCRATCH A temporary file is opened If it already exists it is erased and not restarted After the job has finished the file is no longer existent DELETE Same effect as SCRATCH Note that RESTART is now the default for all permanent files All temporary files are usually allocated automatically where needed I O buffers are allocated at the t
273. ed This variant is invoked using the CIPT2 directive h20 mscaspt2 opt cor 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 156 CIPT2 In this case all excitations solely from active orbitals are treated by MRCI while the remaining excitations involving inactive closed shell orbitals are treated by second order perturbation theory Both methods are coupled by minimizing an appropriate energy functional Of course this method is much more expensive that MRPT2 The cost is comparable to the cost for an MRCI without correlating the inactive orbitals 22 9 Further options for CASPT2 and CASPT3 Other options can be set using the OPTION command These options are mainly used for testing purposes and should be used with care It should be noted that the only option that can be modified in the RS2C program is IFDIA all others only work with RS2 RS3 OPTION codel value code2 value Of relevance for the CASPT2 3 program are the following options IPROCS 0 Default Calculation uses uncontracted singles with RS2 IPROCS 1 Non interacting singles are projected out during update This is an approximate procedure which should be used with care IPROCS 2 The singles are fully internally contracted in RS2 This is achieved via a projection operator during the coefficient update and may be inefficient G IPROCS 3 Only singles with one or two holes in the closed shells are in ternally contracted in RS2 using a projection
274. ed directly in the SCF and MCSCF programs but in this case no orbital contributions are printed SRevision 2006 0 h20 properties geometry 0o hl o r h2 0 r hl theta r l ang theta 104 gexpec dm qm hf multi state 2 natorb state 1 1 natorb state 2 1 Z matrix geometry input bond length bond angle Expos global request of dipole and quadrupole DD Sel com do scf calculation do full valence CASSCF compute natural orbitals for state 1 1 compute natural orbitals for state 2 1 32 PROPERTIES AND EXPECTATION VALUES 208 32 2 Distributed multipole analysis Any density matrix can be analysed using the distributed multipole analysis described by Stone Chem Phys Letters 1981 83 233 The multipole moments arising from the overlap of each pair of primitives are calculated with respect to the overlap centre and then shifted to the nearest of a number of multipole sites By default these comprise all atoms specified in the integral input However the list of multipole sites can be modified by deleting and or adding sites and also by restricting the rank of multipole which may be transferred to any given site The atomic charges are stored in the MOLPRO variable ATCHARGE The i th element in ATCHARGE corresponds to the i th row of the Z matrix input Options may appear in any order except DENSITY which must be first if given The present version does not allow generally contracted AO basis sets 32 2 1 Ca
275. ed in Theor Chim Acta 78 1990 175 are also available Electronically excited states can be computed as described in Theor Chim Acta 84 95 1992 Multireference second order and third order perturbation theory MR PT2 MR PT3 as described in Mol Phys 89 645 1996 and J Chem Phys 112 5546 2000 Mgller Plesset perturbation theory MPPT Coupled Cluster CCSD Quadratic config uration interaction QCISD and Brueckner Coupled Cluster BCCD for closed shell systems as described in Chem Phys Lett 190 1992 1 Perturbative corrections for triple excitations can also be calculated Chem Phys Letters 227 1994 321 ii Open shell coupled cluster theories as described in J Chem Phys 99 1993 5219 Chem Phys Letters 227 1994 321 Full Configuration Interaction This is the determinant based benchmarking program de scribed in Comp Phys Commun 54 1989 75 Analytical energy gradients for SCF DFT state averaged MCSCF CASSCF MRPT2 CASPT2 MP2 and QCISD T methods Analytical non adiabatic coupling matrix elements for MCSCF Valence Bond analysis of CASSCF wavefunction and energy optimized valence bond wavefunctions as described in Int J Quant Chem 65 439 1997 One electron transition properties for MCSCF MRCI and EOM CCSD wavefunctions CASSCF and MRCI transition properties also between wavefunctions with different or bitals Spin orbit coupling as described in Mol Phys 98 1823 2000
276. efined by 2 atoms ANGLE Bond angle defined by 3 atoms angle 1 2 3 DIHEDRAL Dihedral angle defined by 4 atoms angle between the planes formed by atoms 1 2 3 and 2 344 respectively OUTOFPLANE Out of plane angle defined by 4 atoms angle between the plane formed by atoms 2 3 4 and the bond 1 4 DISSOC A dissociation coordinate defined by two groups of atoms CARTESIAN Cartesian coordinates of an atom For all types except DI SSOC and CARTESIAN atoms are given as ATOMS a1 a2 a3 where the number of atoms required varies with type as specified above and the atomic names al a2 a3 can be either atomic tag names from the Z matrix input or integers corresponding to Z matrix rows Note that the square brackets are required here and do not indicate optional input For DISSOC the specification is as follows DISSOC GROUP 1 a1 a2 GROUP 2 b1 b2 The corresponding internal coordinate is the distance between the centres of mass of the two groups For CARTESIAN the definition is CARTESIAN Z atom where J can be one of X Y Z or 1 2 3 and atom can be a z matrix atom name or an integer referring to the z matrix row With this definition the constraints are defined as CONSTRAINT VALUE value unit F AC TOR fac prim F AC TOR fac prim where value is the value imposed to the constraint and prim is either the name of the primitive defined before this constraint or an explicit definition and
277. elative to reference geometry Dont use extra symmetries Use orbitals for j 1 as reference for j 2 3 Use diabatic orbitals Save MRCI for displaced geometries examples h2s diab2 com Save adiabatic energies for use in ddr Save adiabatic energies for table printing Compute transition densities at R2 DR j Save transition densities on this record Compute transition densities between R2 DR j this racord and R1 IQ amp Qawa transition dancitiec on 35 QUASI DIABATIZATION 226 The calculation produces the following table Mixing angles and non adiabatic coupling matrix elements for H2S R MIXCI MIXTOT DCHI NACMECI 24 95 15 2694 15 2644 9 2226 5 2365 2 60 27 8740 27 8772 3 4702 3 4794 Diabatic energies for H2S obtained from CI vectors R El E2 H11CI H22CI H21CI 2 55 398 64572746 398 63666636 398 64509901 398 63729481 0 00230207 2 60 398 64911752 398 63771802 398 64662578 398 64020976 0 00471125 Diabatic energies for H2S obtained from CI vectors and orbital correction R El E2 H11 H22 H21 2 55 398 64572746 398 63666636 398 64509941 398 63729441 0 00230139 2 60 398 64911752 398 63771802 398 64662526 398 64021027 0 00471160 As expected the coupling matrix elements obtained from the 3 point DDR calculation NACMECI and by differentiating the mixing angle DCHI are in close agreement 36 THE VB PROGRAM CASVB 221 36 THE VB PROGRAM CASVB CASVB is a general program for valenc
278. elements Under certain conditions it may happen that biorthogo nalization is not possible and then an error message will be printed 21 3 15 Saving the density matrix DM record ifil idip 21 THE CI PROGRAM 140 The first order density matrices for all computed states are stored in record record on file ifil If idip is not zero the dipole moments are printed starting at iteration idip See also NATORB In case of transition moment calculation the transition densities are also stored provided both states involved have the same symmetry 21 3 16 Natural orbitals NATORB RECORD record ifil PRINT nprint CORE 2natcor Calculate natural orbitals The number of printed external orbitals in any given symmetry is nprint default 2 nprint 1 suppressed the printing If record is nonzero the natural orbitals and density matrices for all states are saved in a dump record record on file ifil If record ifil is specified on a DM card see above this record is used If different records are specified on the DM and NATORB cards an error will result The record can also be given on the SAVE card If CORE is specified core orbitals are not printed Note The dump record must not be the same as savecp or saveco on the SAVE card or the record given on the PROJECT 21 3 17 Miscellaneous options OPTION codel value code2 value Can be used to specify program parameters and options If no codes and values are specified active
279. elf consistent procedure by Hamprecht et al This functional needs to be mixed with 0 21 exact exchange See reference for more de tails K e pa pg Pa 0 E Pg 0 Ao Ain d i 42 n d 21 Y e ps 0 Bo Bin Xs 7 A2 Bo n x 22 84 3 8 34 Y ps Co Cin 2 3 C n xs 23 where d V2 Xa 1 2 xg 85 ue 0 n 0 4 1 10 86 e a B a B tres n v mana e r a b 73 Us Va Wa Xs Ys Ps o E a B 1 G o B C HelrG B s Us Mss ers T Yi A Cs GEB gt 87 3 3 1 r a B 1 4 3473 Foray 88 a p 2 8 89 a p C DENSITY FUNCTIONAL DESCRIPTIONS 327 a I 1 2 2 z 2 253 90 1 e r t u v w x y p 2t 1 ur ln Me Imion 91 c 1 709921 92 T 0 031091 0 015545 0 016887 93 U 0 21370 0 20548 0 11125 94 V 7 5957 14 1189 10 357 95 W 3 5876 6 1977 3 6231 96 X 1 6382 3 3662 0 88026 97 Y 0 49294 0 62517 0 49671 98 P 1 1 1 99 A 0 955689 0 788552 5 47869 100 B 0 0820011 2 71681 2 87103 101 C 0 789518 0 573805 0 660975 102 and A 0 006 0 2 0 004 103 To avoid singularities in the limit p 0 G Ps0 Bo Bin Xs A2 B2 n Xs A2 7 31813428 YT ps 43 Co Cin xs 3 Ca n xs 3 Ue C 10 B97 Density functional part of B97
280. em CPU time in seconds for last program called Elapsed time in seconds for last program called The variable names for properties are the same as used on the EXPEC input cards OV EKIN POT DELTA DEL4 DARWIN MASSV EREL XX YY ZZ XY XZ XY XXX XXY XXZ QMXX OMYY QMZZ DMX DMY DMZ Overlap Kinetic energy Potential Delta function v4 Darwin term of relativistic correction Mass velocity term of relativistic correction Total relativistic correction Dipole moments Second moments XYY XYZ XZZ YYY YYZ YZZ ZZZ Third moments QMXY QMXZ QMXY Quadrupole moments 8 VARIABLES 50 EFX EFY EFZ Electric field FGXX FGYY FGZZ FGXY FGXZ FGXY Hlectric field gradients D DX D DY D DZ Velocity LSX LSY LSZ One electron spin orbit LL Total angular momentum squared L LX LY LZ Electronic angular momentum LXLX LYLY LZLZ LXLY LXLZ LYLZ Two electron angular momentum By default only the dipole moments are computed and defined The values of other properties are only stored in variables if they are requested by EXPEC cards If more than one state is computed e g in state averaged MCSCF corresponding arrays PROP istate are returned If properties are computed for more than one center the center number is appended to the name e g EFX1 EFX2 etc If transition properties are computed their values are stored in corresponding variables with prefix TR e g TRDMX TRDMY
281. ems node1 defaults to the local host name and there is no default for node2 and higher On Cray T3E and IBM SP systems and on systems running under the PBS batch system if N is not specified nodes are obtained from the system in the standard way tasks1 tasks2 etc may be used to control the number of tasks on each node as a more flexible alternative to n tasks per node If omitted they are each set equal to n tasks per node userl 2 RUNNING MOLPRO 4 user2 etc give the username under which processes are to be cre ated Most of these parameters may be omitted in favour of the usually sensible default values G global memory memory Some parts of the program make use of Global Arrays for holding and communicating temporary data structures This op tion sets the amount of memory to allocate in total across all pro cessors for such activities 3 DEFINITION OF MOLPRO INPUT LANGUAGE 5 3 DEFINITION OF MOLPRO INPUT LANGUAGE 3 1 Input format MOLPRO s execution is controlled by an input file In general each input record begins with a keyword which may be followed by data or other keywords Molpro input contains commands directives options and data The commands and directives are sequentially executed in the order they are encountered Furthermore procedures can be defined anywhere in the input which can include any number of commands and directives They are only executed when called which may be before or after the de
282. epresentation or if localized orbitals define the CASSCF wavefunction Note that the specified transformation must always be orthogonal 36 10 5 Symmetry relations between orbitals In general for a VB wavefunction to be symmetry pure the orbitals must form a representa tion not necessarily irreducible of the symmetry group Relations between orbitals under the symmetry operations defined by S YMELM may be specified according to ORBREL i i2 labell label2 Orbital i is related to orbital i by the sequence of operations defined by the label specifications defined previously using S YMELM The operators operate right to left Note that 7 and i may coincide Only the minimum number of relations required to define all the orbitals should be provided an error exit will occur if redundant ORBREL specifications are found 36 10 6 The SYMPROJ keyword As an alternative to incorporating constraints one may also ensure correct symmetry of the wavefunction by use of a projection operator NO SYMPROJL irrep irrepa The effect of this keyword is to set to zero coefficients in unwanted irreducible representations For this purpose the symmetry group defined for the CASSCF wavefunction is used always a subgroup of D 5 The list of irreps in the command specifies which components of the wave function should be kept If no irreducible representations are given the current wavefunction symmetry is assumed In a state average
283. ere key can be IDIR If starting at a transition state or near a transition state determine where to take the first step If IDIR 0 is chosen the first step will be towards the transition state This is the default If IDIR 1 is given in the input the first optimization step will be along the transition vector i e the hessian eigenvector to the smallest eigenvalue which points down towards the minimum If using a larger IDIR parameter the first step will be larger if using a negative value the first step will be in the opposite direction STPTOL If using an updated hessian matrix this parameter determines what update to take If the step size between two subsequent points on which the steepest decent lines are puzzled together is smaller than stptol 1 e if we are close to a minimum the BFGS update is used otherwise it is Powell update The default value of stptol is 1 d 6 SLMAX This option is only valid with the old version of the reaction path fol lowing algorithm i e METHOD SRSTEEP In this algorithm s1max determines the frequency of the recalculation of the numerical hes sian If the total step size of the last steps exceeds slmax the hessian will be recalculated otherwise it will be updated By default simax is two times the maximum step size of the optimization step steplength see STEP section 39 2 12 If you are using METHOD QSD the SLMAX option is obsolete and the NUMHES command see above should be used instead
284. es start with orbitals from referenc save orbitals to record 2141 2 generate diabatic orbitals by maximizing the loverlap with the orbitals at the reference geometry geometry CI for 2 states wavefunction saved to record paran nacme com Compute overlap and transition density lt R R DR gt Save transition density to record 8100 2 repeat at r dr same CASSCF as above Two 1A1 states start with orbitals from referenc save orbitals to record 2142 2 generate diabatic orbitals by maximizing the loverlap with the orbitals at th geometry geometry referenc ICI for 2 states wavefunction saved to record 6002 2 Compute overlap and transition density lt R R DR gt Save transition density to record 8200 2 compute NACME using 2 point formula forward difference store result in variable nacmelp compute NACME using 2 point formula store result in variable nacmelm backward difference compute NACME using 3 point formula lorbital records for R R DR R DR transition density records for R R DR R DR store result in variable nacme2 lend of loop over differend bond distances laverage the two results forward and backward differences print a table with results title for table 35 QUASI DIABATIZATION 221 This calculation produces the following table Non adiabatic couplings for LiF R NACME1P NACME1M NACMEAV NACME2 10 0 0 22828936 0 22328949 0 22578942 0 22578942 1
285. es 281 FROZEN 16 113 FULL 235 Full CI 199 G1 152 Gaussian 73 GDIRECT GENERAL 208 GEOMETRY 70 Geometry files 74 Molpro 92 style 73 Writing CRD files 73 Writing Gaussian input 73 Writing MOLDEN input 73 Writing XMol files 73 XYZ input 72 Z matrix geometry 70 geometry optimization 251 automatic 251 conical intersection 264 convergence criteria 252 counterpoise correction DIIS method energy variables quadratic steepest descent method 251 rational function method 251 255 saddle point transition state GEXPEC B6 GOPENMOL 215 GOTO BT GPARAM 40 GPRINT gradients 244 GRADTYP 244 RID I01 RIDPRINT 104 RIDSAVE 103 RIDSYM 103 RIDTHRESH 00000 INDEX GROUP 109 234 GTHRESRH 4 GUESS 230 Help 21 HESSELEM 260 HESSIAN 258 hessian elements 260 model numerical HF options HF SCF Hints IF F blocks INACTIVE 258 INCLUDE ndexed Variables INDIVIDUAL 210 INIT 289 input format 5 input structure Integral direct integrals INTOPT 127 Intrinsic functions 9 intrinsic reaction coordinate 256 263 Introductory examples IPOL IPRINT 125 IRC TRREPS 232 Isotope mass 75 ITERATIONS 122 Keywords 18 KS 99 KS SCF 99 LABEL BI LATTICE LIBMOL 85 libmol 78 library LIMIT 208 LINEAR 208 LINESEARCH 263
286. es Other methods such as the Newton Raphson procedure or the Augmented Hessian procedure are also implemented and can be selected using the ITERATIONS directive for state averaged calculations only the non linear optimization method can be used For CASSCF calculations the CI problem is solved in a basis of Slater determinants unless a CONF IG card is given Some procedures may be disabled using the DONT directive 20 6 1 Selecting the CI method CONF IG key key may be DET or CSF and defaults to CSF If no CONFIG or SELECT card is given the default is determinants CASSCF 20 6 2 Selecting the orbital optimization method The ITERATIONS directive can be use to modify the defaults for the optimization method It consists of a sequence of several cards which should be enclosed in a pair of curly brackets 20 THE MCSCF PROGRAM MULTI 123 ITERATIONS DO method l iterl TO iter2 DONT method2 iter3 TO iter4 method can be one of the following DIAGCI Diagonalize hamiltonian in the beginning of the specified iter ations This is the default for iteration 1 INTERNAL Optimize internal orbitals at the beginning of the specified iter ations This is default for second and subsequent iterations WERNER use Werner Meyer Knowles non linear optimization method for the specified iterations This is the default for all iterations AUGMENT Use step restricted Augmented Hessian method for the speci fied iterations NEWTON
287. es elements R 2 to R amp R but R 2 denotes a single element of R e Vector scalar operations R R 2 multiplies each element of R by 2 Instead of the number 2 also scalar one dimensional variables or expressions can be used e g R R ANG converts all elements of R from ngstr m to bohr or Z2 R COS THETA creates a vector Z with elements Z i R i COS THETA All other algebraic operators can be used instead of e Vector vector operations If A and B are vectors of the same length then A x Bis also a vector of this length Here x stands for any algebraic operator and the operation is done for each pair of corresponding elements For instance A B adds the vectors A and B andA B multiplies their elements Note that the latter case is not a scalar product If an attempt is made to connect two vectors of different lengths by an algebraic operator an error occurs e Intrinsic functions Assume THETA 100 110 120 130 to be a vector of angles in degrees In this case X 2 COS THETA is also a vector containing the cosines of each element of THETA multiplied by two i e X i 2 COS THETA 1 MAX THETA or MIN THETA return the maximum and minimum values respectively in array THETA Vector operations can also be nested e g MAX ABS THETA returns the maximum value in array ABS THETA At present vector operations are not supported with string variables 8 8 Special variables 8 8 1 Variables se
288. es this directory and then with lower priority some potential system directories for libraries relevant to the hardware including that specified by a p3 p4 athlon amd64 em64t command line option For Intel and AMD Linux systems we recommend the following BLAS libraries mkl The Intel Math Kernel Library mkl version 8 0 or higher http www intel com cd software products asmo na eng perflib mkl atlas The Atlas library http math atlas sourceforge net You must use the atlas library specific to your processor Pentium III Linux PIIISSEl Pentium 4 Xeon Linux PA4SSE2 AMD Athlon Linux ATHLON AMD Opteron Linux HAMMER64SSE2 2 64 bit When using atlas MOLPRO will automatically compile in the extra lapack subroutines which do not come by default with the package and so the liblapack a which comes with Atlas is sufficient The appropriate linker options are L blasdir lcblas 1f77blas latlas acml For Opteron systems then ACML http developer amd com acml aspx is the preferred blas library SGI Altix can use the scs1 library is preferred HP platforms can use the m1ib math library IBM Power platforms can use the ess1 package 5 configure prompts for the destination directory INSTBIN for final installation of the MOLPRO executable This directory should be one normally in the PATH of all users A INSTALLATION OF MOLPRO 309 who will access MOLPRO and its specification will depend on whether the installation
289. etry optimizations are par ticularly attractive for weakly bound systems since virtually BSSE free structures are obtained see section 28 9 8jand Refs 21 23 For reasons of efficiency itis strongly advisable to use the DF LMP2 Gradient 17 for all geometry optimizations Setting SCSGRD 1 on the DF LMP2 command or DF IT directive activates the gradient with respect to Grimmes SCS scaled MP2 energy functional see also section DFIT Analytical energy gradients are not yet available for the multipole approximation of distant pairs and therefore MULTP cannot be used in geometry optimizations or frequency calculations In geometry optimizations the domains are allowed to vary in the initial optimization steps When the stepsize drops below a certain threshold default 0 01 the domains are automatically frozen In numerical Hessian or frequency calculations the domains are also frozen It is there fore not necessary to include SAVE and START options 28 LOCAL CORRELATION TREATMENTS 194 Particular care must be taken in optimizations of highly symmetric aromatic systems like e g benzene In Dg symmetry the localization of the m orbitals is not unique i e the localized orbitals can be rotated around the Cg axis without changing the localization criterion This redundancy is lost if the symmetry is slightly distorted which can lead to sudden changes of the localized orbitals If now the domains are kept fixed using the SAVE and START option
290. ets HLSTRANS 2 At present symmetry adaption can only be performed for triplet states where the following notation is used to indicate the symmetry adapted spin functions S Ms 5 05 Ms T IS Ms S Ms 7 US Ms E IS Ms If only singlet and triplet states are considered the spin orbit matrix is blocked according to double group symmetry and the eigenvalues for each each block are printed separately In all other cases the HLSTRANS option is ignored MATEL If the entire SO matrix is calculated using HLSMAT the individual matrix elements are normally not shown When the option MATEL 1 is given the individual matrix elements and the contributions of the internal and external configuration classes are printed 37 6 Examples 37 6 1 SO calculation for the S atom using the BP operator 37 SPIN ORBIT COUPLING SRevision 2006 0 SO calculation for the S atom geometry s basis spd s vtz irhf oQt 3 2 2 2 Wf 16 4 2 multi wf l6 4 2 Wf l6 6 2 wf 106 7 2 W l6 1 0 state 3 wf 16 4 0 wf 16 6 0 wf 16 7 0 ci wf ci wf 1 ci wf 1 ci wf 1 ci wf 1 ci wf l ci wf 1 ci wf 1 ci wf 1 ci wf 1 16 1 0 save 4010 ed energy 1 es energy 3 16 4 2 save 4042 1 ep energy 16 6 2 save 4062 l state 3 noexc 1 noexc 1 noexc 1 noexc 1 noexc 1 noexc 1 noexc 1 state 3 L 242 use uncontracted basis Irhf for 3P state cassc
291. excitation level N using LEVEL CCn methods only for ground states CC3 7 3 CC4 7 4 CCN 7 N Specify excitation level N using LEVEL CC n 3 methods CCSDT 3 8 3 CCSDTO 3 8 4 CC N 3 8 N Specify excitation level N using LEVEL 172 27 THE MRCC PROGRAM OF M KALLAY MRCC 173 Examples Closed shell ground state calculations for H20 xxx mtc calculations for h2o memory 8 m gthresh energy 1 d 8 geometry o hl o r h2 O0 r hl theta theta 104 r l ang basis vdz hf mrcc method cc3 method 1 program e 1 energy ccsd t method 2 CCSD T MOLPRO e 2 energy mrcc method ccsd t method 3 CCSD T MRCC e 3 energy mrcc method ccsdt dir mrecdir method 4 program e 4 energy CC3 calculation Ithe final energy is returned in variable energy ICCSD T calculation using Molpro examples CCSD T calculation using MRCC log mice coin CCSDT calculation run in directory mrccdir mrcc method ccsdt q restart 1 dir mrccdir CCSDT Q calculation method 5 program e 5 energy lrestart with previous amplitudes mrcc method CC n excitation 4 restart 1 dir mrccdir CCSDTQ calculation method 6 program e 6 energy table method e This yields METHOD CC3 76 CCSD T MOLPRO 76 CCSD T MRCC 76 CCSDT 76 CCSDT Q es CCSDTQ 76 Excitation energies for H20 a 23912734 23905150 23905150 23922746 23976632 239
292. extension in LCCSD The default value depends on FITDOM CCSD Connectivity criterion for fitting domain extension in LCCSD The default value depends on FITDOM_CCSD Distance criterion for fitting domain extension in CPHF de fault 3 0 Distance criterion for fitting domain extension in gradients de fault 5 0 Switches the DF LMP2 analytic gradient to Grimmes SCS scaled MP2 energy functional default 0 11 1 5 Miscellaneous control options There is a rather large number of parameters Many of these should normally not be changed and therefore only a subset is described here A full list can be obtained using HELP CFIT 12 GEOMETRY SPECIFICATION AND INTEGRATION 69 12 GEOMETRY SPECIFICATION AND INTEGRATION Before starting any energy calculations MOLPRO checks if the one and two electron integrals are available for the current basis set and geometry and automatically computes them if neces sary Itis therefore not necessary any more to call the integral program explicitly as was done in older MOLPRO versions using the INT command The program also recognizes automatically if only the nuclear charges have been changed as is the case in counterpoise calculations In this case the two electron integrals are not recomputed Before any energy calculation the geometry and basis set must be defined in GEOMETRY and BASIS blocks respectively 12 1 Sorted integrals By default two electron integrals are evaluated once and
293. ey apply The general order of these cards is WF or REF RESTRICT optional SELECT optional CON optional If a RESTRICT cards precedes the WF card it applies to all reference symmetries Note that RESTRICT also affects the spaces generated by SELECT and or CON cards 21 THE CI PROGRAM 135 21 2 9 Explicitly specifying reference configurations CON n n2 na n4 Specifies an orbital configuration to be included in the reference function n m2 etc are the occupation numbers of the active orbitals 0 1 or 2 Any number of CON cards may follow each other but they must all appear directly after a SELECT card 21 2 10 Defining state numbers STATE nstate nroot 1 nroot 2 nroot nstate nstate is the number of states treated simultaneously nroot i are the root numbers to be calcu lated These apply to the order of the states in the initial internal CI If not specified nroot i i Note that it is possible to leave out states 1 e STATE 1 2 calculates second state STATE 2 1 3 calculates first and third state All states specified must be reasonably described by the internal configuration space It is pos sible to have different convergence thresholds for each state see ACCU card It is also possible not to converge some lower roots which are included in the list nroot i see REFSTATE card For examples see REFSTATE card 21 2 11 Defining reference state numbers REFSTATE nstatr nrootr I
294. f 11D and 1S states 3P states save casscf wavefunctions using mrci mrci calculations for 1D 1S states save energy for 1D state in variable e save energy for 1S state in vari temples mrci calculations for 3P states TTA save energy for 3P state in variable ep mrci calculations for 3P states mrci calculations for 3P states compute so integrals Only triplet states casscf Only triplet states ci wf 16 7 2 save 4072 1 text only triplet states casscf lsint text 3P states casscf ci hlsmat 1s 3042 1 3062 1 3072 1 text 3P states mrci ci shiemat ls 4042 1 4062 1 4072 1 text 3P 1D 1S states casscf ci hlsmat 1s 3010 1 3040 1 3060 1 3070 1 3042 1 3062 1 3072 1 text only triplet states use mrci hlsdiag d d S d d d P Ep pl nergies and casscf SO matrix elements set variable hlsdiag to mrci energies ci hlsmat 1s 3010 1 3040 1 3060 1 3070 1 3042 1 3062 1 3072 1 37 6 2 SO calculation for the I atom using ECPs mrci All states casscf 37 SPIN ORBIT COUPLING SRevision 2006 0 kc I r memory 5 M gprint orbitals civector bas XS gthresh energy 1 d 8 coeff 1 geometry I basis 1 Iodine ECP 1 ecp 1 46 4 3 Q8 Dirac Fock with SO coupling 15527 1 00000000 0 00000000 20727 3 50642001 83 09814545 2 1 74736492 5 06370919 4 2 2 99860773 1 3 81 88444526 2 3 01690894 2 3 83 41280402 25 1 59415934 1 3 2 32392477 2
295. f ftcflag adds a token to the specifiers for the Fortran preprocessor ftc largefiles nolargefiles controls whether large file gt 2Gb support is wanted This option is not relevant or used on all architectures All modern Linux distributions should support large files p3 p4 athlon amd64 em64t specifically identifies a particular hardware in order to force appropriate run time libraries where possible These options are supported only on Linux systems If any of these options is given the MOLPRO executable will be named molpro p3 exe molpro p4 exe ormolpro athlon exe in the mpp case e g molpro p3 tcgmsg exe It is possible to install different plat form variants simultaneously in the same MOLPRO tree see section A 3 4 Configuration of multiple executables in the same MOLPRO tree On Linux systems it may be desirable to have optimized versions for different hardware archi tectures like p3 p4 athlon or x86 64 see section A 3 3 Provided the compiler options are the same i e neither p4 nor athlon specific the different versions differ only by the use of specific BLAS libraries It is then possible to install different executables for each case in the same MOLPRO tree without the need to recompile the program To do so one first needs to run A INSTALLATION OF MOLPRO 310 configure for each case and specify the appropriate libraries when configure prompts for them These library paths are
296. f the make command in the MOLPRO base directory The file format consists of expressions which must be separated by a blank line Expressions consist of a quantity and value and the syntax is given by quantity value The syntax of value is a maple expression and quantity may take any name the user chooses with the exception of the special quantity names listed in table 8 18 5 Examples The following shows the use of both non self consistent and self consistent DFT 18 THE DENSITY FUNCTIONAL PROGRAM blurb Text to document the functional ref Alias for reference contained in doc references xml title Text to appear as a heading for the functional documentation Table 8 ACG special quantity names and definitions of their values geometry c n c r r 1 1 angstrom df b lyp rhf method 1 program dft edf 1 dftfun uhf method 2 program dft edf 2 dftfun uks method 3 program edf 3 dftfun dft method 4 program edf 4 dft fun table dftname dftfuns table method edf 107 examples endft com 19 ORBITAL LOCALIZATION 108 19 ORBITAL LOCALIZATION Localized orbitals are calculated according to the Boys or Pipek Mezey criteria Localization takes place within each symmetry species separately If complete localization is desired no symmetry should be used All subcommands can be abbreviated by three characters The localization program is invoked by the LOCALI command LOCALI method The keyword metho
297. fac is a factor of the corresponding primitive in the constraint If fac is omitted it is taken to be 1 39 GEOMETRY OPTIMIZATION OPTG 269 If value is specified in Angst rom or Radian unit must be given Examples for H20 in C symmetry Constraining the bond angle to 100 degrees constraint 100 deg angle atoms h1 0 h2 which is equivalent to primitive al angle atoms hl o h2 constraint 100 al Keeping the two OH distances equal constraint 0 bond atoms h1 0 1 bond atoms h2 0 which is equivalent to primitive bl bond atoms h1l 0 primitive b2 bond atoms h2 0 constraint 0 b1 1 b2 39 3 2 Defining internal coordinates By default SLAPAF optimizes in force constant weighted normal coordinates that are deter mined automatically However the user can define his own coordinates The definition of internal coordinates similar to constraints is based on primitive coordinates The input is INTERNAL NAME name FACTOR fac prim F ACTOR fac prim FIX NAME name FACTOR fac prim F ACTOR fac prim Internal coordinates that are specified using INTERNAL are varied and those using FIX are fixed to their initial values An important point for the definition of internal coordinates is that their total number must be equal to the number of degrees of freedom of the molecule Otherwise an error message is generated Only symmetry independent coordinates need to be given 3
298. factor Takes the column vectors v and v2 from matrix and adds their outer product to result v and v2 must be given in the form icol isym e g 3 2 means the third vector in symmetry 2 The result is result a b result a b factor v1 a v2 b If result has not been used before it is zeroed before performing the operation 43 13 Forming a closed shell density matrix DENS DENS density orbitals iocc iocc Forms a closed shell density matrix density from the given orbitals The number of occupied orbitals in each symmetry i must be provided in iocc 43 MATRIX OPERATIONS 298 43 14 Computing a fock matrix FOCK FOCK fid computes a closed shell fock matrix using density d The result is stored in f 43 15 Computing a coulomb operator COUL COUL J d computes a coulomb operator J d using density d 43 16 Computing an exchange operator EXCH EXCH K d computes an exchange operator K d using density d 43 17 Printing matrices PRINT PRINT name ncol 1 ncol 2 prints matrix name ncol isym is the number of columns to be printed for row symmetry isym if not given all columns are printed For printing orbitals one can also use ORB 43 18 Printing diagonal elements of a matrix PRID PRID name prints the diagonal elements of matrix name 43 19 Printing orbitals PRIO PRIO name n nN2 N3 Ng prints orbitals name The first n orbitals are printed in symmetry i If n
299. fault values by THRMAX DKEXT THREST DKEXT 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 62 with the restriction that the initial values cannot be smaller than the final ones For historical reasons many options have alias names The following tables summarize the default values for all options and thresholds and also gives possible alias names Table 6 Default values and alias names for direct options Parameter Alias Default value SCREEN 1 MAXRED 7 VARRED 1 d 7 SWAP 1 SWAP _DFOCK SWAP DMP2 DTRAF 1 PAGE DTRAF PAGE 1 SCREEN DTRAF SCREEN MAXSHLO1 DTRAF NSHLO1 32 MINSHLO1 DTRAF 0 MAXSHLO2_DTRAF NSHLQ2 16 MINSHLQ2 DTRAF 0 MAXCEN DTRAF 0 PRINT DTRAF 1 SWAP_DTRAF SWAP DKEXT DRVKEXT 3 SCREEN DKEXT SCREEN MAXSIZE DKEXT 0 MINSIZE DKEXT 5 MAXCEN_DKEXT 1 PRINT_DKEXT 1 SWAP DKEXT SWAP MXMBLK DKEXT depends on hardware B parameter on molpro command 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 63 Table 7 Default thresholds and alias names for direct calculations Parameter Alias Default value THREST THRAO min AE 1 d 2 1 d 9 THRINT THRSO min AE 1 d 2 1 d 9 THRPROD THRP min AE 1 d 3 1 d 10 THRMAX 1 d 8 THREST DSCF THRDSCF lt 1 d 10 depending on accuracy and basis set THRMAX_DSCF THRDSCF MAX THRMAX THR DTRA
300. figuration spaces as the MRCI i e only the doubly external configurations are internally contracted A new version of the program has been implemented in which also subspaces of the singly ex ternal and internal configuration spaces are internally contracted see reference given above This program which is called using the keyword RS2C is more efficient than RS2 in particu lar for large molecules with many closed shell inactive orbitals It is recommended to use this program for normal applications of second order multireference perturbation theory CASPT2 RASPT2 Note that it gives slightly different results than RS2 due to the different contraction scheme It should also be noted that neither RS2 or RS2C are identical with the CASPT2 of Roos et al J Chem Phys 96 1218 1992 since certain configuration subspaces are left uncontracted However the differences are normally very small The last point that should be mentioned is that the calculation of CASPT2 RASPT2 density matrices and therefore molecu lar properties is presently possible only with the RS2 command and not with RS2C The results of multireference perturbation theory may be sensitive to the choice of the zeroth order Hamiltonian This dependence is more pronounced in second order than in third order Several options are available which will be described in the following sections It may also happen that A 0 E 0 in the basis of the configuration state function
301. finition in the input file The input file can be written in free format The following conversions take place comma move to next tab stop i e this delimits input fields semicolon end of record i e a new record is started exclamation mark ignore rest of input line useful for comments three dashes end of file rest of input is ignored Input may be given upper or lower case The input processor converts all characters to upper case All integers are appended with only floating point numbers are read by the program Several logical input records can actually be typed on one line and separated by semicolons i e a given input line may contain many actual commands separated by semicolons or just one as you prefer These basic command units records delimited by semicolons are also frequently referred to as cards throughout this manual Exception to these general rules are KER first data line always INCLUDE include other input file FILE definition of named files TEXT prints text TITLE defines a title for the run or a table CON specifies orbital configurations 35 last line of input These commands always occupy a whole line Using INCLUDE it is possible to open secondary input files If an INCLUDE command is encountered the new input file is opened and read until its end Input is then continued after the include card in the first file INCLUDE s may be nested A MOLPRO input record
302. fitting can be invoked by prepending the command name by DF e g DF LMP2 DF LCCSD TO etc In density fitting calculations an additional auxiliary basis set is needed Details about choosing such basis sets and other options for density fitting are described in sections 28 10 and 11 The general input for local coupled LMP2 or coupled cluster calculations is LMP2 options Local MP2 calculation LCCSD options Local CCSD calculation LCCSD TO options Local CCSD TO calculation The same options as on the command line can also be given on subsequent LOCAL and MULTP directives Instead of using the MULTP directive the MULTP option on the command line can also be used In the following we will first give a summary of all options and directives These will be described in more detail in the subsequent sections For new users it is recommended to read section at the end of this chapter before starting calculations 28 3 Summary of options Many options can be specified on the command line For all options appropriate default values are set and so these options must usually be modified only for special purposes For convenience and historical reasons alias names are available for various options which often correspond to the variable name used in the program Table 11 summarizes the options aliases and default values In the following the parameters will be described in more detail 28 LOCAL CORRELATION TREATMENTS 179 Table
303. following records A job title A brief description of the file contents I5 3F12 6 number of atoms coordinates of grid origin bohr 15 3F12 6 number of grid points n step vector for first grid dimension 15 3F12 6 number of grid points n2 step vector for second grid dimension 15 3F12 6 number of grid points n3 step vector for third grid dimension 32 PROPERTIES AND EXPECTATION VALUES 215 I5 4F12 6 atomic number charge and coordinates one such record for each atom 6E13 5 n X n records of length n3 containing the values of the density or orbital at each grid point In the case of a number of orbitals m the record length is m x n3 with the data for a single grid point grouped together In the case of the density gradient there is first a record of length n3 containing the density then one of length 353 containing the gradient with the three cartesian components contiguous For the laplacian there is a further record of length n3 32 7 GOPENMOL calculate grids for visualization in gOpenMol GOPENMOL filename iflag nj n2 n3 The syntax and sub options are exactly the same as for CUBE except that the files produced are in a format that can be used directly in the gOpenMol visualization program The following should be noted e Only the base name up to the last in filename is used and is appended by different suffices to create several different files crd A CHARMm CRD fo
304. for computing relativistic corrections TB ida oe die a RR Bava dow ark Bk ae Oo we NR UE RET RUE ds PA Eu P eno i ne me ee eM Ee TTE TIPP LA DATAN seo sce hme mo Rok EORR ox Roo wow e eo uoo obo dee E 7 5 Assigning punch files PUNCH seen 7 0 MOLPRO system parameters GPARAM le 8 VARIABLES 8 1 Setting yariablesj ee 8 2 Indexed variables lt s esor 22e bat Sen E pda dde ae di A age ek eee So Bake tes da de lo a we Oe A aA TO TTD 8 6 Indexed Variables Vectors 22e se a a A a SGcSo a aa a a 8 8 1 Variables set by the program oo eee ips ee cn arta Mes ANLE ERR TR gee ee ae a 8 9 1 The SHOW COMA leen dead a a A A ye ete ae A dec rd 9 TABLES AND PLOTTING ACE A 9 27 Dlotting a case a a ad deed no edes 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 10 1 Example for integral direct calculations o o 11 1 Options for density fitting o o e e iia siete pd a dica das A 11 1 3 Parameters to enable local fitting CONTENTS 11 1 4 Parameters for fitting domains 000 11 1 5 Miscellaneous control options o o 12 GEOMETRY SPECIFICATION AND INTEGRATION 12 1 Sortedintegrals llle 12 2 Symmetry specification e 12 3 Geometry specifications ee 12 3 1 Z matrxinput ss 12 3 2 XYZimnput 5 2x9 aly RR ok rU or Row P Rx
305. for first monomer mp2 for first monomer save mp2 energy in variable CP calculation for HF2 MONOMER ehf2 energy text dummy hf accu 16 mp2 edimer energy first hf is now dummy scf for second monomer mp2 for second monomer save mp2 energy in variable DIMER CALCULATION reset dummies scf for dimer mp2 for dimer save mp2 energy in variable tot edimer ehf2 ehfl ehflinf ehf2inf total BSSE corrected energy optg numerical variable etot gradient 1 d 4 startcmd label loptimize text compute optimized monomer energy vrhf rnr1 278 examples hfdimer cpcoptl num geometry 39 GEOMETRY OPTIMIZATION OPTG 279 In the last example the monomer structures are kept fixed and the interaction energy is opti mized examples hfdimer_cpcopt2 test Revision 2006 0 HF dimer MP2 CP optimization without monomer relaxation basis avtz gthresh energy 1 d 8 INITIAL VALUES OF GEOMETRY VARIABLES RFF Sees HETA1 HETA2 111 geomet ry x noorient fl 2 Fy BEE hl f1 1 74764059 f2 thetal h2 2 1 74764059 fl theta2 hl 180 using fixed HF distances of isolated HF label text CP calculation for HF1 MONOMER dummy 2 h2 second hf is now dummy hf accu 16 Iscf for first monomer mp2 mp2 for first monomer ehfl energy save mp2 energy in variable forces compute mp2 gradient for first monomer scale 1 multiply gradien
