Home

MOLPRO

1. 232 32 2 6 Transition state saddle point optimization ROOT 233 32 2 7 Saving optimization information SAVE 233 32 2 8 Restarting a geometry optimization START 234 32 2 9 Setting a maximum step size STEP o 234 32 2 10 Number of point used in hessian update UPDATE 234 32 2 11 Redefining the trust ratio TRUST o o 235 32 2 12 Setting a cut parameter CUT 2 2 ooo oo 235 32 2 13 Line searching LINESEARCH o 0000 ee eae 235 32 2 14 Numerical gradients NUMERICAL 2 235 32 2 15 Numerical Hessian NUMHES 235 32 2 16 Hessian starting guess from a frequency calculation HSTART 236 LA de Ged we ee es 237 32 2 18 Optimizing energy variables VARIABLE 238 PP ees 238 32 2 20 Printing options PRINT 2 2 0 2000002002 ee 243 32 2 21 Conical Intersection optimization CONICAL 243 32 5 Bxamples o sos a eb ka fa dee Rae EE ee Ge eRe ES 246 CONTENTS eat pb ek A 32 3 2 Allene in natural internal coordinates Sb Hae be a he ae doe a 32 3 4 Calleme XYZ ee es 32 3 5 Transition State of Bicyclo 1 1 0 butane ring opening 32 3 6 Reaction path of the HCN HNC isomerization 33 VIBRATIONAL FREQUENCIES FREQUENCIES 33 1 Numerical hessian using energy variables VARIABL
2. 14 6 Expectation values 14 7 Miscellaneous options o e 14 7 1 Level shifts 14 7 2 Maximum number of iterations 14 7 3 Convergence threshold o ooo o 14 7 4 Print options 14 7 5 Interpolation 14 7 6 Reorthonormalization of the orbitals 14 7 7 Direct SCF 14 7 8 Options 15 THE DENSITY FUNCTIONAL PROGRAM 15 1 Density Functionals 15 1 1 B86 Xa 70 70 70 70 71 71 72 72 74 76 TT 78 78 78 79 79 80 81 81 81 82 82 83 83 83 84 84 84 84 85 86 86 86 87 87 87 87 88 88 88 88 88 88 89 CONTENTS x eat genes 91 15 1 3 B86R Xa y Re optimised a dG he ee HA ee 91 er er ee ep re e ee a 91 PS Ae ee SSE ES E Al 92 15 1 6 B95 Becke s 1995 Correlation Functional 93 e es 93 A A ala iaa 94 15 1 9 BR Becke Roussel Exchange Functional 94 LAS a o See ee Go ae hace a 94 15 1 11 BW Becke Wigner Exchange Correlation Functional 94 15 1 12 CS Colle Salvetti correlation functionall 95 15 1 13 G96 Gill s 1996 Gradient Corrected Exchange Functional 95 Red CO Oo ea ONE Pak ees Res aes 96 WS VAS ACT AU 20s a ba cor HOA a ee et he oe i 96 WS TO HCTH UATE Pe ecg ok Soe oe we eee Soe a be aa he GS Be dae a 96 bob a ha 2a Sob eRe Bae ee 97 15 1 18 LYP Lee Yang and Parr Correlation Function
3. geomt yp xyZ geomet ry nosym 4 FORMALDEHYDE Cc 0 0000000000 0 0000000000 0 5265526741 O 0 0000000000 0 0000000000 0 6555124750 H 0 0000000000 0 9325664988 1 1133424527 H 0 0000000000 0 9325664988 1 1133424527 NE accu 14 ct p2 m optg coord 3n frequencies thermo sym c2v print thermo optg coord 3n frequencies analytic hermo sym c2v print thermo examples form_freq com 33 VIBRATIONAL FREQUENCIES FREQUENCIES 255 Phosphorous pentafluoride Vibrational Frequencies memory 1 m basis 3 21G geomt yp xyZ geomet ry nosym 6 PES Hy A AY y y 0 rhf optg frequencies print low E Oe SO Oa 0 00000 00000 00000 00000 60400 60400 thermo sym d3h temp 200 400 50 use cartesian coordinates xmol style geometry input 00000 11100 52800 41700 00000 00000 O O O PURO optimize geometry don t oOoOOrFOrF oO use symmetry 00000 12400 40100 52500 00000 00000 examples pf5_freq com calculate vibrational frequencies print frequenciestmodes of zero frequencies calculate thermodynamical properties 200 temperature rang 400 K 34 ORBITAL MERGING 256 34 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 perfor
4. RADIAL method m scale no n1 n2 n3 Specify the details of the radial quadrature scheme Four different radial schemes are available specified by method 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 i x r 2 r w Zg 250 15 THE DENSITY FUNCTIONAL PROGRAM 111 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 251 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 0 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 1 x 1 x 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 the whole Bragg Slater radius was used and setting scale to a val
5. procedures enables the automatic inclusion of default procedure files the de fault noprocedures disables such inclusion gl 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 log 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 B G stand for 210 270 and 2 0 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 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 L library directory specifies the directory where the basis set library files LIBMOL are found 1 file 1 directory directory
6. 17 5 2 Rotating pairs of initial orbitals ROTATE orb1 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 17 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 2 16 Therefore 1f 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 17 5 4 Saving the CI vectors and information for a gradient calculation Old form obsolete SAVE cidump refsav grdsav New form SAVE CI cidump REF refsav GRD 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 pre
7. CCSD T text cp calculation for Ar dummy Oo h hf ecsd t e_ar energy Imake OH dummy REF for OH for OH save energy in variable e_oh scf for Ar for Ar save energy in variable e_ar CCSD T text separate calculation for OH roh Igeometry for OH alone geomet ry 0 H 0 rhf oceo 3 1 1 wf 9 1 1 rccsd t e_oh_inf energy CCSD T text separate calculation for Ar Igeometry for OH alone Iscf for Ar for Ar save energy in variable e_ar_inf geomet ry AR hf ccsd t e_ar_inf energy de e_ohar e_oh CCSD T inf e_ar_inf tocm de_cp e_ohar oh ar tocm bsse_oh e_oh e oh_inf tocm bsse_ar e_ar e ar_inf tocm bsse_tot bsse_oh bsse_ar REF for OH examples ohar_bsse com for OH save energy in variable e_oh_inf compute uncorrected interaction energy compute counter poise corrected interaction energy BSSE BSSI Itotal BSSE for OH for Ar For performing counterpoise corrected geometry optimizations see section 32 2 19 10 BASIS INPUT The basis set 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 10 1 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 t
8. Ty 2y 1 00000000 0 00000000 Lp Li 3 50642001 83 09814545 2 1 74736492 5 06370919 4 2 2 99860773 1 3 81 88444526 2 3 01690894 2 3 83 41280402 24 1 59415934 1 3 2 32392477 2 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 2 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 Li 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 2 0 49970082 2 7 0 27936581 2 0 79638982 2 7 0 70184261 Todine basis s 1I 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 d 1 0 2 0 4 fy Ly 0 35 ht Oce Lili arto do scf for 2Pz multi oce 1 1 1 13 casscf with minmal active space WELL ME o kL laverage 2P states ci wf 7 2 1 noexc save 5000 2 save casscf vector for 2Px state ci wf 7 3 1 noexc save 5100 2 save cass
9. 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 9 GEOMETRY SPECIFICATION AND INTEGRATION 66 MASS NOORIENT PLANEXZ 9 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 the C gt 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 r p2 p3 B J or alternatively group atom p1 X y Zz where group atom P2 P3 atomic group number optional Can be us
10. 17 9 Coupled perturbed MESCEF Loco 131 17 9 1 Gradients for SA MCSCH o o e e o 132 17 9 2 Difference gradients for SA MCSCF o o o o o ooo 132 17 9 3 Non adiabatic coupling matrix elements for SA MCSCF 132 A D de 133 Eo Goes ay de as ae ae eee e st a 133 Ve esas E Boe abuts Bas 133 135 Leek AS Aaa taa is e aa A E 133 ad aio Gee GA al ld ole 135 E O aa ae a ees ae 135 18 2 2 Frozen core orbitals 2 ee ee 135 18 2 3 Closed shell orbitals 2 0 202 200 2002 02 000000 136 eee a So oh a we RAR a se E so ee ae a 136 18 2 5 Defining the state symmetry o o o 136 DR ma BES A Pe eee 137 petine is ee EE a ee eR ee 137 tetas bh ee ede a ea ke dae ee 138 reer errors 138 fob Bi One ee Aes a a eres Boe eS 139 be tds de e oxen te div ee ee ge eS 139 ee ee o a eee 140 18 3 7 Restriction of classes of excitations o o 140 Pewee be he eae RE bas bea ee Eek PLS a ee 140 hg deh ew wa a b amp b 140 18 4 2 Coupled Pair Functional ACPF AQCC o 140 sh ee eh ae Me at ek a 141 Ue POE ae a a aa io 141 EOE Pee ee eon a ees a a 141 18 4 6 _ Level sh ifts 2 2 0 ee 141 18 4 7 Maximum number of iterations o 141 eee ee 142 18 4 9 Selecting the primary configuration set 142 on EM PS Se NG A Gee Gerd me 142 CONTENTS xiii o e a Be a ee ee 142 i a ein de E Ue ee eee E A 143 SEP RAR ELEM ERE
11. 179 GEOMETRY OPTIMIZATION 246064 92 282800 92 334149 92 1 1 1828200 4074500 0500000 55 6 1 1976978 1 2228083 3210587 78 1 1 1936923 1 0422291 3660334 118 16 1645479 9885425 0 0819778 159 21 1594176 Tx 9828926 0 8343282 179 293268 336380 1834846 3993195 L 2003447 1809654 1870887 1 0249059 1623854 9862022 0 247544 92 92 9 2 2 6613279 60 7 2505762 12 85 2669557 127 17 2729194 164 22 1596834 9831460 9969372 249550 92 303710 92 337934 92 3 1893895 3377010 2156762 63 8 1 2020825 1408173 1291242 92 13 17935205 1 0084811 0 8629833 140 18 1611436 Les 9847864 0 9184482 170 251 252805 92 257800 311460 92 319520 339027 92 339543 4 5 1920069 1947440 3025012 1 2644292 4020361 67 3580592 9 10 1 2019915 1983793 1047430 0727752 6038298 101 1239958 14 15 1706286 1 1673774 9953467 0 9917554 1807944 146 5637757 19 20 1599052 1 1595435 9834921 0 9830566 6662651 175 3627943 92 264640 92 327790 92 339 713 33 VIBRATIONAL FREQUENCIES FREQUENCIES 252 33 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 m
12. Distributed Multipole Analysis A J Stone Automatic geometry optimization as described in J Comp Chem 18 1997 1473 e 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 MRCT 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 as in Chem Phys Lett 290 143 1998 J Chem Phys 111 5691 1999 and J Chem Phys 113 9443 2000 and references therein Analytical energy gradients for LMP2 as described in J Chem Phys 108 1998 5185 Parallel execution on distributed memory machines as described in J Comp Chem 19 1998 1215 At present SCF DFT MRCI MP2 LMP2 CCSD T energies and SCF DFT gradients are parallelized when running with conventional integral evaluation integral direct SCF DFT and LMP2 are also parallel The program is written mostly in standard Fortr
13. Pi Pp 209 Pes 20 s 210 P Vi 2 Vi 2 Sua Ogg E 211 Saa Ogg 2 Gaa One y aa BB i aa 2 212 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 2 3 2 5 3 1 9 3 2 5 3 4 2 7 6 4 3 3 2 5 3 213 15 THE DENSITY FUNCTIONAL PROGRAM u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 v 0 0 0 0 1 1 1 1 2 2 2 2 0 0 0 0 0 0 0 0 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 and o 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 15 1 38 THGFCO D J Tozer N C Handy and W H Green Chem Phys Lett 273 183 1997 106 214 215 216 217 Density and gradient dependent first row exchange correlation functional of the form FCFO but fitted to a training both set of open and closed shell systems o 0 962998 0 860233 1 54092 0 381602 0 210208 0 391496 0 107660 0 0 105324 0 00837384 0 0617859 0 0383072 0 00526905 0 00381514 0 0321541 0 0568280 0 0288585 0 368326 0 328799 1 22595 1 36412 15 1 39 VSxXC T Van Voorhis and G E Scuseria J Chem Phys 109 400 1998 K F X Z P3 93 13 3 U3 V3 03 Pa Pp Pa 0 e pg 0 Ba Vil PEL zs Pp1 q1 11 t1 41 V1 04 ds Ps 0 F Xs 25 P2 92 12 12 U2 V2 02 where O Gey Ts Zs 573 Cf
14. 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 6 VARIABLES ENERGC FTFUN FTFUNS ifun FTNAME ifun GOs Gh a FTFAC ifun DFTEXFAC PROP istate PROGRAM ITERATIONS CPUSTEP SYSSTEP WALLSTEP 49 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 System 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 DELT DEL4 DARWIN MASSV EREL DMX DM
15. The purpose of this experimental option is to reduce the basis set sensitivity of the Boughton Pulay BP method for domain selection 23 LOCAL CORRELATION TREATMENTS 176 MAXBP maxbp MULLIKEN option PIPEKAO option NONORM value LMP 2ALGO value OLDDEF value Thresholds THRPIP thresh THRORB thresh THRLOC thresh THRMP 2 thresh Only basis functions with angular momentum up to Jmax I are in cluded when computing the overlap of the approximate and exact or 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 DOMSEL must often be reduced to 0 97 or so This option should only be used with care If maxbp 1 the atoms are ranked according to their contribution to the Boughton Pulay overlap default this should normally give the smallest and best orbital domains If maxbp 0 the atoms are ranked according to Mulliken charges In both cases atoms with Mulliken charges greater than 0 6 are always included and atoms with the same Mulliken charges are added as groups Determines method to determine atomic charges option 0 diagonal 1 5 elements of S2C are used option 1 Mulliken gross If option gt 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