306. for the optimized energy Compute Hessian for variable varname This implies numerical cal culation of the Hessian from energies The default is to use the same variable as for the energy and gradient Use central gradient differences for computing Hessian only effective if gradients are available 39 GEOMETRY OPTIMIZATION OPTG 255 HESSFORW Use forward gradient differences for computing Hessian only effec tive if gradients are available This effectively computes the Hessian at a slightly displaced geometry but needs only half the number of displacements This is the default UPDATE BFGS IBFGS CGRD PMS POWELL MS NONE Hessian update method to be used See section 39 2 9 for details MAXUPD maxupd Max number of Hessian updates The count is reset to zero each time a Hessian is computed 39 1 8 Miscellaneous options VARSAVE Save Cartesian gradients in variables GRADX GRADY GRADZ NONUC Do not compute gradients at lattice points DEBUG Set debug print options PRINT iprint Print option for optimization SAVEXY Z file Save optimized coordinates in an xyz file In case of reaction path following one file is written for each step SAVEACT file Save optimized variables in given file In case of reaction path fol lowing the variables are saved in each step The file can be read later using the READVAR command SAVEGRD file In case of reaction path following write in each step the Cartesian coordinates and gra
307. g information for CIGPS print vectors GS at exit CIGPS print matrices GP at exit CIGPS print paging information for CIGPI print total GP in orthogonal basis print matrices GP and TP print paging information for CIGIP print GI at exit CIGIP print paging information for CIGSS print vectors GS at exit CIGSS print paging information for CIGSI print GS at exit CIGSI print paging information for CIGIS print GI at exit CIGIS print intermediate information in internal CI print coupling coefficients a P O print coupling coefficients B P O print coupling coefficients y P Q print coupling coefficients for pair internal interactions not yet used 21 THE CI PROGRAM DSI LOG CC 0 CC 1 TEST 0 TEST 1 TEST 2 CPU ALL 21 6 Examples SRevision 2006 0 Single reference CISD r 0 957 angstrom theta 104 6 degree geometry 0 Hi Opty H2 0 r H1 theta 144 not yet used At end of each iteration write summary to log file Delete at end of job if LOG 0 print address lists for coupling coefficients print coupling coefficients print internal first order density print internal second order density print internal third order density print first second and third order densities print first order transition densities print second order transition densities print first and second order transition densities print list of non redundant pairs print list of all pairs print summa
308. g of the diabatic states in the adiabatic wavefunctions In principle this mixing can be obtained by integration of the non adiabatic coupling matrix elements Often it is much easier to use an approximate method in which the mixing is determined by inspection of the CI coefficients of the MCSCF or CI wavefunctions This method is applicable only if the orbital mixing is negligible For CASSCF wavefunctions this can be achieved by maximizing the overlap of the active orbitals with those of a reference geometry at which the wavefunctions are assumed to be diabatic e g for symmetry reasons The orbital overlap is maximized using using the new DIAB command in the MCSCF program This procedure works as follows first the orbitals are determined at the reference geometry Then the calculations are performed at displaced geometries and the diabatic active orbitals which have maximum overlap with the active orbitals at the reference geometry are obtained by adding a DIAB directive to the input Old form Molpro96 obsolete DIAB orbref orbsav orbl orb2 pri New form DIAB orbref TYP E orbtype STATE state SP IN spin MS22ms2 SAVE orbsav ORB1 0rb1 ORB2 0rb2 PRINT pri Here orbref is the record holding the orbitals of the reference geometry and orbsav is the record on which the new orbitals are stored If orbsav is not given recommended the new orbitals are stored in the default dump record 2140 2 or the one given
309. g the NUMHES m command see section 39 2 7 If the Hessian matrix is recalculated in every optimiz ation step NUMHES 1 a algorithm different to the one with updated Hessians is used which is very accurate Using the PRINT OPT card this algorithm prints in every optimization step a reaction path point r which is different from the point where the energy and the gradient is calculated but closer to the real reaction path for further details of the algorithm see J Sun and K Ruedenberg J Chem Phys 99 5257 1993 For further input options of the QSD reaction path following see OPTION section 39 2 16 SRSTEEP Old Version of QSDPATH 39 GEOMETRY OPTIMIZATION OPTG 257 39 2 2 Optimization coordinates COORD It is possible to use various coordinate types and algorithms for the optimization This can be controlled by additional subcommands as described in this and the following subsections COORD opt_space opt_coord NOROT These options choose the optimization space and the coordinate system in which the optimiza tion takes place opt space defines the parameters to be optimized By default if the geometry input is given in Z matrix format all variables on which the Z matrix depends are optimized Subsets of the variables on which the Z matrix depends can be chosen using the ACTIVE or INACTIVE subdirectives If the Z matrix depends on no variables or xyz input is used all 3N cartesian coordinates are optimized opt
310. given in section An error will result if the added perturbation is not totally symmetric symmetry 1 FIELD adds to any existing field otherwise any previous field is removed Note that FIELD does currently not modify core polarization potentials CPP If CPPs are present only DIP and QUAD should be used 32 4 4 Examples The first examples shows various possibilities to add perturbations to the one electron hamilto nian Revision 2006 0 H2O finite fields memory 4 m R 0 96488518 ANG THETA 101 90140469 geometry H1 O H1 R H2 O R H1 THETA Iscf without field hf wf 10 1 f 0 05 dip f hf field dmz f hf quad f hf field qmzz f hf add dipole z Ido scf with modified add dipole z same result as previ do scf with modified ladd quadrupole Ido scf with modified ladd quadrupole field field to ho to ho HO qmzz qmzz ous ho ho xampl field to field to same result as previous Ido scf with modified h0 field zz f xx 0 5 f yy 0 5 f hf ladd general field Ido scf with modified h0 xampl same result examples ng field com ho as quad above field zz f same as before with separate field commands field xx 0 5 f field yy 0 5 f hf field hf Ido scf with modified h0 remove field scf without field The second example shows how to compute dipole moments and pol
311. gram LMP3 calls closed shell local MP3 program LMP 4 calls closed shell local MP4 program 4 GENERAL PROGRAM STRUCTURE 21 LCISD calls closed shell local CISD program LCCSD calls closed shell local coupled cluster program Explicitly correlated methods DF MP2 R12 MP2 R12 program with density fitting DF MP2 F12 MP2 F12 program with density fitting DF LMP2 R12 Local MP2 R12 program with density fitting DF LMP2 F12 Local MP2 F12 program with density fitting Orbital manipulation LOCALI calls orbital localization program MERGE calls orbital manipulation program Properties and wavefunction analysis POP calls population analysis program DMA calls distributed multipole analysis program PROPERTY calls properties program DIP adds dipole field to h QUAD adds quadrupole field to h LATTICE read or disable lattice of point charges Gradients and geometry optimization FORCES calls gradient program OPTG performs automatic geometry optimization MIN performs energy minimization with respect to some parameters PUT print or write geometry to a file HESSIAN calculate Hessian FREQUENCY calculate vibrational frequencies MASS define atomic masses DDR evaluates approximate non adiabatic coupling matrix elements The command names for single reference coupled cluster methods QCISD CCSD LQCISD LCCSD can be appended by T and then a perturbative correction for triple excitations will be computed e g CCS
312. gular momentum up to max 1 are in cluded when computing the overlap of the approximate and exact or 28 LOCAL CORRELATION TREATMENTS 183 PIPEKAO option NONORM value LMP 2ALGO value OLDDEF value Thresholds THRPIP fhresh THRORB thresh THRLOC thresh THRMP 2 thresh THRCOR thresh bitals For example MAXANG 2 means to omit all contributions of d f and higher angular momentum functions To obtain reasonable domains the value of THRBP must often be reduced to 0 97 or so This option should only be used with care If option 0 the orbitals are localized my maximizing the coefficients of basis functions of a given type at a given atom Normally this is only useful to uniquely define degenerate orbitals in atoms For in stance when this option is used to localize the orbitals for a dimer like Ar at a very long distance clean s px py and p atomic or bitals will be obtained It is not recommended to use this option for molecular calculations Determines if projected functions are normalized not recommended value 1 projected orbitals are normalized before redundancy check value 0 projected orbitals are normalized after redundancy check default value 1 projected orbitals are normalized in redundancy check af terwards unnormalized value 2 projected orbitals are never normalized default in gradient calculations If nonzero use low order scaling method in LMP2 iterations Values can be 1 2 or
313. h at convergence Using the BCCD T com mand the contributions of connected triples are also computed by perturbation theory Nor mally no further input is needed if the BCCD card follows the corresponding HF SCF Other wise occupancies and orbitals can be specified as in the CI program BRUECKNER parameters can be modified using the BRUECKNER directive The Brueckner orbitals and approximate density matrix can be saved on aMOLPRO dump record using the SAVE option The orbitals are printed if the PRINT option is given TYPE can be used to specify the type of the approximate density to be computed TYPE REF Compute and store density of reference determinant only default This corresponds to the BOX Brueckner orbital expectation value method of Chem Phys Lett 315 248 1999 TYPE TOT Compute and store density with contribution of pair amplitudes lin ear terms Normally this does not seem to lead to an improvement TYPE ALL Compute and store both densities Note The expectation variables are stored in variables as usual In the case that both densities are made the variables contain two values the first corresponding to REF and the second to TOT e g DMZ 1 and DMZ 2 If TYPE REF or TYPE TOT is give only the corresponding values are stored For avoiding error exits in case of no convergence see CCSD T 24 3 1 The BRUECKNER directive BRUECKNER orbbrk ibrstr ibrueck brsfak This directive allows the modification
314. hamiltonian matrix more sophisticated use is possible but not documented here The program is interfaced to free standing versions such as supplied in the CPC program library by use of the DUMP option The program is called with the command FCI 30 1 Defining the orbitals ORBIT name file name file specifies the record from which orbitals are read The default is the set of orbitals from the last SCF MCSCF or CI calculation 30 2 Occupied orbitals OCC ni n2 ng ni specifies numbers of occupied orbitals including CORE in irreducible representation number i If not given the default is the complete basis set 30 3 Frozen core orbitals CORE ni n ng nj is the number of frozen core orbitals in irrep number i These orbitals are doubly occupied in all configurations i e not correlated If no CORE card is given the program uses the same core orbitals as the last CI calculation if there was none then the atomic inner shells are taken as core To avoid this behaviour and correlate all electrons specify CORE 30 4 Defining the state symmetry The number of electrons and the total symmetry of the wavefunction are specified on the WF card WF elec sym spin where elec is the number of electrons sym is the number of the irreducible representation spin defines the spin symmetry spin 2S singlet 0 doublet 1 triplet 2 etc 30 THE FULL CI PROGRAM 200 30 5 Printing options PRINT code value
315. he GEXPEC card expectation values are computed in all subsequent programs if applicable For a number of operators it is possible to use generic operator names e g DM for dipole mo ments which means that all three components DMX DMY and DMZ are computed Alternatively individual components may be requested The general format is as follows G EXPEC opname icen x y z where opname operator name string either generic or component icen z matrix row number or z matrix symbol used to determine the origin x y z must not be specified If icen 0 or blank the origin must be specified in x y z Several GEXPEC cards may follow each other or several operators may be specified on one card Examples GEXPEC OM computes quadrupole moments with origin at 0 0 0 GEXPEC QM1 computes quadrupole moments with origin at centre 1 GEXPEC QM O1 computes quadrupole moments with origin at atom O1 GEXPEC QM 1 2 3 computes quadrupole moments with origin at 1 2 3 The following table summarizes all available operators Expectation values are only nonzero for symmetric operators parity 1 Other operators can be used to compute transition quantities spin orbit operators need a special treatment By default the dipole moments are computed 6 13 1 Example for computing expectation values The following job computes dipole and quadrupole moments for H20 6 PROGRAM CONTROL SRevision 2006 0 h20 p
316. he ORBITAL card is not needed because the location of the or bitals is stored in the MCSCF dump record 38 1 4 MCSCF gradients MCSCF MCSCF record file Triggers code for MCSCF gradient record file specifies the location of information dumped from the MCSCF program using a SAVE GRD recmc filmc card This card is not needed if the FORCE command appears directly after the corresponding MCSCF input since the program automatically remembers where the MCSCF information was stored The same is true if OPTG is used 38 1 5 State averaged MCSCF gradients with SEWARD SA MCSCF gradients can be computed using segmented or generally contracted basis sets using SEWARD and the RS2 gradient program The NOEXC directive has to be used in the RS2 input but no CPMCSCE card is required in MULTI The RS2 gradient program does the CP MCSCF automatically Example compute SA CASSCF gradients for II and E state of OH geometry o h o r r 1 83 multi wf 9 2 1 wf 9 23 1 wf 9 1 1 state averaged casscf for X 2PI and A rs2 noexc wf 9 1 1 compute A 2SIGMA energy forces lenergy gradient for A 2SIGMA state rs2 noexc wf 9 2 1 compute A 2PI energy forces lenergy gradient for A 2PI state Without the NOEXC directive the RS2 CASPT2 gradient would be evaluated using the state averaged orbitals 38 1 6 State averaged MCSCF gradients with CADPAC Normally no further input is required for computing gradients for state averaged MCSCF
317. he VBDUMP directive o 36 3 Other wavefunction directives o a a a A ans a pie pis eco d a e dao ples derit Read ek we EUN MAUS i ee 2 San he cadens bce sees hoe ee da Led anes oe ie alee ae aes dl 36 7 2 Guess for structure coefficients 0 00084 36 7 3 Read orbitals or structure coefficients 0 368 Permu ngorbital 2 ee ee 36 9 Optimization control eA xvi 208 208 200 200 200 200 200 200 210 210 210 210 210 210 211 211 212 212 213 213 213 213 214 214 214 214 214 214 215 216 218 218 221 CONTENTS xvii b kaob Rok dba shE x 80e ob dA AS BUE Re 231 369 2 Number of iterations ee 231 paros cda 231 36 9 4 Saddle pointop mization o o o 231 M 232 c 232 A a Me ah doi uti N oue 232 O E TN ae 232 36 10 2 The IRREPS Keyword o o e e 232 36 10 3 The COEFFS keyword Re 233 36 10 4 The TRANS keyword e 233 TUTTI 233 36 10 60 The SYMPROJ keyword len 233 a PCI 234 TOME 234 enema a anaes 234 desti uec cas ok Soaks Got oes apes d soe Be A d 234 Pia et ROAD bbe vate ae bbe E dU ha A 235 Rade iced PEE Re el a 235 eer eee 235 OP 235 36 12Controlling the amount of output len 236 36 13 Further facilities sas ccs em rA 236 36 14Service mode eA 236 ATE DUI 237 239 eee ee a a ee eee 239 2E ee ep tonit
318. he format This gives the type A F or D for each column sensible defaults are normally used DIGITS digl dig2 dig3 Give the number of digits after the decimal points to be printed for each column sensible defaults are normally used TYPE Specify a data format for the table The default is TEXT which gives a plain text file Other possibilities are CSV comma separated fields suitable for a spreadsheet LATEX a IATEX table environment MATHEMATICA Mathematica code that assigns the table to an array MATLAB Matlab code that as signs the table to an array MAPLE Maple code that assigns the table to an array HTML an HTML TABLE construction and XML an XML document containing a tree representing the table The actual formatis XHTML 9 TABLES AND PLOTTING 55 SAVE file status Specify a file on which the table will be written If status is NEW the file is rewound otherwise it is appended If file has a suffix that is one of txt csv tex m mpl html xml and a TYPE command is not specified then the type will be set to that which is conventionally appropriate for the suffix TITLE title Specify one line of a title several TITLE cards may follow each other Note that titles are only displayed in the SAVE file if the SAVE command is given before the TITLE card SORT coll col2 Sort rows according to increasing values of the given columns The columns are sorted in the order they are specified PRINT
319. he next largest charge Orbitals are reordered separately within each localization group First all orbitals are sorted so that the primary centres are in the order of the given centrelist Orbitals with primary centres which are not in centrelist come last 2 Within each group of orbitals found for a given primary centre those containing only one centre lone pairs are included first The remaining ones are ordered so that the secondary atoms are in the order of centrelist Orbitals with secondary centres which are not in centrelist come last 3 If REVERT is given the order in each localization group is reverted 4 If GROUP is given only the orbitals in the given group are reordered igrp is 2 for closed shells and inactive orbitals 1 for open shells in single reference methods and 3 for active orbitals in CASSCF calculations 5 If THREIG is given only orbitals with energies larger than the given value are reordered eps must be negative The remaining orbitals come last first if REVERT is given Note that core orbitals are neither localized nor reordered 19 7 3 Defining reference orbitals REFORB REFORB record file specifications The localized orbitals are reordered such that the overlap with the reference orbitals read from record file is maximized This is useful for local correlation treatments for keeping the order of the localized constant for different geometries A state specific orbital set can be selected usi
320. he number of coincident centres If the centres option is used an atom list should be given enclosed by square brackets The domains of all orbitals located exclusively at these atoms will be merged and the resulting merged domains will be used for all these orbitals One may also give a record number from a previously saved local calculation The domain list contained in the record will be matched to the current one and orbital domains augmented merged to include both sets This domain definition should then be adequate for calculations on both points and all those in between This procedure can be repeated to include more geometries In this way domains can be defined that are appropriate for a whole range of ge ometries e g a reaction path and if these domains are used in all calculations a strictly smooth potential energy surface is obtained 28 8 7 Energy partitioning for molecular cluster calculations ENEPART The local character of occupied and virtual orbitals in the local correlation treatment also of fers the appealing possibility to decompose the intermolecular interaction energy of molecular clusters into individual contributions of different excitation classes This allows to distinguish between intramolecular dispersive and ionic components of the correlation contribution to the interaction energy cf M Sch tz G Rauhut and H J Werner J Phys Chem 102 5197 1998 The energy partitioning algorithm is activated
321. hed key Specify record key If omitted all keys are searched type Specify entry type i e s p If omitted all types are searched format One of text default molpro MOLPRO input format table tabular or htm1 html table to govern the output for mat A more convenient way of browsing the basis library is through a web based interface The CGI script nolpro basis presents a graphical and forms based interface for performing searches It may be installed locally but is also normally available at 13 BASIS INPUT 79 http www molpro net current molpro basis 13 4 Default basis sets If a basis is not specified at all for any unique atom group then the program assumes a global default Presently this default is VDZ but may be overridden using BASIS basis or BASIS basis basis is looked up in the file 1ib defbas which generates an appropriate request for a com plete contracted set together in some cases with an ECP from the library This mapping includes the following commonly used basis sets All of the Dunning correlation consistent sets through the use of either the standard name of the basis set e g aug cc pVDZ or an abbreviation e g AVDZ The older segmented Dunning Hay double zeta sets for the first row DZ and DZP The Roos ANO basis sets ROOS The Stuttgart ECPs and associated basis sets e g ECP 1 OMWB The Hay ECPs and corresponding basis sets ECP 1 and ECP2 Some of the Ka
322. his may optionally be appended without blank by an integer which can act as sequence number e g C1 H2 etc Dummy centres with no charge and basis functions are denoted either Q or X optionally appended by a number e g O1 note that the first atom in the z matrix must not be called X since this may be confused with a symmetry specification use Q instead atom to which the present atom is connected This may be ei ther a number n where n refers to the n th line of the Z matrix or an alphanumeric string as specified in the atom field of a pre vious card e g C1 H2 etc The latter form works only if the atoms are numbered in a unique way Distance of new atom from p This value is given in bohr unless ANG has been specified directly before or after the sym metry specification 12 GEOMETRY SPECIFICATION AND INTEGRATION 72 p2 A second atom needed to define the angle amp po p p2 The same rules hold for the specification as for pj a Internuclear angle 0 po pi p2 This angle is given in degrees and must be in the range 0 lt a lt 180 P3 A third atom needed to define the dihedral angle B po pi p2 pa Only applies if J 0 see below p Dihedral angle B po p1 p2 p3 in degree This angle is de fined as the angle between the planes defined by po pi p2 and pi po pa 180 lt B lt 180 Only applies if J 0 see below J If this is specified and nonzero the new position is specif
323. his can take long time for larger molecules The calculation of the MP2 hessian finite differences of analytical gradients examples hcn_ccsd_ts com Revision 2002 10 HCN lt gt NHC Transition State Optimization and Frequencies rcn 1 18 ang rnh 1 40 ang alpha 55 degree basis vtz geometry examples N 1 rcn hen ccesd ts com H 2 rnh 1 alpha hf ccsd t optg root 2 hessproc runmp2 Transition state optimization for ccsd t using mp2 hessian frequencies ICCSD T frequencies using numerical second derivatives runmp2 hf mp2 procedure definition The last example shows how to do a MRCI Q MRCI with Davidson correction optimization with an CASPT2 hessian As for CCSD T the MRCI Q gradient as computed numerically while the CASPT2 hessian is obtained using finite differences of analytical CASPT2 gradients 39 GEOMETRY OPTIMIZATION OPTG 274 lexamples hcn mrci ts com Revision 2002 10 HCN lt gt NHC Isomerization Transition State Optimization and Frequencies print orbitals civector rcn 1 18 ang rnh 1 40 ang alpha 55 degree basis vtz geometry C N 1 rcn examples H 2 rnh 1l alpha hcn mrci ts com closed 4 global setting for casscf inactive space hf HF SCF multi mrci optg root 2 variable energd hessproc runrs2 loptimize mrci q transition state and caspt2 for runrs2 multi rs2 procedure definition for caspt2 39 4 6 Reaction path of the HCN HNC isomerization The fol
324. hosen By default the last density computed is evaluated on the grid and written to filename This behaviour can be modified by one or more of the following subcommands 32 6 1 STEP setting the point spacing STEP stepx stepy stepz stepx stepy stepz specify the point spacing in each of three axis directions By default the value of stepx stepy stepz is determinated by the number of grid points the bragg radii of the atoms and some related parameters 32 6 2 DENSITY source of density DENSITY density source GRADIENT density source LAPLACIAN density source Compute the density and optionally its gradient and laplacian density source may be a record number containing the required density and may contain further qualification such as set number in the usual way By default the last computed density is taken 32 6 5 ORBITAL source of orbitals ORBITAL orbital list orbital source jorbital list is a list of one or more orbital numbers of the form number symmetry or keywords chosen from HOMO LUMO OCC ALL If nothing is specified the default is HOMO orbital source may be a record number containing the required density and may contain further qual ification such as set number in the usual way By default the last computed orbitals are taken Note that the CUBE file format precludes simultaneous orbital and density dumps but that this may be achieved in the GOPENMOL format see 3
325. iables may be indexed but only one dimensional arrays vectors are supported The index may itself be a variable For instance METHOD 1 PROGRAM E 1 ENERGY are valid variable definitions provided 1 PROGRAM and ENERGY are also defined variables Indices may be nested to any depth Different elements of an array can be of different type either real or logical However only one unit can be assigned to an array String variables have no associated value and cannot be mixed with the other variable types Therefore a given variable name can only be used either for a string variable or a real logical variable Vectors arrays can be conveniently defined using square brackets 8 VARIABLES 46 R 1 0 1 2 1 3 ANG This defines an array with three elements which can be accessed using indices for instance R 2 has the value 1 2 ANG A repeat specifier can be given in front of the left bracket 5 0 is equivalent to 0 0 0 0 0 Brackets can even be nested for instance 2 1 2 2 2 1 3 1 is equivalent to 1 2 2 1 3 1 2 1 3 1 1 2 2 1 3 1 2 1 3 1 Arrays can be appended from a given position just by entering additional elements for instance R 4 1 4 1 5 ANG or R 4 1 4 1 5 ANG extends the above array to length 5 Previously defined values can be overwritten For instance R 2 1 25 1 35 1 45 modifies the above vector to 1 0 1 25 1 35 1 45 1 5 If no index is given on the left hand side of the e
326. ically from finite energy differences Normally the last computed ground state energy is used But the VARIABLE directive or option can be used to optimize e g Davidson corrected energies excited states or counterpoise corrected energies 39 1 Options Most parameters can be given as options on the OPTG command line as described in this section Alternatively directives can be used which will be described in section 39 1 1 Options to select the wavefunction and energy to be optimized By default the last computed energy is optimized and all commands on which the last en ergy calculation depends are automatically executed For certain purposes e g optimization of counter poise corrected energies or Davdison corrected energies the following options can be used to alter the default behaviour STARTCMD command Specifies a start command In each geometry optimization step all in put beginning with command to the current OPTG is processed This input must not include numerical gradient or Hessian calculations If numerical gradients are needed these will be computed for the final energy or specified variable by OPTG It is assumed that these com mands have been executed before entering the OPTG program PROC procname specifies a procedure to be executed in each geometry optimization step This must define a complete energy calculation orbital opti mization and correlation treatment and must not include numerical 39 GEOMETRY O
327. ich specific for CASPT2 3 and are described below 22 2 Excited state calculations There are two possibilities to perform excited state calculations 1 One can calculate each state separately This is done using the card STATE l root where root is the desired root i e 2 for the first excited state In this case the Fock operator used in the zeroth order Hamiltonian is computed using the density for the given state 2 Alternatively two or more states can be computed simultaneously using STATE n rootl root2 rootn 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 148 where n is the number of states to be computed The default is to compute the lowest n roots Optionally this default can be modified by specifying the desired roots rooti as shown One should note that this does not correspond to the multi state CASPT2 as described in section 22 3 In the case that several states are computed simultaneously the fock operator employed in the zeroth order Hamiltonian is computed from a state averaged density matrix and the zeroth order Hamiltonians for all states are constructed from the same fock operator By default equal weights for all states are used This default can be modified using the WEIGHT directive WEIGHT wl w2 wn If a REFSTATE card is given see section 21 2 11 the state averaged fock operator is made for all reference states and the WEIGHT card refers to the corresponding states 22 3
328. id confusion with program specific PRINT cards The syntax is GPRINT key1 valuel key2 value2 NOGPRINT keyl key2 Normally value can be omitted but values gt 0 may be used for debugging purposes giving more information in some cases The default is no print for all options except for DISTANCE ANGLES default 0 and VARIABLE NOGPRINT key is equivalent to PRINT key 1 key can be one of the following BASIS DISTANCE ANGLES ORBITAL CIVECTOR PAIRS CS CP REF PSPACE MICRO CPU Print basis information Print bond distances default Print bond angle information default If gt 0 dihedral angles are also printed Print orbitals in SCF and MCSCF Print CI vector in MCSCF Print pair listin CI CCSD Print information for singles in CI CCSD Print information for pairs in CI CCSD Print reference CSFs and their coefficients in CI Print p space configurations Print micro iterations in MCSCF and CI Print detailed CPU information 6 PROGRAM CONTROL 36 IO Print detailed I O information VARIABLE Print variables each time they are set or changed default 6 13 One electron operators and expectation values GEXPEC The operators for which expectation values are requested are specified by keywords on the global GEXPEC directive The first letter G is optional but should be used to avoid confusion with program specific EXPEC cards which have the same form as GEXPEC For all operators specified on t
329. ied by two bond angles rather than a bond angle and a dihedral angle If J 1 Bis the angle B po pi p3 If J 1 the triple vector product pi po p1 p2 x p p3 is positive while this quantity is negative if J 1 X Y Z Cartesian coordinates of the new atom This form is assumed if p lt 0 if p lt 0 the coordinates are frozen in geometry optimizations All atoms including those related by symmetry transformations should be specified in the Z matrix Note that for the first atom no coordinates need be given for the second atom only p1 r are needed whilst for the third atom p3 P J may be omitted The 6 missing coordinates are obtained automatically by the program which translates and re orients the molecule such that the origin is at the centre of mass and the axes correspond to the eigenvectors of the inertia tensor see also CHARGE option above Once the reorientation has been done the program then looks for symmetry Dzn and sub groups unless the NOSYM option has been given It is possible to request that reduced symme try be used by using appropriate combinations of the options X Y Z XY XZ YZ XYZ These specify symmetry operations the symbol defining which coordinate axes change sign under the operation The point group is constructed by taking all combinations of specified elements If symmetry is explicitly specified in this way the program checks to see that the group requested
330. ies as the number of states treated in the subsequent spin orbit calculation use CLEAR HLSDIAG to clear any previous values in the variable It is the user s responsibility that the order of the energies in HLSDIAG is correct 37 SPIN ORBIT COUPLING 241 37 5 1 Print Options for spin orbit calculations PRINT option value option3 value gt where option can be HLS HLS 1 only the SO energies and transition matrix elements between ground and excited states are printed default HLS gt 0 The SO matrix is printed HLS gt 1 The property matrices are printed HLS gt 2 The individual matrix elements are printed same as OPTION MATEL HLS gt 3 Debugging information is printed VLS VLS 1 No print of eigenvectors default VLS gt 0 The eigenvectors are printed 37 5 2 Options for spin orbit calculations Some options can be set using the OPTION directive in any order OPTIONS WIGNER value HLSTRANS value MATEL value where WIGNER This option determines whether the Wigner Eckart theorem should be used when the SO matrix is determined WIGNER 1 default uses the theorem WIGNER 0 calculates each SO matrix element individ ually This option is needed for test purposes only HLSTRANS This option determines whether a SO matrix calculation should be performed in the not spin symmetry adapted basis set HLSTRANS 0 in the spin symmetry adapted basis set HLSTRANS 1 default or with both basis s
331. ike ORBITALS CORRESPOND TO DIFFERENT GEOMETRY are ignored If IGNORE_ERROR is specified MPn or triples calculations can be forced with other than canon ical orbitals Note that this can lead to meaningless results If any of the above options are given they must be obeyed strictly i e the program aborts if the request cannot be fulfilled Examples ORBITAL 2100 2 lUse SCF orbitals ORBITAL 2140 2 lUs state averaged MCSCF orbitals ORBITAL 2140 2 CANONICAL luse canonical MCSCF orbitals ORBITAL 2140 2 NATURAL STATE 2 1 use natural MCSCF orbitals for second state in sym 4 12 Summary of keywords known to the controlling program This is a summary of all keywords presently implemented in the controlling program Each module knows further keywords which are described in the chapters about the individual pro grams For detailed information about the use of the commands listed below consult the fol lowing chapters 1 4 GENERAL PROGRAM STRUCTURE 19 MEMORY UNCH ILE START NCLUDE ASIS a o H fal ro L1 EOMETRY FU AR D p p E p HRESH GTHRESH IRECT GDIRECT EXPEC GEXPEC NDDO 7RHAREAER OA a ial x le m H H H z H d E o A us d H a LSEIF Ez zi Jg HA rrj LABEL Ol G1 o O HI p O s ELETE ERA Z n eg ATROP wa Jo Jo ao C m Ol H El ARTESIAN PHERICAL USER Variables indicates start of a new calculation
332. il usually the wfu file 2 VARIABLE variable Name of a variable for which the hessian is computed COORD UNIQUE Use symmetry unique displacements in the numerical calculation of the hessian default COORD 3N Don t use symmetry unique displacements not recommended using finite differences 40 1 Numerical hessian using energy variables VARIABLE VARIABLE name Defines a variable name which holds the energy value to be used for computing the hessian using finite differences By default this is ENERGY 1 as set by the most recent program For other other variables which can be used see section Note that numerical hessians cannot be computed when dummy atoms holding basis functions are present 40 2 Thermodynamical properties THERMO It is also possible to calculate the thermodynamical properties of the molecule Since MOLPRO can only handle Abelian point groups it is necessary to give the point group of the molecule in the input file THERMO SYM pointgroup 40 VIBRATIONAL FREQUENCIES FREQUENCIES pointgroup has to be the Schoenflies Symbol e g C3v for ammonia linear molecules have to be C v or D h respectively If no point group card is given rotational degeneracy will be set to 1 eventually causing deviations in the rotational entropy If no other input card is given the zero point vibrational energy and the enthalpy H t H 0 kJ mol heat capacity C J mol K and entropy S J mol K are calculated for s
333. iler options Normally CONFIG will not need changing but you should at the least examine it and change any configuration parameters which you deem necessary For further information see any comments in the CONF IG file The configure procedure may be given command line options and normally additionally prompts for a number of parameters 1 On certain machines it is possible to compile the program to use either 32 or 64 bit in tegers and in this case configure may be given a command line option i4 or i8 respectively to override the default behaviour Generally the 64 bit choice allows larger calculations files larger than 2Gb more than 16 active orbitals but can be slower if the underlying hardware does not support 64 bit integers e g some IBM RS6000 hardware Note that if i 4 is used then large files greater than 2Gb are supported on most systems but even then the sizes of MOLPRO records are restricted to 16 Gb since the internal ad dressing in MOLPRO uses 32 bit integers If i8 is used the record and file sizes are effectively unlimited A INSTALLATION OF MOLPRO 307 2 In the case of building for parallel execution the option mpp or mppx must be given on the command line For the distinction between these two parallelism modes please refer to the user manual section 2 At present Molpro supports several different cases the GA library can be either built on top of tcgmsg mpi or myrinet on the IBM SP platform
334. imizing the energy average of the particular states It is also possible to force convergence to specific states by choosing a subset of configurations as primary space PSPACE The hamiltonian is constructed and diagonalized explicitly in this space the coeffi cients of the remaining configurations are optimized iteratively using the P space wavefunction as zeroth order approximation For linear molecules another possibility is to use the LOUANT option which makes it possible to force convergence to states with definite A quantum number i e X IL A etc states 20 1 Structure of the input All sub commands known to MULTI may be abbreviated by four letters The input commands fall into several logical groups within each group commands may appear in any order but the groups must come in correct order a The program is invoked by the command MULTI or MCSCF b cards defining partitioning of orbitals spaces OCC FROZEN CLOSED c general options most commands not otherwise specified here d a WF card defining a state symmetry e options pertaining to that state symmetry WEIGHT STATE LQUANT 20 THE MCSCF PROGRAM MULTI 113 f configuration specification for that state symmetry SELECT CON RESTRICT g definition of the primary configurations for that state symmetry PSPACE h further general options Stages d through to h may be repeated several times this is the way in which you specify an average energy of several sta
335. in Table 1 Normally MOLPRO determines the symmetry automatically and rotates and translates the molecule accordingly However explicit symmetry specification is sometimes useful to fix the orientation of the molecule or to use lower symmetries 4 GENERAL PROGRAM STRUCTURE 15 Table 1 The symmetry generators for the point groups Generators Point group null card C i e no point group symmetry X orYorZ C XY C5 XYZ Ci X Y Cay XY Z Con XZ YZ D X Y Z Don Table 2 Numbering of the irreducible representations in Do Dan No Name Function 1 Ag KY 2 B3 X 3 Boy y 4 Big xy 5 Biu 4 6 B XZ 7 Bag yz 8 Ay Xyz The irreducible representations of each group are numbered 1 to 8 Their ordering is important and given in Tables B A Also shown in the tables are the transformation properties of products of x y and z s stands for an isotropic function e g s orbital and for these groups this gives also the transformation properties of x y and z Orbitals or basis functions are generally referred to in the format number irrep i e 3 2 means the third orbital in the second irreducible representation of the point group used 4 9 Defining the wavefunction In all program modules where such information is required the total symmetry of the N electron wavefunction is defined on WF wavefunction cards in the following way WE nelec irrep spin or alternatively 4 GENERAL PROGRAM STRUCT
336. in center of mass o ql ro h ql rh o 180 geometry of OH ar ql rar o theta h 0 geometry of Ar roh 1 8 OH bond length rar 7 5 distance of Ar from center of mass theta 0 langle OH Ar ro roh 16 17 distance of O from center of mass rh roh 1 17 distance of H from center of mass basis avdz basis set text calculation for complex rhf occ 8 23 3wf 27 1 1 RHF for total system rccsd t CCSD T for total system e ohar energy save energy in variable e ohar text cp calculation for OH dummy ar make Ar a dummy center rhf occ 3 1 1 wf 9 1 1 RHF for OH rccsd t ICCSD T for OH e_oh energy save energy in variable e oh examples text cp calculation for Ar ohar bsse com dummy o h make OH dummy hf Iscf for Ar ccsd t CCSD T for Ar e ar energy save energy in variable e ar text separate calculation for OH geometry 0 H 0 roh geometry for OH alone rhf occoc 3 1 1 wf 9 1 1 RHF for OH recsd t cCSD T for OH e_oh_inf energy save energy in variable e oh inf text separate calculation for Ar geometry AR geometry for OH alone hf scf for Ar ccsd t CCSD T for Ar e ar inf energy save energy in variable e ar inf de e ohar e oh inf e ar inf tocm compute uncorrected interaction energy de cp e ohar oh ar tocm compute counter poise corrected interaction energy bsse oh e oh e oh inf tocm IBSSE for OH bsse ar e ar e ar inf tocm BSSE for Ar bsse_tot bsse_oh bsse_ar Itotal BSSE For
337. ing because the coupling is then zero Three sets of FORCE commands only two for Singlet Triplet intersection follow the MULTI input They will be like FORCE SAMC record n file CONICAL record4 file NODC where record file is one of the records containing CPMCSCF info and record4 file points to a free record used for internal storage by the CONICAL code record4 file must be the same on all the CONICAL directives Furthermore the present implementation works properly only if file 1 on the CONICAL directive The optional keyword NODC must be used in case of different spins e g S T crossing when DC is not needed The actual optimization is performed using OPTG STARTCMD MULTI The example below optimizes the conical intersection in Li ground and excited states are both doublets 39 GEOMETRY OPTIMIZATION OPTG 266 lexamples 1lih2_D0D1 com Revision 2002 10 ERA ilz basis sto 3g print orbitals civector geometry x luse only molecular plane Both states must be in the same symmetry Li Ely Eheu h2 Li r hl theta r 3 0 theta 35 e hf wf 4 1 0 multi occ 6 1 wf 5 1 1 state 2 state averaged casscf CPMCSCF NACM 1 1 2 1 accu 1 0d 7 record 5100 1 cpmcscf for non adiabatic coexamiples CPMCSCF GRAD 1 1 spin 0 5 accu 1 0d 7 record 5101 1 gradient for state 1 lih2 DODI com CPMCSCF GRAD 2 1 spin 0 5 accu 1 0d 7 record 5102 1 gradient for state 2 Force SAMC 5100 1 compute
338. ing the limit In most cases the choice of active orbitals or of the optimized states is not appropriate see introduction of MULTT 20 8 4 Test options TEST il i2 i3 Activate testing options numbered il i2 Please do not use unless you know what you are doing 20 8 5 Special optimization parameters The following parameters can also be given as options on the MULTI command line STEP radius trustl tfacl trust2 tfac2 Special parameters for augmented hessian method For experts only GOPER igop Use G operator technique in microiterations Default If igop lt O do not use G operators COPT ciacc copvar maxci cishft icimax icimx1 icimx2 icstrt icstep Special parameters for the direct CI method For experts only ciacc grad threshold for CI diagonalization copvar start threshold for Clroptimization maxci max number of CI optimizations per microiteration cishft denominator shift for q space icimax max number of CI optimizations in first macroiteration icimxl max number of Cl optimizations in second and subsequent Iterations icimx2 max number of CI optimizations in internal absorption step icstrt first microiteration with CI optimization icstep microiteration increment between Cl optimizations 20 THE MCSCF PROGRAM MULTI 127 INTOP T maxito maxitc maxrep nitrep iuprod Special parameters for internal optimization scheme For experts only NONLINEAR itmaxripri drmax drdamp gfak1 gfak2 gfak3 irdamp n