16. ci wf 16 6 0 save 3060 1 noexc ci wf 16 7 0 save 3070 1 noexc ci wf 16 4 2 save 3042 1 noexc ci wf 16 6 2 save 3062 1 noexc ci wf 16 7 2 save 3072 1 noexc examples ci wf 16 1 0 save 4010 1 state 3 Imrci calculations for 1D 18 sotches ed energy 1 save energy for 1D state in variable ed es energy 3 save energy for 1S state in variable es ci wf 16 4 2 save 4042 1 mrci calculations for 3P states ep energy save energy for 3P state in variable ep ci wf 16 6 2 save 4062 1 mrci calculations for 3P states ci wf 16 7 2 save 4072 1 mrci calculations for 3P states text only triplet states casscf lsint compute so integrals text 3P states casscf ci hlsmat 1s 3042 1 3062 1 3072 1 Only triplet states casscf text 3P states mrci ci hlsmat 1s 4042 1 4062 1 4072 1 Only triplet states mrci text 3P 1D 1S states casscf ci hlsmat 1s 3010 1 3040 1 3060 1 3070 1 3042 1 3062 1 3072 1 All states casscf text only triplet states use mrci energies and casscf SO matrix elements hlsdiag ed ed es ed ed ed ep ep ep set variable hlsdiag to mrci energies ci hlsmat 1s 3010 1 3040 1 3060 1 3070 1 3042 1 3062 1 3072 1 30 6 2 SO calculation for the I atom using ECPs 30 SPIN ORBIT COUPLING RA T memory 5 M gprint orbitals civector basis gthresh energy 1 d 8 civector 1 d 8 geomet ry 1I basis 1 Todine ECP Dirac Fock with SO coupling ecp 1 46 4 3
17. sets variables or numbers to their inverse obsolete sets variable arrays obsolete clears variables 2 GENERAL PROGRAM STRUCTURE 16 CLEARALL GETVAR SHOW TABLE INT INTS INTE or INTEGRAL INTD LSINT SORT CPP HF or UHF SCF clears all variables recovers variables from file displays the values of variables pints tables Wave function optimization calls the machine default integral program This is optional and needs not to be given calls SEWARD integral program calls Pitzer s integral program flags direct computation of integrals obsolete please use GD IRECT instead calls the spin orbit integral program calls two electron sorting program compute core polarization potential integrals HF RHF HF SCF or RHF SCF calls spin restricted Hartree Fock program open or closed shell calls spin unrestricted Hartree Fock program calls the density functional program call the Kohn Sham spin restricted density functional program call the Kohn Sham spin unrestricted density functional program MULTI MCSCF or CASSCF calls MCSCF CASSCF program CASVB CI MRCI or CI PRO ACPF AQCC CEPA RS2 RS3 MP2 MP3 MP4 CISD CCSD BCCD OCT OCSID UCCSD RCCSD FCI or FULLCI Orbital manipulation LOCALI MERGE calls the CASVB valence bond program calls internally contracted MRCI program calls internally contracted MR ACPF program calls single reference CEPA program cl
18. 8554 1997 Re optimised B of B8 6 used in part 3 of Becke s 1997 paper B 0 00787 14 15 1 4 B88C A D Becke J Chem Phys 88 1053 1988 Correlation functional depending on the B86MGC exchange functional with empirical atomic parameters and u B8 6MGC should be used in conjunction with B88C 15 THE DENSITY FUNCTIONAL PROGRAM In 1 K 08pappq 1 mito 0 01 Loa 1 a Z where q t xa xg 2 1 3 l xo 0 5 pa A 1 xa i Bp xp 0 5 p 1 2x3 t 0 63 z 2uX 274 3 1 BX5Ps Xs 0 5P cps pee A u 0 96 _ Oss 4ps 3 3 1 3 SON B 0 00375 d Ts and 0 007 15 1 5 B88 A D Becke Phys Rev A 38 3098 1988 K Y pi c Bus F i 1 6Bx arcsinh x where B43 1 3 2 an and B 0 0042 92 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 15 THE DENSITY FUNCTIONAL PROGRAM 93 15 1 6 B95 Becke s 1995 Correlation Functional A D Becke J Chem Phys 104 1040 1995 t dependent dynamical correlation functional E FE ps K a 7 30 1 1 12 x s H 1 vxX2 where E 8 Pa Pp Pa 0 e pg 0 31 l 0 0031 32 Oss A 33 H 3 0m p3 5 34 v 0 038 35 and a B is the correlation energy per particle of the Local Spin Density Approximation PW92C 15 1 7 B97 B97 B97DF 0 1943 Exact Exchange A D Becke J
19. AS jstrucl TO jstruc2 F ROM record READ ALL F ROM 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 29 THE VB PROGRAM CASVB 211 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 29 8 Permuting orbitals ORBPERM 1 slmacti 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 for this to be possible 29 9 Optimization control 29 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 Leas Pve sida aga and the latter simply minimizes the energy A Sp Pvg Yve J 29 9 2 Number of iterations MAXITER Nirer Specifie
20. Chem Phys 107 8554 1997 B97DF is given by Pa Pp Pa 0 7 e pg 0 Ao Ain d A1 Aon d Lel Ps 0 Bo Bin xZ A2 Bon X7 A2 36 33N 4 3 2 2 42 3 4r Ps Co Cin x5 A3 Cn x5 43 37 where d x3 Xp 2 38 2 oe 7 8 u 0 39 0 9454 0 7471 4 5961 40 0 1737 2 3487 2 4868 41 0 8094 0 5073 0 7481 42 A 0 006 0 2 0 004 43 and B is the correlation energy per particle of the Local Spin Density Approximation PW92C 15 THE DENSITY FUNCTIONAL PROGRAM 94 15 1 8 B97R B97 Re parameterized by Hamprecht et al B97R B97R 0 21 Exact Ex change F A Hamprecht A J Cohen D J Tozer and N C Handy J Chem Phys 109 6264 1998 Re parameterization of the B97 functional in a self consistent procedure by Hamprecht et al B97RDF is given by A 0 955689 0 788552 5 47869 44 B 0 0820011 2 71681 2 87103 45 C 0 7895 18 0 573805 0 660975 46 15 1 9 BR Becke Roussel Exchange Functional A D Becke and M R Roussel Phys Rev A 39 3761 1989 1 K Ue 47 72050 47 where Us 1 e xe 2 b 48 ee 49 STPs Sn and x is defined by the nonlinear equation 2x 3 22 39 xe TI Ps l 50 x 2 30 where Os Vs 2yDs 6 51 Oss Sg 2 5E ap ne and y 1 53 15 1 10 BRUEG Becke Roussel Exchange Functional Uniform Electron Gas Limit A D Becke and M R Roussel Phys
21. DEFAULT name atoml name1 atom2 name2 Here the basis sets namel name2 overwrite the default basis set name for specific atoms atoml 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 If no further primitive basis set specifications follow one can also use the one line form BASIS DEFAULT VTZ O AVTZ H VDZ or 10 BASIS INPUT 75 BASIS VTZ O AVTZ H VDZ Both of these are equivalent to BASIS DEFAULT VTZ O AVTZ H VDZ END Note that any new BASTS 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 i e the union of atom specific and primitive basis set definitions is used for the atom Examples BASIS DEFAULT VTZ l use cc pVTZ basis as default H VDZ use cc pVDZ for H atoms END This could also be
22. DMSZY DMSXZ DMSYZ DMSZZ diamagnetic shielding tensor LOP 1 LX LY LZ Angular momentum operators Le E Da LOP2 1 LXLX LYLY LZLZ one electron parts of products of LXLY LXLZ LYLZ angular momentum operators The symmetric combinations L R 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 5 FILE HANDLING 40 5 FILE HANDLING 5 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 1t 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 top of the dynamic mem ory and the available memory decre
23. Geometry definition set symmetry 2 3 1 spin 1 2Pix 2Piy and 2Sigma states oce 5 2 2 1 4 sigma 2 pi 1 delta occupied hf do scf calculation examples sym 1 1 1 1 1 2 Sth orbital in symmetry 1 has extra symmetry 2 oHerhhmtci4 com runcas SA CASSCF for all three states set symmetry 2 1 remove Piy runmrci MRCI for 2Pix and 2Sigma Note that this calculation is quite expensive 3 8 3 MP2 geometry optimization The following input performs an MP2 geometry optimization for water 3 INTRODUCTORY EXAMPLES 26 H20 basis vtz geometry 0 H1 O R H2 0 R H1 THETA R 0 96 Ang Theta 104 optmp2 show R Theta title use cc pVTZ basis Z matrix for water examples start bond distance h2 GO o_optmp2 com start bond angle Ido MP2 geometry optimization Ishow optimized geometry parameters At the end of the output the following summary of results is printed RESULTS FOR BASIS VTZ METHOD STATE S ENERGY DIPX DIPY DIPZ MP2 D 1 1 0 0 76 31865774 0 0 0 0 0 76152434 R ANG 0 95906396 THETA DEGREE 103 51638614 The next calculation optimizes the geometry at the MP2 level and subsequently performs MP4 and CCSD T calculations at the optimized geometry geomet ry 0 H1 0 Re H2 0 R H1 THETA basis vtz R 0 96 Ang Theta 104 optmp2 runmp4 runccsdt Z matrix for water use VTZ basis Istart bond distance st
24. 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 variables MASSV DARW and EREL respectively 25 5 1 Example ar2 geometry arl ar2 arl1 r Igeometry 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 Ishow individual contribution and their sum dkroll 1 luse douglas kroll one electron integrals examples hf lrelativistic scf calculation ar2 relcom 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 25 PROPERTIES AND EXPECTATION VALUES 193 25 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 is the unix path name of the file to be written and its specification is mandatory iflag If iflag is negative defaul
25. Rev A 39 3761 1989 As for BR but with y 0 8 15 1 11 Bw Becke Wigner Exchange Correlation Functional P A Stewart and P M W Gill J Chem Faraday Trans 273 183 1995 Hybrid exchange correlation functional comprising Becke s 1998 exchange and Wigner s spin polarised correlation functionals 1 4 3 2 Pape d 4 3 BPs Xs 4 pe Yap 4 gt p a APS 1 6Bys arcsinh on 15 THE DENSITY FUNCTIONAL PROGRAM 95 where pape a o B 0 0042 56 c 0 04918 57 and d 0 349 58 15 1 12 CS 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 where ta oo 60 p 2 2 61 E yO 62 and the constants are a 0 04918 b 0 132 c 0 2533 d 0 349 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 CS1 and CS2 are identical but small numerical differences appear with finite integration grids CS is an alias for CS1 15 1 13 G96 Gill s 1996 Gradient Corrected Exchange Functional P M W Gill Mol Phys 89 433 1996 4 3 1 38 K E A 2 Ps o 137 63 3 331U3 a 3 37 64 where 15 THE DENSITY FUNCTIONAL PROGRAM 96 15 1 14 HCTH93 F A Hamprecht A J Cohen
26. and v2 from matrix and adds their outer product to result vl 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 x v1 a x v2 b If result has not been used before it is zeroed before performing the operation 35 13 Forming a closed shell density matrix DENS DENS density orbitals iocc ioccz 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 35 MATRIX OPERATIONS 267 35 14 Computing a fock matrix FOCK FOCK fid computes a closed shell fock matrix using density d The result is stored in f 35 15 Computing a coulomb operator COUL COUL J d computes a coulomb operator J d using density d 35 16 Computing an exchange operator EXCH COUL K d computes an exchange operator K d using density d 35 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 35 18 Printing diagonal elements of a matrix PRID PRID name prints the diagonal elements of matrix name 35 19 Printing orbitals PRIO PRIO name n nN2 N3 Ng prints orbitals name The first n orbitals are printed in symmetry i If n 0 all orbitals of that symmetry
27. convergence threshold for energy see also ACCU card convergence threshold for coefficients see also ACCU card omit coefficient changes which are smaller than this value 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 PRINT codel value code2 value Print options Generally the value determines how much intermediate information is printed 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 JOP 0 JOP 1 JOP 2 print orbitals print operator list print coulomb operators in MO basis print coulomb operators in AO and MO basis as JOP for internal exchange operators print paging information for CIKEXT 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
28. dont allow automatic reorientation s hl s rl h2 s r2 hl theta Z matrix geometry input gprint orbitals civector Iglobal 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 orbital 2140 2 Isave orbitals to 2140 2 examples E 2140 2 reforb 0 h2s_diab com 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 Iset 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 Isave new orbitals to record diab reforb compute diabatic orbitals using reference orbitals stored on record reforb reforb 2141 2 set variable reforb to the new orbitals enddo See section 28 for the automatic generation of diabatic energies 17 6 Selecting the optimization methods By default MULTI uses the non linear optimization method developed by Werner Meyer and Knowles 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 proble