339. ingl to a new variable string3 stringl lequivalent to previous case Sstring4 mystring define a string variable Since mystring is not given in quotes it will be converted to upper case string5 mystring string5 will not be a string variable since is missing 3 DEFINITION OF MOLPRO INPUT LANGUAGE 11 yields SETTING STRING1 This is a string SETTING STRING2 This is a string SETTING STRING3 This is a string SETTING STRING4 MYSTRING VARIABLE MYSTRING UNDEFINED ASSUMING 0 SETTING STRING5 0 00000000 For more information concerning strings and string variables see section 8 3 3 10 Procedures 3 10 1 Procedure definition Procedures are sequences of commands and or options They can be defined anywhere in the input as PROC procname 1 command blocks directives or PROC procname command blocks directives ENDPROC In order to avoid unexpected results procname must differ from all known command names Procedures must not contain geometry blocks Note that procedures are not executed when encountered in the input but only when called Procedure definitions must not be nested Procedures can contain procedure calls up to a nesting level of 10 3 10 2 Procedure calls Procedures can be called anywhere in the input The syntax is the same as for commands cf section 3 2 except that the procedure name replaces the command name PROCEDURE No options ar
340. interaction energy terms are stored in millihartree in distinct variables and may be collected in arrays for producing potential energy surfaces For example the input geometry nosym hel he2 hel r basis avtz wf records ca 2101 2 cb 2102 2 distances dist 4 5 5 0 5 5 5 0 06 05 6 5 71 0 do i 1 dist r dist i monomer A dummy he2 hf save ca Sapt monomerA monomer B dummy he1 hf start atdens save cb sapt monomerB linteraction contributions sapt intermol ca ca cb cb elst i Elpol exch i Elex ind i E2ing exind i E2exind disp i E2disp exdisp i E2exdisp etot 1 E12tot data truncate ca enddo table dist elst exch ind exind disp exdisp etot ftyp d d d d d d d d d plot yields the plot 4 5 5 5 5 6 6 5 T DIST 31 SYMMETRY ADAPTED INTERMOLECULAR PERTURBATION THEORY 203 Currently SAPT only accepts single determinant wave functions for the monomers i e from Hartree Fock or Kohn Sham DFT see next section calculations No point group symmetry can be exploited in a SAPT calculation 31 5 DFT SAPT It is of crucial importance to account for the intramolecular correlation effects of the individual SAPT terms since Hartree Fock theory often yields poor first and second order electrostatic properties While this can be done using many body perturbation theory 1 in a double pertur bation theory ansatz a more effi
341. ion 8 8 1 which can be used for further processing A new feature of MOLPRO2002 is that most system variables are write protected and cannot be overwritten by the user The input is automatically checked before the job starts and should a system variable be set in the input the job will stop immediately with an error message Only in some exceptions see section 8 4 system variables can be modified using the SET command but not with the simple NAME value syntax Note that due to the changed usage and syntax of the SET command compatibility with MOLPRO92 input syntax is no longer maintained S 1 Setting variables A variable can be defined using variablel valuel variable2 value2 A variable definition is recognized by the equals sign in the first field of the input card For example THRESH ENERGY 1 d 8 GRADIENT 1 d 5 does not define variables here ENERGY and GRADIENT are options for the THRESH directive Variables can have different types 8 VARIABLES 42 Numbers The value is a number or an expression The general form of value is expression unit unit is an optional string which can be used to associate a unit to the value ANG STROM DEGREE HARTREE are examples Undefined variables in expressions are assumed to be zero and defined to be zero at the same time Logicals The value can be TRUE or FALSE T and F also work or a logical expression Internally TRUE is stored as 1 and FALSE as ze
342. ion theory MP2 using local and density fitting approximations J Chem Phys 118 8149 2003 12 M Sch tz and F R Manby Linear scaling local coupled cluster theory with density fitting Part I 4 external integrals Phys Chem Chem Phys 5 3349 2003 13 Polly H J Werner F R Manby and Peter J Knowles Fast Hartree Fock theory using local density fitting approximations Mol Phys 102 2311 2004 14 H J Werner and M Sch tz Low order scaling coupled cluster methods LCCSD T with local density fitting approximations in preparation LMP2 Gradients and geometry optimization 15 A El Azhary G Rauhut P Pulay and H J Werner Analytical energy gradients for local second order M ller Plesset perturbation theory J Chem Phys 108 5185 1998 16 G Rauhut and H J Werner Analytical Energy Gradients for Local Coupled Cluster Meth ods Phys Chem Chem Phys 3 4853 2001 17 M Sch tz H J Werner R Lindh and ER Manby Analytical energy gradients for local second order Moller Plesset perturbation theory using density fitting approximations J Chem Phys 121 737 2004 LMP2 vibrational frequencies 18 G Rauhut A El Azhary F Eckert U Schumann and H J Werner Impact of Local Ap proximations on MP2 Vibrational Frequencies Spectrochimica Acta 55 651 1999 19 G Rauhut and H J Werner The vibrational spectra of furoxan and dichlorofuroxan a comparative theoretical study using den
343. ional directives can be given before the PLOT directive NOSPLINE Prevents spline interpolation of data points NSPLINE number Number of interpolation points default 20 COLOR icolorl icolor2 Colour map to be used for columns 1 2 zero means to use default values colors black blue red green cycle SYMBOL isymbl isymb2 Symbol types to be used for columns 1 2 1 means no sym bols zero means to use default values 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 56 10 INTEGRAL DIRECT CALCULATIONS GDIRECT References Direct methods general M Sch tz R Lindh and H J Werner Mol Phys 96 719 1999 Linear scaling LMP2 M Sch tz G Hetzer and H J Werner J Chem Phys 111 5691 1999 All methods implemented in MOLPRO apart from full CI FCD and perturbative triple excitations T can be performed integral direct i e the methods are integral driven with the two electron integrals in the AO basis being recomputed whenever needed avoiding the bottleneck of storing these quantities on disk For small molecules this requires significantly more CPU time but reduces the disk space requirements when using large basis sets However due to efficient prescreening techniques the scaling of the computational cost with molecular size is lower in integral direct mode than in conventional mode and therefore integral direct calculations for extended molecules may even be less expensive than conventional ones The brea
344. ions 8 VARIABLES METHODC METHODT 1 METHODT 2 METHODT 3 ENERGC FTFUN FTFUNS ifun FTNAME ifun U cU y FTFAC ifun DFTEXFAC PROP istate PROGRAM ITERATIONS CPUSTEP SYSSTEP WALLSTEP 49 holds MP4 SDQ energy in MP4 calculations The MP4 SDTQ en ergy is stored in variable ENERGY String variable holding name of the methods used for ENERGC e g CCSD BCCD QCI String variable holding name of the methods used for ENERGT 1 e g CCSD T BCCD T QCI T String variable holding name of the methods used for ENERGT 2 e g CCSD T BCCD T OCI T String variable holding name of the methods used for ENERGT 3 e g CCSD T BCCD T QCI T Total energy excluding perturbative triples correction set only in QCI or CCSD with triples correction enabled total value of density functional in DFT or KS value of ifun th component of density functional in DFT or KS name of ifun th component of density functional in DFT or KS factor multiplying ifun th component of density functional in DFT or KS factor multiplying exact exchange in KS computed property for state istate See below for the names PROP of various properties last program called as specified in the input e g HF CCSD T etc Number of iterations used Set negative if no convergence or max number of iterations reached User CPU time in seconds for last program called Syst
345. iple see section 31 4 no calculation of the dimer wave function is required Therefore SAPT is free from a zeroth order basis set superposition error which occurs in the supermolecular approach interaction terms namely electrostatic E 1 induction E References General Symmetry adapted perturbation theory and many body SAPT 1 B Jeziorski R Moszynski and K Szalewicz Chem Rev 94 1887 1994 DFT SAPT 2 G Jansen and A HeBelmann J Phys Chem A 105 646 2001 3 A HeBelmann and G Jansen Chem Phys Lett 357 464 2002 4 A HeBelmann and G Jansen Chem Phys Lett 362 319 2002 5 A HeBelmann and G Jansen Chem Phys Lett 367 778 2003 6 A HeBelmann and G Jansen Phys Chem Chem Phys 5 5010 2003 Density fitting DFT SAPT DF DFT SAPT 7 A Hefelmann G Jansen and M Sch tz J Chem Phys 122 014103 2005 31 2 First example A typical input for SAPT has the following form r 5 6 geometry nosym hel he2 hel r basis avqz wf records ca 2101 2 cb 2102 2 monomer A dummy he2 hf save Sca sapt monomerA monomer B dummy hel hf start atdens save cb sapt monomerB linteraction contributions sapt intermol ca ca cb cb 31 SYMMETRY ADAPTED INTERMOLECULAR PERTURBATION THEORY 202 Here the sapt nonomerA B store some informations about the two monomers which are needed in the subsequent SAPT calculation invoked by sapt intermol The individual
346. irective must be used to select the desired record If analytical gradients are not available for the last wavefunction the gradient is computed numer ically For more details regarding numerical energy gradients see section 38 2 38 1 1 Adding gradients ADD ADD factor NOCHECK If this card is present the current gradient and energy are added to the previous ones using the given factor This is useful for the optimization of counterpoise corrected energies cf 39 4 7 By default the program will stop with an error message unless NOORIENT has been specified in the geometry input This behaviour can be disabled by the NOCHECK option This option should only be given if all gradients which are added are evaluated at exactly the same nuclear geometry otherwise wrong results could result due to unintended rotations of the system 38 1 2 Scaling gradients SCALE SCALE factor 38 ENERGY GRADIENTS 245 If this card is present the current gradient and energy are scaled by the give factor This is sometimes useful for the optimization of counterpoise corrected energies cf 39 4 7 38 1 3 Defining the orbitals for SCF gradients ORBITAL ORBITAL record file In the SCF case record file specifies the location of the orbitals which are used for construct ing density matrices etc This card is only needed if the SCF for which the gradient is to be computed was not the most recent energy calculation For MCSCF wavefunctions t
347. is where appropriate essential for C2 and C2 In that case the possibilities are null card C i e no point group symmetry Z Cs XY Ca XYZ C X Y Co XY Z Con XZ YZ D X Y Z Don Note that Abelian point group symmetry only is available so for molecules with degenerate symmetry an Abelian subgroup must be used e g C2 or D for linear molecules See section 4 8 for more details of symmetry groups and ordering of the irreducible represen tations Also see section 12 3 1 for more information about automatic generation of symmetry planes 12 3 Geometry specifications The geometry may be given in standard Z matrix form XYZ form or cartesian and polar coor dinate MOLPRO92 format The geometry specifications are given in the form geometry options atom specifications The following are permitted as options Any valid combination of symmetry generators as described in the previous section NOSYM Disable use of symmetry ANGSTROM Bond lengths specified by numbers or variables without asso ciated units are assumed to be in A CHARGE Orient molecule such that origin is centre of charge and axes are eigenvectors of quadrupole moment 12 GEOMETRY SPECIFICATION AND INTEGRATION 71 MASS NOORIENT ZSIGNX PLANEXZ 12 3 1 Z matrix input Orient molecule such that origin is centre of mass and axes are eigenvectors of inertia tensor default Disable re orientation of molecule For
348. itals Using the defaults described above the following input is sufficient in most cases DIAB orbref Using Molpro98 is is not necessary any more to give any GEOM and DISPL cards The displacements and overlap matrices are computed automatically the geometries are stored in the dump records along with the orbitals The diabatic orbitals have the property that the sum of orbital and overlap contributions in the non adiabatic coupling matrix elements become approximately zero such that the adiabatic mixing occurs only through changes of the CI coefficients This allows to determine the mixing angle directly from the CI coefficients either in a simple way as described for instance in J Chem Phys 89 3139 1988 or in a more advanced manner as described by Pacher Ceder baum and K ppel in J Chem Phys 89 7367 1988 Recently an automatic procedure as described in J Chem Phys 102 0000 1999 has been implemented into MOLPRO This is available in Version 99 1 and later and is described in section 35 Below we present an example for the first two excited states of H2S which have B and A symmetry in Cz and A symmetry in Cs We first perform a reference calculation in Czy sym metry and then determine the diabatic orbitals for displaced geometries in Cs symmetry Each subsequent calculation uses the previous orbitals as reference One could also use the orbitals of the Ca calculation as reference for all other calculations I
349. ith angular momentum projectors j are supplemented by a local term for l lia 14 EFFECTIVE CORE POTENTIALS 86 the first giving the expansion length n in max e a number of cards specifying V max Nmax mj 2 yr Vimax mE y cjr i e j l and the following n ones giving the parameters in the form max m V1 C13M2 V2 C25 e a number of cards specifying the scalar relativistic semi local terms in the order 0 1 lax 1 For each of these terms a card with the expansion length n in ni O lma yr o cr j l has to be given and immediately following n cards with the corresponding parameters in the form m sem cs ud e analogously a number of cards specifying the coefficients of the radial potentials AV of the SO part of Vps 14 3 Example for explicit ECP input xxx CU SCF d10s1 gt d9s2 excitation energy of the Cu atom using the relativistic Ne core pseudopotential and basis of the Stuttgart Koeln group gprint basis orbitals geometry cu basis ECP 1 10 3 ECP input 1 NO LOCAL POTENTIAL aloe A eei S POTENTIAL 0 22 355 770158 2 13 19 70 865357 P POTENTIAL 3 13 233 891976 2 13 22 53 947299 D POTENTIAL examples 38 42 31 272165 2 13 26 2 741104 cu ecp explicit com 8s7p6d 6s5p3d BASIS SET 1 27 69632 13 50535 8 815355 2 380805 952616 112662 040486 01 231132 656811 545875 504327 16 285464 5 994236 2 536875 897934
350. itial iterations using 2x2 rotation method final convergence using NR method DELETE Delete the last ndel basis functions of each angular momentum type for each atom in PM localization This can be useful to achieve proper localization with diffuse augmented basis sets MAXDL If ndel 0 delete functions only up to angular momentum maxdl ORDER If iorder 1 order final orbitals according to increasing diago nal fock matrix elements If iorder 2 order final orbitals according charge centres de fault THRESH Localization threshold same as on THRESH directive STEP Max step size in NR method default 0 1d0 19 10 Printing options PRINT PRINT ORBITAL pri CHARGE CENTRES TEST TRAN If ORB ITAL is given the localized orbitals are printed If CHA RGE or CEN TRES is given the charge centres of the localized orbitals are printed If TRAN is given the transformation matrix is printed Boys only If TEST is given intermediate information is printed 20 THE MCSCF PROGRAM MULTI 112 20 THE MCSCF PROGRAM MULTI MULTI is a general MCSCF CASSCF program written by P J Knowles and H J Werner 1984 Bibliography H J Werner and P J Knowles J Chem Phys 82 5053 1985 P J Knowles and H J Werner Chem Phys Lett 115 259 1985 All publications resulting from use of this program must acknowledge the above See also H J Werner and W Meyer J Chem Phys 73 2342 1980
351. ities exist For more information see section 7 The source distribution of MOLPRO which consists of a base compressed tar archive with a file name of the form molpro 2006 1 tar gz together possibly with one or more module archives with file names of the form molpro module 2006 1 tar gz The modules contain code which is not generally distributed or features which are not always required to install the code An example is the program developers kit mod ule develop The archives can be unpacked using gunzip and tar All archives must be unpacked in the same directory It is essential that the base archive is unpacked first and advisable that any modules are unpacked before further installation Under some circumstances MOLPRO is delivered as a single tar file with a name of the form molpro all1 2006 1 tar This archive contains all necessary base and mod ule compressed tar archives together with a shell script unpack which performs the unpacking described above A 3 3 Configuration Once the distribution has been unpacked identify the root directory that was created normally molpro2006 1 In the following description all directories are given relative to this root Having changed to the root directory you should check that the directory containing the Fortran compiler you want to use is in your PATH Then run the command configure which creates the file CONF IG This file contains machine dependent parameters such as com p
352. ities that were new in MOLPRO98 MOLPRO98 has the full functionality of MOLPRO96 but in order to make the code more mod ular and easier to use and maintain a number of structural changes have been made In particu lar the number of different records has been significantly reduced The information for a given wavefunction type like orbitals density matrices fock matrices occupation numbers and other information is now stored in a single dump record Even different orbital types e g canonical natural or localized orbitals are stored in the same record and the user can subsequently access individual sets by keywords on the ORBITAL directive New facilities allow the use of start ing orbitals computed with different basis sets and or different symmetries for SCF or MCSCF calculations The default starting guess for SCF calculations has been much improved which is most useful in calculations for large molecules The use of special procedures for computing non adiabatic couplings or diabatization of orbitals has been significantly simplified We hope that these changes make the program easier to use and reduce the probability of input errors However in order to use the new facilities efficiently even experienced MOLPRO users should read the sections RECORDS and SELECTING ORBITALS AND DENSITY MATRICES in the manual It is likely that standard MOLPRO96 inputs still work but changes may be required in more special cases involving particular reco
353. ive otherwise this directive is ignored If the START directive is given the domain information as well as the amplitudes of the previous calculation are used for restart It is possible for instance to start a local CCSD calculation with the amplitudes previously saved for a local QCISD calculation but of course it is not possible to use a record saved for a non local CCSD or QCISD calculation If it is intended only to use the domain information but not the amplitudes for a restart the START option on the command line or LOCAL directive must be used cf section 28 3 28 8 5 Correlating subsets of electrons REGION In large molecules it may be sufficient to correlate only the electrons in the vicinity of an active group and to treat the rest of the molecule at the SCF level This approach can even be extended different correlation levels may be used for different sections of the system The REGION directive allows the specification of a subset of atoms REGION METHOD method DEF AUL T default method TYPE INCLUSIVE EXCLUSIVE atoml atom2 The orbitals located at these atoms will be treated at the level specified in method The remaining orbitals will be treated as defined in default If not given by the user the latter option will be set to HF The orbital selection can be done in two ways If type is set to INCLUSIVE any orbital con taining one of the atoms in its domain centre list will be included If type is se
354. ively If nprim is specified the first nprim exponents only are taken from the library If nprim is negative or ndel is given the last nprim ndel basis functions from the library set are deleted Associated with the library basis may be a set of default contraction coefficients which may be accessed in subsequent contraction cards type can include several types e g SPD or DF This usually makes sense only with default contractions i e such cards should be followed only by C without any other specifications for contractions b Explicit basis input type atom expl exp2 expn expn l General specification of exponents continuation onto subsequent cards separated by semi colon is permitted as shown the first card can hold up to 19 exponents each following card 20 exponents The exponents and other numerical parameters described below such as numbers of functions and contraction coefficients can be given as general input expressions possibly involving vari ables It is important to note however that these expressions are evaluated typically just once 13 BASIS INPUT 83 at the same time as the complete basis set is parsed This generally happens the first time that the basis set is required perhaps before the first SCF calculation can be done If the variables on which the basis depends are altered this will not be noticed by the program and the new basis set will not be used for subsequent stages of the
355. ives is as follows ORBITAL RECORD Jrecord TY PE orbtype STATE state SYM METRY symmetry SPIN spin MS2 ms2 N ELEC nelec SET iset OVL NOCHECK IGNORE ERROR DENSITY RECORD record TY PE dentype STATE state SYM METRY symmetry SPIN spin MS2 ms2 N ELEC nelec SET iset where the optional specifications can be used to select specific orbitals if several different orbital sets are stored in the same record The meaning of the individual specifications are as follows orbtype Orbital type This can be one of CANONICAL canonical or pseudo canonical orbitals NATURAL natural orbitals LOCAL localized orbitals LOCAL PM localized Pipek Mezey orbitals LOCAL BOYS localized Boys orbitals PROJECTED projected virtual orbitals used in local calculations Without further specification the most recently computed orbitals of the specified type are used If the orbital type is not specified the program will try to find the most suitable orbitals automatically For instance in MRCI calculations NATURAL orbitals will be used pref erentially if available MRPT CASPT2 calculations will first search for CANONICAL orbitals and local calculations will first look for 4 GENERAL PROGRAM STRUCTURE 18 state dentype symmetry spin ms2 nelec iset LOCAL orbitals Therefore in most cases the orbital type nee
356. jobs directory 27 2 Running MRCC The MRCC program is invoked by the command MRCC options directives The available options summarized in Table 9l For a detailed description please refer to the MRCC manual of M Kallay file manual the mrcc directory In MOLPRO the method to be used can be given as a string option METHOD string The avail able methods and the corresponding MRCC input parameters see MRCC manual as specified in Table 10 Directives are usually not necessary but the CORE OCC ORBITAL MAXIT directives work as in the MOLPRO CCSD program In addition the number of states can be given on a STATE directive and this has the same meaning as the EOM NSTATES option 27 THE MRCC PROGRAM OF M KALLAY MRCC 171 Table 9 Options for MRCC Option Alias Default value Meaning METHOD CALC CC n Computational method See Table 10 EXCITATION LEVEL 1 Excitation level in cluster operator RESTART CC RESTART 0 Restart option If 1 restart with previous amplitudes DIRECTORY DIR T Subdirectory in which MRCC runs necessary for restart jobs EOM NSING NSING 1 Number of excited singlet states in closed shell case EOM_NTRIP NTRIP 0 Number of excited triplet states in closed shell case EOM NSTATES NDOUB 1 Number of states in open shell case SYMM SYMMETRY 1 Symmetry of excited states DENSITY IDENS 0 Parameter for density calculation HF 1 O for UHF or ROHE 1 for closed shell NACTO 0 Number of active
357. k even point depends strongly on the size of the molecule the hardware and the basis set Depending on the available disk space calculations with more than 150 200 basis functions in one symmetry should normally be done in integral direct mode Integral direct calculations are requested by the DIRECT or GDIRECT directives If one of these cards is given outside the input of specific programs it acts globally i e all subsequent calcu lations are performed in integral direct mode On the other hand if the DIRECT card is part of the input of specific programs e g HF CCSD it affects only this program The GDIRECT directive is not recognized by individual programs and always acts globally Normally all cal culations in one job will be done integral direct and then a DIRECT or GDIRECT card is required before the first energy calculation However further DIRECT or GDIRECT directives can be given in order to modify specific options or thresholds for particular programs The integral direct implementation in MOLPRO involves three different procedures 1 Fock matrix evaluation DFOCK ii integral transformation DTRAF and iii external exchange operators DKEXT Specific options and thresholds exist for all three programs but it is also possible to specify the most important thresholds by general parameters which are used as defaults for all programs Normally appropriate default values are automatically used by the program and in m
358. ke and H J Werner J Chem Phys 111 4523 1999 CONTENTS Contents 1 HOW TO READ THIS MANUAL 2 RUNNING MOLPRO ZO Options ose uuo kr Es RR mere de e Ug eae rmn 2 0 2 Running MOLPRO on parallel computers 5 Input formato edis e serra wack ker a ha Oo aa 3 2 Qommands 44 44 Gea ake isa a XU oe a a SR B 3 Directiyes a ma 27h Fes Ro AR owned ROSSO Xe use RO how ee xus 3 4 Global directives 2 aa 3 5 Optio 2 go RUE ook A BE eh ghe ae ee 0d B 6 Data ea i eho ee o eee ee he de me Re qo koe o Ub eos s ahhh se al a eos ds o de EH lo e ee la E 3 8 Intrinsic functions 2 2l les 5 9 Vanables s vau ec cei ke Bb ER RA ke geb ek Rom Redi t 39 1 Setting varlables lees CT 3 10 Procedures E E x 3 10 2 Procedure calls 2 e 4 Input structure E A ko A A ew onse beum X e Ros BR Ede dos Eso Y qued d 44 Restart a E a E E a A E O 4 5 Data set manipulation e 4 6 Memory allocation es 4 7 Multiple passesthroughtheinputl lens 4 0 Symietty uc ba x cw be xexe wee eee ee xg CR 4 9 Defining the wavefunction 0 000000 00005 4 10 Defining orbital subspaces lene 4 11 Selecting orbitals and density matrices ORBITAL DENSITY 4 12 Summary of keywords known to the controlling program 4 13 MOLPRORBelp re mo Romo Rom RR Rome xm EUR e 5 INTRODUCTORY EXAMPLES 5
359. l PBEX PBE Exchange Functional PW86 PW91Ct Perdew Wang 1991 GGA Correlation Functional W91X Perdew Wang 1991 GGA Exchange Functional Perdew Wang 1992 GGA Correlation Functional Test for number of electrons H1 Tozer and Handy 1998 EE U H i Hi GI E Co N Hy Hi r HS GECEO GFCO 3 Hi L O GA Fr O H r E G1 nj El 18 4 1 Alias density functionals Additional functional keywords are also defined as convenient aliases The following table gives the translations 18 THE DENSITY FUNCTIONAL PROGRAM 106 functionals Ref B88x m B97 0 1943d0 EXACT B97R 0 21d0 EXACT B97RDF BECKE 1 BH LYP 0 5d0 EXACT 0 5d0 B88 LvP cs gt E HFB B88 1 HFS LDA T LSDAC 2 LSDC LYP88 PBE PBEX PBEC PBEO 0 25d0 EXACT 0 7540 PBEx 2w91c PBEREV PBEXREV PW91 PW91X 5 S DIRAC S VWN vws SLATER VS99 VWN VWN80 18 4 ACG documentation The automatic code generation ACG program 6 is used to implement new density function als into Molpro In order to work the program requires the maple mathematics program and the xsltproc xml parser The program requires a file with extension df containing all of the infor mation about the new functional All density functional files are placed in the directory 1ib df and are automatically activated on the next instance o
360. lable operators If no operators are specified the dipole polarizabilities are computed Presently this is working only for closed shell without direct option 17 9 Miscellaneous directives All commands described in this section are optional Appropriate default values are normally used 17 9 1 Level shifts SHIFT shifta shiftb nitord nitcl nitocc A level shift of shifta and shiftb hartree for and B spin orbitals respectively is applied This can improve convergence but has no effect on the solution shifta 0 2 to 0 3 are typi cal values The defaults are shifta O and shifta 0 3 in closed and open shell calculations respectively and shiftb 0 In open shell calculations the orbitals are reordered after each iteration to obtain maximum overlap with the orbitals from the previous iteration This takes only effect after nitord iterations The default is nitord maxit 4 if no start card is present and nitord 1 if a START card is found Starting with iteration nitocc the occupation pattern is kept fixed The default depends on the quality of the starting guess If the iteration count is smaller than nitcl only the closed shell part of the Fock matrix is used default nitcl 0 17 9 Maximum number of iterations MAXIT maxit sets the maximum number of iterations to maxit The default is maxit 30 17 9 3 Convergence threshold ACCU accu The convergence threshold is set to 10 accu This applies to the squar
361. larization integrals are effective at the moment 15 2 Example for ECP CPP SRevision 2006 0 Na2 Potential curve of the Na2 molecule using 1 ve ECP CPP gprint basis orbitals rvec 2 9 3 0 3 1 3 2 3 3 ang do i 1 frvec rNa2 rvec 1 geometry na na na rNa2 basis ecp na ecplO0sdf ecp input s na even 8 3 5 basis input examples p na even 6 3 2 na2 ecp cpp com d na 12 03 cpp init 1 CPP input na 1 9947 62 hf ehf i energy cisd core eci i energy enddo table rvec ehf eci 16 RELATIVISTIC CORRECTIONS There are three ways in MOLPROto take into account scalar relativistic effects 1 Use the Douglas Kroll relativistic one electron integrals 2 Compute a perturbational correction using the Cowan Griffin operator see section 6 13 3 Use relativistic effective core potentials see section 14 16 1 Using the Douglas Kroll Hess Hamiltonian For all electron calculations the prefered way is to use the Douglas Kroll Hess DKH Hamil tonian which is available up to arbitrary order in MOLPRO It is activated by setting DKROLL 1 somewhere in the input before the first energy calculation If no further input is specified the standard second order Douglas Kroll Hess Hamiltonian DKH2 is used 16 RELATIVISTIC CORRECTIONS 89 Starting with this release 2006 1 MOLPRO does however also provide the DKH Hamiltonian up to any arbitrary order of decoupling
362. lculation at a neighbouring geometry are available these should be used as starting guess 17 4 3 Starting with a previous density matrix START DENSITY record file specifications A density matrix is read from the given dump record and used for constructing the first fock matrix A specific density matrix can be specified as described in section It is normally not recommended to use the DENSITY option 17 5 Rotating pairs of orbitals ROTATE orb sym orbz sym angle Performs a 2 x 2 rotation of the initial orbitals orb and orb in symmetry sym by angle degrees With angle 0 the orbitals are exchanged See MERGE for other possibilities to manipulate orbitals In UHF only the B spin orbitals are rotated 17 6 Using additional point group symmetry Since MOLPRO can handle only Abelian point groups there may be more symmetry than ex plicitly used For instance if linear molecules are treated in C5 instead of C the 8 2 52 orbitals appear in symmetry 1 Aj In other cases a linear geometry may occur as a special case of calculations in Cs symmetry and then one component of the x orbitals occurs in sym metry 1 A The program is able to detect such hidden extra symmetries by blockings in the one electron hamiltonian A and the overlap matrix S Within each irreducible representation an extra symmetry number is then assigned to each basis function These numbers are printed at the end of the integral output Usually the ex
363. lden com 74 20_scf com 22 20 scf vtz com 22 79 20 scf vtz explicit com INDEX h20o scfopt 631g com 23 h2o sto3gstartl com 4 h20_sto3gstart2 com 95 h2o table com 23 h20 vqz fp com h2o0_vqz_fp_explicit com h20_xyzinput com 72 h20p_mrci_trans com 144 h2s_diab com 121 216 h2s_diabl com 222 h2s_diab2 com 224 hen_ccsd_ts com 273 hcn_isomerization com 274 hcn mp2 ts com 272 hcn mrci ts com 73 hf eom conv com 168 hf eom pes com 166 hf eom prop com 167 hfdimer cpcoptl com 275 hfdimer_cpcopt1_num com 277 hfdimer_cpcopt2 com 279 i_ecp com 242 lif_mr_mscaspt2 com 150 lif nacme com 219 lif sr mscaspt2 com 149 lih24 SOTO com 1ih2_D0D1 com 265 natrop com 299 natropfield com 300 2 rasscf com 130 n n n n2f2 ccsd com n n n a2 ecp cpp com o mergel com 290 o_merge2 com 291 o2_mrec com 175 oh_macros com oh samcforce com D45 ohar bsse com p 5 freq com D82 s_so com 241 EXCHANGE 100 EXPEC 36 97 124 139 EXPEC2 124 Expectation values 36 Explicit correlation 196 Explicitly correlated methods 196 Expressions EXTRA 288 FCI 199 FIELD PI FIELD 211 358 FILE B9 Files 12 FIXORB D34 F IXSTRUC 234 FOCK 110 138 FORCE 244 FREEZE 114 FREQUENCIES 280 frequencies 280 energy variabl
364. le Optionally a specific orbital set can be specified as described in section The specified dump record may correspond to a different geometry basis set and or symmetry than used in the present calculation Using starting orbitals from a different basis set can be useful if no previous orbitals are available and the ATDENS option cannot be used see above The following example shows how to change the symmetry between scf calculations Of course this example is quite useless but sometimes it might be easier first to obtain a solution in higher symmetry and then convert this to lower symmetry for further calculations r1 1 85 r2 1 85 theta 104 set geometry parameters geometry 0 z matrix geometry input HL Ozrl H2 0 r2 H1 theta basis vdz hf Iscf using c2v symmetry orbital 2100 2 save on record 2100 2 b l Mox EE examples hf h20 c2v cs start com start 2100 2 start with previous orbitals from c2v symmetry orbital 2101 2 save new orbitals set zsymel x y hf start 2101 2 start with orbitals from cs symmetry orbital 2102 2 save new orbitals 17 THE SCF PROGRAM 96 Note however that this only works well if the orientation of the molecule does not change Sometimes it might be helpful to use the noorient option Note also that a single dump record cannot hold orbitals for different basis dimensions Using save 2100 2 in the second calculation would therefore produce an error If orbitals from a corresponding SCF ca
365. lling the DMA program DMA DMA This command initializes the DMA program 32 2 2 Specifying the density matrix DENSITY DENSITY record file specifications The density matrix to be analysed is that found in record record on file file If omitted record file defaults to current orbital record If specified DENSITY must appear first in the input Density matrices for specific states can be selected using specifications as explained in section 32 2 3 Linear molecules LINEAR GENERAL GENERAL default invokes the normal program which copes with any geometry LINEAR invokes a faster program which can be used when all the atoms are arranged parallel to the z axis and only the m 0 components of the multipoles are required 32 2 4 Maximum rank of multipoles LIMIT LIMIT name lmax Imax is the highest rank of multipole that is to be calculated by the program Default and maximum is 10 for the general program and 20 for the linear one If name is specified the limit applies only to multipole site name 32 2 5 Omitting nuclear contributions NONUCLEAR NONUCLEAR The nuclear contributions to properties are not to be evaluated 32 PROPERTIES AND EXPECTATION VALUES 209 32 2 6 Specification of multipole sites ADD DELETE ADD name x y z Imax radius Add a new site at x y z with the name specified The multipole rank is limited to max if a value is specified otherwise the value of Imax specified by the L
366. lowing input first optimizes the transition state and then performs reaction path calcula tions in both directions The results are plotted examples hcn_isomerization com Revision 2002 10 HCN lt gt NHC Isomerization Reaction Path memory 1 m basis 3 21G rcn 1 18282 ang Starting geometry is transition state rnh 1 40745 ang alpha 55 05 degree geomet ry x Cs Symmetry C N 1 rcn H 2 rnh 1 alpha int rhf optg root 2 saveact hcn ts Find the TS examples optg method qsdpath dir 1 numhess 5 hesscentral saveact hcn path Os fare in pos cn isomerization col readvar hcn ts Reset geometry to TS optg method qsdpath dir 1 numhess 5 hesscentral saveact hcn path append find IRC in negat readvar hcn path alpha alpha pi 180 convert angle to radian table irc rcn rnh alpha e opt tabulate results table irc e opt plot energy profile as function of irc plot file hcn_eopt plot table irc rcn rnh alpha plot distances and angle as function of irc plot file hcn_dist plot 39 GEOMETRY OPTIMIZATION OPTG 275 This produces the plots 92 24 92 26 92 28 m 92 3 m E OPT 92 34 m I I 92 36 3 E E IRC e e RCN W nm RNH ALPHA IRC 39 4 7 Optimizing counterpoise corrected energies Geometry optimization of counterpoise corrected energies is possible by performing for the total system a
367. ls including respecifying the default to be used see the specification of the BASIS subcommand below 13 6 Primitive set definition A group of basis functions is defined by a data card specifying a set of primitive gaussians optionally followed by one or more cards specifying particular contractions of primitives to be included in the final basis see section for specification of contractions When all contraction definitions have been read delimited by the next data card other than a contraction definition the remaining primitives in the set which have not been included in any contraction set are added uncontracted to the basis set There are four different input forms as explained below under a to d In case that options e g SCALE NPRIM are specified they can be given in any order but no value without option key must be given after an option In all cases type defines the angular symmetry S P D F G H or I type can include several types e g SPD or DF This usually makes sense only with or default library contractions or no contractions The basis is loaded for all atoms with tag name atom in the geometry input If atom is an integer it refers to a z matrix row a Library basis sets type atom key scale2 nprim Or type atom key S CALE scale SCALE2 scale2 NPRIM nprim DELETE ndel Load basis named key from the library If scale or scale2 is present all exponents are scaled by scale or scale 2 respect
368. lt values are used If both n and first last are omitted default values for both are used d nC first last record file orb sym n contracted functions taken from orbitals orb orb 4 1 orb n 1 of symmetry sym on molpro file record file The first nonzero coefficient in the specified orbital corresponds to the first associated basis function first last specifies the range of primitives to be contracted If first last is omitted all coefficients from the specified orbitals are used Example 205 1 12 2T71004 2 1 1 generates two contractions using the first 12 coefficients from orbitals 1 1 and 2 1 The orbitals are read from record 2100 2 13 8 Examples This shows the use of default basis sets for HO H20 basis VQZ f p R 0 95 ANG THETA 104 DEGREE geometry 0 H1 O R H2 O R H1 THETA hf do closed shell SCF This is equivalent to the explicit input form H20 R 0 95 ANG THETA 104 DEGREE geometry 0 H1 O R H2 O R H1 THETA basis spdf o vqz c sp h vqz c hf do closed shell SCF 14 EFFECTIVE CORE POTENTIALS Pseudopotentials effective core potentials ECPs may be defined at the beginning of BASIS blocks The general form of the input cards is ECP atom ECP specification examples h20 vqz fp com examples h20 vqz fp explicit c 14 EFFECTIVE CORE POTENTIALS 85 which defines a pseudopotential for an atom specified either by a chemical symbol
369. ly based on the latest MCSCF calculation however STATE and WEIGHT information will not be restored in such a case 36 3 Other wavefunction directives The definitions of the CASSCF wavefunction may also be specified manually using some or all of the directives OCC Occupied orbitals CLOSED Closed shell orbitals FROZEN Frozen core orbitals WE Wavefunction card STATE Number of states for this wavefunction symmetry WEIGHT Weights of states For the exact definition of these cards see sections 20 2 and 20 3 These commands may also be used to modify the values defined in VBDUMP The information given on these cards should correspond to the CI vector saved in the CASSCF calculation The cards and their ordering should therefore coincide with those used in MULTI except for the WEIGHT cards which may differ At present the VB wavefunction must correspond to a well defined number of electrons and total spin Other states may be present but an error condition will occur if non zero weights are specified for wavefunction symmetries with varying values of elec or spin 36 4 Defining the valence bond wavefunction 36 4 1 Specifying orbital configurations The number of core and active orbitals mcore mact active electrons Nact and the value of the total spin will be identical to that defined for the CASSCF wavefunction The spatial VB 36 THE VB PROGRAM CASVB 229 configurations are defined in terms of the active orbitals only
370. ly be used if third order is not required note that RS3C is not available Options can be the following Gn Use modified zeroth order Hamiltonian see section 22 4 SHIFT value Level shift see section MIX nstates Invokes multi state MS CASPT2 treatment using nstates states See section for more details ROOT ioptroot Root number to be optimized in geometry optimization This refers to the nstates included in the MS CASPT2 See section for more details SAVEH record Record for saving the effective Hamiltonian in MS CASPT2 calculations If this is not given a default record will be used recommended INIT logical Initializes a MS CASPT2 with single state reference functions see section 22 3 IGNORE logical Flags an approximate gradient calculation without CP CASP 2 see section for details In addition all valid options for MRCI can be given see Sect 21 22 1 Introduction Multireference perturbation calculations are performed by the MRCI program as a special case For RS2 CASPT2 RASPT2 only matrix elements over a one electron operator need to be computed and therefore the computational effort is much smaller than for a corresponding MRCI For RS3 CASPT3 the energy expectation value for the first order wavefunction must be computed and the computational effort is about the same as for one MRCI iteration The 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 147 RS2 and RS3 programs use the same con
371. m luse C2v symmetry sct for h2 mcscf for h2 linear geometry for F H2 examples h2f merge com rhf orbitals for F atom 2L starting at 4 1 Imove orbitals 1 1 move all remaining hf orbitals for H2 move these to fr positions save merged orbitals Irhf for F H2 rhf orbitals for F atom move orbitals 1 1 2 1 move orbital 3 1 to 4 1 move all remaining starting at 6 1 mcscf orbitals for H2 move these to fr positions save merged orbitals casscf for F H2 using valence space This example merges the SCF orbitals of N and O to get a full valence space for NO In the simplest case the atomic calculations are performed in the individual separate basis sets but using the same symmetry C2 as the molecular calculation 42 ORBITAL MERGING Revision NO merge r 2 1 geometry x y n EERE OCG Sy 151 wl TS orbital 2110 2 geomet ry x y o tr ocer ydy ly wf 8 4 2 orbital 2120 2 geometry n o n r MERGE ORBITAL 21 OV OV OV OV OV OV OV OV OV ORB OV ROT mn N W ND t aru he WNrRrRrR Oo WUN WNE e NON BBP H WN Hn 5 0UNRoN RIRS AL 2120 2 1 1 0 4 1 4 1 45 Ho m n om onm eo ea eee tr X Wwe ROT 5 1 6 1 45 PRINT 1 ORTH 6 2 2 save 2150 2 multis occ 67272 wf l5 2 1l wf 15 3 1 start 2150 2 2006 0
372. m in particular for strongly conjugated sys tems or when diffuse basis sets are used This is caused by localization tails due to the over lapping diffuse functions The problem is particularly frequent in calculations of systems with short bonds e g aromatic molecules It can be avoided using directive PIPEK DELETE n with n 1 or 2 This means that the contributions of the n most diffuse basis functions of each angular momentum type are ignored in the localization This often yields much better localized orbitals when diffuse basis sets are used For aug cc pVTZ n 2 has been found to work very well while for aug cc pVDZ n 1 In rare cases it might also happen that the localization procedure does not converge It is them possible to choose a second order Newton Raphson localization scheme using the directive PIPEK METHOD 2 DELETE n 28 9 4 Orbital domains The orbital domains are determined automatically using the procedure of Boughton and Pulay J Comput Chem 14 736 1993 and J Chem Phys 104 6286 1996 For higher accuracy the domains can be extended and in this way the canonical result can be systematically approached cf Ref 1 and section 28 6 2 Details are described in section 28 6 In most cases the domain selection is uncritical for saturated molecules Nevertheless in partic ular for delocalized systems it is recommended always to check the orbital domains which are printed in the beginning of each l