29. expr1 INT expr1 expr2 expr2 Note all trigonometric functions use or produce angles in degrees 2 6 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 ordered tables in user supplied format Furthermore there are up to 9 binary MOLPRO files available each one known to the program simply by its number 1 to 9 By default they 2 GENERAL PROGRAM STRUCTURE 8 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 w
30. in variable method 2 examples h2o0_table com casscf Ido casscf calculation e 3 energy save scf energy in variable e 3 method 3 program save the string CASSCF in variable method 3 mrci Ido mrci calculation e 4 energy save scf energy in variable e 4 method 4 program save the string MRCI in variable method 4 table method e print a table with results title Results for H20 basis Sbasis title for the table 3 INTRODUCTORY EXAMPLES 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 3 INTRODUCTORY EXAMPLES 23 3 7 Procedures You could simplify this job by defining a procedure SAVE_E as follows 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 1 energy save scf energy in variable e i method i program save the present method in variable method i endproc lend of procedure AZO A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input Hl Orr h2 0 r H1 theta basis vtz luse VTZ basis hf closed shell scf save_e call procedure save results ccsd t Ido ccsd t calculation save_e call procedure save results casscf Ido casscf calculation save_e call procedure save results mrci Ido mrci calculation save_e call procedure save results table method
31. or frozen core orbitals must be taken from a closed shell SCF calculation The present version does not work 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 DFT MCSCF MP2 LMP2 and QCISD It does not work with state averaged SA MCSCF wave functions The ALASKA gradients are default for general contracted basis sets In all other cases the CADPAC gradients are default On most architectures these are faster than ALASKA gradients However it is possible to choose the gradients 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 B2 2 19 Note that for comput ing gradients for state averaged MCSCF a CPMCSCF is required in the MCSCF calculation see CPMCSCF 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 info
32. ral orbitals and print the configurations and their associated co efficients This has the same effect as the GPRINT CIVECTOR directive see section 4 12 By default only configurations with coefficients larger than 0 05 are printed This threshold can be modified using the THRESH see section 17 8 2 or GTHRESH see section 4 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 density for all states of the given spin is used 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 pri
33. the MOLPRO executable This directory should be one normally in the PATH of all users who will access MOLPRO and its specification will depend on whether the installation is private or public configure prompts for the destination directory INSTLIB for installation of ancil lary files which are required for program execution configure prompts for the destination directory for documentation This should nor mally be a directory that is mounted on a worldwide web server 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 k key specifies the licence key i8 l 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 011121314 specifies system BLAS level as described above mpp nompp controls whether compilation is to be for MPP parallelism see above ifcl pgf controls whether the Intel ifc or Portland pgf compiler is be used on Linux 1A32 systems Note that appropriate environment variables must
34. the sequence of input cards starting with command and ending with FORCE is used to compute the energies For example in order to compute numerical gradients for ccsd t it is necessary to do a HF and a CCSD T calculation at each geometry The input would read hf ccsd t forces numerical startcmd hf The program will then automatically repeat HF and CCSD T at as many geometries as needed for evaluating the gradient The keyword NUMERICAL implies that also STARTCMD must be given otherwise an error results 31 ENERGY GRADIENTS 226 31 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 3N or CART Compute the gradients for all 3N nuclear coordinates This is the default if the z matrizx 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_t
35. 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 7 TABLES AND PLOTTING 54 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 head 1 head2 FORMAT format FTYP typl typ2 typ3 DIGITS digl dig2 dig3 SAVE file status TITLE title SORT coll col2 PRINT keyl key2 NOPRINT keyl key2 NOPUNCH RANGE start end STATISTICS 7 2 Plotting Specify a heading for each column By default the names of the variables are used as headings 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 Simplified form to modify the format This gives the type A F or D for each column sensible defaults are normally used Give the number of digits after the decimal points to be printed for each column sensible defaults are normally used Specify a file on which the table will be written If status is NEW the file is
36. 1 1 d0 219474 63067d0 HARTREE ER Z 1 d0 6 5796838999d15 HARTREE G K K C C HZ 1 H A ANGSTROM 1 d0 0 529177249d0 BOHR TOC d0 27 2113961d0 HARTREE d0 219474 63067d0 HARTREE d0 6 5796838999d15 HARTREE 1 d0 0 529177249d0 BOHR OEV 27 2113961d0 EV TOK OKELVIN 3 157733d5 K 3 157733d5 K 219474 63067d0 CM 1 OHERTZ 6 5796838999d15 HZ TOHZ 6 5796838999d15 HZ TOKJ 2625 500d0 KJ MOL OKJOULE 2625 500d0 KJ MOL TOKCAL 627 5096d0 KCAL MOL TOA TOANG 0 529177249d0 ANGSTROM TODEBYE 2 54158d0 DEBYE 0 529177249d0 ANGSTROM Further variables which are set during execution of the program 6 VARIABLES INTYP INTDONE CARTESIAN SCFDONE UMVAR TATUS HARGE ELEC Nn ZaQan AA PIN ORBITAL LASTORB LASTSYM LASTSPIN LASTNELEC ENERGR istate ENERGY istate ENERGD istate ENERGP istate ENERGT 1 ENERGT 2 ENERGT 3 METHODT 1 METHODT 2 METHODT 3 48 defines integral program to be used Either INTS Seward or INTP Argos has the value t rue if the integrals are done for the current geom etry 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 las
37. 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input Hl OEL H2 0 r2 H1 theta basis vdz hf Iscf using c2v symmetry orbital 2100 2 save on record 2100 2 set zsymel x examples hf h20_c2v_cs_start com start 2100 2 Istart 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 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 calculation at a neighbouring geometry are available these should be used as starting guess 14 3 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 14 4 Rotating pairs of orbitals ROTATE orb sym orbz 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 See MERGE for other possibilities t
38. 1 HOW TO READ THIS MANUAL 1 1 2 1 Running MOLPRO o o h e 1 A ee Ere ee 1 naar pa 3 2 2 Input format see ci a cd a ae o a ee A a 3 2 3 Input structure i 24 442 cado e oa a a a 6 Ca gaiii Gotha Gh a eho ae eR ee ae Pe 6 2 Intrinsic functions 2 ee 7 2 0 lesion oa a ee be a ba RS Ee E eR baw 7 ANT oe ge dea Se ad ws ee a de See A de ee a a dc A 8 A AO E aed Oe ad ae 9 a desta tes eas gee eat a ee do 9 A 9 Z1 Variables 2 5 6 a9 ia A AN 9 2 12 Multiple passes through the Input oo o 9 2 13 SYMMEIY y o oros cr a Re we Ee ee hae ewe 10 E aw ale aw Beak ale ag Bee E 10 pa a amp beac g duals Geis Som ee ae a a 12 2 16 Selecting orbitals and density matrices ORBITAL DENSITY 12 Saa 14 Sd ad aa aa eo oben 17 bg OG A a o ao RS ae 17 opere nena 18 La a A e E a wd a 19 2 19 MOLPRO help os co coe mesa cana e e 19 3_INTRODUCTORY EXAMPLES 19 3 1 Using the molpro command o o o 19 st he eA we ei ge ls Ena os ye 20 ieee Gn ho nh a ee ee eee nd data 20 34 CCOSD T 2 ot wk eed BAER RA Ree ee aw oe Sd es 21 3 5 CASSCF and MRCI o co ca ee ee a 21 AE s i a aia a ee a eee A e a ee Bw tagiga al ee ie ai a 21 3 7 Procedures vu cr be eR aw eR a aea ee la 23 fees Med NN a 23 3 8 1 RCCSD T for different states 0 2 2008 23 3 8 2 SA CASSCF and MRCI o o e e 24 A a ee 25 Li A a AE bw ae ee 26 wom ak
39. 116 117 118 119 120 121 122 123 124 125 15 THE DENSITY FUNCTIONAL PROGRAM 100 15 1 25 PW86 J P Perdew and Y Wang Phys Rev B 33 8800 1986 1 K 5 E 2ps 126 where EA E n 2 3 n4 3F S 127 F S 141 2968 145 0 286 128 and x 129 2 672 15 1 26 PW91 PW91 PW91X PW91C J P Perdew J A Chevary S H Vosko K A Jackson M R Pederson and C Fiolhais Phys Rev B 46 6671 1992 15 1 27 PWw91C Perdew Wang 1991 GGA Correlation Functional J P Perdew J A Chevary S H Vosko K A Jackson M R Pederson and C Fiolhais Phys Rev B 46 6671 1992 K p e Pa Pp H d Pa Pp gt 130 where E on Ge d 131 4u Pa Pp 3p7 vey u ot B 1 6 0 B 1 6l B 2 132 H d 0 B L d a B J d a B 133 Po Pp 1 d A a B d MAB oN rara OS uf papp J a a B v 0 r a B K 3Z 7 W Pappe 0 135 __2ve o48 AlB gt i 1 136 1 0 09 137 A VK 138 31 1 3 v 16 i 139 0 004235 140 Z 0 001667 141 15 THE DENSITY FUNCTIONAL PROGRAM 101 o r O r Z 142 2 mile i000 1 mA A 097 i we E 23 266 144 0 007389 145 A 8 723 146 Y 0 472 147 and a B is the correlation energy per particle of the Local Spin Density Approximation PW92C 15 1 28 Pw91X Perdew Wang 1991 GGA Exchange Functional J P Perdew J A Chevary
40. 2 rlif r 1 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 6001 2 dm 8100 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 nacmelm i nacme ddr 2 dr orbital 2140 2 2141 2 2142 2 density 8000 2 8100 2 8200 2 nacme2 1 nacme end do define basis define bond distances define increment define geometry first calculation at R 3 SCF CASSCF 3 inactive orbitals Two 1A1 states dump orbitals to record 2140 2 loop over geometries set bond distance ICASSCF 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 increment bond distance by dr same CASSCF as above Two 1A1 states start with orbitals from referenc save orbitals to record 2141 2 generate diabatic orbitals by maximizing the overlap with the orbitals at the reference geometry geometry CI for 2 states wavefunction saved to record 6001 2 examples r gp at at rd lif_nacme com same CASSCF as above Two 1A1 states start with orbitals from referenc save orbitals to record 2142 2 Igenerate diabatic orbitals by maximizing the overlap with the orbitals at the reference geometry geo
41. 2 CON 2 0 0 0 2 2sigma 2 lpi_y 2 17 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 17 THE MCSCF PROGRAM MULTI 121 A P space threshold 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 car
42. 2 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 are printed Example for a state averaged calculation for CN X and B 25 states and A 2 211 states averaged A full valence CASSCF calculation is performed cn r 2 2 define bond length geometry c n c r cht ode 5 1 1 wt 13 17 1 RHF calculation for sigma state orbital 2100 2 save orbitals to record 2100 2 default multi occ 6 2 2 closed 2 Define active and inactive orbitals start 2100 2 IStart with RHF orbitals from above examples save ref 4000 2 Save configuration weights for CI in EP SALAS GOR wf 13 1 1 state 2 wf 13 2 1 wf 13 3 1 Define the four states natorb ci print Print natural orbitals and associated ci coefficients tran 1z Compute matrix elements over LZ expec2 1zz 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 Dzn symmetry CSF method used ERAN geometry N1 N2 N1 r geometry input r 2 2 bond length hf occ 3 1 1 2 wf 14 1 save 2100 2 scf calculation martireca ly lpg syle Corey Tyrer ly 2100 29 Define occupied orbitals Define frozen core scf orbitals config lUse CSF method examples w 14 1 Define sta
43. 29 JAR xamples o acis u aia a oe WK wd ae Ba a ewe a ee 217 CONTENTS xvi 30_SPIN ORBIT COUPLING 218 aaa eee LO a III 218 iat aera ae a E Ang Hee as ee le ee 218 30 3 Calculation of individual SO matrix elements 218 oa 219 E E io E 219 das E S i pt 220 NN 220 SAR E A a ee Be ye a a 220 ile Bid 4 220 oral 221 223 a de Eos a a dada 223 31 1 1 Adding gradients ADD o o e e 223 31 1 2 Scaling gradients SCALE o o e 223 31 1 3 Defining the orbitals for SCF gradients ORBITAL 224 31 1 4 MCSCF gradients MCSCF o o e ee 224 31 1 5 State averaged MCSCF gradients SAMC o o 224 31 1 6 Non adiabatic coupling matrix elements NACM 225 31 1 7 Difference gradients for SA MCSCF DEMC 225 do ee a id ba ee AS 225 o ds SE ee ee 225 31 2 1 Choice of coordinates COORD 1 2 ee ee 226 cde bet BES PL RE a baan 226 edb aed ea wa da A 226 228 32 1 Geometry optimization step OPT o o 0 00000 oe 228 32 2 Automatic geometry optimization OPTG o o 228 32 2 1 Optimization coordinates COORD o o 230 32 2 2 Defining active geometry parameters ACTIVE 231 32 2 3 Defining inactive geometry parameters INACTIVE 231 32 2 4 Selecting the optimization method METHOD 231 32 2 5 Approximating hessian matrix elements HESSIAN