373. m pop Mulliken population analysis using mcscf density individual give occupations of individual basis functions If specified the Mulliken populations of each individual basis function are printed 32 4 Finite field calculations Dipole moments quadrupole moments etc and the corresponding polarizabilities can be ob tained as energy derivatives by the finite difference approximation This is most easily done with the DIP QUAD or FIELD commands An error will result if the added perturbation is not totally symmetric symmetry 1 Note that the orbitals must be recomputed before performing a correlation calculation 32 4 1 Dipole fields DIP DIP xfield yfield zfield DIP xfield yfield zfield Add a finite dipole field to the one electron Hamiltonian and the core energy The field strength is given by xfield yfield zfield DIP adds to any existing field otherwise any previous field is removed 32 4 5 Quadrupole fields QUAD QUAD xxfield yyfield zzfield xyfield xzfield yzfield QUAD xxfield yyfield zzfield xyfield xzfield yzfield Exactly as the DIP command but adds a quadrupole field 32 PROPERTIES AND EXPECTATION VALUES 32 4 3 General fields FIELD FIEI FIEI D operl facl oper2 fac2 D operl facl oper2 fac2 211 Adds one electron operators operl oper2 with the corresponding factors facl fac2 to the one electron hamiltonian The available operators are
374. m2 0 nam2 naml If nam 0 all records are copied from file ifil to file ifil2 7 5 Assigning punch files PUNCH PUNCH filename REWIND Opens punch file named filename If this file already exists it is appended unless the REWIND or NEW option is specified in that case any previous information on the punch file is overwrit ten See FILE for machine dependent interpretation of filename The punch file contains all important results geometries energies dipole transition moments etc It can be read by a separate program READPUN which can produce tables in user supplied format Example PUNCH H20 PUN allocates punch file H20 PUN Note that the file name is converted to lower case on unix machines 7 6 MOLPRO system parameters GPARAM The GPARAM card allows to change MOLPRO system parameters This should only be used by experts GPARAM option value The following options can be given in any order NOBUFF if present disable system buffering LSEG disk sector length INTREL number of integer words per real word should never be modified IBANK number of memory banks Default is 2 which should always be o k IVECT O scalar 1 vector machine MINVEC minimum vector length for call to mxmb LTRACK page size in buffer routines must be multiple of seg 8 VARIABLES 41 LENBUF length of integral buffer file 1 NTR length of integral records must be multiple of 3 track LTR disk sector length a
375. max K in steps of tstep K 5265526741 6555124750 1133424527 1133424527 examples form freq com 40 VIBRATIONAL FREQUENCIES FREQUENCIES 283 Phosphorous pentafluoride Vibrational Frequencies memory 1 m basis 3 21G geomtyp Xxyz geometry nosym 6 PF5 P 0 00000 F 0 00000 F 0 00000 F 0 00000 F 1 60400 F 1 60400 rhf optg frequencies print low thermo sym d3h temp 200 400 50 use cartesian coordinates xmol style geometry input 00000 11100 52800 41700 00000 00000 O O O H PrO optimize geometry OOrFOrF Oo don t use symmetry 00000 12400 40100 52500 examples 00000 pf5 freq com 00000 calculate vibrational frequencies print frequenciestmodes of zero frequencies calculate thermodynamical properties temperature rang 400 K 41 THE COSMO MODEL 284 41 THE COSMO MODEL The Conductor like Screening Model COSMO A Klamt and G Sch rmann J Chem Soc Perkin Trans II 799 805 1993 is currently available for HF RHF UHF and DFT RKS UKS energy calculations and the corresponding gradients The COSMO model is invoked by the COSMO card COSMO option value option2 valuez where option can be NPPA size of the underlying basis grid The value must satisfy value 10 x 3 x 4 2 default 1082 type integer NSPA number of segments for non hydrogen atoms The value must satisfy values 10 x 3 x 4 2 default 92 t
376. mber of core electrons which are replaced by the pseudopotential X denotes the reference system used for generating the pseudopotential X S single valence electron ion X M neutral atom and Y stands for the theoretical level of the reference data Y HF Hartree Fock Y WB quasi relativistic Y DF relativistic For one or two valence electron atoms X S Y DF is a good choice while otherwise X M Y WB or Y DF is recommended For light atoms or for the discussion of relativistic effects the corresponding Y HF pseudopotentials may be useful Additionally spin orbit SO potentials and core polarization potentials CPP are available to be used in connection with case b ECPs but these are not currently contained in the library so explicit input is necessary here cf below In both cases a and b the same keywords refer to the pseudopotential and the corresponding basis set with a prefix MBS in case a 14 2 Explicit input for ECPs For each of the pseudopotentials the following information has to be provided e acard of the form ECP atomMcore lnax NS where Ncore is the number of core electrons replaced by the pseudopotential Vps Imax is the number of semi local terms in the scalar relativistic part of Vps while I is the max corresponding number of terms in the SO part Z Necore lmax 1 Umax ME Vps Vias Y Vi Vin Y AVRT se 1 0 l 1 the semi local terms w
377. ments and A is the interaction matrix of the screening charges on the segments This solution is exact for an electric conductor For finite dielectrics the true dielectric screening charges can be approximated very well by scaling the charge density of a conductor with f q f q fle e 1 0 5 In every SCF step the screening charges q have to be generated from the potential and then added to the Hamiltonian as external point charges The total energy of the system is 1 Eio Es Pea Lge za where Eo is the bare self energy of the system and Egje the dielectric energy Cavity construction First a surface of mutually excluding spheres of radius R rsolv is constructed where the R are the radii of the atoms defined as element specific radii and rsolv is some radius representing a typical maximum curvature of a solvent molecular surface rsolv should not be misinterpreted as a mean solvent radius nor modified for different solvents Every atomic sphere is represented by an underlying basis grid of nppa points per full atom Basis grid points which intersect a sphere of a different atom are neglected In a second step the remainder of the basis grid points are projected to the surface defined by the radii R As a third step of the cavity construction the remaining basis grid points are gathered to segments which are the areas of constant screening charges in the numerical solution Finally the intersection seams
378. ments are printed anyway HISTORY prints the complete set of previous geometries gradients and energies GRADIENT prints extended gradient information OPT prints detailed information about the optimization process mainly for debugging Several print options can be specified with one PRINT command 39 2 19 Conical Intersection optimization CONICAL To optimize a conical intersection between two electronic states having the same spin three vectors must be evaluated at SA CPMCSCF level 1 Non Adiabatic Derivative Coupling DC 2 Gradient of the lower state LSG 3 Gradient of the upper state USG 39 GEOMETRY OPTIMIZATION OPTG 265 This requires three different CPMCSCF directives in the MULTI input CPMCSCF NACM Sj Sj ACCU 1 0d 7 record recordl file CPMCSCF GRAD S SPIN Spin of state Si ACCU 1 0d 7 record record2 file CPMCSCF GRAD Sj SPIN Spin of state Sj ACCU 1 0d 7 record record3 file where S S are the electronic states in the usual format istate istsym and record n file specifies the name and the file number where CPMCSCF solutions should be stored Parameter SPIN is half of the value in the WF card used to define the electronic state Things to remember i Specify always three different record file on the CPMCSCF directives ii Evaluate the CPMCSCF for USG always last iii Skip the DC evaluation if the conical intersection involves states with different spin e g a Singlet Triplet cross
379. mmy command sometimes useful in conjunction with GOTO 6 PROGRAM CONTROL 32 6 8 Procedures PROC ENDPROC Procedures can be defined at the top of the input in the default file molproi rc or in INCLUDE files as follows PROC name Statements ENDPROC Alternatively one can use the form PROC name statements In the latter case it is required that the left curly bracket appears on the same line as PROC but statements can consist of several lines If in the subsequent input name is found as a com mand in the first field of a line it is substituted by the statements Example PROC SCF IF SPIN EQ 0 OR MOD SPIN 2 NE MOD NELEC 2 SET SPIN MOD NELEC 2 IF SPIN EQ 0 THEN HF ELSE RHF ENDIF ENDPROC Alternatively this could be written as PROC SCF IF SPIN EQ 0 OR MOD SPIN 2 NE MOD NELEC 2 SET SPIN MOD NELEC 2 IF SPIN EQ 0 THEN HF ELSE RHF ENDIF Procedures may be nested up to a depth of 10 In the following example SCF is a procedure PROC CC SCF IF SPIN EQ 0 THEN CCSD ELSE RCCSD ENDPROC Note Procedure names are substituted only if found in the first field of an input line Therefore they must not be used on one line IF statements please use IF ENDIF structures instead If as first statement of a procedure ECHO is specified the substituted
380. modeified using a global THRESH GRID option GRIDMAX gridmax In the initial iterations the grid accuracy is min gridmax tar get coarsefac COARSEFAC coarsefac Factor for initial grid accuracy see above The default is 1000 DFTFAC facl fac2 Factors for each functional The number of given values must agree with the number of functionals EXFAC factor Fraction of exact exchange added to the functional The default depends on the functional TOLORB value Threshold for orbital screening current default 1 d 15 MATRIX matrix Option to select integrator matrix 0 use old slow integrator matrix 1 Use new matrix driven integrator default In addition all options valid for HF see section 17 1 can be given 18 THE DENSITY FUNCTIONAL PROGRAM 100 18 2 Directives The following options may be used to control the operation of the DFT modules In the Kohn Sham case these may come in any order before or after directives for the SCF program as described in Section 18 2 1 Density source DENSITY ODENSITY DENSITY orbc filec ODENSITY orbo fileo For non self consistent DFT calculations specifies the source of the density matrix The total density is read from orbc filec with further options specifying density sets in the standard way as described in Section 4 11 ODENSITY can be used to specify the spin density The defaults are the densities last written by an SCF or MCSCF program 18 2 2 Threshol
381. molprop 2006 0 14 procname libname xc As described above the different executables can then be chosen on a specific machine by setting the environment variable MOLPRO RCFILE to molprop 2006 0 i4 procname libname rc Note that if MOLPRO_RCFILE is not set molpro rc will be used by default which will correspond to the last molprop 2006 0 14 procname libname rc generated A 3 5 Compilation and linking After configuration the remainder of the installation is accomplished using the GNU make com mand Remember that the default make on many systems will not work and that it is essential to use GNU make cf section A 3 2 Everything needed to make a functioning program together with all ancillary files is carried out by default simply by issuing the command make in the MOLPRO base directory Most of the standard options for GNU make can be used safely in particular j can be used to speed up compilation on a parallel machine The program can then be accessed by making sure the bin directory is included in the PATH and issuing the command molpro A INSTALLATION OF MOLPRO 311 A 3 6 Adjusting the default environment for MOLPRO The default running options for MOLPRO are stored in the file bin molpro rc After pro gram installation either using RPMs or from source files this file should be reviewed and ad justed if necessary Particular attention should be payed to some or all of the following see User s manual for full di
382. mputed when needed are now available for all kinds of wavefunctions with the exception of perturbative triple excitations in MP4 and CCSD T calculations This allows the use of significantly larger basis sets than was possible before The direct option can be selected globally using the GDI RECT command or for a specific program using the DIRECT directive See section INTEGRAL DIRECT METHODS in the manual for details Note that the DIRECT module is optional and not part of the basic MOLPRO distribution Local electron correlation methods have been further improved In combination with the integral direct modules which implement efficient prescreening techniques the scaling of the compu tational cost with molecular size is dramatically reduced approaching now quadratic or even linear scaling for MP2 and higher correlation methods This makes possible to perform cor related calculations for much larger molecules than were previously feasible However since these methods are subject of active current research and still under intense development we decided not to include them in the current MOLPRO release They will be optionally available in one of the next releases C DENSITY FUNCTIONAL DESCRIPTIONS 321 C Density functional descriptions C 1 ALYP Lee Yang and Parr Correlation Functional See reference for more details ApappZ K JE ABo 1 18 Papp 47 75 0 2 3p c 10 ABo oo s 225 p 9 5 2 1188 o
383. ms include the frozen core orbitals in the canonicalization and transformation Because of core valence mixing this leads to slightly different energies Neither of the two methods can be regarded as better or more justified it is just a matter of definition However the method in MOLPRO is more efficient since the subsequent integral transformation involves only valence orbitals and no core orbitals 27 THE MRCC PROGRAM OF M KALLAY MRCC 170 27 The MRCC program of M Kallay MRCC An interface exists to use the MRCC program of M Kallay and J Gauss within Molpro The license and source code of the MRCC program must be obtained from Mihaly Kallay http www mrcc hu Currently only single reference methods with RHF reference functions are supported Perturbative methods and CCn methods are only available for closed shell Furthermore only serial execution is supported under MOLPRO i e the mpp version cannot be used 27 1 Installing MRCC A file mrcc tar gz will be provided by by M Kallay This file should be copied into di rectory MRCC under the main MOLPRO directory In this directory a Makefile exists and typing make will compile and install the MRCC program The executables will be copied into the MOLPRO bin directory and are automatically called by MOLPRO Orbitals and other input information are communicated via external files transparent to the user Once the program is installed please run make mrcctest in test
384. n 2006 0 h20 distributed multipole analysis geometry 0o hl o r h2 0 r hl theta Z matrix geometry input r 1 ang bond length theta 104 bond angle examples basis 6 311g h20 dma com hf do scf calculation dma limit 4 results for total multipoles are 32 3 Mulliken population analysis 32 3 1 Calling the population analysis program POP POP Invokes Mulliken analysis program which analyses any density matrix into its contributions from s p d f basis functions on each atom The density matrix is taken from the last dump 32 PROPERTIES AND EXPECTATION VALUES 210 record unless overridden with the DENSITY card The subcommands may be abbreviated by the first four characters The atomic charges are stored in the MOLPRO variable ATCHARGE The i th element in ATCHARGE corresponds to the i th row of the Z matrix input 32 3 2 Defining the density matrix DENSITY DENSITY record file specifications Take density matrix to be analysed from record record on file file Density matrices for specific states can be selected using specifications as explained in section Note that the density matrices are stored in the same record as the orbitals 32 3 3 Populations of basis functions INDIVIDUAL INDIVIDUAL 32 3 4 Example h20 population analysis geometry 0o hl o r h2 0 r hl theta Z matrix geometry input r 1 ang bond length theta 104 bond angle basis 6 311g examples hf Ido scf calculation h20 pop co
385. n OPEN card 17 2 3 Specifying open shell orbitals OPEN orbj symy orb3 sym Orb SyMy This optional card can be used to specify the singly occupied orbitals The number of singly occupied orbitals must be equal to spin and their symmetry product must be equal to sym see WF card If the OPEN card is not present the open shell orbitals are selected automatically The algorithm tries to find the ground state but it might happen that a wrong state is obtained 1f there are several possibilities for distributing the open shell electrons among the available orbitals This can also be avoided using the CLOSED card If orb sym is negative this orbital will be occupied with negative spin only allowed in UHF 17 3 Saving the final orbitals ORBITAL record file SAVE record file The optimized orbitals and the corresponding density matrix fock matrix and orbital energies are saved on record file SAVE is an alias for ORBITAL If this card is not present the defaults for record are 17 THE SCF PROGRAM 94 RHF 2100 UHF 2200 holds both amp and B spin orbitals and related quanti ties These numbers are incremented by one for each subsequent calculation of the same type in the same input Note that this holds for the sequence number in the input independently in which order they are executed see section 4 3 The default for file is 2 17 4 Starting orbitals The START directive can be used to specify the initial orbitals
386. n be performed using density fitting for the necessary integrals Currently the available Ans tze are restricted to the 2A type Meth ods are available in local DF LMP2 R12 DF LMP2 F12 and canonical DF MP2 R12 DF MP2 F 12 versions detailed below Symmetry is not implemented for any of these meth ods and therefore the NOSYM option must be given in the geometry block For DF MP2 F12 the correlation factor is a frozen expansion f 2 of Gaussian type geminals By default the geminal is built from six Gaussian functions and the exponents and coefficients are optimized to obtain the best least squares fit to fj exp 12 using a suitable weight function If correlation consistent basis sets are used a suitable density fitting DF basis is automatically chosen In the case of R12 methods the default for the RI basis is the AO basis set while for F12 methods Hartree Fock JK fitting bases are used by default e g VTZ JKFIT is used for orbital basis VTZ In general only the F12 methods are recommended since these lead to much more accurate results and converge better with respect to the AO DF and RI basis sets than the R12 methods Options for canonical and local versions DF BASIS basis Select the basis for density fitting see section LI for details basis can either refer to a set name defined in the basis block or to a default MP2 fitting basis e g DF BASIS VTZ generates the VTZ MP2FIT basis See section L1 for more
387. n one dump record If no DM directive is given the first order density matrix is saved in single state calculations and only the stage averaged density matrix in state averaged calculations 20 8 Miscellaneous options All commands described in this section are optional Appropriate default values are normally used Note that printing of the orbitals and civectors can also be requested using the global GPRINT command or by giving NATORB or CANORB options 20 THE MCSCF PROGRAM MULTI 125 20 8 Print options ORBPRINT nvirt requests the occupied and nvirt virtual orbitals in each symmetry to be printed default nvirt 0 By default the program does not print the orbitals unless the ORBPRINT directive or a global GPRINT ORBITALS see section command is present Specific orbital sets can be printed using the PRINT option on a NATORB CANORB or LOCORB card see section 20 5 5 To print additional information at the end of the calculation use PRINT keyl key2 Printing is switched on for key1 key2 To print information in each iteration use IPRINT keyl key2 Possible print keys are MICRO print details of microiterations useful for finding out what s going wrong if no convergence REF print summary of configuration set CSFs only REF1 print list of configuration set CSFs only COR print summary of intermediate spaces used in CSF calculation COR1 print list of intermediate configur
388. n this case one would have to take out the second last input card which sets reforb 2141 2 20 THE MCSCF PROGRAM MULTI 122 SRevision 2006 0 H2S diabatic A states basis VDZ luse cc pVDZ basis set geometry x luse Cs symmetry planeyz fix orientation of the molecule noorient dont allow automatic reorientation s hl s rl h2 s r2 hl theta Z matrix geometry input gprint orbitals civector global print options text reference calculation for C2V theta 92 12 r1 2 3 r2 2 3 reference geometry hf occ 7 2 wf 18 1 scf calculation for ground state multi occ 9 2 closed 4 1 define active and inactive spaces wf 18 2 state 2 two A states 1B1 and 1A2 in C2v examples orbital 2140 2 save orbitals to 2140 2 h2s diab com reforb 2140 2 text calculations at displaced geometries rd 2 4 2 5 2 6 define a range of bond distances do i 1 rd loop over displaced geometries r2 rd i set r2 to current distance multi occ 9 2 closed 4 1 same wavefunction definition as at reference geom wf 18 2 state 2 orbital 2141 2 save new orbitals to record diab reforb compute diabatic orbitals using reference orbitals Istored on record reforb reforb 2141 2 set variable reforb to the new orbitals enddo See section B5 for the automatic generation of diabatic energies 20 6 Selecting the optimization methods By default MULTI uses the non linear optimization method developed by Werner Meyer and Knowl
389. nalizes a matrix OPRD Forms an outer product of two vectors DENS Forms a closed shell density matrix FOCK Computes a closed shell fock matrix COUL Computes a coulomb operator EXCH Computes an exchange operator PRINT Prints a matrix PRID Prints diagonal elements of a matrix PRIO Prints orbitals ELEM Assigns a matrix element to a variable READ Reads a square matrix from input WRITE Writes a square matrix to a file SET Assigns a value to a variable Note that the file name appearing in above commands is converted to lower case on unix ma chines See the following subsections for explanations 43 1 Calling the matrix facility MATROP The program is called by the input card MATROP without further specifications MATROP 43 MATRIX OPERATIONS 294 It can be followed by the following commands in any order with the restriction that a maximum of 50 matrices can be handled The first entry in each command line is a command keyword followed by the name of the result matrix If the specified result matrix result already exists it is overwritten otherwise a new matrix is created All matrices needed in the operations must must have been loaded or defined before unless otherwise stated If a backquote is appended to a name the matrix is transposed 43 2 Loading matrices LOAD All matrices which are needed in any of the subsequent commands must first be loaded into memory using the LOAD command Depending on the matrix type the
390. namout file Saves the current output set to record namout file The current output set must be complete and will be Schmidt orthonormalized before it is saved If the SAVE directive is not supplied the output vectors will be saved after all valid commands have been processed to the record specified on the MERGE card 42 12 Printing options PRINT PRINT iprint ideb Specifies print options iprint 0 no print iprint gt 1 orthonormalized orbitals specified on ORTH card are printed iprint gt 2 orbitals are also printed before this orthonormalization iprint gt 3 all final vectors are printed ideb 0 the overlap matrices are printed at various stages 42 ORBITAL MERGING 42 13 Examples 42 13 1 H F This example merges the orbitals of Hz and F SRevision 2006 0 example for merg print orbitals basis rh2 1 4 rhf 300 basis vdz geometry x y F text F irhf wf 9 17 1 0 c0 23 1 1 0fbital Z2130 2 text H2 geometry x y H1 H2 H1 rh2 hf orbital 2100 2 multi occ 2 orbital 2101 2 text FH2 geometry F H1 F rhf H2 H1 rh2 F 180 merge orbital 21 move 1 1 2 move 3 1 0 orbital 21 move l1 save 2131 2 Ehf roce 4a lobe start 213172 orbital 2132 2 merge orbital 2130 2 move 1 1 2 4 1 f move 3 1 3 1 4 1L move 4 1 0 4 6 1 orbital 2101 2 move 1 1 0 4 save 2141 2 Imultizocc 5 1 l1 5tart 2141 2 42 13 2 NO 290 luse C2v symmetry rhf for f ato
391. nd block i e no curley brackets are used STRICTCHECK 0 The input checker tolerates ambiguous directives if they a are followed by a non ambiguous directive which is valid for the current command STRICTCHECK 1 The input checker does not tolerate any ambiguous directives STRICTCHECK 2 The input checker does not tolerate any directives outside curley brackets The default is STRICTCHECK 0 which gives the maximum possible compatibility to previous Molpro versions 3 DEFINITION OF MOLPRO INPUT LANGUAGE 7 As already mentioned the use of curly brackets is normally compulsary if more than one input line is needed In the case of one line commands curley brackets are needed as well if the next command or procedure has the same name as a directive valid for the current command Note DIRECT and associated options cannot be specified on command lines any more 3 3 Directives Directives serve to specify input data and special options for programs They start with a key word followed by data and or options The general format is DIRECTIVE data options The format of data and options is specified in the subsequent sections Data must always be given before any options Examples for directives are OCC CORE CLOSED WF LOCAL DFIT 3 4 Global directives Certain directives can be given anywhere in the input i e either inside or outside command blocks If they are given inside of command blocks the specified options ar
392. nd of the geometry optimization 28 9 8 Intermolecular interactions Local methods are particularly useful for the calculation of weak intermolecular interactions since the basis set superposition error BSSE is largely reduced 1 13 14 and counterpoise cor rections are usually not necessary provided the BSSE of the underlying Hartree Fock is small However one must be careful to define the domains properly and to include all intermolecular pairs at the highest computational level A convenient way to define appropriate domains and pair lists is to use the option INTERACT 1 If this option is given individual molecules are identified automatically and all intermolecular pairs are automatically treated as strong pairs and included in the LCCSD Similarly appropriate triples lists are generated for LCCSD T calcu lations It is required that all orbital domains are located on individual molecules Note however that the inclusion of the intermolecular pairs strongly increases the number of strong pairs and triples and therefore high level calculations can become very expensive For calculations of interaction potentials of weakly interacting systems the domains of the sub systems should be determined at a very large distance and saved using the SAVE record option on the LOCAL or MULTP directive or the SAVE directive see section 8 8 3 If the asymptotic energy is not needed it is sufficient to do this initial calculation using option DOMONLY 1
393. ndex integrals in the AO default 1 d 10 THRPROD Product screening threshold for first half transformation de fault 1 d 8 Analogous thresholds for specfic programs can be set by appending the above keywords by the following specifications _SCF Coulomb and exchange fitting in DF HF DF KS COUL Coulomb fitting in DF HF DF KS EXCH Exchange fitting in DF HF DF KS CPHF Coulomb and exchange fitting in CPHF SCFGRD Coulomb and exchange fitting in DF HF DF KS gradients The default values are the same as for the general thresholds Further thresholds THR2HLF Threshold for second half transformation in exchange fitting default THRAO SCF THRASM SCF Threshold for local assembly of exchange matrix default THRAO_SCF THRAO FOCK Threshold for Coulomb fitting in DF KS default MIN THRAO SCF 1 d 2 1 d 12 11 1 3 Parameters to enable local fitting Local fitting as described in H J Werner F R Manby and P J Knowles J Chem Phys 118 8149 2003 Polly H J Werner F R Manby and Peter J Knowles Mol Phys 102 2311 2004 and M Sch tz H J Werner R Lindh and F R Manby J Chem Phys 121 737 2004 can be activated by setting LOCFIT 1 By default local fitting is disabled because under certain circumstances it can lead to unacceptable errors For instance local fitting must not be used in counter poise calculations since the lack of fitting functions at the dummy atoms can lead to wrong results
394. needed A 3 11 Installation of documentation The documentation is available on the web at http www molpro net info users It is also included with the source code The PDF user s manual is found in the di rectory molpro2006 1 doc manual pdf with the HTML version in the directory molpro2006 1 doc manual top level file is manual htm1 The documentation can be copied to its final destination as specified in the CONF IG file generated by the configure command To install the documentation and interactive basis set tool issue make install in the doc directory Numerous example input files are included in the manual and can alter natively be seen in the directory molpro2006 1 examples B RECENT CHANGES 316 B Recent Changes B 1 New features of MOLPRO2006 1 There are very many new features and enhancements in MOLPRO version 2006 1 most notably efficient density fitting methods explicitly correlated methods local coupled cluster methods and several new gradient programs following 1 2 10 11 12 13 14 15 16 17 18 19 20 More consistent input language and input pre checking More flexible basis input allowing to handle multiple basis sets New more efficient density functional implementation additional density functionals Low order scaling local coupled cluster methods with perturbative treatment of triples excitations LCCSD T and variants like LQCISD T Efficient density fit
395. neighbouring geometry and should then specify the SCF orbitals calculated at the present geometry If a subsequent gradient calculation is performed with this wavefunction record file is mandatory and must specify closed shell SCF orbitals at the present geometry Note that record must be larger than 2000 If the FROZEN card is omitted then the numbers of core orbitals are taken from the most re cent MCSCF calculation or otherwise no orbitals are frozen If the FROZEN card is given as 20 THE MCSCF PROGRAM MULTI 114 FROZEN record file then the orbitals corresponding to atomic inner shells are taken i e 1s for Li Ne 1s2s2p for Na Ar etc A FROZEN card without any specification resets the number of frozen core orbitals to zero 20 2 3 Closed shell orbitals CLOSED n no ng ni is the number of closed shell orbitals in irrep number i inclusive of any FROZEN orbitals These orbitals do not form part of the active space i e they are doubly occupied in all CSFs In contrast to the core orbitals see FROZEN these orbitals are fully optimized If the CLOSED card is omitted then the data defaults to that of the most recent MCSCF calcu lation or else the atomic inner shells as described above for FROZEN 20 2 4 Freezing orbitals FREEZE orb sym The specified orbital will not be optimized and will remain identical to the starting guess orb sym should be an active or closed shell orbital If orb sym is a frozen core orbital thi
396. neralized transformation needed in each CCSD iteration If given this is used as a default for THREST_DCCSD THRINT_DCCSD and THRPROD_DCCSD described below Prescreening threshold for DCCSD transformation Defaults THR_DCCSD THREST_DTRAF THR DTRAF THREST default Integral threshold for DCCSD transformation Defaults THR DCCSD THRINT_DTRAF THR DTRAF THRINT default Product threshold for DCCSD transformation Defaults THR DCCSD THRPROD DTRAF THR DTRAF THRPROD default Initial value for THREST DCCSD in CCSD calculations The threshold will be reduced to THREST DCCSD once a certain accuracy has been reached see VARRED or latest after MAXRED iterations The initial thresholds THRINT_DCCSD and THRPROD_DCCSD are obtained by multiplying their input or default values by THRMAX_DCCSD THREST_DCCSD with the restriction that the initial values cannot be smaller than the final ones Specific options for direct MP2 DMP 2 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 60 DMP2 THR DMP2 THREST_DMP2 THRINT_DMP2 THRPROD DMP2 Selects the transformation method for direct MP2 DMP2 1 automatic selection depending on the available memory DMP 220 use fully direct method for DMP2 min two integral evaluations possibly multipassing no disk space DMP 2 1 use semi direct method for DMP 2 one to four inte gral evaluations depending on PAGE DTRAF DMP 222 use DKEXT to compute exchange operators in
397. ng specifications as explained in section 4 11 19 7 4 Selecting the fock matrix FOCK FOCK record file This specifies a record holding a Fock operator to be used for ordering the orbitals Note that only SCF dump records hold fock operators Default is the Fock operator from the energy calculation which produced the input orbitals 19 ORBITAL LOCALIZATION 111 19 7 5 Selecting a density matrix DENSITY DENSITY record file specifications This specifies a record holding a density matrix for construction of a fock operator used for ordering the orbitals This can be used if no fock operator is available and has only an effect for MCSCF localizations By default the state averaged MCSCF density is used A state specific density matrix can be selected using specifications as described in section 4 11 19 8 Localization thresholds THRESH THRESH thresh eorder thresh is a threshold for localization default 1 d 12 If eorder is nonzero default 1 d 4 the orbitals whose energy difference is smaller then eorder are considered to be degenerate and reordered according to the position of their charge centres see section 19 7 19 9 Options for PM localization PIPEK Some special options exist for Pipek Mezey localization all optional PIPEK METHOD method DELETE ndel MAXD L maxdl T HRESH thresh ORDER iorder STEP step METHOD method 1 use 2x2 rotation method default method 2 use Newton Raphson method method 3 In
398. ng to atomic charges In both cases atoms with charges greater than CHGMAX are always included and atoms with the same charges are added as groups Determines the method to determine atomic charges MULLIKEN 0 default squares of diagonal elements of S C are used L wdin charges MULLIKEN 1 Mulliken gross charges are used The first choice is less basis set dependent and more reliable with diffuse basis sets If number is greater than zero all orbital domains containing number or more atoms in common are merged number 1 is treated as num ber 2 default 0 This is particularly useful for geometry optimiza tions of conjugated or aromatic systems like e g benzene In the lat ter case MERGEDOM 1 causes the generation of full t domains i e the domains for all three x orbitals comprise all carbon basis func tions Note that the merged domains are generated after the above print of orbital domains and information about merged domains is printed separately See section 28 9 7 for further discussion of geom etry optimizations There are some other options which should normally not be modified DELBAS ibaso This parameter determines the method for eliminating redundant functions of pair domains ibaso 0 The space of normalized eigenvectors of SU which correspond to small eigenvalues is eliminated default Any other 28 LOCAL CORRELATION TREATMENTS 185 value is not recommended and not further documented DELCOR nshell
399. ngle reference calculations like HF MP2 CCSD RCCSD the AUTO option can be safely used and is recommended However it should be noted that SYMM AUTO cannot be used for MRCI calculations since the MRCI energy is slighly different with and without symmetry this is due to first order interacting space restrictions and can be avoided using REF cards see secion 21 2 6 Furthermore certain input which depends on orbital occupations or symme try labels cannot be used in frequency calculations with symmetry for instance the use of RESTRICT SELECT REF PROJECT LOCAL state averaged MCSCF will lead on an error unless the calculation is performed in Cj symmetry NOSYM option in the geometry input If the energy second derivatives of a given wavefunction have been calculated numerically or analytically in a previous FREQUENCIES run the frequency calculation can be restarted from a given frequencies record rec on file fil using the command FREQUENCIES START irec ifil If no irec ifil is given information is recovered from the latest FREQUENCIES calculation By default frequency information is saved in record 5300 on file 2 After completion of the frequency calculation the normal modes and frequencies are dumped to record 5400 on file 2 This default record can be changed using the DUMP option The normal modes stored in this record can be visualized using MOLDEN see PUT command section 12 4 By default imaginary and low frequency modes a
400. nsity for all states of the given spin is used 20 THE MCSCF PROGRAM MULTI 120 SAVE record Request to save the civector s to the specified record ORBITAL record Request to save the orbitals to the specified record same effect as specifying record as first agrument see above PRINT nvirt Request to print nvirt virtual orbitals in each symmetry By de fault the orbitals are not printed unless the ORBPRINT option see section 20 8 I is present or the global GPRINT ORBITALS see section 6 12 directive has been given before The PRINT option on this card applies only to the current orbitals Several NATORB CANORB and LOCORB cards for different states may follow each other In contrast to earlier versions of MOLPRO the different orbital sets can all be stored in one dump record but different records still work See section 4 11 for information about dump records and how specific orbital sets can be requested in a later calculation 20 5 6 Pseudo canonical orbitals CANORB record options or CANONICAL record options Request to canonicalize the final orbitals and writing them to record record All options have the same effect as described for NATORB 20 5 7 Localized orbitals LOCORB record options or LOCAL record options Request to localize the final orbitals and writing them to record record All options have the same effect as described for NATORB Note LOCAL is interpreted by MULTI but LOC
401. nt MCSCF or CI calculations this also applies across restarts Furthermore nelec defaults to the sum of the nuclear charges irrep to 1 and spin to 0 or 1 thus in many cases it is not necessary to specify a WF card at all 4 10 Defining orbital subspaces In the SCF MCSCF and CI programs it may be necessary to specify how many orbitals in each symmetry are occupied or internal in CI and which of these are core or closed shell doubly occupied in all CSFs This information is provided on the OCC CORE and CLOSED cards in the following way OCC m m mg CORE Cc04 CO2 C08 CLOSED cl ch clg FROZEN fri fro frg where m is the number of occupied orbitals including core frozen and closed co the number of core orbitals and cl is the number of closed shell orbitals including the core orbitals in the 4 GENERAL PROGRAM STRUCTURE 17 irreducible representation i In general m gt cl and cl gt co It is assumed that these numbers refer to the first orbitals in each irrep FROZEN only exists in the MCSCF program and denotes frozen core orbitals that are not optimized note that in older MOLPRO versions frozen core orbitals were denoted CORE Note that the OCC and CLOSED cards have slightly different meanings in the SCF MCSCF and CI or CCSD programs In SCF and MCSCE occupied orbitals are those which occur in any of the CSFs In electron correlation methods CI MPn CCSD etc however OCC denotes the
402. nt of planar molecules in C symmetry But note that neither the multipole program nor the density fitting programs support symmetry at all so choose always C symmetry for DF calculations or with the MULTP option To turn off symmetry specify NOSYM as the first line of your geometry input e g geometry nosym O1 H1 01 roh H2 01 roh h1 hoh Alternatively add SET ZSYMEL NOSYM before the geometry block 2 Use NOORIENT We recommend to use the NOORIENT option in the geometry input to avoid unintended rotations of the molecule when the geometry changes This is particularly im portant for geometry optimizations and for domain restarts in calculations of interaction energies see section 28 9 8 28 9 3 Localization By default Pipek Mezey localization is used and performed automatically in the beginning of a local correlation calculation Thus df hf Hartree Fock with density fitting d 1mp2 ILMP2 using the Pipek Mezey LMOs is equivalent to df hf Hartree Fock with density fitting locali pipek Orbital localization using the Pipek Mezey criterion d 1mp2 ILMP2 using the Pipek Mezey LMOs Boys localization can be used as well but in this case the localization must be done beforehand e g df hf Hartree Fock with density fitting locali boys Orbital localization using the Boys criterion df 1mp2 ILMP2 using the Boys LMOs 28 LOCAL CORRELATION TREATMENTS 192 Poor localization is sometimes an intrinsic proble
403. nted Hessian procedure This refers to the scaled parameter space default 0 5 The initial step size is stepmx see STEP card 39 2 14 Setting a cut parameter CUT CUT threshold Specifies a threshold for ortho normalization used in conjugate gradient update of hessian de fault 1 d 3 see also UPDATE 39 GEOMETRY OPTIMIZATION OPTG 263 39 2 15 Line searching LINESEARCH LINESEARCH iflag thrimin thrlmax Interpolate the geometry of the stationary point minimum or saddle point by a quartic poly nomial between the current and the previous geometry If iflag 0 or no iflag is set the next optimization step will be taken from the interpolated geometry using the interpolated energy and gradient If iflag 1 the energy and gradient will be recalculated at the interpolated geometry before taking the new optimization step Note though that the additional effort of recalculating the energy and gradient is usually not met by the increase of the convergence rate of the optimiz ation thrimin and thrimax are min and max thresholds for the recalculation of the energy and the gradient in case iflag 1 Le the recalculation just takes place if the interpolated geometry isn t too close to the actual geometry thrlmin and isn t too remote from the actual geometry thrlmax Default values are thrimin 0 001 and thrimax 0 05 in the scaled parameter space of the optimization 39 2 16 Reaction path following options OP TION OPTION Kkey param wh
404. nts and transition Hamiltonian between CASSCF and MRCI wavefunc tions with different orbitals 14 Douglas Kroll Hess Hamiltonian up to arbitrary order 15 16 17 18 19 20 iv A new spin orbit integral program for generally contracted basis sets Improved procedures for geometry optimization and numerical Hessian calculations in cluding constrained optimization Improved facilities to treat large lattices of point charges for QM MM calculations in cluding lattice gradients An interface to the MRCC program of M Kallay allowing coupled cluster calculations with arbitrary excitation level Automatic embarrassingly parallel computation of numerical gradients and Hessians mppx Version Additional parallel codes e g DF HF DF KS DF LCCSD T partly including triples Future enhancements presently under development include Automatic calculation of anharmonic vibrational spectra using vibrational CI Coupling of DFT and coupled cluster methods Open shell local coupled cluster methods Explicitly correlated local coupled cluster methods Local response methods CC2 EOM CCSD for computing excitation energies and tran sition properties in large molecules Analytical energy gradients for CCSD T Analytic second derivatives for DFT These features will be included in the base version at later stages The above list is for infor mation only and no representation is made that any of the above will
405. o export MPI opt mpich or equivalent export PATH SPATH MPI bin export MPI_LIB SMPI lib export MPI_INCLUDE SMPI include export LIBMPI 1mpich make TARGET FC USE MPI yes The details will vary from system to system Whenrunning configure mpp mpptype mpi you should specify the location of the GA libraries and mpirun when prompted When asked for the location of the MPI library Please give both th L and 1 loader options needed to access the MPI library it is necessary to give both the directory and library name even if the library would be found automatically by the linker for example L opt mpich lib lmpich where the directory opt mpich lib will vary between platforms If any extra li braries are needed to link in the MPI library then they should not be specified here but manually added to the LIBS entry in CONF IG After configure you should see some thing similar to this in your CONF IG file MPI LIB L opt mpich lib lmpich MPPNAME mpi MPITYPE mpich MPIBASEDIR opt mpich A INSTALLATION OF MOLPRO 308 3 If any system libraries are in unusual places it may be necessary to specify them explicitly as the arguments to a L command line option 4 configure asks whether you wish to use system BLAS subroutine libraries MOLPRO has its own optimised Fortran version of these libraries and this can safely be used On most machines however it will be advantageous to use a system tuned