44. 8 Restart Information from the permanent files is automatically recovered in subsequent calculations This can be controlled using the RESTART directive 2 9 Data set manipulation It is possible to truncate files and rename or copy records using the DATA command Sev eral standard 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 2 10 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 2 11 Variables The program maintains a set of internal variables These may be used in place of floating point numbers anywhere in the input Before their use variables must be defined as described in detail in Section 6 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 They can also be used to change record names automatically when several geometries are calculated in one run It is thus possible to save the information for each ge
45. 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 10 String specifying how the sorted integrals are written Possi ble values are molpro standard MOLPRO record 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 9 GEOMETRY SPECIFICATION AND INTEGRATION 65 9 2 Symmetry specification If standard Z matrix input is used MOLPRO determines the symmetry automatically by de fau
46. 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 CORE all doubly occ frozen 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 TO 2 END and or disable the simultaneous optimization of internal orbitals amp CI e g ITERATIONS DONT INTERNAL 1 TO 2 END 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 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 17 12 Examples The simplest input for a CASSCF calculation for H20 Cz symmetry is simply geometry o0 h1 0 r h2 0 r h1 theta Z matrix geometry input r 1 ang bond length theta 104 bond angle hf Ido 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 CORE 1 specify frozen core orbitals WF 10 1 define wavefunction symmetry examples h20_casscf com 17 THE MCSCF PROGRAM MULTI 134 START 2100 2 read guess orbitals from record 2100 file
47. Chem Phys 82 5053 1985 P J Knowles and H J Werner Chem Phys Lett 115 259 1985 See also H J 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 ACQQ 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 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 Phys 94 6708 1991 MP2 and LMP2 gradients A
48. GEOMETRY OPTIMIZATION 248 32 3 3 Allene MP2 optimization xxx Allene geometry optimization using Z Matrix memory 2 m basis vdz rcc 1 32 ang rch 1 08 ang acc 120 degree Geometry Cl Cel POS Ol el rec 2 45 C3 2 rec cl 180 q1 0 hl 01 reh 2 acc q1 0 he ol Pen Ez ace Ll L80 h3 C3 ren e7 ace ML 00 h4 c3 rch c2 acc h2 90 Z matrix input examples allene_optmp2 com optmp2 luse default procedure optmp2 32 3 4 Caffeine XYZ CAFFEINE cartesian coordinates XYZ format memory 1 m basis sto 3g geomt yp XyZz geometry nosym 24 CAFFEINE CARTESIAN COORDINATES c 0 8423320060 0 3654865620 0 0000000000 c 0 2841017540 1 1961236000 0 0000000000 N 2 0294818880 1 1042264700 0 0000000000 N 0 0774743850 2 5357317920 0 0000000000 N 1 6472646000 0 6177952290 0 0000000000 c 1 4531962870 2 3678913120 0 0000000000 e 0 6373131870 1 1735112670 0 0000000000 c 1 7812691930 0 7688916330 0 0000000000 N 0 6771444680 1 6306355000 0 0000000000 o 1 6106752160 1 9349693060 0 0000000000 o 2 9202890400 1 2510058880 0 0000000000 c 0 9202462430 3 1094501020 0 0000000000 c 2 8623938560 1 4824503660 0 0000000000 examples c 3 4552156930 0 6811094280 0 0000000000 caffeine_opt_diis com H 2 0878150460 3 2451913360 0 0000000000 H 1 4989252090 3 4222116470 0 8897886280 H 1 4989252090 3 4222116470 0 8897886280 H 0 0071905670 3 7148499490 0 0000000000 H 3 490
49. 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 16 ORBITAL LOCALIZATION 115 16 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 16 7 2 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 using specifications as explained in section 2 16 16 7 3 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 ene
50. INTRODUCTORY EXAMPLES 21 3 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 x h20 A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input BIO he H2 0 r H1 theta examples basis VTZ luse VIZ basis h2o0_ccsdt_vtz com hf closed shell scf ccsd t Ido ccsd t calculation 3 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 MRCTI reference function x h20 A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input hil 0 53 h2 0 r H1 theta basis vtz use VTZ basis examples hf closed shell scf h20_mrci_vtz com ccsd t Ido ccsd t calculation casscf Ido casscf calculation mrci Ido mrci calculation 3 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 x h20 A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input BLOE h2 0 r H1 theta basis vtz luse VIZ basis hf closed shell scf e 1 energy save scf energy in variable e 1 method 1 program save the string HF in variable method 1 ccsd t Ido ccsd t calculation e 2 energy save ccsd t energy in variable e 2 method 2 program save the string CCSD T
51. If this card is not present the thresholds for all states are the default values or those specified on the THRESH card 18 4 6 Level shifts SHIFT shiftp shifts shifti Denominator shifts for pairs singles and internals respectively 18 4 7 Maximum number of iterations MAXITER maxit maxiti maxit maximum number of macroiterations maxiti maximum number of microiterations internal CD 18 THE CI PROGRAM 142 18 4 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 18 4 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 are 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 18 4 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 18 4 11 Saving the wavefunction SAVE savecp saveco idelcg or SAVE CIVEC savecp CONF IG s