406. ocal calculation For such checking the option DOMONLY 1 can be used to stop the calculation after the domain generation The orbital domains consist of all basis functions for a subset of atoms These atoms are selected so that the domain spans the corresponding localized orbital with a preset accuracy alterable with option THRBP A typical domain output here for water looks like this Orbital domains Orb Atom Charge Crit 2s 1 01 Lez 0 00 3 H2 0 84 1 00 344 1 O1 202 1 00 4 1 1 01 1 96 1 00 5 1 101 1 17 0 00 2 H1 0 84 1 00 This tells you that the domains for orbitals 2 1 and 5 1 comprise the basis functions of the oxygen atom and and one hydrogen atom while the domains for orbitals 3 1 and 4 1 consist of the basis function on oxygen only The latter ones correspond to the oxygen lone pairs the former to the two OH bonds and so this is exactly what one would expect For each domain of AOs corresponding projected atomic orbitals PAOs are generated which span subspaces of the virtual space and into which excitations are made Options which affect the domain selection are described in section 28 6 Improper domains can result from poorly localized orbitals see section 28 9 3 or a forgotten NOSYM directive This does not only negatively affect performance and memory requirements but can also lead to unexpected results The default for the selection criterion THRBP is 0 98 This works usually well for small basis sets like cc pVDZ
407. occupation numbers in each symmetry Normally this works well in closed shell cases but sometimes wrong occupations are obtained or the wavefunction alternates between different orbital spaces In such cases the OCC directive must be used to force convergence to the desired state The default behaviour can be modified either by options on the command line or by directives In open shell cases we recommend to use the WF OCC CLOSED or OPEN cards to define the wavefunction uniquely Other commands frequently used are START and ORBITAL or SAVE to modify the default records for starting and optimized orbitals respectively The SHIFT option or directive allows to modify the level shift in the RHF program and EXPEC to calculate expectation values of one electron operators see section 6 13 17 1 Options In this section the options for HF RHF UHF are described For further options affecting Kohn Sham caluculations see section 18 For compatibility with previous MOLPRO versions options can also be given on subsequent directives as described in later sections 17 1 1 Options to control HF convergence ACCU RACY accu Convergence threshold for the density matrix square sum of the density matrix element changes Tf accu gt 1 a threshold of 10 accu is used The default depends on the global ENERGY threshold ENERGY thrden The convergence threshold for the energy The default depends on the global ENERGY threshold START record Re
408. ociated to HO or HO1 is internally stored 43 MATRIX OPERATIONS 295 43 2 6 Loading the kinetic or potential energy operators LOAD name EKIN LOAD name EPOT loads the individual parts of the one electron hamiltonian in the AO basis EPOT is summed for all atoms The nuclear energy is associated to EPOT and internally stored The keyword EKIN EPOT needs not to be given if name EKIN EPOT 43 2 7 Loading one electron property operators LOAD name OPER opname isym x y Z loads one electron operator opname where opname is a keyword specifying the operator a component must be given See section 6 13 for valid keys isym is the total symmetry of the operator default 1 and x y z is the origin of the operator If the operator is not available yet in the operator record it is automatically computed The nuclear value is associated internally to name and also stored in variable OPNUC this variable is overwritten for each operator which is loaded but can be copied to another variable using the SET command Note that the electronic part of dipole and quadrupole operators are multiplied by 1 43 2 8 Loading matrices from plain records LOAD name TRIANG record isym LOAD name SQUARE record isym Loads a triangular or square matrix from a plain record not a dump record or operator record If isym is not given 1 is assumed 43 3 Saving matrices SAVE SAVE name record type At present type can be DENSITY
409. of commonly used orbital basis sets is available which can be extended as required There is a comprehensive users manual which includes installation instructions The manual is available in PostScript PDF and also in HTML for mounting on a Worldwide Web server New methods and enhancements in Version 2006 1 include 1 More consistent input language and input pre checking 2 More flexible basis input allowing to handle multiple basis sets 3 New more efficient density functional implementation additional density functionals 4 Low order scaling local coupled cluster methods with perturbative treatment of triples excitations LCCSD T and variants like LOCISD T 5 Efficient density fitting DF programs for Hartree Fock DF HF Density functional Kohn Sham theory DF KS Second order Mpller Plesset perturbation theory DF MP2 as well as for all local methods DF LMP2 DF LMP4 DF LQCISD T DF LCCSD T 6 Analytical QCISD T gradients 7 Analytical MRPT2 CASPT2 and multi state CASPT2 gradients using state averaged MCSCF reference functions 8 Analytical DF HF DF KS DF LMP2 and DF SCS LMP2 gradients 9 Explicitly correlated methods with density fitting DF MP2 R12 2A DF MP2 F12 2A as well as the local variants DF LMP2 R12 2 A loc and DF LMP2 F12 2 A loc 10 Multi state MRPT2 MS CASPT2 11 Coupling of multi reference perturbation theory and configuration interaction CIPT2 12 DFI SAPT 13 Transition mome
410. oided using the NOCHECK option on the CCSD command see also CCSD T 25 1 Options for EOM Normally no further input is needed for the calculation of excitation energies EOM CCSD amplitudes can be saved using SAVE record ifil The vectors will be saved after every refreshing of the iteration space and at the end of the calculation The calculation can be restarted from the saved vectors if START record ifil is specified The set of vectors to be computed can be different in old and restarted calculations However if both cards SAVE and START are specified and the records for saving and restarting are identical the sets of vectors should be also identical otherwise chaos The identical SAVE and START records can be useful for potential energy surfaces calculations see section 25 4 1 By default only excitation energies are calculated since the calculation of properties is about two times as expensive as the calculation of energies only The one electron properties and transition moments expectation type as defined in J F Stanton and R J Bartlett J Chem Phys 98 7029 1993 can be calculated by adding TRANS 1 to EOM card The CCSD ground state is treated as a special case If RELAX option is specified in EXPEC card also the relaxed one electron density matrix is calculated for the ground state Currently the relaxed CCSD density matrix is available for all electron calculations only By default dipole moments are calculated
411. ol Phys 78 1993 997 Each type of grid specifies a family of which the various members are characterized by a sin gle quantum number spherical harmonics up to degree are integrated exactly min and Imax i 0 1 2 3 specify allowed ranges of for hydrogen helium first row second row and other elements respectively For the Lebedev grids if the value of is not one of the set imple mented in MOLPRO 3 5 7 9 11 13 15 17 19 23 29 41 47 53 then is increased to give the next largest angular grid available In general different radial points will have different and in the absence of any moderation described below will be taken from 7 crowd is a parameter to control the reduction of the degree of quadrature close to the nucleus where points would otherwise be unnecessarily close together larger values of crowd mean less reduction thus larger grids A very large value of this parameter or conventionally setting it c to zero will switch off this feature acca is a target energy accuracy It is used to reduce for a given radial point as far as possible below 1 but not lower than The implementation uses the error in the angular integral of the kernel of the Slater Dirac exchange functional using a sum of approximate atomic densities If acca is zero the global threshold is used instead or else it is ignored 18 3 4 Atom partitioning of integration grid VORONOI VORONO1 My Controls Becke Voronoi
412. ol ot ht eek at 32 2 8 Notesandreferences 2 2 2 LORI RU NIE MU RR SU RE EUER IE 32 3 Mulliken population analySiS lens 32 3 1 Calling the population analysis program POP 32 3 2 Defining the density matrix DENSITY 32 3 3 Populations of basis functions INDIVIDUAL aaa a a ac a4 32 4 Finite field calculations ee 32 4 1 Dipolefielis DIP ln 324 2 Quadrupole fields QUAD o e 3243 General fields FIELD gt s ser eam ernes er eere a Era a ah HO EG aea ega apa A Se 32 5 Relativistic corrections a a ss s s sssr d ra dadam aaa ana a PME a e he a e e e Dm 32 6 CUBE dump density or orbital values 32 6 1 STEP setting the point spacing ooo 32 6 2 DENSITY source of density leen 32 6 3 ORBITAL source of orbital a aa 32 6 5 BRAGG spatialexentofgrid llle 32 6 6 ORIGIN centroidofgrid llle 32 67 TITLE user defined itle llle 32 6 8 DESCRIPTION user defined description 32 69 Formatofcubefile o 2 200002 00000 32 7 GOPENMOL calculate grids for visualization in gOpenMol 33 DIABATIC ORBITALS 34 NON ADIABATIC COUPLING MATRIX ELEMENTS 34 1 The DDR procedure es 35 QUASI DIABATIZATION 36 1 Structure of the Input eA TTD 36 2 1 T
413. on save scf energy for this geometry Ido CCSD T calculation save CCSD energy save CCSD T energy lend of do loop ith lend of do loop irl end of do loop ir2 examples h20 pes ccsdt com ccsd eccsdt produce a table with results modify column headers for table save the table in file h2o tab basis Sbasis title for table sort table The next example shows how to loop over many methods SRevision 2006 h20 benchmark 0 method hf fci ci cepa 0 cepa 1 cepa 2 cepa 3 mp2 mp3 mp4 qci ccsd bccd qci t ccsd t bccd t casscf mrci acpf basis dz Double zeta basis set geometry 0o hl o r h2 0 r hl theta Z matrix for geometry r 1 ang theta 10 do i 1 method Smethod i e i energy enddo escf e 1 efci e 2 table method e 4 Title for table title Results for H20 Ser fci Geometry parameters Loop over all requested methods call program save energy for this method examples h20 manymethods co scf energy fci energy print a table with results basis basis R r Ang Theta Stheta degr 6 7 Branching F ELSE F ENI F IF blocks and IF ELSEIF blocks can be constructed as in FORTRAN 6 7 1 IF statements IF blocks have the same form as in Fortran 6 PROGRAM CONTROL 31 IF logical expression THEN Statements ENDIF If only one statement is needed the one line form IF logical expression statemen
414. on criterion NONORM 2 determines whether projected functions are normalized LMP2ALGO MP2ALGO 3 if nonzero use low order scaling method in LMP22 iterations OLDDEF 0 allows to revert to older defaults T1DISK 10 maximum disk space in GByte for T1 caching algorithm Thresholds THRBP 0 98 Threshold Boughton Pulay method THRPIP 1 d 12 Threshold for Pipek Mezey localization THRORB 1 d 6 Threshold for eliminating projected orbitals with small norm THRLOC 1 d 6 Threshold for eliminating redundant projected orbitals THRCOR 1 d 1 Threshold for eliminating projected core orbitals THRMP2 1 d 8 Threshold for neglecting small fock matrix elements in the LMP2 iteration 28 LOCAL CORRELATION TREATMENTS 181 28 4 Summary of directives The same standard directives as in the canonical programs e g OCC CLOSED CORE WF ORBITAL are also valid in the local methods In addition there are some directives which only apply to local calculations LOCAL Invokes local methods and allows to specify the same options as on the command line MULTP As LOCAL but multipole approximations are used for distant pairs DOMAIN Define domains manually not recommended MERGEDOM Allows to merge domains REGION Allows to select regions of a molecule to be treated at a certain level of theory ENEPART Analysis of pair energies SAVE Save domains and LCCSD amplitudes START Restart with domains and LCCSD amplitudes from a previous calcu
415. ong form of above Print information about the initial approximate hamiltonian ipr 2 print the approximate hamiltonian used to find the first approx imation Print information about effective Hamiltonian ipr 2 print columns of effective hamiltonian and overlap matrix in each iteration Print information about residual vectors ipr 1 no print in iteration ipr 0 print energy values residuum norm squared for each itera tion default ipr 1 also print warning about complex eigenvalue and a warning when no new vectors is added to the trial space due to the too small norm of the residuum vector ipr 2 also print how many vectors are left ipr 1 prints overlaps of sample and tested vectors in each iteration if FOLLOW card is present Increasing ipr switches on more and more printing mostly related to the local EOM CCSD method if ipr 1 do a population analysis of the singles part of the rhs EOM CCSD wave function By default the L wdin method is used The Mulliken analysis can be forced by adding MULLPRINT 1 to EOM card Note that a more correct but more expensive approach is to cal culate and analyse the EOM CCSD density matrix see section 5 1 Print intermediates dependent on ground state CCSD amplitudes ipr 0 no print default ipr 1 print newly created intermediates ipr 2 also print more intermediates related information 25 4 1 PES for lowest excited states for hydrogen fluride This example shows how to calc
416. op of the dynamic mem ory and the available memory decreases by the size of the buffers The MEMORY card must therefore be presented before the first FILE card Examples FILE 1 H20 INT allocates permanent file 1 with name H20 INT Previous information on the file 1s recovered FILE 2 H20 WFU NEW allocates permanent file 2 with name H20 WFU All previous infor mation on the file is erased Note that filenames are converted to lower case on unix machines 7 3 DELETE DELETE filel file2 Deletes the specified files file refers to the logical MOLPRO file numbers as specified on the FILE card 7 3 ERASE ERASE filel file2 Erases the specified files file refers to the logical MOLPRO file numbers as specified on the FILE card 7 FILE HANDLING 40 7 4 DATA The DATA command can be used to modify the MOLPRO binary files UNIT Alias for NPL should never be used RENAME recl rec2 used to rename rec to rec2 recl and rec2 must be given in the form name ifil where ifil is the number of a MOLPRO binary file alias for NAME TRUNCATE nen used to truncate files after nen 1 records alias for NEN TRUNCATE rec used to truncate before record rec rec must be given in the form name ifil where ifil is the number of a MOLPRO binary file COUNT Alias for NRE presently not used COPY recl rec2 Copies record rec to rec2 recl and rec2 must be given in the form naml ifill nam2 ifil2 If na
417. operator IPROCI 0 Default Calculation uses uncontracted internals with RS2 IPROCI 1 Internals with two holes in the inactive space are internally con tracted in RS2 using a projection operator IPROCS 3 IPROCI 1 This combination of options reproduces with RS2 the RS2C result using projection operators This requires lot of memory and disk space and it is feasible only for small molecules IFDIA 0 Default All off diagonal elements of the effective Fock ma trix are included IFDIA 1 The internal external block of the Fock matrix is neglected This eliminates the single pair coupling IFDIA 2 All off diagonal elements of the Fock matrix are neglected This corresponds to CASPT2D of Andersson et al Note in this case the result is not invariant to rotations among active orbitals IHINT 0 Default Only one electron integrals are used in the zeroth order Hamiltonian for all interactions IHINT 1 The all internal two electron integrals are used in the zeroth order Hamiltonian for the internal internal and single single in teractions IHINT 2 The all internal two electron integrals in the zeroth order Hamil tonian are used for the internal internal single single and pair pair interactions Using IHINT 2 and IDFIA 1 corresponds to Dyall s CAS A method for the case that CASSCF references 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 157 NOREF 1 NOREF 0
418. operators Examples 3 44 yields 1 4 3 4 4 yields 1 4 3 4 4 yields 3 4 4 3 4 4 yields 3 4 4 3 4 4 yields 3 4 4 Expressions including numbers may contain variables Examples for the use of data geometry and basis input LATTICE OCC CLOSED CORE WF directives In some cases several lines of data are needed for a certain command or directive in such cases the data must follow directly the corresponding command directive and must be enclosed in square brackets COMMAND options data Normally the input format of data is MOLPRO style i e numbers are separated by commas and variables as well as expressions can be used If data are included using external files the input format of data is free format no commas are needed but no variables and expressions can be used 3 7 Expressions In any input field data can be entered in the form of expressions Numbers and variables are special cases of expressions An expression is typed in Fortran style and may contain any num ber of nested parenthesis The standard intrinsic functions are also available see next section 3 DEFINITION OF MOLPRO INPUT LANGUAGE 9 MOLPRO understands both arithmetic and logical expressions The result of an arithmetic ex pression is a real double precision number Internally all integers are also converted to real numbers The result of a logical expression is either TRUE or FALSE Internally TRUE is sto
419. or a group number The ECP specification may consist either of a single keyword which references a pseudopotential stored in the library or else of an explicit definition extending over several input cards cf below 14 1 Input from ECP library The basis set library presently contains the pseudopotentials and associated valence basis sets by a the Los Alamos group P J Hay and W R Wadt J Chem Phys 82 270 1985 and following two papers and b the Stuttgart K ln group e g A Nicklass M Dolg H Stoll and H Preu J Chem Phys 102 8942 1995 for more details and proper references see the web page http www theochem uni stuttgart de pseudopotentials Pseu dopotentials a are adjusted to orbital energies and densities of a suitable atomic reference state while pseudopotentials b are generated using total valence energies of a multitude of atomic states Library keywords in case a are ECP1 and ECP2 ECP2 is used when more than one pseudopo tential is available for a given atom and then denotes the ECP with the smaller core definition For Cu e g ECP1 refers to an Ar like 18e core while ECP2 simulates a Ne like 10e one with the 3s and 3p electrons promoted to the valence shell For accurate calculations including electron correlation promotion of all core orbitals with main quantum number equal to any of the valence orbitals is recommended Library keywords in case b are of the form ECPnXY n is the nu
420. orbital analysis of the expectation values is printed the density matrix must also be provided If record file is omitted the last dump record is used This is only meaningful for diagonal density matrices SCF or natural orbitals Natural orbitals for specific states can be selected using specifications as explained in section 32 1 4 Specification of one electron operators The required operators are specified by code words Optionally the geometry or the nuclear centre at which the operator is computed can be specified For each operator an input card of the following form is required code centre x y z factor code specifies the property The available operators are given in section 6 13 The other parameters have the following meaning centre row number of Z matrix or atomic symbol defining the centre at which property shall be calculated if centrez 0 you need not read in coordinates XYZ cartesian coordinates of the point only if centre 0 32 PROPERTIES AND factor EXPECTATION VALUES 207 the operator is multiplied by this factor The default is factor 1 except for REL In this cases proper factors for relativistic corrections are used unless factor is given The two commas before factor are needed to preserve compatibility with Molpro 96 32 1 5 Printing options PRINT print This card is used to control output mainly for debugging purposes print 0 no test output default print gt 0 Operators are p
421. orithm turns out to be computationally more expensive than the one selected with DTRAF 1 Note that neither DTRAF 1 nor DTRAF 2 work in the context of LMP2 gradients General threshold for generation of 2 external integrals in lin ear scaling LMP2 If given this is used as a default for all LMP 2 thresholds described below Prescreening threshold for generation of 2 external integrals Defaults THR LMP2 THREST DTRAF THR_DTRAF THREST default Threshold used in the first quarter transformation Defaults THR_LMP2 THRPROD DTRAF THR DTRAF THRPROD default 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 61 THRO2_LMP2 THRAO ATTEN Threshold used in the second and subsequent quarter transfor mations Defaults THR_LMP2 THRINT_DTRAF THR DTRAF THRINT default Special threshold for prescreening of attenuated integrals uu vv Default THREST_LMP2 Options for integral direct computation of external exchange operators DKEXT DKEXT SCREEN DKEXT MAXSIZE_DKEXT MINSIZE_DKEXT MAXCEN_DKEXT SCREEN DKEXT PRINT_DKEXT SWAP _DKEXT MXMBLK_DKEXT Selects driver for DKEXT DKEXT 1 use paging algorithm minimum memory This is automatically used if in core algorithm would need more than one integral pass DKEXT 0 use in core algorithm no integral triples DKEXT 1 use in core algorithm and integral triples DKEXT 2 use in core algorithm and integral triples if at least t
422. ost cases no parameters need to be specified on the DIRECT directive However in order to guaran tee sufficient accuracy the default thresholds are quite strict and in calculations for extended systems larger values might be useful to reduce the CPU time The format of the DI RECT directive is DIRECT keyl valuel key2 value2 The following table summarizes the possible keys and their meaning The default values are given in the subsequent table In various cases there is a hierarchy of default values For in stance if THREST D2EXT is not given one of the following is used THR D2EXT THREST_DTRAF THR DTRAF THREST default The list in brackets is checked from left to right and the first one found in the input is used default is a default value which depends on the energy threshold and the basis set the threshold is reduced if the overlap matrix contains very small eigenvalues General Options apply to all programs THREST Integral prescreening threshold The calculation of an integral shell block is skipped if the product of the largest estimated in tegral value based on the Cauchy Schwarz inequality and the largest density matrix element contributing to the shell block is 10 INTEGRAL DIRECT CALCULATIONS GDIRECT 57 smaller than this value In DTRAF and DKEXT effective density matrices are constructed from the MO coefficients and ampli tudes respectively THRINT Integral prescreening threshold This applies to the p
423. ot be specified and HO is used by default Since eigenvectors of h are often a very poor starting guess it is recommended to generate the starting orbitals using a small basis like STO 3G see section 17 4 2 below Example 17 THE SCF PROGRAM 95 r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input H1 0 r H2 0 r H1 theta basis STO 3G first basis set hf scf using STO 3G basis basis 6 311G second basis set hf Iscf using 6 311G basis set The second calculation uses the optimized orbitals of the STO 3G calculation as starting guess This is done by default and no START card is necessary The explicit use of START and SAVE cards is demonstrated in the example in the next section The following input is entirely equivalent to the one in the previous section examples h20 sto3gstartl com r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input H1 0 r H2 0 r H1 theta basis STO 3G first basis set hf Iscf using STO 3G basis start atdens luse atomic density guess _ examples save 2100 2 save orbitals to record 2100 2 h2o sto3gstart2 com basis 6 311G second basis set hf scf using 6 311G basis set start 2100 2 start with orbitals from the previous STO 3G calculation save 2101 2 save optimized orbitals to record 2101 2 17 4 2 Starting with previous orbitals START RECORD Jrecord file specifications reads previously optimized orbitals from record record on file fi
424. ou put the procedure SAVE E in a file molproi rc or SHOME molproirc it would be automatically included in all your jobs molproi rc is searched first if this file does not exist molpro looks for SHOME molproirc If this also does not exist molpro uses the default file in the system directory 5 8 Do loops Now you have the idea that one geometry is not enough Why not compute the whole surface DO loops make it easy Here is an example which computes a whole potential energy surface for H20 5 INTRODUCTORY EXAMPLES 26 SRevision 2006 0 H20 potential geometry x luse cs symmetry o z matrix hl o rl i h2 0 r2 i hl theta i basis vdz define basis set angles 100 104 110 list of angles distances 1 6 1 7 1 8 1 9 2 0 list of distances i 0 linitialize a counter do ith 1 tangles loop over all angles H1 0O H2 do irl 1 distances loop over distances for O H1 do ir2 1 ir1 loop over O H2 distances rl ge r2 i i 1 lincrement counter rl i distances irl save rl for this geometry A f examples r2 i distances ir2 save r2 for this geometry h2o pes cesdt com theta 1 angles ith save theta for this geometry hf do SCF calculation escf 1 energy save scf energy for this geometry Coen 2 do CCSD T calculation eccsd i energc save CCSD energy eccsdt i energy save CCSD T energy enddo lend of do loop ith enddo end of do loop irl enddo lend of do loop ir2 table rl r2 theta escf
425. ous thresholds 2 2 0 o ees 142 bast deos a had Sal waa re gi we ett ed eet dd ead as bd ewe as 142 TT Mcr 144 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 146 CONTENTS 22 1 Introduction 22 2 Excitedstatecalculations le 22 3 Multi State CASPT2 cles 22 3 1 Performing SS SR CASPT2 calculations 22 3 2 Performing MS MR CASPI2calculaations ln 22 4 Modified Fock operators in the zeroth order Hamiltonian 225 Tevel shifts io museo e ie th SS acea Gad dL b tissimus 22 6 Integral direct calculations es 22 7 CASPT2 gradients o e o Ra ke 22 8 Coupling MRCI and MRPT2 The CIPT2 method 229 Further options for CASPT2 and CASPT3 llle 23 1 Expectation values for MP2 0 0 000002 e 23 2 Density fitting MP2 DF MP2 RI MP2 o o 23 3 Spin component scaled MP2 SCS MP2 lens 24 THE CLOSED SHELL CCSD PROGRAM 24 1 Coupled cluster CCSD 2 sn 24 2 Quadratic configuration interaction QUCI 2l llle 24 3 Brueckner coupled cluster calculations BCCD 24 3 1 The BRUECKNER directive lens ooo a PLE 245 The DIIS directive a A Da Pa a Dra a Ss ad a ana TORET MM OE 24 7 Saving the density matrix ees 24 8 Naturdlorbital Q 25 1 Options for EOM e 2a ea Ba wa m OR e duode m e a
426. p 300 C DENSITY FUNCTIONAL DESCRIPTIONS 341 C 29 PW86 GGA Exchange Functional See reference for more details K 12E 2ps Em where E n 3 4 3V1 In PF S 302 F S 1 12965 145 0 25 15 303 and X56 VIZ 368 To avoid singularities in the limit p 0 G 1 2E 2p 305 C 30 PW91C Perdew Wang 1991 GGA Correlation Functional See reference for more details K p e Po pg H d Pa Pp gt 306 where Js a 03 d M12 307 u Pa pg V x 1p Td u a B 1 2 1 5 0 8 1 2 1 G o 7 308 H d o B L d o B J d o B 309 342 l A a p a 1 L d a B 1 2 u pa pg A In 2 FA o Br A a 335 gt G10 400 pang st 311 J d 0 B v 6 r o B x 3 7Z u pa pp de Vee x 9 ea A 0 B 220 nog 7 2 312 1 0 09 313 VK 314 v 16 Vm 315 T C DENSITY FUNCTIONAL DESCRIPTIONS 342 x 0 004235 316 Z 0 001667 317 o r 8 r Z 318 al 1000 1 ers a on eae E 23 266 320 d 0 007389 321 A 8 723 322 Y 0 472 323 a B e r B Ti U1 V1 Wi X1 Y1 P1 e r os B T3 Us V3 Ws X Ys P3 0 G a B 1 E B a2 e r o B Ta Us Va Wa Xo Ya Ps e r o B Ti U1 Vi W1 X1 Y1 P1 6 oc B G o5 B sia a 1 r a B 1 4 434 ACEROS 325 _a B C a D a p 326 1 cd a
427. p erator or spin orbit pseudopotentials ECPs The state interacting method is employed which means that the spin orbit eigenstates are obtained by diagonalizing A Aso in a basis of eigen functions of A The full Breit Pauli SO operator can be used only for MCSCF wavefunctions For MRCI wavefunctions the full BP operator is used for computing the matrix elements be tween internal configurations no electrons in external orbitals while for contributions of ex ternal configurations a mean field one electron fock operator is employed The error caused by this approximation is usually smaller than 1 cm The program allows either the computation of individual spin orbit matrix elements for a given pair of states or the automatic setting up and diagonalization of the whole matrix for a given set of electronic states In the latter case matrix elements over one electron operators are also computed and transformed to the spin orbit eigenstates by default the dipole matrix elements are computed other operators can be specified on the GEXPEC or EXPEC cards see section 6 13 Since it may be often sufficient to compute the spin orbit matrix elements in a smaller basis than the energies it is possible to replace the energy eigenvalues by precomputed values which are passed to the spin orbit program by the MOLPRO variable HLSDIAG 37 2 Calculation of SO integrals The one and two electron spin orbit integrals over the BP Hamiltonian can b
428. partitioning of space The algorithm of C W Murray N C Handy and G J Laming Mol Phys 78 1993 997 is used with m defined by equation 24 The default value is 10 18 3 5 Grid caching GRIDSAVE NOGRIDSAVE NOGRIDSAVE disables the disk caching of the grid i e forces the recalculation of the grid each time it is needed GRIDSAVE forces the use of a grid cache where possible 18 3 6 Grid symmetry GRIDSYM NOGRIDSYM NOGRIDSYM switches off the use of symmetry in generating the integration grid whereas 18 THE DENSITY FUNCTIONAL PROGRAM 104 GRIDSYM forces the use of any point group symmetry 18 3 7 Grid printing GRIDPRINT GRIDPRINT key value controls printing of the grid which by default is not done At present the only possible value for key is GRID and value should be specified as an integer GRID 0 causes the total number of integration points to be evaluated and reported GRID 1 additionally shows the number of points on each atom GRID 2 causes the complete set of grid points and weights to be printed 18 4 Density Functionals In the following Pa and pg are the Q and spin densities the total spin density is p The gradients of the density enter through Saa Vpa Vpa Ogg VPg VPg Sup Opa VPa Vpg O Saa Opp 20d vV Oaa v Opp Xo am XB c 6 Pa Pg Va V Pa Up V pg V Va Vp 7 Additionally the kinetic energy density for a set of Kohn Sham orbitals
429. patches Normally the distribution when downloaded is fully up to date and initial patching is not nec essary However bug fixes and updates may be desired subsequently The mechanism for updating MOLPRO source code with bug fixes and new features is through the provision of self contained patch files which when applied replace or add files and store the replaced code in order to allow later reversion to the original Those patches that are available can be seen at http www molpro net patch 2006 1 whilst a list of those already installed is printed when running the program Patch files automatically outdate any targets that need rebuilding as a result of the patch for example relevant object files are removed Thus after all patches have been applied it is usually necessary to rebuild the program using make The order in which patches are applied and removed is important Some patches are prerequi sites of others and some patches are parents of one or more children the parent and child patches have one or more files in common but the parent is older than the child Individual patch scripts will themselves refuse to apply or revert if rules based on these considerations would be violated In order to deal with this issue smoothly a program pat cher is provided to manage the application and removal of one or more patches patcher attempts to sort the order in which patches are applied or reverted so as to avoid such conflicts
430. pecify your username and password through command line options or else the program will prompt for them They are then remembered in the file CONF IG in the cache directory In case of problems first consult the file patcher 1log which contains the output from indi vidual patch applications and reversions The following options can be given cache directory cd location of cache directory verbose v Increase amount of information printed Multiple verbose options can be used noverbose Decrease amount of information printed url URL of web server user uu Username for web server password p p Password for web server noaction n No applications or reversions are actually done Useful for seeing what would happen without doing it local Don t attempt to access the web server but use only local files token k Download your licence key ssl s Use SSL when contacting the webserver nossl i Turn off SSL use Examples patcher Applies all patches that are available but not yet installed This is the normal use of the utility in bringing the copy of the source tree up to date with all available updates patcher 1 Lists installed and available patches patcher r xx yy A INSTALLATION OF MOLPRO 315 Reverts patches xx and yy patcher n Loads all uninstalled patches into the cache for later use patcher local Applies all patches in the cache no network connection
431. pied FROZEN 1 first sigma orbital is doubly occupied and frozen WF 6 1 6 electrons singlet Sigma state SELECT triggers configuration input CON 2 2 2sigma 2 3sigma 2 CON 2 1 1 2sigma 2 3sigma 4sigma CON 2 0 2 2sigma 2 4sigma 2 CON 2 0 0 2 k 2sigma 2 lpi_x 2 CON 2 0 0 0 2 2sigma 2 lpi_y 2 20 THE MCSCF PROGRAM MULTI 117 20 4 4 Selecting the primary configuration set PSPACE thresh The hamiltonian is constructed and diagonalized explicitly in the primary configuration space which can be selected with the PSPACE card The coefficients of the remaining configurations Q space are optimized iteratively using the P space wavefunction as zeroth order approxima tion If thresh is nonzero it is a threshold for automatically selecting all configurations as P space configurations which have energies less then emin thresh where emin is the lowest energy of all configurations Further P space configurations can be specified using CON cards which must follow immediately after the PSPACE card These are merged with the ones selected according to the threshold Automatic selection can be avoided by specifying a very small threshold There is a sensible default value for thresh 0 4 so you usually don t need a pspace card in your input Furthermore if the number of configurations in the MCSCF is less than 20 all configurations go into the P space unless you give a PSPACE card in the input A P space threshol
432. present TWOINT Threshold for the neglect of two electron integrals default 1 d 12 PREFAC Threshold for test of prefactor in TWOINT default 1 d 14 LOCALI Threshold for orbital localization default 1 d 8 EORDER Threshold for reordering of orbital after localization default 1 d 4 6 PROGRAM CONTROL 35 ENERGY GRADIENT STEP ORBITAL CIVEC COEFF PRINTCI PUNCHCI SYMTOL GRADTOL THROVL THRORTH Convergence threshold for energy default 1 d 6 Convergence threshold for orbital gradient in MCSCF default 1 d 2 Convergence threshold for step length in MCSCF orbital optimization default 1 d 3 Convergence threshold for orbital optimization in the SCF program default 1 d 5 Convergence threshold for CI coefficients in MCSCF and reference vector in CI default 1 d 5 Convergence threshold for coefficients in CI and CCSD default 1 d 4 Threshold for printing CI coefficients default 0 05 Threshold for punching CI coefficients default 99 no punch Threshold for finding symmetry equivalent atoms default 1 d 6 Threshold for symmetry in gradient default 1 d 6 Threshold for smallest allowed eigenvalue of the overlap matrix de fault 1 d 8 Threshold for orthonormality check default 1 d 8 6 12 Global Print Options GPRINT NOGPRINT Global print options can be set using the GPRINT command outside the individual programs the first letter G is optional but should be used to avo
433. qe 2 327 w z 3333 327 1 e r t u v w x y p 2t 1 ur ln 1 2 O ET 328 c 1 709921 329 C d o B K Q o B M Q o B 330 M d a B 0 5v 6 r 0 8 x 372z dhe mer 331 1 d N a p d ae em 51992 K d a B 0 250000000022 In H2 N o B 20 1 la 1 p 333 C DENSITY FUNCTIONAL DESCRIPTIONS G N 235 6 V n 1p7 6 Q 1 2 T 0 031091 0 015545 0 016887 U 0 21370 0 20548 0 11125 V 7 5957 14 1189 10 357 W 3 5876 6 1977 3 6231 X 1 6382 3 3662 0 88026 Y 0 49294 0 62517 0 49671 and P 1 1 1 To avoid singularities in the limit p 0 G p ps 0 t C 0 Ps 0 j C 31 PW91X Perdew Wang 1991 GGA Exchange Functional See reference for more details K Y INE 2p where ue 3 3 3 72 E n ya VON Ren F S T 7 1 623 S 1 12 Va and 1 0 19645 Sarcsinh 7 7956 S 0 2743 0 1508 g s Pigs S 1 0 19645 Sarcsinh 7 7956 S 0 004 54 To avoid singularities in the limit p 0 G 1 2E 2p 343 334 335 336 337 338 339 340 341 342 343 344 345 346 347 C DENSITY FUNCTIONAL DESCRIPTIONS 344 C 32 PW92C Perdew Wang 1992 GGA Correlation Functional Electron gas correlation energy See reference 2 for more details K pe PoP 348 where e a B e r a B T U1 Vi W X1 Y1 P1 e
434. qual sign an existing variable of the same name is replaced by the new values and all old values are lost For instance THETA 100 110 120 130 set four values THETA 1 104 replace THETA 1 by anew value THETA 2 4 are unchanged THETA 140 150 old variable THETA is replaced THETA 3 4 are deleted Square brackets can also be used to define an array of strings e g METHOD INT HF CASSCF MRCT These could be used as follows DO I 1 4 SMETHOD 1 ENDDO The above input would be equivalent to INT HF CASSCF MRCI The current length of an array can be accessed by preceding to the variable name For instance in the above examples R and METHOD have the values 5 and 4 respectively If a variable is not defined zero is returned but no error occurs This can be used to test for the existence of a variable for example IF SPIN EQ 0 AND NELEC EQ 1 SET SPIN MOD NELEC 2 This defines variable SP IN if it is unknown and if NELEC is a scalar one dimensional variable 8 VARIABLES 47 8 7 Vector operations The following simple vector operations are possible e Copying or appending a vector to another vector For instance S R copies a vector R to a vector S S 3 R copies R to S 3 S 4 S S 1 R appends vector R to vector S It is also possible to access a range of subsequent elements in a vector S R 2 4 copies elements 2 to 4 of R to S 1 S 2 S 3 Note that R 2 denot
435. r case Also note that quotes are compulsory if the string contains blanks Example str a b 4 This is an example for strings yields STR 1 A STR 2 B 4 STR 3 This is an example for strings As a general rule string variables are replaced by their value only if they are preceded by a dollar exceptions in variable definitions on SHOW cards and in logical expressions on IF cards the dollar is optional This is a precaution to avoid commands which have the same name as a variable being interpreted as variables Variables may also appear on TEXT or TITLE cards or in strings but must be preceded by in these cases Example SMETHOD MCSCF R 1 5 TEXT Smethod results for R R Bohr prints MCSCF results for R 1 5 Bohr String variables can be concatenated with strings or other string variables in the following way Assume that variable PROGRAM has the value MRCT Setting METHOD SPROGRAM Q sets METHOD to MRCI Q Alternatively if we would also have a variable VERSION with value Q we could write METHOD SPROGRAM SVERSION Again the value of METHOD would be MRCI 0 Note that the quotes are necessary in these cases Substring operations are not implemented 8 VARIABLES 44 8 4 System variables As mentioned above most system variables cannot be written by the user In some exceptions it is possible to redefine them using the SET command SET variable expression
436. r coordinates This is the default if the z matrix does not depend on variables or if the xyz input format is used If this option is used and the original geometry is given in z matrix form the z matrix is lost The specification of displacement type is optional and only affects the numerical calculation of the gradient for 3N coordinates It can also be given using DISPLACE displacement type displacement type can be one of the following 38 ENERGY GRADIENTS 249 SYM Use symmetrical displacements This yields the minimum number of displace ments and always preserves the symmetry of the wavefunction This is the default and only recommended option CART Displacements are generated for all 3N Cartesian coordinates This is normally not recommended since in cases in which molecular symmetry is present it gener ates far more displacements than needed Also the wavefunction symmetry is not preserved and the calculation must be done in C1 symmetry UNIQUE As CART but symmetry equivalent displacements are eliminated Not recom mended either 38 2 2 Numerical derivatives of a variable Numerical derivatives of the value of a variable can be computed using VARIABLE name The default is to compute the gradient of the current energy 38 2 3 Step sizes for numerical gradients By default the numerical step sizes are 0 01 bohr for distances or Cartesian coordinates and 1 degree for angles These defaults can be changed using RSTEP
437. r size Naturally such approximation can introduce some errors and therefore the user has to be more careful than with standard black box methods On the other hand the low order scaling makes it possible to treat much larger systems at high levels of theory than it was possible so far This section summarizes some important points to remember when performing local correlation calculations 28 9 1 Basis sets For numerical reasons it is useful to eliminate projected core orbitals since these may have a very small norm By default projected core orbitals are eliminated if their norm is smaller then 0 1 this behaviour can be changed using the DELCOR and THRCOR options For local calculations we recommend the use of generally contracted basis sets e g the correlation con sistent cc pVnZ sets of Dunning and coworkers For these basis sets the core basis functions are uniquely defined and will always be eliminated if the defaults for DELCOR and THRCOR are used 28 LOCAL CORRELATION TREATMENTS 191 The correlation consistent basis sets are also recommended for all density fitting calculations since optimized fitting basis sets are available for each basis 28 9 2 Symmetry and Orientation 1 Turn off symmetry Otherwise you won t get appropriately localized orbitals local orbitals will tend to be symmetry equivalent instead of symmetry adapted Symmetry can be used only if all atoms are symmetry unique This allows the local treatme
438. r this to be possible 36 9 Optimization control 36 9 1 Optimization criterion CRIT method Specifies the criterion for the optimization method can be OVERLAP or ENERGY OVERLAP is default The former maximizes the normalized overlap with the CASSCF wavefunction Pas V vg TE ene and the latter simply minimizes the energy an Spt eat Pvg Yve 36 9 2 Number of iterations MAXITER Niter3 Specifies the maximum number of iterations in the second order optimizations Default is Niter 50 36 9 3 CASSCF projected structure coefficients NO CASPROJ With this keyword the structure coefficients are picked from the transformed CASSCF CI vector leaving only the orbital variational parameters For further details see the bibliography This option may be useful to aid convergence 36 9 4 Saddle point optimization SADDLE n Defines optimization onto an n order saddle point See also T Thorsteinsson and D L Cooper Int J Quant Chem 70 637 50 1998 36 THE VB PROGRAM CASVB 232 36 9 5 Defining several optimizations More than one optimization may be performed in the same CASVB deck by the use of OPTIM keywords OPTIM F INOPTIM The subcommands may be any optimization declarations defined in this section as well as any symmetry or constraints specifications described in section Commands given as arguments to OPTIM will be particular to this optimization step whereas commands specified outside
439. range of bond distances loop over displaced geometries set r2 to current distance same wavefunction definition as at reference geom save new orbitals to record compute diabatic orbitals using reference orbitals stored on record reforb set variable reforb to the new orbitals 34 NON ADIABATIC COUPLING MATRIX ELEMENTS 218 34 NON ADIABATIC COUPLING MATRIX ELEMENTS Non adiabatic coupling matrix elements can be computed by finite differences for MCSCF or CI wavefunctions using the DDR program For state averaged MCSCF wavefunctions they can also computed analytically cf section 20 9 2 Note that present numerical procedure has been much simplified relative to Molpro96 No GEOM and DISPL input cards are needed any more and the three necessary calculations can be done in any order 34 1 The DDR procedure In order to compute the coupling matrix elements by finite differences one has to compute and store the wavefunctions at two first order algorithm or three second order algorithm slightly displaced geometries The order of these calculations is arbitrary The typical strategy is as follows 1 Compute the wavefunction at the reference geometry The wavefunctions for both states have to be stored using the SAVE command of the CI program If the matrix elements are computed for MCSCF wavefunctions it is necessary to recompute the wavefunction with the CI program using the NOEXC option The transition density matri
440. rations are read in from the speci fied record If ref is not specified the program assumes that the configurations are read from subsequent CON cards see CON rec2 gt 2000 Additional configurations are read from the spec ified record If rec2 is negative all records between rec and abs rec2 are read All configurations found in this way are merged ref2 rec2 file refthr Selection threshold for configurations read from disc records recl rec2 This applies to the norm of all CSFs for each or bital configuration refstat Specifies from which state vector the configurations are se lected This only applies to the case that the configurations were saved in a state averaged calculation If refstat is not spec ified the configurations are selected from all states mxshrf max number of open shells in the selected or generated con figurations 20 4 3 Specifying orbital configurations CON N1 N2 N3 N4 Specifies an orbital configuration to be included in the present symmetry The first CON card must be preceded by a SELECT card nj n2 etc are the occupation numbers of the active orbitals 0 1 or 2 For example for OCC D 2 2 CLOSED 27 ley Los n is the occupation of orbital 3 1 number sym n is the occupation of orbital 4 1 n3 of 5 1 n4 of 2 2 and ns of 2 3 Any number of CON cards may follow each other Example for the BH molecule OCC 4 1 ls four sigma one pi orbitals are occu
441. rbitals records for saving CI wavefunction like SAVE card in MCSCF Variables recognized by the CI CCSD program CHARGE NELEC SPIN CISYM METRY SYMMETRY CISTATE STATE Total charge of the molecule can be given instead of nelec number of electrons spin multiplicity minus one wavefunction symmetry If this is an array only SYMMETRY 1 is used as CISYMM only used if CISYMM is not present number of states in CI as CISTATE only used if CISTATE is not present 8 VARIABLES 52 CISELECT records from which configurations can be selected SELECT as CISELECT only used if CI SELCT is not present CIRESTRICT defines occupancy restrictions RESTRICT as RESTRICT only used if CIRESTRICT is not present CIOC C number of occupied orbitals in each symmetry OCC as CIOCC only used if CIOCC is not present CICL OSED number of closed shell orbitals in each symmetry CLOSED as CICLOSED only used if CICLOSED is not present CICO RE number of core orbitals in each symmetry CORE as CICORE only used if CICORE is not present CIORB record of orbitals used in CI CISAVE records for saving CI wavefunction like SAVE card in CI CISTART records for restarting with previous CI wavefunction like START card in CI Variables recognized by the DFT KS program DF ifun orDFTNAME ifun name of ifun th component of density functional DFTFAC ifun factor multiplying i fun th component of density functional DF
442. rdered tables in user supplied format Furthermore there are up to 9 binary MOLPRO 4 GENERAL PROGRAM STRUCTURE 13 files available each one known to the program simply by its number 1 to 9 By default they are temporary files usually allocated dynamically by the program but they can be connected to permanent files with the FILE command Each file is direct access and word addressable word 64 bit usually but is organised in records of any length The name address and length of each record is held in a directory at the start of the file File 1 is the main file holding basis set geometry and the one and two electron integrals By default file 2 is the dump file and used to store the wavefunction information i e orbitals CI coefficients and density matrices File 3 is an auxiliary file which can be used in addition to file 2 for restart purposes Often files 1 and 2 and 3 are declared as permanent files see FILE to enable restarts Storing the wavefunction information on file 2 is useful since the integral file is overwritten at each new geometry while the orbitals and CI coefficients of one calculation can be used as a starting guess for the next calculation at a neighbouring geometry Files 4 to 8 are used as scratch space e g for sorting the integrals storage of transformed integrals and of the CI vectors These files should normally not be made permanent Note that the file name appearing in molpro input is always converte
443. rds for orbitals density matrices or operators B RECENT CHANGES 320 All one electron operators needed to compute expectation values and transition quantities are now stored in a single record Operators for which expectation values are requested can be selected globally for all programs of a given run using the global GEXPEC directive or for a specific program using the EXPEC directive All operators are computed automatically when needed and the user does not have to give input for this any more See section ONE ELECTRON OPERATORES AND EXPECTATION VALUES of the manual for details Due to the changed structure of dump and operator records the utility program MATROP has a new input syntax MOLPRO96 inputs for MATROP do not work any more In addition to these organizational changes a number of new programs have been added An alytic energy gradients can now be evaluated for MP2 and DFT wavefunctions and harmonic vibrational frequencies intensities and thermodynamic quantities can be computed automati cally using finite differences of analytical gradients Geometry optimization has been further improved and new facilities for reaction path following have been added An interface to the graphics program MOLDEN has been added which allows to visualize molecular structures orbitals electron densities or vibrations Integral direct calculations in which the two electron integrals in the AO basis are never stored on disk but always reco
444. re added if either required by select or if configurations are found which are degenerate to the last p space configura tion A minimum number of npspace is automatically deter mined from the state specifications 21 3 10 Canonicalizing external orbitals FOCK n1 N2 5 External orbitals are obtained as eigenfunctions of a Fock operator with the specified occupation numbers n Occupation numbers must be provided for all valence orbitals 21 3 11 Saving the wavefunction SAVE savecp saveco idelcg or SAVE CIVEC savecp CONF I G saveco DENS IT Y dumprec NATORB dumprec F ILES savecp record name for save of wavefunction If negative the wave function is saved after each iteration else at the end of the job In case of coupled cluster methods CCSD QCISD BCCD the wavefunction is saved in each iteration in any case presently only implemented for the closed shell case 21 THE CI PROGRAM 139 saveco record name for save of internal configurations and their maxi mum weight over all states for subsequent use as reference in put see SELECT card If the record already exists the record name is incremented by one until a new record is created idelcg if nonzero or FILES is specified don t erase icfil and igfil holding CI and residual vectors at the end of the calculation dumprec Dump record for saving density matrix and natural orbitals Only one dump record must be given In any case the den sity mat
445. re computed all values are stored in the respective variable arrays with the bra states running fastest 37 4 Calculation and diagonalization of the entire SO matrix HLSMAT type recordl record2 record3 Computes the entire SO matrix and diagonalizes it using all states which are contained in the records recordl record2 record3 All records must have been generated using the SAVE directive of the MRCI program type may be either LS for Breit Pauli calculations or ECP for ECP LS calculations By default the eigenvalues and dipole transition matrix elements between the ground and excited states are printed As with the TRANLS card the HLSMAT is recognized only by the MRCI program and must be preceded by a CI card Also the OCC and CLOSED cards must be the same for all states used in a HLSMAT calculation 37 5 Modifying the unperturbed energies Often it may be sufficient to compute the spin orbit matrix elements in a smaller basis or at a lower computational level than the energies It is therefore possible to replace the energy eigen values by precomputed values which are passed to the spin orbit program by the MOLPRO variable HLSDIAG The energy values in HLSDIAG must be in exactly the same order as the states in the records given on the HLSMAT card Before any spin orbit calculation the variable HLSDIAG must either be undefined or cleared then the original energies are used or must con tain exactly the number of energ
446. re not stored By specifying DUMPALL rather than DUMP all modes are written out 40 VIBRATIONAL FREQUENCIES FREQUENCIES 281 By default all computed frequencies including low and imaginary ones are printed The fol lowing options can be used to modify the print level PRINT HESSIAN print the force constant matrix hessian i e the second derivative matrix of the energy and the mass weighted hessian matrix PRINT LOW print low vibrational frequencies i e the 5 or 6 frequencies belong ing to rotations and translations and their normal modes default PRINT LOW 1 suppresses the print PRINT IMAG print imaginary vibrational frequencies and their normal modes de fault PRINT IMAG 1 suppresses the print Imaginary frequen cies appear at transition states The normal mode of an imaginary frequency represents the transition vector of that state The threshold for low vibrations default 150 cm7 can be changed using THRESH LOW value where value is the threshold in cm Other subcommands of FREQUENCIES are STEP rstep determines the step size of the numerical differentiation of the energy Default step size rstep 0 001 bohr NOPROJECT don t project translations and rotations out of the hessian SAVE irec ifil Save information of numerical frequency calculation to record irec By default frequencies are saved on record 5300 2 START jrec ifil Restart numerical frequency calculation from record irec on file f
447. red as a one 1 0 and FALSE as zero 0 0 Expressions may contain any number of variables The following standard operations can be performed expr expr Addition expr expr Subtraction expr expr Multiplication expr expr Division expr OR expr Logical OR expr AND expr Logical AND expr XOR expr Exclusive OR NOT expr Logical NOT expr GT expr Greater Than expr EQ expr Equal expr LT expr Less Than expr GE expr Greater Equal expr LE expr Less Equal expr NE expr Not Equal expr expr Exponentiation expr expr Exponentiation expr Parenthesis no effect expr Change sign expr Keep sign no effect 3 8 Intrinsic functions Expressions may contain the following intrinsic functions ABS expr Absolute value MAX exprexpr Largest value of arbitrary number of numbers or expressions MIN exprexpr Smallest value of arbitrary number of numbers of expressions EXP expr Exponential LOG expr Natural Logarithm LOG10 expr Common Logarithm SQRT expr Square Root NINT expr Next nearest integer INT expr Truncate to integer SIN expr Sine 3 DEFINITION OF MOLPRO INPUT LANGUAGE 10 COS expr Cosine TAN expr Tangent ASIN expr Arcsine ACOS expr Arccosine ATAN expr Arctangent COSH expr Hyperbolic cosine SINH expr Hyperbolic sine TANH expr Hyperbolic tangent MOD exprl expr2 Remainder expr1 INT expr1 expr2 expr2 Note
448. reference symmetry 3 Or NT pep dd define wavefunction symmetry 2 REF 1 define additional reference symmetry 1 REF 2 define additional reference symmetry 2 Each REF card may be followed by RESTRICT SELECT and CON cards in the given order 21 2 7 Selecting configurations SELECT ref1 ref2 refthr refstat mxshrf This card is used to specify a reference configuration set other than a CAS which is the default Configurations can be defined using CON cards which must appear after the SELECT card Alternatively if ref is an existing MOLPRO record name the configurations are read in from that record and may be selected according to a given threshold The select card should normally be placed directly after the WF or REF card s or if present the RESTRICT cards The general order of these cards is 21 THE CI PROGRAM WF or REF RESTRICT optional SELECT optional CON optional refl recl file ref2 rec2 file refthr refstat mxshrf 134 rec gt 2000 The configurations are read in from the specified record See section p0 5 4 about how to save the configurations in the MCSCF calculation If ref is not specified the program assumes that the configurations are read from subsequent CON cards see CON rec27 2000 additional configurations are read from the spec ified record If rec2 is negative all records between rec and abs rec2 are read All configurations found in this way
449. relativistic corrections ar2 geometry arl ar2 arl r geometry definition r 2 5 ang bond distance hf non relativisitic scf calculation expec rel darwin massv compute relativistic correction using Cowan Griffin operator e nrel energy save non relativistic energy in variable enrel show massv darwin erel show individual contribution and their sum examples dkroll 1 luse douglas kroll one electron integrals ar2 rel com hf relativistic scf calculation e dk energy save relativistic scf energy in variable e dk show massv darwin erel show mass velocity and darwin contributions and their sum show e dk e nrel show relativistic correction using Douglas Kroll 17 THE SCF PROGRAM 90 17 THESCF PROGRAM The Hartree Fock self consistent field program is invoked by one of the following commands HF or RHF calls the spin restricted Hartree Fock program UHF or UHF SCF options calls the spin unrestricted Hartree Fock program In contrast to older versions of MOLPRO the HF and RHF directives have identical functionality and can both be used for closed shell or open shell calculations Other aliases are HF SCF or RHF SCF Often no further input is necessary By default the number of electrons is equal to the nuclear charge the wavefunction is assumed to be totally symmetric symmetry 1 and the spin mul tiplicity is 1 singlet for an even number of electrons and 2 doublet otherwise The Aufbau principle is used to determine the
450. res e CPPADD factor e CPP SETI fcppl CPPINIT ncentres gt abs lt ncentres gt further cards will be read in the following format lt atomtype gt lt ntype gt lt Ag gt lt 04 gt lt Ba gt lt cutoff gt lt atomtype gt corresponds to the recognition of the atomic centres in the integral part of the program ntype gt fixes the form of the cutoff function choose 1 for Stoll Fuentealba and 2 for Mueller Meyer lt Qq gt is the static dipole polarizability lt Q gt is the static quadrupole polarizability lt Ba gt is the first non adiabatic correction to the dipole polarizability and lt cutoff gt is the exponential parameter of the cutoff function When lt ncentres gt is lower than zero only the integrals are calculated and saved in the record 1490 1 Otherwise the ho matrix records 1200 1 and 1210 1 and the two electron integrals record 1300 1 will be modified 16 RELATIVISTIC CORRECTIONS 88 CPP ADD lt factor gt With this variant previously calculated matrix elements of the polarization matrix can be added with the variable factor lt factor gt default lt factor gt 1 to the ho matrix as well as to the two electron integrals In particular CPP ADD 1 can be used to retrieve the integrals without the polarization contribution CPP SET lt fcpp gt normally not necessary but may be used to tell MOLPRO after a restart with what factor the po
451. residuals If KEEPCL 2 all close pairs are fully included in the LCCSD this does not affect the triples list This option is not yet implemented as effi ciently as it could and can therefore lead to a significant increase of the CPU time Setting a distance criterion to zero means that all pairs up to the corresponding class are treated as strong pairs For instance RCLOSE 0 means that strong and close pairs are fully included in the LCCSD in this case KEEPCL 1 has no effect Note however that setting RCLOSE 0 increases the length of the triples list 28 8 Directives 28 8 1 The LOCAL directive The LOCAL directive can be used to specify options for local calculations If this directive is inside the command block of a local calculation the options are used only for the current calculation and this is entirely equivalent as if they were specified on the command line The LOCAL directive can also be given outside a command block and in this case the options are used for all subsequent local correlation calculations in the same input Example DF LMP2 THRBP 0 985 is equivalent to DF LMP2 LOCAL THRBP 0 985 In the following example the LOCAL directive is global and acts on all subsequent local calcu lations i e both calculations will use THRBP 0 985 LOCAL THRBP 0 985 DF LMP2 local MP2 calculation OPTG geometry optimization using the DF LMP2 energy DF LCCSD T local coupled cluster at the optimized structure
452. rical gradients can be computed with respect to variables on which the Z matrix depends or with respect to Cartesian coordinates In the latter case it is most efficient to use symmetrical displacement coordinates These do not change the symmetry of the molecule and the number of displacements is minimal Alternatively mainly for testing purpose the gradients can be computed using symmetry unique Cartesian displacements or all 3N Cartesian displacements In these cases the symmetry of the molecule can be reduced by the displacements and using such displacements is normally not recommended 39 GEOMETRY OPTIMIZATION OPTG 254 DISPLACE ZMAT SYMM UNIQUE CART Displacement coordinates to be used for numerical gradient The de fault is ZMAT if the geometry is given as a zmatrix which depends on variables and SYMM symmetrical displacement coordinates other wise The use of UNIQUE or CART is not recommended SYMMETRY AUTO NOSYM Symmetry to be used in wavefunction calculations of numerical AUTO NOSYM RSTEP rstep DSTEP dstep ASTEP astep FOURPOINT NUMERICAL gradients This option is only relevant if DISPLACE UNIQUE CART If AUTO is given the maximum possible symmetry is used for each displacement This implies that the energy is independent of the sym metry used Note that this often not the case in MRCI or CASPT2 calculations The option can also not be used in local correlation cal culations logical Same
453. rinted 32 1 6 Examples The following example computes the dipole quadrupole moments of water and prints an orbital analysis By default the origin is at the centre of mass and this is taken as origin for the quadrupole moments SRevision 2006 0 h20 properties geometry 0o hl o r h2 0 r hl theta r l ang theta 104 hf property orbital density dm qm multi state 2 dm natorb state 1 1 natorb state 2 1 property orbital state 1 1 density state 1 1 dm am property orbital state 2 1 density state 2 1 dm am Z matrix geometry input bond length bond angle do scf calculation call property program lread scf orbitals read scf density matrix compute dipole moments and print orbital contributions compute quadrupole moments and print orbital contributi do full valence CASSCF compute natural orbitals for state 1 1 compute natural orbitals for state 2 1 examples h20 property com call property program read casscf natural orbitals for state 1 1 lread casscf density matrix for state 1 1 compute dipole moments and print orbital contributions compute quadrupole moments and print orbital contribut call property program read casscf natural orbitals for state 2 1 lread casscf density matrix for state 2 1 compute dipole moments and print orbital contributions compute quadrupole moments and print orbital contributi Alternatively the dipole and quadrupole moments can be comput
454. rix and the natural orbitals are saved See also DM or NATORB cards 21 3 12 Starting wavefunction START readcl irest readcl record name from which the wavefunction is restored for a restart In the case of coupled cluster methods CCSD QCISD BCCD the amplitudes are read from record readcI and used for restart presently only implemented for closed shell meth ods irest If nonzero the CI coefficients are read and used for the restart otherwise only the wavefunction definition is read in 21 3 13 One electron properties EXPEC oper opero opera After the wavefunction determination calculate expectation values for one electron operators oper See section for the available operators and their keywords In multi state calculations or in projected calculations also the transition matrix elements are calculated 21 3 14 Transition moment calculations TRANS readcl readc2 BI ORTH oper oper2 opers Instead of performing an energy calculation only calculate transition matrix elements between wavefunctions saved on records readc and readc2 See section for a list of available operators and their corresponding keywords If no operator names are specified the dipole transition moments are calculated If option BIORTH is given the two wavefunctions may use different orbitals However the number of active and inactive orbitals must be the same in each case Note that BIORTH is not working for spin orbit matrix
455. rk with generally contracted basis functions The ALASKA gradient program is based on the SEWARD integral routines by R Lindh It allows the calculation of gradients of generally contracted basis functions for closed shell SCF open shell RHF UHF RKS UKS MCSCF MP2 LMP2 DF LMP2 QCISD QCISD T and RS2 CASPT2 Gradients for state averaged MCSCF wave functions can be evaluated using the RS2 gradient program see section 38 1 5 For details about CASPT2 gradients see section 22 7 By default the program uses ALASKA gradients whenever possible However it is possible to force the use of a particular gradient program by defining the variable GRADTYP before calling the gradient program GRADTYP ALASKA GRADTYP CADPAC The gradient program is called using the FORCE command FORCE Normally the FORCE command is not needed since geometry optimizations should be per formed using the OPTG procedure An exception is the optimization of counterpoise corrected energies which requires several force calculations cf section 39 4 7 If no further data cards are given the default is to evaluate the gradient for the last optimized wavefunction In this case no further input is needed for ordinary gradient cases the program remembers the records on which the wavefunction information is stored An exception is the unusual case that several different CPMCSCF calculations have been formed in a previous MC SCF calculation In this case the SAMC d
456. rmat file containing the coordinates is al ways produced and may be used in the invocation of gOpen Mol rungOpenMol ifilename crd density plt If DENSITY is given then the file filename density plt is produced and contains the density grid in gOpenMol internal format orbital number symmetry plt If ORBITAL is given then for each orbital num ber symmetry specified the file filename orbital number symmetry p1t is produced and contains the orbital grid in gOpenMol internal format The default is not to produce any orbitals or densities and so only the atomic coordinates are dumped The default is to use unformatted binary files and this should not normally be changed The ORIGIN and AXIS commands should not be used e If INTERACT is given in the input when all the grids have been calculated an attempt is made to start gOpenMol by executing the Unix command rungOpenMol If rungOpenMol is not in PATH then nothing happens Otherwise gOpenMol should start and display the molecule Any plt files produced can be added to the display by following the Plot Contour menu item The name of the Unix command may be changed from the default rungOpenMo1 by specifying it as the first argument to the INTERACT directive By default gOpenMol is not started and this is equivalent to giving the command BATCH 33 DIABATIC ORBITALS 216 33 DIABATIC ORBITALS In order to construct diabatic states it is necessary to determine the mixin
457. ro Strings The value can either be a string enclosed in quotes or a string variable See section 8 3 for more details 8 2 Indexed variables Variables can be indexed but only one dimensional indexing is available Indexed variables can be defined either individually e g R 1 1 0 ANG R 2 1 2 ANG R 3 1 3 ANG or as a vector of values enclosed by square brackets R 1 0 1 1 1 2 ANG Subranges can also be defined e g 1 1 1 2 ANG leads to the same result as the above two forms The type of each element depends on the type of the assigned value and it is possible to mix types in one variable Example geometry he hf result program energy status gt 0 yields RESULT 1 HF SCF RESULT 2 2 85516048 AU RESULT 3 TRUE In this example the variables PROGRAM ENERGY and STATUS are system variables which are set by the program see section 8 4 8 VARIABLES 43 8 3 String variables As explained already in section 8 I string variables can be set as other variables in the form variable string variable string variable Strings must be enclosed by quotes Otherwise the string is assumed to be a variable and if this is undefined it is assumed to be zero Alternatively if the name of the variable is preceded by a dollar all values is assumed to be a string This can a string variable a quoted string or an unquoted string Note that unquoted strings are converted to uppe
458. ro after MP2 but are not eliminated from the pair list and not skipped in any loop skipdist 0 No pairs are deleted from pair list but weak and distant pairs are skipped in the loops were appropriate skipdist 1 Very distant pairs are neglected from the beginning Dis tant pairs are eliminated after MP2 skipdist 2 As skipdist 1 but also weak pairs are eliminated after MP2 skipdist 3 As skipdist 2 but distant pairs are eliminated from the operator list in case of LMP2 with multipole approximations for dis tant pairs This is the default Experimental test parameter Enables the use of asymmetric domains for distant pairs The asymmetric domain approximation supplements the multipole approximation for distant pairs as it suppresses the treatment of configurations for which no integrals can be computed by multipole expansion This leads to computational savings and im proved numerical stability jiterm 0 Disable asymmetric domains jiterm 1 Enable asymmetric domains default jiterm 2 Enable a variation of the asymmetric domain formalism Exchange operators will initially be projected to the asymmetric do main instead of simply packed If locsing ne O the single excitations use the full space i e they are not treated locally This is only works for LOCAL 1 The purpose of this experimental option is to reduce the basis set sensitivity of the Boughton Pulay BP method for domain selection Only basis functions with an
459. ro configure which will download the key from the molpro website and place it in usr local lib molpro mpptype arch token Other configuration options as described in section may also be specified in the script file usr local bin molpro A 3 Installation from source files A 3 1 Overview There are usually four distinct stages in installing MOLPRO from source files Configuration A shell script that allows specification of configuration options is run and creates a configuration file that drives subsequent installa tion steps Compilation The program is compiled and linked and other miscellaneous utilities and files including the default options file are built The essential resulting components are 1 The molpro executable which is a small front end that parses options performs housekeeping functions and starts the one or more processes that do computation A INSTALLATION OF MOLPRO 305 2 The molpro exe executable which is the main back end For parallel computation multiple copies of molpro exe are started by a single instance of molpro using the appropriate system utility e g mpirun parallel poe etc 3 Themolpro rc file which contains default options for molpro cf section A 3 6 4 The molproi rc file which contains MOLPRO script proce dures 5 Machine ready basis set and other utility libraries Validation A suite of self checking test jobs is run to provide assurance that
460. rocessor systems including workstation clus ters under the control of the Global Arrays parallel toolkit There are also some parts of the 2 RUNNING MOLPRO 3 code that can take advantage of shared memory parallelism through the OpenMP protocol although these are somewhat limited and this facility is not at present recommended It should be noted that there remain some parts of the code that are not or only partly parallelized and therefore run with replicated work Additionally some of those parts which have been par allelized rely on fast inter node communications and can be very inefficient across ordinary networks Therefore some caution and experimentation is needed to avoid waste of resources in a multiuser environment Molpro can be compiled in three different ways 1 Serial execution only In this case no parallelism is possible at run time 2 MPP a number of copies of the program execute simultaneously a single task For example a single CCSD T calculation can run in parallel with the work divided between the processors in order to achieve a reduced elapsed time 3 MPPX a number of copies of the program run in serial executing identical indepen dent tasks An example of this is the calculation of gradients and frequencies by finite difference for the initial wavefunction calculation the calculation is replicated on all processes but thereafter each process works in serial on a different displaced geomet
461. roduct of the exact i e computed integral value and a density ma trix This threshold is only used in DTRAF and DKEXT A shell block of integrals is skipped if the product of the largest in tegral and the largest element of the effective density matrix contributing to the shell block is smaller than this threshold If it set negative no computed integrals will be neglected THRPROD Prescreening threshold for products of integrals and MO coefficients DTRAF or amplitudes DKEXT Shell blocks of MO coeffi cients or amplitudes are neglected if the product of the largest integral in the shell block and the largest coefficient is smaller than this value If this is set negative no product screening is performed THRMAX Initial value of the prescreening threshold THREST for DFOCK and DKEXT in iterative methods SCF CI CCSD If nonzero it will also be used for DKEXT in MP3 and MP4 SDQ calcu lations The threshold will be reduced to THREST once a cer tain accuracy has been reached see VARRED or latest after MAXRED iterations In CI and CCSD calculations also the ini tial thresholds THRINT DKEXT and THRPROD DKEXT are in fluenced by this value For a description see THRMAX DKEXT If THRMAX O the final thresholds will be used from the begin ning in all methods SCREEN Enables or disables prescreening SCREEN 0 full screening enabled SCREEN 0 THRPROD is unused No density screening in direct SCF SCREEN lt 1 THRINT is
462. ron operators With EXPEC only expectation values are calculated oper is a codeword for the operator The available operators and their associated keywords are given in section 20 7 2 Matrix elements over two electron operators EXPEC 2 oper oper2 0pern TRAN2 oper oper2 0pern Calculate transition matrix elements for two electron operators This is presently only useful for angular momentum operators With EXPEC2 only diagonal matrix elements will be computed For instance TRAN2 LXX calculates matrix elements for L2 TRAN2 LYY calculates matrix elements for p TRAN2 LXZ calculates matrix elements for 2 UL L Ly TRAN2 LXX LYY LZZ calculates matrix elements for L I and P The matrix ele ments for the sum Z are also printed 20 7 3 Saving the density matrix DM spindens If the DM directive is given the first order density matrix in AO basis is written to the dump record specified on the ORBITAL card default 2140 2 If no ORBITAL card is present but a record is specified on a NATORB CANORB or LOCORB card the densities are saved to the first record occurring in the input In a state averaged calculation the SA density as well the individual state densities are saved See section 4 11 for information about how to recover any of these densities for use in later programs Of spindens is a number greater than zero the spin density matrices are also saved Note that a maximum of 50 density matrices can be saved i
463. roperties geometry 0o hl o r h2 0 r hl theta r 1 ang theta 104 gexpec dm sm qm methods hf multi ci do i 1 methods Smethods i e i energy dip i dmz quadxx 1 qmxx quadyy i qmyy quadzz 1 qmzz smxx i xx smyy 1 yy smzz i zz enddo table methods dip smxx smyy smzz table methods e quadxx quadyy quadzz This Job produces the following tables ETHODS DIP SMXX HF 0 82747571 5 30079792 ULTI 0 76285513 5 29145148 CI 0 76868508 25432191822 ETHODS E QUADXX HF 76 02145798 1 69070039 ULTI 76 07843443 1 60318949 CI 16523369021 1 60150114 37 Z matrix geometry input bond length bond angle compute dipole and quarupole moments Ido hf casscf mrci loop over methods run energy calculation examples save dipole moment in variable dip lido sexpec com save quadrupole momemts save second momemts print table of first and second moments print table of quadrupole moments SMYY SMZZ 3 01408114 4 20611391 3 aL 1397 4 25941000 3 15540500 4 28542917 QUADYY QUADZZ 1 73937477 0 04867438 1 65831677 0 05512728 1 64826869 0 04676756 6 13 2 Example for computing relativistic corrections rar2 geometry arl ar2 arl r geometry definition r 2 5 ang bond distance hf expec rel darwin massv e nrel energy show massv darwin erel dkroll 1 hf e dk energy show massv darwin erel show e dk e nrel luse douglas kroll one elec
464. rr 295 43 3 Saving matrices SAVE aaaea 295 43 4 Adding matrices ADD o a 296 43 5 Trace of a matrix or the product of two matrices TRACE 296 43 6 Setting variables SET rs 296 43 7 Multiplying matrices MULTA e 296 43 8 Transforming operators TRAN les 297 43 9 Transforming density matrices into the MO basis DMO 297 43 10Diagonalizing a matrix DIAG o o o 0020000000 297 43 11Generating natural orbitals NATORB o 297 43 12Forming an outer product of two vectors OPRD 297 43 13Forming a closed shell density matrix DENS o 297 43 14Computing a fock matrix FOCR les 298 43 15Computing a coulomb operator COUL o ooo 298 43 16Computing an exchange operator EXCH o 298 43 17Printing matrices PRINT ls 298 43 18Printing diagonal elements of a matrix PRID 298 43 19Printing orbitals PRIO 22s 298 43 20 Assigning matrix elements to a variable ELEM 298 43 2 Reading a matrix from the input file READ 299 43 22 Writing a matrix to an ASCII file WRITE oaaao 299 4d3 23bxamples ac 4 duke PR ow um a c or ommo aoa di RS 299 CONTENTS 43 24Exercise SCF program 22e Bibliography A Installation of MOLPRO B A l Obtaining the distribution material llle
465. rslruhe basis sets SV TZV and for some elements SVP TZVP TZVPP TZVPPP The Binning Curtiss sets for Ga Kr BINNING SV BINNING SVP BINNING VTZ and BINNING VTZP Most of the Pople basis sets using their standard names e g 6 31G 6 311 G D P etc Note that specially in this case the mechanism described below using parenthesized modifiers to restrict the basis set is disabled to allow the full range of standard basis sets to be specified Example BASIS VTZ generates valence triple zeta basis set for all atoms Thus the input x x h20 cc pVTZ basis A title r 1 85 theta 104 set geometry parameters geometry 0 z matrix geometry input H1 0 r examples H2 0 r H1 theta h20 scf vtz com basis VTZ luse VTZ basis hf closed shell scf is entirely equivalent to h20 cc pVTZ basis A title r 1 85 theta 104 set geometry parameters geometry 0 z matrix geometry input ql 070 H2 0 r H1 theta examples basis h20 scf vtz explicit c spdf o vtz c spd h vtz c hf 13 BASIS INPUT 80 Default basis sets can be defined anywhere in the input before the energy calculation to which it should apply using a single BASIS cards The default basis set applies to all types of atoms but can be superceded by different basis sets for specific atoms as explained later Some restrictions concerning the maximum angular momentum functions to be used or the number of contracted functions are possible as follows The
466. ry At present this is implemented only for numerical gradients and Hessians Which of these three modes is available is fixed at compilation time and is reported in the job output The options described below for selecting the number and location of processors are identical for MPP and MPPX Specifying parallel execution The following additional options for the molpro command may be used to specify and control parallel execution n tasks tasks tasks per node smp threads tasks specifies the number of Global Array processes to be set up and defaults to 1 tasks per node sets the number of GA processes to run on each node where appropriate The default is installation dependent In some environments e g IBM running under Loadleveler PBS batch job the value given by n is capped to the maximum allowed by the environment in such circumstances it can be useful to give a very large number as the value for n so that the control of the number of processes is by the batch job specification smp threads relates to the use of OpenMP shared memory parallelism and specifies the maximum number of OpenMP threads that will be opened and defaults to 1 Any of these three components may be omitted and appropriate combinations will allow GA only OpenMP only or mixed paral lelism N task specification userl nodel tasksl user2 node2 tasks2 nodel node2 etc specify the host names of the nodes on which to run On most par allel syst
467. ry of internal configurations N N 1 and N 2 electron print internal configurations N N 1 N 2 print summary of reference configurations print reference configurations and their coefficients print p space configurations print diagonal elements for internals print diagonal elements for singles various levels of intermediate information in pair orthogonal ization routine print information at each subroutine call print in addition information about I O in LESW SREIBW print also information about I O in FREAD FWRITE print analysis of CPU and I O times print everything at given level be careful and CEPA 1 for water z matrix geometry input ximpleg h20 cepal com hf wf 10 1 TOTAL SCF ENERGY 76 02680642 ci occ 3 1 1 core 1 wf 10 1 ITOTAL CI SD ENERGY 76 22994348 cepa 1 0cc 3 1 1 core 1 wf 10 1 ITOTAL CEPA 1 ENERGY 76 23799334 21 THE CI PROGRAM 145 SRevision 2006 0 Valence multireference CI for X and A states of H20 gthresh energy 1 d 8 r 0 957 angstrom theta 104 6 degree geometry 0 z matrix geometry input HO H2 0 r H1 theta hf wf 10 1 TOTAL SCF ENERGY 76 02680642 multi occ 4 1 2 closed 2 freeze 1 wf 9 2 1 wf 9 1 1 tran ly IMCSCF ENERGY 75 66755631 examples IMCSCF ENERGY 75 56605896 hZop irei Gans com ci occ 4 1 2 closed 2 core 1 wf 9 2 1 save 7300 1 T
468. s Additional output formats for tables XHTML I4TEX Maple Mathematica Matlab and comma separated variables orbitals and basis sets XML and an optional well formed XML output stream with important results marked up B RECENT CHANGES 317 B 2 New features of MOLPRO2002 6 Relative to version 2002 1 there are the following changes and additions 1 Ph N B 3 Support for IA 64 Linux systems HP and NEC and HP UX 11 22 for IA 64 Itanium2 Support for NEC SX systems Support for IBM power4 systems Modified handling of Molpro system variables The SET command has changed see sections B and 8 4 The total charge of the molecule can be specified in a variable CHARGE or on the WF card see section Improved numerical geometry optimization using symmetrical displacement coordinates see sections and 39 Improved numerical frequency calculations using the symmetry AUTO option see section 40 New features of MOLPRO2002 Relative to version 2000 1 there are the following principal changes and additions 1 Modules direct and local are now included in the base version This means that integral direct procedures as described in M Sch tz R Lindh and H J Werner Mol Phys 96 719 1999 linear scaling local MP2 as described in G Hetzer P Pulay and H J Werner Chem Phys Lett 290 143 1998 M Schiitz G Hetzer and H J Werner J Chem Phys 111 5691 1999
469. s a large error in the energy might result On the other hand if the domains are not kept fixed their size and quality might change during the optimization again leading to spurious energy changes and divergence of the optimization The best way to avoid this problem is to use the MERGEDOM 1 option see section 28 6 If this option is given the domains for the 7 orbitals will comprise the basis functions of all six carbon atoms and the energy will be invariant with respect to unitary transformations among the three T orbitals Note that this problem does not occur if the symmetry of the aromatic system is lowered by a substituent Redundant orbital rotations can also lead to convergence difficulties of the Pipek Mezey localization This can be overcome by using PIPEK METHOD 2 With this option the second derivatives of the localization criterion with respect to the orbital rotations is computed and diagonalized and rotations corresponding to zero eigenvalues are eliminated Finally we note that the LMP2 gradients are quite sensitive to the accuracy of the SCF con vergence as is also the case for MP2 If very accurate structures are required or if numerical frequencies are computed from the gradients the default SCF accuracy might be insufficient We recommend in such cases to add an ACCU 14 directive possibly even ACCU 1 6 after the HF command Indicative of insufficient SCF accuracy are small positive energy changes near the e
470. s card has no effect 20 3 Defining the optimized states Each state symmetry to be optimized is specified by one WF card which may optionally be followed by STATE WEIGHT RESTRICT SELECT CON and or PSPACE cards All cards belonging to a particular state symmetry as defined on the WF card must form a block which comes directly after the WF card The cards can be in any order however 20 3 1 Defining the state symmetry The number of electrons and the total symmetry of the wavefunction are specified on the WF card WF elec sym spin where elec is the number of electrons sym is the number of the irreducible representation spin defines the spin symmetry spin 25 singlet 0 doublet 1 trip let 2 etc Note that these values take sensible defaults if any or all are not specified see section 4 8 The input directives STATE WEIGHT LQUANT SELECT PUNCSF always refer to the state symmetry as defined on the previous WF card If such a directive is found before a WF card has been given the current state symmetry is assumed either from a previous calculation or from variables MC SYMMETRY 1 and MC SPIN 1 if these are defined If any of these cards or a WF card is given the variables STATE WEIGHT LQUANT SELECT are not used and the number of state symmetries defaults to one regardless of how many symmetries are specified in variable MC SYMMETRY 20 THE MCSCF PROGRAM MULTI 115 20 3 2 Defining the number of
471. s use Ci project 3000 3 1 Ci project 3000 3 Ci project 3000 3 21 3 4 Transition matrix element options TRANH option If option gt 1 this forces calculation of transition hamiltonian matrix elements in a TRANS or PROJECT calculation If option lt 1 this forces calculation of one electron transition properties 21 3 5 Convergence thresholds ACCU istate energy coeff Convergence thresholds for state istate The actual thresholds for the energy and the CI coeffi cients are 10 energy and 10 coeff If this card is not present the thresholds for all states are the default values or those specified on the THRESH card 21 3 6 Level shifts SHIFT shiftp shifts shifti Denominator shifts for pairs singles and internals respectively 21 THE CI PROGRAM 138 21 3 7 Maximum number of iterations MAXITER maxit maxiti maxit maximum number of macroiterations maxiti maximum number of microiterations internal CI 21 3 8 Restricting numbers of expansion vectors MAXDAV maxdav maxvi maxdav maximum number of external expansion vectors in macroitera tions maxvi maximum number of internal expansion vectors in internal CI 21 3 9 Selecting the primary configuration set PSPACE select npspac select energy criterion for selecting p space configurations If nega tive a test for p space H is performed npspac minimum number of p space configurations Further configu rations a
472. s MASSV DARW and EREL respectively 32 5 1 Example ar2 geometry arl ar2 arl r geometry definition r 2 5 ang bond distance hf non relativisitic scf calculation expec rel darwin massv compute relativistic correction using Cowan Griffin operator e nrel energy save non relativistic energy in variable enrel show massv darwin erel show individual contribution and their sum examples dkroll 1 luse douglas kroll one electron integrals ar2 rel com hf relativistic scf calculation e dk energy save relativistic scf energy in variable e dk show massv darwin erel show mass velocity and darwin contributions and their sum show e dk e nrel show relativistic correction using Douglas Kroll 32 PROPERTIES AND EXPECTATION VALUES 213 32 6 CUBE dump density or orbital values CUBE filename iflag n n2 n3 calls a module which dumps the values of various properties on a spatial parallelopipedal grid to an external file The purpose is to allow plotting of orbitals densities and other quantities by external programs The format of the file is intended to be the same as that produced by other programs filename 1s the unix path name of the file to be written and its specification is mandatory iflag If flag is negative default a formatted file will be written otherwise unformatted fortran i o will be used nj n n3 specify the number of grid points in each of three dimensions If not specified sensible defaults are c
473. s appear with finite integration grids C 15 CS2 Colle Salvetti correlation functional R Colle and O Salvetti Theor Chim Acta 37 329 1974 C Lee W Yang and R G Parr Phys Rev B 37 785 1988 CS2 is defined through 2bp 5 Pat fg phy e P k ep Pata perg ptw e dao 1 dp where Ta Va la 140 a 2 8 140 Tp p t 141 B 2 8 141 lo 1 tw 5 5 142 w 8p 2 142 and the constants are a 0 04918 b 0 132 c 0 2533 d 0 349 C 16 DIRAC Slater Dirac Exchange Energy Automatically generated Slater Dirac exchange See reference for more details K 2 6 py 143 where c 3 8 34 P m1 144 C 17 G96 Gill s 1996 Gradient Corrected Exchange Functional See reference for more details 1 K Bla q ps o i07 Ja 145 where a 3 8 34 P VT 146 To avoid singularities in the limit p 0 G p a 912 147 89 a c 147 C DENSITY FUNCTIONAL DESCRIPTIONS 331 C 18 HCTH120 Handy least squares fitted functional See reference for more details K Ppa Pp Pa 0 pg 0 Ao Am d 24 A2 n 4 3 43 n 4 4 A4 n d 24 Y e ps 0 Bo Bin 72 Bo n s A2 E Bs n s 7 A2 g 148 Ba n Q5 m 3 8 3428 Vx ps Co Cin x 2 34 n xs 23 C3 n xs A3 3 Ca n xXs 7 A3 where d 1 2 Xo V2 xg 149 n 0 u D 150
474. s available such output may be especially useful for plotting of orbitals 36 13 Further facilities Calculations can also be performed for various types of direct product wavefunctions and or with strictly localized orbitals Details are available from the authors These facilities will be documented in a later release 36 14 Service mode SERVICE This keyword takes precedence over any others previously defined to CASVB It provides simple facilities for retrieving orbital coefficients and VB structure coefficients It should not be used during a run of CASVB that has been invoked from inside MULTI START record file Coefficients are taken from record file The default value is 2700 2 WRITE write Vectors in the symmetry orbital basis are written to channel iabs iwrite The default action is 36 THE VB PROGRAM CASVB 237 to write these vectors to the standard output If write is negative then the vectors are instead written to a binary file as a single record SPECIAL idiml idim2 idim3 idima If present this keyword must come last The program attempts to retrieve from record file a vector of length idim idim2 idim3 after first skipping idim4 elements The vector is written according to the setting of iwrite Default idim values are zero 36 15 Examples HARES dO Al singlet state geometry angstrom c hiycpledly h2 c 1 117 h1 102 4 int hf multi occ 4 1 2 closed 1 6 in 6 CASSCF natorb ci save 3
475. s becomes nearly singular This is known as intruder state problem and can cause convergence problems or lead to a blow up of the wavefunction Often such problems can be eliminated by including more orbitals into the reference wavefunction but of course this leads to an increase of the CPU time The use of modified Fock operators see below or level shifts as proposed by Roos and Andersson Chem Phys Lett 245 215 1995 may also be helpful Presently only real level shifts have been implemented With no further input cards the wavefunction definition core closed and active orbital spaces symmetry corresponds to the one used in the most recently done SCF or MCSCF calculation By default a CASSCF reference space is generated Other choices can be made using the OCC CORE CLOSED WF SELECT CON and RESTRICT cards as described for the CI program The orbitals are taken from the corresponding SCF or MCSCF calculation unless an ORBI TAL directive is given For a CASPT2 calculation the zeroth order Hamiltonian can be brought to a block diagonal form when pseudo canonical orbitals are used This leads to fastest convergence It is there fore recommended that in the preceding MULTI calculation the orbitals are saved using the CANONICAL directive note that the default is NATORB Most options for MRCI calculations like STATE REFSTATE etc apply also for RS2 C and RS3 and are not described here again Some additional options wh
476. s for H O SRevision 2006 0 h20 test memory 1 m allocate 1 MW dynamic memory geometry 0o hl o r h2 0 r hl theta Z matrix geometry input basis vtz cc pVTZ basis set r 1 ang bond length theta 104 bond angle hf do scf calculation text examples for single reference correlation treatments f examples ci ICISD using MRCI code i h2o0 ccsd com cepa 1 cepa 1 using MRCI code mp2 Second order Moeller Plesset mp3 Second and third order MP mp4 Second third and fourth order MP4 SDTQ mp4 notripl IMP 4 SDQ cisd ICISD using special closed shell code ccsd t coupled cluster CCSD T qci t quadratic configuration interaction QCISD T becd t Brueckner CCD T calculation 24 6 2 Single reference correlation treatments for N F 24 THE CLOSED SHELL CCSD PROGRAM 163 SRevision 2006 0 N2F2 CIS GEOMETRY C2h rnn 1 223 ang rnf 1 398 ang alpha 114 5 geometry N1 N2 N1 rnn F1 N1 rnf N2 alpha F2 N2 rnf N1 alpha F1 180 basis vtz define N N distance define N F distance define FNN angle examples cc pVTZ basis set n2f2 ccsd com method hf cisd ccsd t qcisd t bccd t lall methods to use do i 1 method method i e i energy enddo table method e title Results for n2f2 basis basis This calculation produces the following table Results for n2f2 METHOD CISD 308 BCCD T 308 CCSD T 308 QCISD T 308 basis VTZ 4634948
477. s of the xc kernel default 0 If O write all dimer amplitudes to file if 1 write 3 index response propagators to file and 1f 2 write 3 index response propagators com pressed to file The latter two variants save disk space but need more CPU time to compute E 0 default 0 exch disp If SAPT_DISK 2 this value determines the compression cutoff de fault 1d 12 If SAPT_DISK gt 0 calculate also uncoupled exchange dispersion en ergies default false Threshold for AO 3 index integrals default 1 d 12 Threshold for MO 3 index integrals default 1 d 8 Threshold for AO 2 index integrals default 1 d 10 Product threshold for first half transformation default 1 d 8 Threshold for Schwarz screening default 1 d 5 The last threshold values for the 2 and 3 index integrals should not be set higher in density fitting calculations as this can cause lower accuracies in the interaction terms In addition SAPT knows the following subcommands MONOMERA MONOMERB INTERMOL CA CB SAPTLEVEL FITLEVEL ICPKS FROZA FROZB NLEXFAC CPKSTHR CPKSMAXIT Stores informations like number of electrons etc about previous monomer A calculation See above Starts the SAPT calculation INTERMOL may have the following subkeywords Record number of wave function for monomer A always needed Record number of wave function for monomer B always needed See above See above See above See above See above Amo
478. s taken as the origin This effect can be avoided by using the charge centre as origin i e specifying CHARGE as first entry in the GEOMETRY input GEOMETRY CHARGE 12 8 Dummy centres DUMMY atoml atom2 Sets nuclear charges on atoms 1 2 etc to zero for doing counterpoise calculations for ex ample atoml atom2 can be Z matrix row numbers or tag names Note that the current setting of dummies is remembered by the program across restarts via the MOLPRO variable 12 GEOMETRY SPECIFICATION AND INTEGRATION 76 DUMMYATOMS Dummies can be reset to their original charges using a DUMMY card with no entries Dummy centres are also reset to their original charges if 1 and INT command is en countered or ii a new geometry input is encountered The program does not recognize automatically if the symmetry is reduced by defining dummy atoms Therefore for a given dummy atom either all symmetry equivalent atoms must also be dummies or the symmetry must be reduced manually as required An error will result if the symmetry is not consistent with the dummy centre definitions 12 8 1 Counterpoise calculations Counterpoise corrections are easily performed using dummy cards One first computes the energy of the total system and then for the subsystems using dummy cards 12 8 2 Example interaction energy of OH Ar 13 BASIS INPUT TI SRevision 2006 0 0H 2Sig Ar linear memory 2 m geometry ql dummy center
479. s well as for each individual fragment separate FORCE calculations The gradients and energies are added using the ADD directive This requires that NOORIENT has been specified in the geometry input in order to avoid errors due to unintended rotation of the system This default can be disabled using the NOCHECK option see ADD above The way a counterpoise corrected geometry optimization works is shown in the following exam ple Note that the total counterpoise corrected energy must be optimized not just the interaction energy since the interaction energy depends on the monomer geometries and has a different minimum than the total energy The interaction energy could be optimized however if the monomer geometries were frozen In any case the last calculation before calling OPTG must be the calculation of the total system at the current geometry in the example below the dimer calculation since otherwise the optimizer gets confused 39 GEOMETRY OPTIMIZATION OPTG lexamples hfdimer cpcoptl test Revision 2006 0 HF dimer MP2 CP optimization with relaxed monomers basis avtz gthresh energy 1 d 8 INITIAL VALUES OF GEOMETRY VARIABL E n RFF DS R1 1 76 R2 12375 HETA1 7 0 HETA2 111 geomet ry x noorient 1 2 EL rff hil f1 rl f2 thetal h2 2 X2 fl theta2 hl 180 label text CALCULATION AT LARGE SEPARATION rff save rff save current rff distance rff
480. scussion of options d dirl dir2 T directory W directory k key m G n N where dirl dir2 is a list of directories which may be used for creating scratch files Each of the directories should be writable by those who will use the program and the directory specification may contain embedded environment variables in shell form for example STMPDIR or tmp SUSER these will be expanded at run time If multiple scratch file systems are available it is advantageous to present a list of directories of which there is one in each file sys tem Some parts of MOLPRO present extreme I O demands and it is therefore important to be careful in optimizing the provision and specification of scratch directories Note that in the building ofbin molpro rc the environment vari ables STMPDIR STMPDIR2 STMPDIR3 are used to construct the list of scratch directories for the d option Thus these envi ronment variables should at make time be filled with the names of directories on each available scratch file system cf section A 3 3 This determines the destination of permanent integral files At run time this file is located in the first directory specified after d i e dirl see above but after completion of the job the file will be copied to the directory given after I Since the integral file can be very large it is normally recommended that directory is identical to dir this is the default Then
481. sed for quite large molecules The states to be computed are specified similarly as for EOM e g hf cis 3 1 1 2 trans 1 26 OPEN SHELL COUPLED CLUSTER THEORIES 169 26 OPEN SHELL COUPLED CLUSTER THEORIES Spin unrestricted RHF UCCSD and partially spin restricted RHF RCCSD open shell coupled cluster theories as described in J Chem Phys 99 1993 5219 see also erratum J Chem Phys 112 2000 3106 are available in MOLPRO In both cases a high spin RHF reference wavefunction is used No coupled cluster methods based on UHF orbitals are implemented in MOLPRO the only correlation method in MOLPRO which uses UHF orbitals is UMP2 In the description that follows the acronyms RCCSD and UCCSD are used but the theories should normally be referred to as RHF RCCSD RHF UCCSD in order to distinguish them from al ternative ans tze based on spin unrestricted orbitals The program will accept either the full or abbreviated acronyms as input commands In the RCCSD theory certain restrictions among the amplitudes are introduced such that the linear part of the wavefunction becomes a spin eigenfunction this is not the case in the UCCSD method even if an RHF reference function is used At present the implementation of RCCSD is only preliminary and no CPU time is saved by as compared to UCCSD However improved algorithms as described in the above publication are currently being implemented and will be available in the near future Th
482. seful to avoid that the next program in a chain is executed STATUS MULTI CI STOP will check the status of the most previous MULTI and CI steps and stop if something did not converge STATUS RHF CLEAR will clear the status flag for last RHF No action even if RHF did not converge Note that the status variables are not recovered in a restart By default the program automatically does the following checks 1 If an orbital optimization did not converge and the resulting orbitals are used in a subse quent correlation calculation an error will result This the error exit can be avoided using the IGNORE ERROR option on the ORBITAL directive 2 Ifa CCSD QCI BCC LMPn calculation did not converge further program steps which depend on the solution e g Triples CPHF EOM will not be done and an error will result This can be avoided using the NOCHECK option on the command line 3 In geometry optimizations or frequency calculations no convergence will lead to immediate error exits 6 11 Global Thresholds GTHRESH A number of global thresholds can be set using the GTHRESH command outside the individual programs the first letter G is optional but should be used to avoid confusion with program specific THRESH cards The syntax is GTHRESH keyl valuel key2 value2 key can be one of the following ZERO Numerical zero default 1 d 12 ONEINT Threshold for one electron integrals default 1 d 12 but not used at
483. shows how the same calculations can be done using numerical gradients In this case first the total counter poise corrected energy is formed and then optimized Note that the ADD command does not work for numerical gradients 39 GEOMETRY OPTIMIZATION OPTG lexamples hfdimer cpcoptl num test Revision 2006 0 HF dimer MP2 CP optimization with relaxed monomers basis avtz gthresh nergy 1 d 8 INITIAL VALUES OF G REF R1 R2 HETA1 HETA2 EOMETRY VARIABL E n geomet ry x noorient label text rff_save rff ds f2 l hele EL h2 2 CALCULATION AT LARGE rff 1000 text dummy 2 h2 hf accu 16 mp2 HF1 ELE rl 2 r2 f1 thetal theta2 hl 180 un EPARATION ehflinf energy text dummy f1 h1 hf accu 16 mp2 HF2 ehf2inf energy einf ehflinftehf2inf rff rff save text dummy 2 h2 hf accu 16 mp2 save current rff distance dimer calculation at large separation second hf is now dummy scf for first monomer mp2 for first monomer save mp2 energy in variable first hf is now dummy scf for second monomer mp2 for second monomer save mp2 energy in variable total energy of unrelaxed momomers reset HF HF distance to current value CP calculation for HF1 MONOMER ehfl energy text dummy f1 h1 hf accu 16 mp2 second hf is now dummy scf
484. sing DO loops this ensures that each calculation will start with the orbitals from the corresponding orbitals from the previous cycle as long as the order of the commands in the input remains unchanged If for instance the first SCF would be skipped in the second cycle using some IF ENDIF structure the second SCF would still use record 2101 2 Thus under 4 GENERAL PROGRAM STRUCTURE 14 most circumstances the program defaults are appropriate and the user does not have to specify the records After a restart this logic will still work correctly if the number and sequence of SCF and MCSCF commands is kept unchanged Thus if you want to skip certain parts of the input after a restart it is recommended to use IF ENDIF structures or the GOTO command rather than to delete or comment certain commands If for some reason this is not possible the START and ORBITAL directives can be used to specify explicitely the records to be used In general we recommend the use of program defaults whenever possible since this minimizes the probability of input errors and frustration After completion of each program step MOLPRO prints a summary of the records on each file 4 4 Restart Information from the permanent files is automatically recovered in subsequent calculations This can be controlled using the RESTART directive 4 5 Data set manipulation It is possible to truncate files and rename or copy records using the DATA command Sev eral stand
485. sity functional theory and Local Electron Correlation Methods Phys Chem Chem Phys 5 2001 2003 20 T Hrenar G Rauhut and H J Werner Impact of local and density fitting approximations on harmonic vibrational frequencies J Phys Chem A 110 2060 2006 Intermolecular interactions and the BSSE problem 21 M Sch tz G Rauhut and H J Werner Local Treatment of Electron Correlation in Molec ular Clusters Structures and Stabilities of H20 n 2 4 J Phys Chem 102 5997 1998 See also 2 and references therein 22 N Runeberg M Sch tz and H J Werner The aurophilic attraction as interpreted by local correlation methods J Chem Phys 110 7210 1999 23 L Magnko M Schweizer G Rauhut M Sch tz H Stoll and H J Werner A Comparison of the metallophilic attraction in X M PHA3 M Cu Ag Au X H CI Phys Chem Chem Phys 4 1006 2002 28 2 Getting started The local correlation treatment is switched on by preceding the command name by an L i e by using the LMP2 LMP3 LMP 4 LOCISD LCCSD or LCISD commands The LOCISD and LCCSD commands can be appended by a specification for the perturbative treatment of triple excitations e g LCCSD TO T Use the default triples method Currently this is TO TO Non iterative local triples This is the fastest triples option It is usually sufficiently accurate and recommended to be used in most cases 28 LOCAL CORRELA
486. ssumed in CI default 1 is reasonable NCACHE machine cache size in bytes IASYN if nonzero use asynchronous I O on CONVEX MXMBLK column row block size for mxma MXMBLN link block size for mxma NCPUS maximum number of cpus to be used in multitasking MINBR1 min number of floating point ops per processor MXDMP highest file number to be treated as dump file with full functionality 1 lt MXDMP 3 The MXDMP option is for experts only This prevents basis and geometry information from being written to dump files with higher file number than the given value and can sometimes be useful for counterpoise corrected geometry optimizations Note that some functionality is lost by giving this option and errors will result unless all input is correct 8 VARIABLES Data may be stored in variables A variable can be of type string real or logical depending on the type of the expression in its definition Any sequence of characters which is not recognized as expression or variable is treated as string In this section we will discuss only real and logical variables String variables will be discussed in more detail in section 8 3 Variables can be used anywhere in the input but they can be set only outside the input blocks for specific programs For example if a variable is used within the input block for HF it must have been set before the HF input block MOLPRO automatically stores various results and data in system variables see sect
487. states in the present symmetry STATE nstate nstate is the number of states in the present symmetry By default all states are optimized with weight 1 see WEIGHT card 20 3 3 Specifying weights in state averaged calculations WEIGHT w 1 w 2 w nstate w i is the weight for the state i in the present symmetry By default all weights are 1 0 See also STATE card If you want to optimize the second state of a particular state symmetry alone specify STATE 2 WEIGHT 0 1 Note however that this might lead to root flipping problems 20 4 Defining the configuration space By default the program generates a complete configuration set CAS in the active space The full space may be restricted to a certain occupation pattern using the RESTRICT option Alter natively configurations may be selected from the wavefunction of a previous calculation using SELECT or explicitly specified on CON cards Note that this program only allows to select or specify orbital configurations For each orbital configuration all spin couplings are always in cluded Possible RESTRICT SELECT and CON cards must immediately follow the WF card which defines the corresponding state symmetry 20 4 1 Occupation restrictions RESTRICT nmin nmax orb orb2 orby This card can be used to restrict the occupation patterns Only configurations containing be tween nmin and nmax electrons in the specified orbitals orbi orbz orb are included in th
488. structure of the hessian changes during the optimization FORWARD Use forward differences default CENTRAL Use the more accurate central differences RSTEP rstep Step length for distances in bohr The default is 0 01 ASTEP astep Step length for angles in degree The default is 0 5 or 1 for angles below and above 90 degree respectively DSTEP dstep Step length for symmetrical displacements in bohr The default is 0 01 VARIABLE varname Use given variable for numerical calculation of the Hessian Note that this disables the use of gradients and Hessian evaluation can be very expensive PROCEDURE procname Procedure to be used for computing Hessian This procedure must be define a complete energy calculation orbital optimization and corre lation treatment A different method can be used than for the opti mized energy For instance an MP2 hessian can be used for CCSD T 39 GEOMETRY OPTIMIZATION OPTG 260 DISPLACE type CALC icalc THRESH thresh optimizations or a CASPT2 hessian for MRCI optimizations By de fault the same procedure is used for the hessian as for the optimized energy type can be one of the following SYMM Use symmetric displacement coordinates default This is the only recommended option CART Use 3N cartesian displacements not recommended This requires many more energy calculations than necessary and does not preserve the molecular symmetry UNIQUE Use symmetry unique cartesian displ
489. subsequently known outside the procedures as well The reason is that procedures are included into the internal input deck at the beginning of the job and not at execution time for the same reason variable substitution of procedure names is not possible e g one cannot use constructs like method scf Smethod this does not work 6 9 Text cards TEXT TEXT XXXXXX will just print xxxxxx in the output If the text contains variables which are preceded by a dollar these are replaced by their actual values e g r 2 1 text Results for R r will print Results for R 2 1 6 10 Checking the program status STATUS STATUS ALL LAST commands IGNORE STOP CRASH CLEAR 6 PROGRAM CONTROL 34 This command checks and prints the status of the specified program steps commands may be a list of commands for wavefunction calculations previously executed in the current job If no command or LAST is specified the status of the last step is checked If ALL is given all program steps are checked If CRASH or STOP is given the program will crash or stop respectively if the status was not o k STOP is default If IGNORE is given any bad status is ignored If CLEAR is specified all status information for the checked program steps is erased so there will be no crash at subsequent status checks Examples STATUS HF CRASH will check the status of the last HF SCF step and crash if it was not o k i e no convergence CRASH is u
490. subtracting the two corresponding monomer correlation energies from the intramolecular correlation component of the complex given in the output Alternatively the following form can be used ENEPART RMAX rl r2 r3 and the program will then print the energy contributions of all pairs in the ranges between the given distances in bohr enclosed by square brackets e g enepart rmax 0 3 5 7 9 11 A second list in which the contributions are given as a function of the number of bonds between the pair domains will also be printed 28 9 Doing it right The local correlation methods in MOLPRO employ localized molecular orbitals LMOs Pipek Mezey localization is recommended but Boys localization is also possible The virtual orbital space is spanned by non orthogonal projected atomic orbitals PAOs The local character of this basis makes it possible to introduce two distinct approximations first excitations are restricted to domains which are subspaces of PAOs that are spatially close to the orbitals from which the electrons are being excited Secondly the orbital pairs are classified according to their importance based on distance or connectivity criteria and only strong pairs are treated at the highest level e g CCSD The remaining weak and distant pairs are treated at the LMP2 level and very distant pairs are neglected These approximations lead to linear scaling of the computational resources as a function of the molecula
491. t but it works 92227207 31376891 73536433 64753557 41652680 77903293 93094231 98812258 00939154 01708679 01988143 02088864 02125263 02138387 3124 4833 5450 5672 5752 5781 5792 5796 5797 5797 OO UQ OS Q OQ G QOO O0 OO OS o ro DIP 17407361 06209922 10199751 79658706 43669203 17616098 05644998 63401784 91637513 76319435 86107911 80513445 83990621 81956198 83202128 82464809 82912805 82646089 82807428 82711046 82769196 82734386 82755355 82742787 302 REFERENCES 303 References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 A D Becke Phys Rev A 38 3098 1988 J P Perdew and Y Wang Phys Rev B 45 13244 1992 J P Perdew K Burke and M Ernzerhof Phys Rev Lett 77 3865 1996 C Adamo and V Barone J Chem Phys 110 6158 1999 J P Perdew J A Chevary S H Vosko K A Jackson M R Pederson and C Fiolhais Phys Rev B 46 6671 1992 R Strange F R Manby and P J Knowles Computer Physics Communications 136 310 2001 B Miehlich A Savin H Stoll and H Preuss Chem Phys Lett 157 200 1989 A D Becke J Chem Phys 85 7184 1986 A D Becke J Chem Phys 107 8554 1997 A D Becke J Chem Phys 84 4524 1986 A D Becke J Chem Phys 88 1053 1998 A D
492. t can be used except if statement is a procedure name ELSEand ELSE IF can be used exactly as in Fortran IF statements may be arbitrarily nested Jumps into IF or ELSE IF blocks are allowed In this case no testing is performed when an ELSE is reached control continues after ENDIF The logical expression may involve logical comparisons of algebraic expressions or of strings Examples IF STATUS LT 0 THEN TEXT An error occurred calculation stopped STOP ENDIF IF Smethod eq HF then ENDIF In the previous example the dollar and the quotes are optional IF METHOD EQ HF then ENDIF 6 7 2 GOTO commands GOTO commands can be used to skip over parts of the input The general form is GOTO command n nrep Program control skips to the n th occurrence of command Default n 1 command must be a keyword in the first field of an input line If n is positive the search is forward starting from the current position If n is negative search starts from the top of the input The GOTO command is executed at most nrep times The default for nrep is 1 if n 0 and infinity otherwise We recommend that GOTO commands are never used to construct loops Alternatively one can jump to labels using GOTO label Since labels must be unique the search starts always from the top of the input It is required that the abel ends with a colon 6 7 3 Labels LABEL LABEL This is a du
493. t by 1 examples text CP calculation for HF2 MONOMER hfdimer cpcopt2 com dummy f1 h1 first hf is now dummy hf accu 16 scf for second monomer mp2 mp2 for second monomer ehf2 energy save mp2 energy in variable forces compute mp2 gradient for first monomer add 1 subtract from previous gradient text DIMER CALCULATION dummy reset dummies hf accu 16 Iscf for dimer mp2 mp2 for dimer edimer energy save mp2 energy in variable forces compute mp2 gradient for dimer add 1 ladd to previous gradient optg gradient d 5 startcmd label find next energy text optimized geometry parameters show rhf rff thetal theta2 text computed interaction energies de edimer ehfl ehf2 tocm ICPC corrected interaction energy with fixed monomers 40 VIBRATIONAL FREQUENCIES FREQUENCIES 280 40 VIBRATIONAL FREQUENCIES FREQUENCIES FREQUENCIES method SYMM flag START rec ifil DUMP dumprec ifil Calculate harmonic vibrational frequencies and normal modes To get reasonable results it is necessary to do a geometry optimization before using the frequency calculation This option uses a hessian matrix calculated numerically from 3N cartesian coordinates Z Matrix coordi nates will be destroyed on this entry The hessian is calculated analytically or numerically by finite differences from the input coordinates In numerical differentiation if analytic gradients are available these are differentiated once to build the hessian otherwis
494. t by the program A number of variables are predefined by the program The following variables can be used to convert between atomic units and other units EV 1 d0 27 2113961d0 HARTREE KELVIN 1 d0 3 157733d5 HARTREE KJOULE 1 d0 2625 500d0 HARTREE KCAL 1 d0 627 5096d0 HARTREE CM 1 d0 219474 63067d0 HARTREE CM 1 1 d0 219474 63067d0 HARTREE H H A A D Z 1 d0 6 5796838999d15 HARTRE ERTZ 1 d0 6 5796838999d15 HARTREE G 1 d0 0 529177249d0 BOHR GSTROM 1 d0 0 529177249d0 BOHR ral OEV 27 2113961d0 EV TOK 3 157733d5 K OKELVIN 3 157733d5 K TOCM 219474 63067d0 CM 1 8 VARIABLES HERTZ2 6 5796838999d HZ 6 5796838999d15 KJ 2625 500d0 KJ MO KJOULE 2625 500d0 K KCAL 627 5096d0 KCA A 0 529177249d0 ANG G 0 529177249d0 A D 48 15 HZ HZ J MOL MOL STROM NGSTROM 00000000 DEBYE 2 54158d0 DEB YE Further variables which are set during execution of the program INTYP INTDONE CARTESIAN SCFDONE NUMVAR STATUS CHARGE NELEC SPIN ORBITAL LASTORB LASTSYM LASTSPIN LASTNELEC ENERGR istate ENERGY istate ENERGD istate ENERGP istate ENERGT 1 ENERGT 2 ENERGT 3 defines integral program to be used Either INTS Seward or INTP Argos has the value true if the integrals are done for the current geom etr
495. t is 1 logical Use 2 point central formula needs 2M energy calculations for M degrees of freedom logical Use forward gradients needs only M 1 energy calcula tions but less accurate logical Use 4 point formula for accurate numerical gradient needs 4M energy calculations 38 ENERGY GRADIENTS 248 NUMERICAL logical Force the use of numerical gradients even if gradients are available VARSAV logical Save gradient in variables GRADX GRADY GRADZ Example hf cosdit forces numerical The program will then automatically repeat HF and CCSD T at as many geometries as needed for evaluating the gradient This is equivalent to hf ccsd t forces numerical startcmd hf or using a procedure forces numerical proc runccsdt runccsdt hf ccsd t 38 2 1 Choice of coordinates COORD By default the numerical gradients are computed relative to all variables on which the z matrix depends If the z matrix depends on no variables or on 3N variables the gradient is computed for all 3N coordinates and symmetrical displacement coordinates are used to evaluate the gradient This yields the minimum computational effort These defaults can be modified using the COORD directive COORD coord type displacement type where coord type can be one of the following ZMAT Compute the numerical gradients for all variables on which the geometry depends default 3NorCART Compute the gradients for all 3N nuclea
496. t scal ing Chem Phys Letters 318 370 2000 4 M Sch tz Low order scaling local electron correlation methods III Linear scaling local perturbative triples correction T J Chem Phys 113 9986 2000 5 M Sch tz and H J Werner Low order scaling local electron correlation methods IV Lin ear scaling local coupled cluster LCCSD J Chem Phys 114 661 2001 6 M Sch tz Low order scaling local electron correlation methods V Connected Triples be yond T Linear scaling local CCSDT 1b J Chem Phys 116 8772 2002 7 M Sch tz A new fast semi direct implementation of Linear Scaling Local Coupled Cluster Theory Phys Chem Chem Phys 4 3941 2002 Multipole treatment of distant pairs 8 G Hetzer P Pulay H J Werner Multipole approximation of distant pair energies in local MP2 calculations Chem Phys Lett 290 143 1998 Linear scaling local MP2 9 M Sch tz G Hetzer and H J Werner Low order scaling local electron correlation meth ods I Linear scaling local MP2 J Chem Phys 111 5691 1999 28 LOCAL CORRELATION TREATMENTS 177 10 G Hetzer M Sch tz H Stoll and H J Werner Low order scaling local electron corre lation methods II Splitting the Coulomb operator in linear scaling local MP2 J Chem Phys 113 9443 2000 Density fitted local methods 11 H J Werner F R Manby and P J Knowles Fast linear scaling second order Moller Plesset perturbat
497. t to EXCLUSIVE the program will only add orbitals whose domains are exclusively covered by the given atoms Any local correlation treatment can be given as method with the restriction that only MP2 and HF can be used as default method Up to two REGION directives may be included in a single calculation ordered according to the correlation level method specified for the region The highest level region should be given last 28 LOCAL CORRELATION TREATMENTS 189 It is advisable to check the region orbital list and the orbital domains printed by the program The use of regions may significantly reduce the computation time and provided the active atoms are sensibly chosen may give still sufficiently accurate results for the active group e g bond lengths and bond angles 28 8 0 Domain Merging MERGEDOM The restriction of the virtual space in local calculations may result in discontinuities for reaction path calculations due to changes of the geometry dependent domains This may be avoided by the use of a MERGEDOM directive MERGEDOM NE 1GHBOUR value CENTERS atoml atom2 RECORD CHECK This directive provides augmented domains which can be saved using option or directive SAVE see section 28 8 3 for later use in reaction paths or in single point calculations in cases where the orbital domain description is unbalanced The use of the neighbour option works in the same way as the local option MERGEDOM with value specifying t
498. ta set see also SAVE command If namout file is omitted and no SAVE card is present the new orbitals are not saved All output orbitals must be supplied via ORBITAL and ADD MOVE EXTRA or PROJECT directives before they can be saved 42 1 Defining the input orbitals ORBITAL ORBITAL namin file specifications Reads an input orbital set from a dump record specifications can be used to select specific orbital sets as described in section Subsets of these orbitals can be added to the output set by the ADD MOVE or EXTRA commands 42 2 Moving orbitals to the output set MOVE MOVE orb1 syml orb2 sym2 orb3 sym3 ioff fac istart iend Moves orbitals orb sym to orb2 sym2 from the input set to the first vector of symmetry sym3 in the output set which is undefined so far The first orb3 1 vectors in the output set are skipped regardless of whether they have been defined before or not If sym2 gt syml sym3 will run from syml to sym2 and the input for sym3 has no effect If orbl sym is negative abs orb1 is the maximum number of orbitals to be moved starting with orbital sym up to orb2 sym2 If orb2 sym2 is negative abs orb2 is the maximum number of vectors to be moved starting at orbl isyml up to the last orbital in symmetry sym2 Orbitals from the input set which have already been moved or added to the output set are gen erally skipped If orb and orb2 are zero the whole input set is moved to the output set In this case the
499. tals saved on 2110 2 c SCF for the O atom with dummy basis on the N atom orbitals saved on 2120 2 d Merge the atomic SCF orbitals Finally obtain the virtual orbitals by projecting the merge orbitals out of the SCF orbitals for NO 42 ORBITAL MERGING 292 SRevision 2006 0 NO merge geometry n o n r r 2 1 irhfrocc b5b 2 1 Wf l5542 1 orbital 2100 2 dummy O rht ocGo 3 1 21 wf 7 4 3 orbital 2110 2 dummy n Prhfroct3yl113 wf 8 4 2 orbital 2120 2 MERGE ORBITAL 2110 2 OVE 1 1 1 1 OVE 2 1 3 1 3 1 OVE OVE ORBI OVE OVE OVE ROT 3 1 ROT 4 1 PRINT 1 ORTH 6 2 2 PROJ 2100 2 SAVE 2150 2 dummy multi occ 6 2 2 WE L5 2p We 75 68521 start 2150 2 rhf for NO 2Pi state save orbitals to record 2100 on file 2 loxygen is dummy rhf nitrogen 14S state save orbitals to record 2110 on file 2 Initrogen is dummy rhf for oxygen 3P state save orbitals to record 2120 on file 2 call merge program read orbitals of N atom move input vector 1 1 to output vector 1 1 move input vectors 2 1 3 1 to output vectors 3 1 and 4 1 move inpu move inpu read orbitals of O atom move input vectors 1 1 to 3 1 to output vectors Loly 5 14 62 move input vector 1 2 to output vector 2 2 move input vector 1 3 to output vector 2 3 rotate 2s orbitals to make bonding and antibonding linear combinations Ctt examples
500. tandard Temperature and Pressure T 298 150 K p 1 atm Subcommands of THERMO are PRINT THERMO SCALE factor TEMP tmin tmax tstep PRESSURE p The FREQUENCIES program sets the variable zpe containing the zero point energy of the har monic vibrations in atomic units If the THERMO option is used the variables htotal and gtotal containing the enthalpy and the free enthalpy of the system in atomic units are also set in calculating the thermodynamical properties use vibrational freq uencies scaled with factor in order to take account of systematic er rors of the wavefunction e g using SCF wavefunctions factor 0 89 is reasonable calculate the thermodynamical properties at a given pressure of p atm 40 3 Examples SRevision formaldehyde freqency calculation memory 8 m basis vdz 2006 0 gthresh energy 1 d 8 geomt yp xyz geomet ry nosym 4 FORMALDEHYDE OO hf accu 14 0 0 0000000000 0 0 0000000000 0000000000 0000000000 optg coord 3n frequencies analytic thermo sym c2v print thermo mp2 optg coord 3n frequencies thermo sym c2v print thermo 0000000000 0000000000 9325664988 9325664988 additional information such as atomic masses partition functions and thermodynamical function in calories is printed to the output calculate the thermodynamical properties at different temperatures starting with tmin K up to t
501. ted the ini tial hessian is assumed to be diagonal with values 1 hartree bohr 2 for all lengths 1 hartree radian 2 for all angles Additional matrix elements of the hessian can be defined using the HESSELEM directive see section 39 2 8 In transition state searches the Hessian is evaluated numerically in the first iteration by default Alternatively if READ is specified a previously computed hessian is used 39 2 7 Numerical Hessian NUMHESS NUMHESS options or NUMHESS hstep options If this directive is present a numerical Hessian is computed using finite differences If analytical gradients are available one can use forward gradient differences needs one gradient calcula tion for each coordinate or central differences more accurate needs two gradient calculations for each coordinate For transition state optimizations it is usually sufficient to use forward differences If analytical gradients are not available for the optimized method the energy is differentiated twice In this case only central differences are possible The following options can be given HSTEP hstep hstep 1 Don t calculate numerical hessian default for minimiza tion hstep 0 Calculate numerical hessian only once at the start of the op timization default for transition state searches hstep n Calculate numerical hessian after each n optimization steps This is useful for difficult transition state optimizations e g if the eigenvalue
502. tep Step length for distances in bohr The default is 0 01 ASTEP astep Step length for angles in degree The default is 0 5 or 1 for angles below and above 90 degree respectively DSTEP dstep Step length for symmetrical displacements in bohr The default is 0 01 CENTRAL Use central differences for gradient default FORWARD Use forward differences not recommended for gradient FOURPOINT Use four point formula for very accurate numerical gradients PROCEDURE procname Use given procedure for numerical calculation of the gradient This procedure must define a complete energy calculation orbital optimiz ation and correlation treatment VARIABLE varname Use given variable for numerical calculation of the gradient DISPLACE type The displacement type Note that the displacement type for gradient and hessian must be the same type can be one of the following SYMM Use symmetric displacement coordinates default This is the only recommended option CART Use 3N cartesian displacements not recommended This requires many more energy calculations than necessary and does not preserve the molecular symmetry UNIQUE Use symmetry unique cartesian displacements not rec ommended 39 GEOMETRY OPTIMIZATION OPTG 262 39 2 11 Transition state saddle point optimization ROOT ROOT root Specifies the eigenvector of the hessian to be followed root 1 specifies a minimization default root 2 specifies a transition state
503. ternal names of the matrices If fac is not given facl I is assumed If fac2 is not given fac2 0 is assumed If a backquote is appended to matl or mat2 the corresponding matrix is transposed before the operation If a backquote is appended to result the resulting matrix is transposed 43 MATRIX OPERATIONS 297 43 8 Transforming operators TRAN TRAN result Op C calculates result 2 C T Op C The strings result C and Op are the internal names of the matrices If a backquote is appended to C or Op the corresponding matrix is transposed before the operation Thus TRAN result Op C computes result C Op C T 43 9 Transforming density matrices into the MO basis DMO DMO result D C calculates result C T S D S C The strings result C and D are internal names 43 10 Diagonalizing a matrix DIAG DIAG eigvec eigval matrix iprint Diagonalizes matrix The eigenvectors and eigenvalues are stored internally with associated names eigvec and eigval respectively arbitrary strings of up to 16 characters The if iprint gt 0 the eigenvalues are printed If iprint gt 1 also the eigenvectors are printed 43 11 Generating natural orbitals NATORB NATORB name dens thresh computes natural orbitals for density matrix dens Orbitals with occupation numbers greater or equal to thresh default 1 d 4 are printed 43 12 Forming an outer product of two vectors OPRD OPRD result matrix orb1 orb2
504. tes of different symmetry Many options can be specified on the MULTI command line MULTI options Selected options MAXIT Max number of iterations default 10 ENERGY Convergence threshold for energy GRADIENT Convergence threshold for gradient STEP Convergence threshold for steplength FAILSAFE logical Use options for more robust convergence Many further options and thresholds which can also be given on the command line are de scribed in section 20 8 5 20 2 Defining the orbital subspaces 20 2 1 Occupied orbitals OCC ni n2 ng ni specifies numbers of occupied orbitals including FROZEN and CLOSED in irreducible representation number i In the absence of an OCC card the information from the most recent MCSCF calculation is used or if there is none those orbitals corresponding to a minimal valence set i e full valence space are used 20 2 2 Frozen core orbitals FROZEN n 7n5 record file nj is the number of frozen core orbitals in irrep number i These orbitals are doubly occupied in all configurations and not optimized Note that in earlier MOLPRO versions this directive was called CORE and has now been renamed to avoid confusion with CORE orbitals in the MRCI and CCSD programs record file is the record name for frozen core orbitals if not supplied taken from orb on START card record file can be specified in any field after the last nonzero nj It should always be given if the orbital guess is from a
505. texp Special parameters for non linear optimization scheme For experts only Old form obsolete THRESH thrpri thrpun varmin varmax thrdiv thrdoub New form THRESH THRPRI thrpri T HRPUN thrpun VARMIN varmin VARMAX varmax THRD1V thrdiv THRDOUB thrdoub thrpri threshold for printing CI coefficients default 0 04 thrpun threshold for writing CI coefficients to the punch file Default 1s no write to the punch file varmin varmax thresholds for non linear optimization scheme For experts only thrdoub threshold for detecting almost doubly occupied orbitals for in clusion into the pseudo canonical set default 0 i e the feature is disabled DIIS disvaraugvar maxdis maxaug idsci igwgt igvec idstrt idstep Special parameters for DIIS convergence acceleration For experts only 20 8 6 Saving wavefunction information for CASVB VBDUMP vbdump For users of the valence bond program CASVB all wavefunction information that may subse quently be required is saved to the record vbdump The default is not to write this information If the keyword is specified without a value for vbdump then record 4299 2 is used This keyword is not needed prior to variational CASVB calculations 20 8 7 Saving transformed integrals TRNINT trnint trnint specifies the record name for integrals in the basis of active CASSCF MOs These are used for example by CASVB see section 36 5 The default value for trnint is 1900 1 2
506. that the definition of structures depends on the value of SPINBASIS Doubly occupied orbitals occur first in all configurations and the spin eigenfunctions are based on the singly occupied orbitals being in ascending order 36 7 3 Read orbitals or structure coefficients The READ keyword can take one of the following forms READ ORB iorb1 TO iorb2 AS jorb1 TO jorb2 FROM record READ STRUC istrucl TO istruc2 AS jstrucl TO jstruc2 F ROM record READ ALL FROM record In this way a subset of orbitals and or structure coefficients may be picked out from a previous calculation Renumbering of orbitals or structures can be done using the AS construct as outlined above If the VB wavefunction was previously saved in the AO basis the orbitals will 36 THE VB PROGRAM CASVB 231 be projected onto the present active space note that it is necessary to specify a record name for the molecular orbitals orb in the START commmand for this to be possible Default for record is the vb record name specified in keyword START if applicable 36 8 Permuting orbitals ORBPERM R Upact Permutes the orbitals in the valence bond wavefunction and changes their phases according to 0 sign i 1 Pabs i The guess may be further modified using the GUESS keyword Also the structure coefficients will be transformed according to the given permutation note that the configuration list must be closed under the orbital permutation fo
507. the whole Bragg Slater radius was used and setting scale to a value other than 1 allows a different amp to be used m is not necessary for this radial scheme AHLRICHS is the radial scheme defined by O Treutler and R Ahlrichs J Chem Phys 102 1995 346 It is based on the transformation their M4 mapping 2 4 with using standard Gauss Chebyshev quadrature of the second kind for the x space integration m is not necessary for this radial scheme 1 0 6 r log 2 x og no n n2 n3 are the degrees of quadrature n see equation 3 of Murray et al for hydro gen helium first row second row and other elements respectively accr as given by the THR command specifies a target accuracy the number of radial points is chosen according to a model instead of using an explicit n The stricter of nj acer is used unless either is zero in which case it is ignored 18 THE DENSITY FUNCTIONAL PROGRAM 103 18 3 3 Angular integration grid ANGULAR ANGULAR method acca crowd LMIN fin pus pm pen T max max max max LMAX pr do cs Specify the details of the angular quadrature scheme The default choice for method is LEBEDEV ie as in A D Becke J Chem Phys 88 1988 2547 which provides angular grids of octahe dral symmetry The alternative choice for method is LEGENDRE which gives Gauss Legendre quadrature in 0 and simple quadrature in Q as defined by C W Murray N C Handy and G J Laming M
508. this chapter For the details of the algorithms used see J Chem Phys 82 5053 1985 Chem Phys Letters 115 259 1985 Advan Chem Phys 59 1 1987 20 12 Examples The simplest input for a CASSCF calculation for H20 C symmetry is simply geometry 0o hl o r h2 0 r hl theta Z matrix geometry input r l ang bond length theta 104 bond angle hf do scf calculation multi do full valence casscf This could be extended for instance by the following input cards occ 4 1 2 specify occupied space CLOSED 2 specify closed shell inactive orbitals FROZEN 1 specify frozen core orbitals WF 10 1 define wavefunction symmetry ORBITAL 2140 2 save final orbitals to record 2140 file 2 NATORB PRINT CI print natural orbitals and diagonalize the hamiltonian for the natural orbitals The largest CI coefficients i START 2100 2 read guess orbitals from record 2100 file 2 i are printed examples h20 casscf com 20 THE MCSCF PROGRAM MULTI 130 Example for a state averaged calculation for CN X and B Y states and A TT 211 states averaged A full valence CASSCF calculation is performed SRevision 2006 0 xxx cn r 2 2 geometry c n c r trhf occ b l l wE l3 l 1 orbital 2100 2 multayoce 6 24 2 0losed 2 start 2100 2 save ref 4000 2 wf l3 1 1 state 2 wf 13 2 1 wf 13 3 1 natorb ci print tran lz expec2 1zz define bond length RHF calculation for
509. ting DF programs for Hartree Fock DF HF Density functional Kohn Sham theory DF KS Second order Mpller Plesset perturbation theory DF MP2 as well as for all local methods DF LMP2 DF LMP4 DF LQCISD T DF LCCSD T Analytical QCISD T gradients Analytical MRPT2 CASPT2 and MS CASPT2 gradients using state averaged MCSCF reference functions Analytical DF HF DF KS DF LMP2 and DF SCS LMP2 gradients Explicitly correlated methods with density fitting DF MP2 R12 2A DF MP2 F12 2A as well as the local variants DF LMP2 R12 2 A loc and DF LMP2 F12 2 A loc Coupling of multi reference perturbation theory and configuration interaction CIPT2 DFT SAPT Transition moments and transition Hamiltonian between CASSCF and MRCI wavefunc tions with different orbitals A new spin orbit integral program for generally contracted basis sets Douglas Kroll Hess Hamiltonian up to arbitrary order Improved procedures for geometry optimization and numerical Hessian calculations in cluding constrained optimization Improved facilities to treat large lattices of point charges for QM MM calculations in cluding lattice gradients An interface to the MRCC program of M Kallay allowing coupled cluster calculations with arbitrary excitation level Automatic embarrassingly parallel computation of numerical gradients and Hessians mppx Version Additional parallel codes e g DF HF DF KS DF LCCSD T partly including triple
510. tion SPARSE value If Non zero use sparse algorithm in second half transforma tion default See section LI for a more general description of density fitting At present expectation values and gradients cannot be computed with DF MP 2 but work with the local variant DF LMP2 23 M LLER PLESSET PERTURBATION THEORY 159 23 3 Spin component scaled MP2 SCS MP2 The spin component scaled MP2 energy as proposed by Grimme J Chem Phys 118 9095 2003 is printed automatically using the default scaling factors 1 2 for antiparallel spin 1 3 for parallel spin These factors can be modified using the options SCSFACS and SCSFACT respectively i e MP2 SCSFACS facs SCSFACT fact The SCS MP2 total energy is stored in the variable EMP 2_SCS Gradients can be computed for SCS MP2 by setting the option SCSGRD 1 This only operational for density fitted MP2 i e using DF MP2 DF BASIS fitbasis SCSGRD 1 SCSFACS facs SCSFACT fact followed by FORCES or OPTG In the latter case the geometry is optimized using the SCS MP2 energy 24 THE CLOSED SHELL CCSD PROGRAM 160 24 THE CLOSED SHELL CCSD PROGRAM Bibliography C Hampel K Peterson and H J Werner Chem Phys Lett 190 1 1992 All publications resulting from use of this program must acknowledge the above The CCSD program is called by the CISD CCSD BCCD or QCI directives CID or CCD can be done as special cases using the NOSINGL directive The code also allows
511. tion of weights in CASVB see T Thorsteinsson and D L Cooper J Math Chem 23 105 26 1998 CIWEIGHTS keyl ey2 Ncont Prints weights of the CASSCF wavefunction transformed to the basis of nonorthogonal VB structures For the key options see VBWEIGHTS above Note that the evaluation of inverse over lap weights involves an extensive computational overhead for large active spaces Weights are 36 THE VB PROGRAM CASVB 236 given for the total CASSCF wavefunction as well as the orthogonal complement to yg The default for the number of configurations requested Neonf is 10 If Noonr 1 all configurations are included 36 12 Controlling the amount of output PRINT i Each number specifies the level of output required at various stages of the execution according to the following convention 1 No output except serious or fatal error messages 0 Minimal output Standard level of output 2 Extra output The areas for which output can be controlled are i Print of input parameters wavefunction definitions etc i Print of information associated with symmetry constraints 13 General convergence progress i4 Progress of the 2nd order optimization procedure is Print of converged solution and analysis ig Progress of variational optimization 17 Usage of record numbers on file 2 For all the default output level is 1 If is 22 VB orbitals will be printed in the AO basis provided that the definition of MOs i
512. tle r 1 85 theta 104 set geometry parameters geometry 0 z matrix geometry input e Oo x3 H2 0 r H1 theta s rampes h20 ccsdt vtz com basis VTZ luse VTZ basis hf closed shell scf ccsd t do ccsd t calculation 5 5 CASSCF and MRCI Perhaps you want to do a CASSCF and subsequent MRCI for comparison The following uses the full valence active space in the CASSCF and MRCI reference function KERNO A title r 1 85 theta 104 set geometry parameters geometry o z matrix geometry input Ki O agts h2 0 r H1 theta examples basis vtz luse VTZ basis ho mravcom hf closed shell scf ccsd t do ccsd t calculation casscf Ido casscf calculation mrci do mrci calculation 5 6 Tables You may now want to print a summary of all results in a table To do so you must store the computed energies in variables 5 INTRODUCTORY EXAMPLES h2o r 1 85 theta 104 geometry o hl 0 4 h2 0 r H1 theta basis vtz hf e 1 energy method 1 program ccsd t e 2 energy method 2 program casscf e 3 energy method 3 program mrci e 4 energy method 4 program table method e A title set geometry parameters z matrix geometry input luse VTZ basis closed shell scf save scf energy in variable e 1 save the string HF in variable method 1 do ccsd t calculation save ccsd t energy in variable e 2 save the string CCSD T in variable method 2 Ido casscf calculation save scf energy in
513. to achieve sufficient accuracy Values of at least 10 bohr have been found to work reasonably well only for F12 IAODOM Connectivity criterion for RI domain extensions Zero means full RI basis default Values greater or equal to 6 should lead to sufficiently accurate results THRAO Screening threshold for integrals in the AO or RI basis THRMO Screening threshold for half transformed integrals THRPROD Product screening threshold in the first half transformation Further options for density fitting are described in section The use of local DF and RI domains is still experimental and is not recommended yet for general use Published work arising from these methods should cite the following F R Manby J Chem Phys 119 4607 2003 for canonical DF MP2 R12 A J May and F R Manby J Chem Phys 121 4479 2004 for canonical DF MP2 F12 H J Werner and F R Manby J Chem Phys 124 054114 2006 for local DF LMP2 R12 F R Manby H J Werner T B Adler and A J May J Chem Phys 124 094103 2006 for local DF LMP2 F12 30 THE FULL CI PROGRAM 199 30 THE FULL CI PROGRAM This module is the determinant full CI program as described in P J Knowles and N C Handy Chem Phys Letters 111 1984 315 P J Knowles and N C Handy Comp Phys Commun 54 1989 75 Published work resulting from the use of this program should cite these references The program in normal use finds the lowest eigenvector of the complete CI
514. to calculate Brueck ner orbitals QCI and CCSD are identical in this case Normally no further input is needed if the CCSD card follows the corresponding HF SCF Optional ORBITAL OCC CLOSED CORE SAVE START PRINT options work as described for the MRCI program in section 21 The only special input directives for this code are BRUECKNER and DIIS as described below The following options may be specified on the command line NOCHECK Ignore convergence checks DIRECT Do calculation integral direct NOSING Do not include singly external configurations MAXIT value Maximum number of iterations SHIFTS value Denominator shift for update of singles SHIFTP value Denominator shift for update of doubles THRDEN value Convergence threshold for the energy THRVAR value Convergence threshold for CC amplitudes This applies to the square sum of the changes of the amplitudes The convergence thresholds can also be modified using THRESH ENERGY thrden COEFF thrvar Convergence is reached if the energy change is smaller than thrden default 1 d 6 and the square sum of the amplitude changes is smaller than thrvar default 1 d 10 The THRESH card must follow the command for the method e g CCSD and then overwrites the corresponding global options see GTHRESH sec 6 11 The computed energies are stored in variables as explained in section 8 8 As well as the energy the T diagnostic T J Lee and P R Taylor Int J Quant
515. tra symmetries are ordered with increasing quantum number of the basis functions This information can be used to determine and fix the extra symmetries of the molecular orbitals by means of the SYM command SYM irrep sym 1 sym 2 sym n sym i are the extra symmetries for the first n orbitals in the irreducible representation irrep For instance if you want that in a linear molecule the orbitals 1 1 to 3 1 are o and 4 1 5 1 6 the SYM card would read calculation done with X Y as symmetry generators SYM 1 1 1 1 2 2 If necessary the program will reorder the orbitals in each iteration to force this occupation The symmetries of occupied and virtual orbitals may be specified By default symmetry contamina tions are not removed If irrep is set negative however symmetry contaminations are removed Note that this may prevent convergence if degenerate orbitals are present 17 THE SCF PROGRAM 97 17 7 Expectation values EXPEC oper opero 0per Calculates expectation values for one electron operators operi oper Opern See section 6 13 for the available operators By default the dipole moments are computed Normally it is recommended to use the GEXPEC directive if expectation values for other operators are of interest See section 6 13 for details 17 8 Polarizabilities POLARIZABILITY oper opero Operg Calculates polarizabilities for the given operators operi oper Opern See section for the avai
516. tron integrals relativistic scf calculation save relativistic scf energy in variable e dk non relativisitic scf calculation compute relativistic correction using Cowan Griffin operator save non relativistic energy in variable enrel show individual contribution and their sum examples ar2 rel com show mass velocity and darwin contributions and their sum show relativistic correction using Douglas Kroll This jobs shows at the end the following variables ASSV AU DARWIN AU EREL AU 14 84964285 11 25455679 3 59508606 6 PROGRAM CONTROL 38 Table 5 One electron operators and their components Generic Parity Components Description name OV 1 Overlap EKIN 1 Kinetic energy POT 1 potential energy DELTA 1 delta function DEL4 1 A DARW 1 one electron Darwin term i e DELTA with appropriate factors summed over atoms MASSV 1 mass velocity term i e DEL4 with appropriate factor REL 1 total Cowan Griffin Relativistic correction i e DARW MASSV DM 1 DMX DMY DMZ dipole moments SM 1 XX YY 22 XY XZ YZ second moments TM 1 XXX XXY XXZ XYY XYZ XZZ YY Y YYZ YZZ ZZZ third moments MLTPn 1 all unique Cartesian products of order n multipole moments OM 1 OMXX OMY Y QMZZ OMXY OMXZ OMYZ quadrupole moments and R QMRR XX YY ZZ QMXX 3 XX RR 2 QMXY 3 XY 2 etc EF 1 EFX EFY EFZ electric field FG 1 FGXX FGYY FGZZ FGXY FGXZ FGYZ el
517. ulate potential energy surfaces for several excited states using restart from a previous calculation 25 EXCITED STATES WITH EQUATION OF MOTION CCSD EOM CCSD 167 PES for several lowest states of hydrogen fluoride memory 2 m basis avdz define basis set geometry h f h r z matrix r 0 8 Ang start from this distance do n 1 100 loop over distances rr n r save distance for table hf do SCF calculation ccsd do CCSD calculation try to restart start 4000 2 save 4000 2 and save final T amplitudes om 2 1 1 2 1 4 start 6000 2 save 6000 2 do EOM CCSD calculation try to restart and save final excited states amplitudes examples hf eom pes com ebase n energy 1 save ground state energy for this geometry 2 n energy 2 energy 1 save excitation energies for this geometry 3 n energy 3 energy 1 n 4 n energy 4 rgy 1 r r 0 01 increment distance enddo end of do loop table rr ebase e2 e3 e4 make table with results 54 9 modify number of digits 2 1 E EXC 1 2 E EXC 1 4 modify headers of table digrus 2 8 5 555 5 9 45 head R Ang EGRST E EXC title of table title EOM CCSD excitation energies for hydrogen fluoride in hartree basis Sbasis save hf eom ccsd tab save table in file This calculation produces the following table EOM CCSD excitation energies for hydrogen fluoride in hartree basis AVD
518. ulations may be performed using one of the following commands DFT calculate functional of a previously computed density RKS or RKS SCF calls the spin restricted Kohn Sham program KS and KS SCF are aliases for RKS UKS or UKS SCF calls the spin unrestricted Kohn Sham program Each of these commands may be qualified with the key names of the functional s which are to be used and further options command keyl key2 key3 options If no functional keyname is given the default is LDA see below Following this command may appear directives specifying options for the density functional modules see section 18 2 or the Hartree Fock program see section 17 1 On completion of the functional evaluation or self consistent Kohn Sham calculation the val ues of the individual functionals are stored in the MOLPRO vector variable DF TFUNS the total is in DFTFUN and the corresponding individual functional names in DFTNAME Energy gradients are available for self consistent Kohn Sham calculations Normally sensible defaults are used to define the integration grid The accuracy can be con trolled using options as described in section or directives as described in section 18 2 More control is provided by the GRID command as described in section 18 1 Options The following options may be specified on the KS or UKS command lines GRID 1target Specifies the grid target accuracy per atom The default is 1 d 6 unless this has been
519. um number of iterations in EOM CCSD default 50 MAXEXTRA maxex Maximum number of extra configurations allowed to be included in initial Hamiltonian default 0 In the case of near degeneracy it is better to include a few extra configurations to avoid a slow conver gence EOMLOCAL eoml If set to O non local calculation default EOMLOCAL 1 switchs on the local module experimental INIMAX ini Number of CSFs included in initial Hamiltonian used only if INISINGL and INIDOUBL are both zero All keywords can be abbreviated by at least four characters 25 3 Options for EOMPRINT card The following print options are mostly for testing purposes and for looking for the convergence problems EOMPRINT key1 valuel key 2 value2 where the following keywords key are possible DAVIDSON ipr Information about Davidson procedure ipr 1 print results of each small diagonalization ipr 2 also print warning information about complex eigenvalues ipr 3 also print hamiltonian and overlap matrix in trial space 25 EXCITED STATES WITH EQUATION OF MOTION CCSD EOM CCSD 166 DIAGONAL ipr PSPACE ipr HEFF ipr RES IDUUM ipr LOCEOM ipr POPUL ipr INTERMED ipr 25 4 Examples Information about configurations ipr 1 print the lowest approximate diagonal elements of the trans formed hamiltonian ipr 2 print orbital labels of important configurations ipr 3 print all approximate diagonal elements ipr 4 also print the l
520. unit This holds for the following variables CHARGE Total charge of the molecule NELEC Number of electrons SPIN Spin quantum number given as 2 M_S integer SCFSPIN Same as SPIN but only for HF MCSPIN Same as SPIN but only for MCSCF CISPIN Same as SPIN but only for MRCI STATE State to be optimized MCSTATE Same as STATE but only for MCSCF CISTATE Same as STATE but only for MRCI SYMMETRY State symmetry SCFSYM METRY Same as SYMMETRY but only for HF MCSYM METRY Same as SYMMETRY but only for MCSCF CISYM METRY Same as SYMMETRY but only for MRCI ZSYMEL Symmetry elements LQUANT Lambda quantum number for linear molecules OPTCONV Geometry optimization convergence criterion PROGRAM Last program name CPUSTEP CPU time of last program step SYSSTEP System time of last program step WALLSTEP Elapsed time of last program step FOCKDONE Indicates if closed shell fock operator is available 8 5 Macro definitions using string variables String variables for which the stored string has the form of an algebraic expression are evaluated to a number if they are preceded by two dollars Example string a b a 3 b 4 text This is string string which evaluates to string prints This is string atb which evaluates to 7 This can be used to define simple macros which can be used at various places in the subsequent input For instance 8 VARIABLES 45 ECORR
521. unt of nonlocal exact exchange in hybrid DFT SAPT calculations Threshold for density matrix convergency in the coupled perturbed Kohn Sham program Maximum number of iterations in the coupled perturbed Kohn Sham program 32 PROPERTIES AND EXPECTATION VALUES 206 32 PROPERTIES AND EXPECTATION VALUES 32 The property program The property program allows the evaluation of one electron operators and expectation values Normally the operators are computed automatically when using the global GEXPEC directive see section 6 13 or the EXPEC or TRAN commands in the SCF MCSCF and CI programs The explicit use of the property program is only necessary in the rare case that the user is interested in an orbital analysis of the properties 32 1 1 Calling the property program PROPERTY PROPERTY invokes the property program 32 1 2 Expectation values DENSITY DENSITY record file specifications If this card is present the density matrix will be read from record record file and property expec tation values will be calculated If the specification record file is omitted the last dump record is used Density matrices for specific states can be selected using specifications as explained in section 11 Note that the density matrices are stored in the same record as the orbitals 32 1 3 Orbital analysis ORBITAL ORBITAL record file specifications If this card is present the orbitals are read from record record file and an
522. update of hessian default Use BFGS update of the inverse hessian Use Conjugate Gradient update see also CUT TRUST Don t do any update In transition state optimizations type may be 39 GEOMETRY OPTIMIZATION OPTG 261 PMS Combined Powell Murtagh Sargent update of hessian default POWELL Use Powell s update of the hessian MS Use update procedure of Murtagh and Sargent NONE Don t do any update 39 2 10 Numerical gradients NUMERICAL NUMERICAL options active step active2 step2 With this directive the gradients are computed by finite differences step is the increment for the active geometry parameter active For active parameters which are not specified the default values are used By default the increment is 0 01 bohr for bond distances and 0 5 or 1 degree for angles less than or greater than 90 degrees respectively These defaults can be modified by specifying RSTEP or ASTEP DSTEP is the length of symmetrical displacements which are used if the optimization is performed in 3N coordinates For each active variable two energy calculations are necessary in each geometry optimiza tion step so numerical optimizations may be expensive In optimizations of 3N coordinates symmetrical displacement coordinates are normally used to minimize the number of energy calculations see section 38 2 1 For optimization of special energies see VARIABLE section 39 2 17 The following options can be given RSTEP rs
523. used in the SCF iteration It is either possible to generate an initial orbital guess or to start with previously optimized orbitals Alternatively one can also use a previous density matrix to construct the first fock operator If the START card is absent the program tries to find suitable starting orbitals as follows First Try to read orbitals from record specified on the ORBITAL or SAVE card or the corresponding default see ORBITAL AII files are searched Second Try to find orbitals from a previous SCF or MCSCF calculation All files are searched Third If no orbitals are found the starting orbitals are generated using approximate atomic densities or eigenvectors of h see below Since these defaults are usually appropriate the START card is not required in most cases 17 4 1 Initial orbital guess An initial orbital guess can be requested as follows START TYPE option The option keyword can be HO Use eigenvectors of h as starting guess ATDEN Use natural orbitals of a diagonal density matrix constructed using atomic occupation numbers The atomic density guess works very well with minimal or generally contracted basis sets for which the first contracted basis functions correspond to the atomic 1s 25 2p orbitals e g Dunning s cc pVnZ sets the STO 3G or the 6 31G bases For such basis sets ATDEN is used by default If a segmented basis set with several contractions for each shell is used ATDEN should n
524. using Z matrix lexamples allene optscf com Revision 2002 10 Allene geometry optimization using Z Matrix memory l m basis sto 3g rcc 1 32 ang rch 1 08 ang acc 120 degree Geometry C1 C2 cl ECO Ol pel rec C3 22 ree hlgcl rch h2 cl rch HE 3 On h4 c3 rch hf optg saveact allene C2 45 cl 180 q1 0 c2 acc ql1 0 6Z2 000 h1 180 pO2 ace i 90 C2 pace 42 90 Z matrix input examples allene optscf com dat savexyz allene xyz default optimization using model hessian Save optimized variables in file allene dat Save optimized geometry in xyz style in in fil 39 4 2 Optimization using natural internal coordinates BMAT 39 GEOMETRY OPTIMIZATION OPTG 271 lexamples allene opt bmat com Revision 2002 10 Allene geometry optimization using natural internal coordinates memory l m basis sto 3g rcc 1 32 ang rch 1 08 ang acc 120 degree Geometry nosym CI Z matrix input UG Gl rege QLclyrec c2 45 3 02 EC aL L80 ql 00 hl cl rch c2 acoc q1 0 42 51 reh c2 001 180 h3 G3 rch cZ2 cc hl 90 h4 cS3 rch c2 acc h2 90 examples allene opt bmat com hf optg default optimization using model hessian coord bmat use natural internal coordinates optg coord bmat same as above 39 4 3 MP2 optimization using a procedure lexamples allene_optmp2 com Revision 2002 10 Allene geometry optimization using Z Matrix memory 2 m basis vdz rcc
525. values are displayed The equal signs may be omitted The following codes are allowed max 7 per card NSTATE see state card NSTATI number of states calculated in internal CI NSTATR see refstat card NCEPA see CEPA card NOKOP if nonzero skip integral transformation ITRDM if ge O transition moments are calculated ITRANS if nonzero perform full integral transformation not yet imple mented IDIP Print dipole moments from iteration number value REFOPT if nonzero optimize reference coefficients otherwise extract reference coefficients from internal CI IAVDEN average HII and HSS denominators over spin couplings if nonzero IDELCG if ne O then destroy files icfil igfil at end IREST 1f nonzero restart NATORB if nonzero natural orbitals are calculated and printed The number of printed external orbitals per symmetry is min natorb 2 WFNAT if nonzero natural orbitals are saved to this record IPUNRF if nonzero punch coefficients of reference configurations NPUPD if nonzero update pairs in nonorthogonal basis otherwise in orthogonal basis 21 THE CI PROGRAM MAXIT MAXITI MAXDAV MAXVI NOSING NOPAIR MXSHRF IKCPS 0 IKCPS 1 IKCPS 2 IOPTGM IOPTGM 0 IOPTGM 1 IOPTGM 2 IOPTOR 141 see maxiter card see maxiter card see maxdav card see maxdav card see nosing card see nopair card see select card In CIKEXT only K
526. variable e 3 save the string CASSCF in variable method 3 do mrci calculation save scf energy in variable e 4 save the string MRCI in variable method 4 print a table with results title Results for H20 basis basis Ititle for the table This job produces the following table Results for H20 basis VTZ METHOD E HF 76 05480122 CCSD T 76 33149220 CASSCF 76 11006259 MRCI 76 31960943 24 examples h2o table com 5 INTRODUCTORY EXAMPLES 25 5 7 Procedures You could simplify this job by defining a procedure SAVE_E as follows SRevision 2006 0 KEK ZO A title proc save e define procedure save if i eq 0 i 0 linitialize variable i if it does not exist i itl lincrement i e i energy save scf energy in variable e i method i program save the present method in variable method i endproc lend of procedure r 1 85 theta 104 set geometry parameters geometry o z matrix geometry input ELORE h2 0 r H1 theta examples basis vtz luse VTZ basis h20 ERU AE hf closed shell scf P save e call procedure save results ccsd t do ccsd t calculation save e call procedure save results casscf Ido casscf calculation save e call procedure save results mrci do mrci calculation save e call procedure save results table method e print a table with results title Results for H20 basis basis Ititle for the table The job produces the same table as before If y
527. velopers of MOLPRO would appreciate re ceiving a report There is a web based mechanism at http www molpro net bug at which as many details as possible should be filled in make test produces a file of the form testjobs report tar gz that contains some details of the MOLPRO installation and the output files of the failing test jobs You should normally attach this file to the bug report Please note that the purpose of such bug reports is to help the developers improve the code and not for providing advice on installation or running A 3 9 Installing the program for production Although the program can be used in situ it is usually convenient to copy only those files needed at run time into appropriate installation directories as specified at configuration time see section A 3 3 and stored in the file CONFIG To install the program in this way do make install The complete source tree can then be archived and deleted If multiple Linux executables have been generated see section A 3 4 they can be installed using make MPPLIB libname ARCH procname install into the same INSTBIN and INSTLIB directories but note that the TNSTLIB directories must be distinct for 14 and i8 versions The overall effect of this is to create in the INSTBIN directory an executable command file of the form name arch mpplib where name is one of molpros molprop corresponding to serial or parallel execution If the file INSTBIN name does not already exist or
528. version instead In the case of BLAS you should enter a number between 1 2 and 3 if for example you specify 2 the system libraries will be used for level 2 and level 1 BLAS but MOLPRO s internal routines will be used for level 3 i e matrix matrix multiplication Normally however one would choose either 0 or 3 If a system BLAS is chosen you will be prompted to enter appropriate linker options e g L usr lib lblas to access the libraries A special situation arises if 64 bit integers are in use 18 since on many platforms the system BLAS libraries only supports 32 bit integer arguments In such cases e g IBM SGI SUN either 0 or 4 can be given for the BLAS level BLAS 0 should always work and means that the MOLPRO Fortran BLAS routines are used On some platforms IBM SGI SUN BLAS 4 will give better performance in this case some 32 bit BLAS routines are used from the system library these are then called from wrapper routines which convert 64 to 32 bit integer arguments Note that this might cause problems if more than 2 GB of memory is used For good performance it is important to use appropriate BLAS libraries in particular a fast implementation of the matrix multiplication dgemm is very important for MOLPRO Therefore you should use a system tuned BLAS library whenever available Specification of BLAS libraries can be simplified by placing any relevant downloaded libraries in the directory blaslibs configure search
529. voked by adding the prefix DF to the command name e g DF HF DF KS DF MP2 and so on Gradients are available for DF HF DF KS and DF LMP2 By default a fitting basis set will be chosen automatically that corresponds to the current orbital basis set and is appropriate for the method For instance if the orbital basis setis VTZ the default fitting basis is VIZ JKFIT for DF HF or DF KS and VTZ MP2FTIT for DF MP2 Other fitting basis sets from the library can be chosen using the DF BASIS option e g BASIS VTZ luse VTZ orbital basis DF HF DF BASIS VQZ luse VQZ JKFIT fitting basis DF MP2 DF_BASIS VOZ luse VOZ MP2FIT fitting basis The program then chooses automatically the set which is appropriate for the method Alterna tively fitting basis sets can be defined in a preceding basis block see 13 and then be refered to with their set names e g DF HF DF BASIS MYJKBASIS DF MP2 DF BASIS MYMP2BASIS where MYJKBASIS and MYMP2BASIS are sets defined in a basis block In this case it is the responsibility of the user to ensure that the basis set is appropriate for the method Further options as fully described in section 11 1 can be added on the command line In this case they are valid only for the current command Alternatively the options can be specifed on a separate DF IT directive If this is given within a command block the options are used only for the current program this is entirely equivalent to the case that the options are
530. when CADPAC is used Note however that a CPMCSCF GRAD state directive is required in the SA MCSCF calculation see Section 20 9 The gradients are then computed automatically for the 2SIGMA examples oh samcforce com state specified on the CPMCSCF card The same is true for difference gradients CPMCSCF DGRAD statel state2 and non adiabatic coupling matrix elements CPMCSCF NACM state 1 state2 It is pos sible to do several coupled perturbed MCSCF calculations one after each other in the same MCSCE In this case FORCE would use the last solution by default The information from the 38 ENERGY GRADIENTS 246 CPMCSCE is passed to the FORCE program in a certain records default5101 1 5102 1 If several CPMCSCF calculations are performed in the same MCSCF several such records may be present and a particular one can be accessed in the FORCE program using the SAMC directive SAMC record An alias for SAMC is CPMC For compatibility with earlier versions one can also use NACM record for non adiabatic couplings or DEMC record for difference gradients Example multi state 3 cpmcscf nacm 1 1 2 1 save 5101 1 do cpmcscf for coupling of states 1 1 2 1 cpmcscf nacm 1 1 3 1 save 5102 1 do cpmcscf for coupling of states 1 1 3 1 cpmcscf nacm 2 1 3 1 save 5103 1 do cpmcscf for coupling of states 2 1 3 1 force samc 5101 1 compute NACME for states 1 1 2 1 force samc 5102 1 compute NACME for states 1 1 3 1
531. whereby an existing output file is saved backup switches it on again V verbose causes the procedure to echo debugging information noverbose selects quiet operation default 2 RUNNING MOLPRO 2 e cho procedures causes the contents of the default procedure files to be echoed at run time noecho procedures selects quiet operation default procedures enables the automatic inclusion of default procedure files the de fault noprocedures disables such inclusion g use logfile causes some long parts of the program output for example dur ing geometry optimizations and finite difference frequency calcu lations to be diverted to an auxiliary output file whose name is derived from the output file by replacing its suffix usually out by 10g nouse logfile disables this facility causing all output to appear in the normal output file m memory memory specifies the working memory to be assigned to the program in 8 byte words The memory may also be given in units of 1000 words by appending the letter k to the value or in units of 1000000 with the key m or 10 with g K M G stand for 2 9 22 and 2 I main file repository directory specifies the directory where the permanent copy of any integral file file 1 resides This may be a pathname which is absolute or relative to the current directory e g would specify the current directory Normally the I directory should be equal to
532. wo integrals of a triple differ DKEXT 3 use in core algorithm and integral triples if all inte grals of a triple differ if given replaces value of SCREEN for DKEXT Largest size of merged shells in DKEXT 0 not used Shells are only merged if their size is smaller than this value 0 not used Maximum number of centres in merged shells 0 no limit Enables of disables screening in DKEXT Print parameter for DKEXT Enables of disables label swapping in DKEXT test purpose only Largest matrix block size in DKEXT only used with DKEXT gt 1 Thresholds for integral direct computation of external exchange operators DKEXT THR DKEXT THREST DKEXT THRINT DKEXT THRPROD_DKEXT THRMAX DKEXT General threshold for DKEXT If given this is used as a default for all DKEXT thresholds described below Prescreening threshold for DKEXT Defaults THR DKEXT THREST default Integral threshold for DKEXT Defaults THR DKEXT THRINT default Product threshold for DKEXT Defaults THR DKEXT THRPROD default Initial value for THREST_DKEXT in CI and CCSD calcula tions If nonzero it will also be used for DKEXT in MP3 and MP 4 SDQ calculations The threshold will be reduced to THREST_DKEXT once a certain accuracy has been reached see VARRED or latest after MAXRED iterations The initial thresh olds THRINT DKEXT and THRPROD DKEXT are obtained by multiplying their input or de
533. x is stored using the DM directive of the CI program 2 Compute the wavefunctions at the positively displaced geometry and store the CI wave function in a second record 3 If the second order three point method is used step 2 is repeated at a negatively dis placed geometry 4 Compute the transition density matrices between the states at the reference geometry and the displaced geometr ies This is done with the TRANS directive of the CI program 5 Finally the DDR program is used to assemble the matrix element Using the first order two point method only a single input line is needed DDR dr orbl orb2 trdm2 where dr is the geometry increment used as denominator in the finite difference method orb is the record holding the orbitals of the reference geometry orb2 is the record holding the orbitals of the displaced geometry and trdm2 is the record holding the transition density matrix computed from the CI vectors at R and R DR If central differences three points are used the input is as follows DDR 2 dr ORBITAL orbl orb2 orb3 DENSITY trdml trdm2 trdm3 where dr orbl orb2 are as above and orb3 is the record holding the orbitals at the negatively displaced geometry trdm1 trdm2 trdm3 are the records holding the transition densities y R R y R R DR and Y R R DR respectively If more than two states are computed simultaneously the transition density matrices for all pairs of states will be
534. y Set to one if Cartesian basis functions are used has the value true if an SCF calculation has been done for the current geometry number of variables presently defined status of last step 1 no error 1 error or no convergence Total charge of the molecule number of electrons in last wavefunction spin multiplicity minus one of last wavefunction record of last optimized orbitals set but never used in the program Type of last optimized orbitals RHF UHF UHFNAT or MCSCF Symmetry of wavefunction for last optimized orbitals 2 Ms for wavefunctions for last optimized orbitals Number of electrons in wavefunction for last optimized orbitals Reference energy for state istate in MRCI and CCSD last computed total energy for state istate for the method specified in the input e g HF MULTI CCSD T or CCSD T Total energy for state istate including Davidson correction set only in CI Total energy for state istate including Pople correction set only in CI Total energy including perturbative triples T correction set only in CCSD T QCI T Total energy including perturbative triples T correction set only in CCSD T QCI T Total energy including perturbative triples t correction set only in CCSD T QCI T holds MP2 energy in MPn CCSD BCCD or QCISD calculations and RS2 energy in MRPT2 CASPT2 calculations holds MP3 energy in MP3 and MP4 calculations and RS3 energy in MRPR3 CASPT3 calculat
535. y p 2t 1 ur ln U RUPEE wri xr lt a 460 C DENSITY FUNCTIONAL DESCRIPTIONS c 1 709921 p 0 98 0 3271 0 7035 q 0 003557 0 03229 0 007695 r 0 00625 0 02942 0 05153 t 0 00002354 0 002134 0 00003394 u 0 0001283 0 005452 0 001269 v 0 0003575 0 01578 0 001296 o 0 001867 0 005151 0 00305 T 0 031091 0 015545 0 016887 U 0 21370 0 20548 0 11125 V 7 5957 14 1189 10 357 W 3 5876 6 1977 3 6231 X 1 6382 3 3662 0 88026 Y 0 49294 0 62517 0 49671 and PLA To avoid singularities in the limit p 0 353 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 G p PF Qt zs pi qi ri fi ui vi 00 dse ps 0 F Xs zs P2 2 12 12 U2 V2 02 476 C DENSITY FUNCTIONAL DESCRIPTIONS 354 C 43 VWN3 Vosko Wilk Nusair 1980 III local correlation energy VWN 1980 III functional The fitting parameters for Ae rs 111 appear in the text shortly after equation 4 4 of the reference See reference for more details K pe 477 where x 114 34 478 TP C B 479 p e 2 ay 1 2 nC 480 E 4 3 9 4 3 481 A h 4 9 1 1 a wee A q k 11 m n1 483 q ko l2 m2 n2 484 a q k3 l3 m3 n3 485 q A p c d A in zi 2carctan 222 Oled X Ded 2x c cp in E25 2 c 2p arctan 222 Q
536. ymmetry under consideration is computed see section 22 2 In a multi state MS MR CASPT2 calculation the state for which the gradient is computed must be specified using the ROOT option default ROOT 1 i e 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 153 RS2 MIX nstates ROOT ioptroot where 1 X ioptroot lt nstates The program works with state averaged MCSCF CASSCF orbitals and no CPMCSCF directive is needed The RS2 gradient program can also be used to compute state averaged MCSCF CASSCF gradients using the NOEXC directive Level shifts can be used By default the exact gradient of the level shift corrected energy is computed For a non zero shift this requires to solve the CASPT2 Z vector equations which roughly doubles the computational effort In single state calculations it is possible to ignore the effect of the level shift on the gradient and not to solve the Z vector equation This variant which is described in the above paper may be sufficiently accurate for many purposes It is invoked using the IGNORE option e g RS2 SHIFT 0 2 IGNORE OPTG Any publications employing the CASPT2 gradients should cite the above paper A citation for MS CASPT2 gradient method is P Celani and H J Werner to be published Example CASPT2 geometry optimizations for H20 22 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 154 SRevision 2006 1 Je se memory 8 m gthresh energy 1 d 10 basis v
537. ype integer CAVITY the intersection seams of the molecular surface are closed 1 or open 0 default 1 type integer EPSILON dielectric permittivity default 1 d0 which means oo type real DISEX distance criteria for the A matrix setup Short range interactions seg ment centre distances DISEX x mean atomic diameter are calcu lated using the underlying basis grid Long range interactions are calculated via the segment centres default 10 d0 type float ROUTE factor used for outer cavity construction The radii of the outer cavity are defined as r r ROUTF x RSOLV default 0 85d0 type float PHSRAN phase offset of coordinate randomization default 0 d0 type float AMPRAN amplitude factor of coordinate randomization default 1 0d 5 type float RSOLV additional radius for cavity construction default 1d0 the optimized H radius is used type float MAXNPS maximal number of surface segments default 1 will be estimated type integer It is recommended to change the default values for problematic cases only By default the program uses optimized radii if existent and 1 17x vdW radius else The op timized radii A are H 1 30 C 2 00 N 1 83 O 1 72 F 1 72 S 2 16 Cl 2 05 Br 2 16 122 32 Own proposals can be given directly subsequent to the cosmo card RAD symbol radius where the radius has to be given in Example cosmo rad O 1 72 rad H 1 3 Output file
538. ys be used for diabatization This basis is too small for real application Reference geometry Displaced geometries Samll displacements for finite difference NACME calculation Orbital dumprecord at reference geometry IMRCI record at reference geometry MRCI record at displaced geometries C2v geometry Ihfjocc 9 2 wf l8 2 4 0rDital 2100 2 multi occ 9 2 closed 4 1 wf 18 2 state 2 natorb reforbl noextra foi oce 9 2 closed 4 1 wf 18 2 0 state 2 orbital reforbl save refci Text Displaced geometries do i 1 r data truncate savcitl reforb reforbl do j 1 3 r2 r i dr 3 multi occ 9 2 closed 4 1 wf 18 2 0 state 2 start reforb orbital 3140 2 j diab reforb noextra reforb 3141 2 ci occ 9 2 closed 4 1 wf 18 2 0 state 2 orbital diabatic save savcitj eadia energy if j eq 1 then el 1 energy 1 e2 1 energy 2 end if ci trans savci j savci tj dm 7000 2 43 ci trans savci j refci cdm 7100 9431 1B1 and 1A2 states Save reference orbitals on reforbl Dont use extra symmetries IMRCI at referenc 11B1 and 1A2 states Use orbitals from previous CASSCF Save MRCI wavefunction geometry Loop over different r values truncate dumpfile after referenc Loop over small displacements for NACME Set current r2 Wavefunction definition Starting orbitals Dumprecord for orbitals Generate diabatic orbitals r

Manual

Contents

Download Pdf Manuals

Related Search

Related Contents