52. PROGRAM 108 15 1 41 Alias functional descriptions Additional functional keywords are also defined as convenient aliases The following table gives the translations LDA S VWN DIRAC S B88 B B88X B LYP88 LYP B LYP B88 LYP88 B3LYP 0 72 B88 0 08 S 0 81 LYP88 0 19 VWN 0 2 exact exchange B97 B97DF 0 1943 exact exchange B9TR B9TRDF 0 21 exact exchange PBEO PBEODF 0 25 exact exchange LSDAC PW92C LSDC PW92C VS99 VSXC 15 2 Options The following options may be used to control the operation of the DFT modules In the Kohn Sham case these may be followed by further options for the SCF program as described in Section 14 Note that DFT and SCF options cannot be intermixed 15 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 2 16 ODENSITY can be used to specify the spin density The defaults are the densities last written by an SCF or MCSCF program 15 2 2 Thresholds THR THR keyl valuel key2 value 2 Sets various truncation thresholds key can be one of the following TOTAL Overall target accuracy of density functional Defaults to the value of the global threshold ENERGY For proper use of this threshold other thresholds should be left at t
53. S H Vosko K A Jackson M R Pederson and C Fiolhais Phys Rev B 46 6671 1992 1 K E 2p s 14 7 LE 2Ps 148 where ON 4 3 E n F S 149 n 3 2 ro 149 Xs S 150 2 672 and 1 0 19645 Sarcsinh 7 7956 S 0 2743 0 1508 eR S F S 151 S 1 0 19645 Sarcsinh 7 7956 S 0 004 S veh 15 1 29 Pw92C Local Spin Density Approximation Correlation Energy J P Perdew and Y Wang Phys Rev B 45 13244 1992 Electron gas correlation energy K pelPa Pp 152 where e a p e e r a B 71 U1 V1 W1 X1 Y1 Pi e r a B T3 U3 V3 W3 X3 Y3 P3 C a B 1 C a B c a e r a B 72 U2 Va W2 X2 Y2 P2 e r amp B Ti U1 V1 W1 X1 Y1 P1 o Z a B a B 153 3 1 3 gt 154 lp gt 154 a p B 155 0 8 a 155 142 4 1 24 2 o z Pto E E 156 2473 2 15 THE DENSITY FUNCTIONAL PROGRAM 1 elr t u v w x y p 2t 1 ur In 1 2t vyr wr xr3 yrPt c 1 709921 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 LSDAC and LSDC are aliased to PW92C 15 1 30 S Slater Dirac Exchange Energy J C Slater Phys Rev 81 385 1951 K cF ps 3 3 1 3 a where 15 1 31 TH1 D J Tozer and N C Handy J Chem Phys 108 2545 1998 Density a
54. SCF CASSCF and MR ACPF calculations At present multiple state MR ACPF calculations are not possible 2 18 2 Procedures for geometry optimizations If the geometry is given as Z matrix and depends on variables only these are optimized and other numerical parameters are kept fixed This behaviour can be modified by setting the variable OPTFULL TRUE which causes the geometry optimization be done in cartesian coordinates for all degrees of freedom In this case the variables are not modified If the geometry is given as XYZ form or the Z matrix does not depend on variables all degrees of freedom are optimized optscf Performs SCF geometry optimization optdft Performs DFT geometry optimization The functional can be speci fied using either the FUNCTIONAL or DFTNAME variable default B3LYP optmp2 Performs MP2 geometry optimization optcas Performs CASSCF geometry optimization The procedure does not support state averaged calculations 3 INTRODUCTORY EXAMPLES 19 2 18 3 Procedures for frequency calculations In all cases the geometry is optimized for all degrees of freedom before computing the fre quencies Variables on which the Z matrix depends are not modified The hessian is computed by finite differences from analytical energy gradients Frequencies intensities and thermody namic quantities are evaluated and printed in the output The summary at the end shows only the energies fregscf Performs SCF frequency calculatio
55. affect the BSSE for the SCF and therefore the basis set should be sufficiently large to make the SCF BSSE negligible Usually diffuse basis functions are important for obtaining accurate intermolecular interactions Unfortunately these spoil the efficiency of prescreening and therefore make direct calculations much more expensive For examples and discussions of these aspects see Refs 6 7 23 LOCAL CORRELATION TREATMENTS 168 23 4 2 Gradients and frequency calculations Geometry optimizations 4 5 and numerical frequency calculations 5 can be performed using analytical energy gradients 4 for local MP2 LMP2 geometry optimizations are particularly attractive for weakly bound systems since virtually BSSE free structures are obtained see sec tion 23 4 1Jand Refs 6 7 It should be noted however that the current implementation is not particularly efficient and nothing has been done so far to achieve low order scaling for large systems 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 A particular problem in LMP2 gradient calculations is the elimination of redundant basis func tions in the domains see Refs 1 4 If redundancies are present gradient calculations require the elimination of individual basis functions option DELBAS 1 which is less unique than the elimination of eigenvectors c
56. approach Default 13 Specifying which integrals to treat by which multipole expansion type RMAIN distance RIONIC distance SUPPRESS distance When the distance between two orbitals is closer than the absolute value of distance multipole corrections in the context of the split approximation will be carried out as monopolar expansions other wise a more sophisticated approach will be used involving four ex pansions for each pair one for the ordinary dispersion block of the exchange matrix for the given pair two for the two ionic exchange blocks and one for the exchange dispersion block see the section on energy partitioning for explanation of these terms If distance is a positive value the distance between the orbitals will be taken to be the distance between the closest atoms of the two orbital domains as in WEAKPAIR DISTPATR etc if it is a negative value the distance of the centroids of the two orbitals will be considered Default 1 When the distance between two orbitals is closer than distance a u the multipole correction of ionic blocks of the exchange operators will be carried out as monopolar expansion otherwise a bipolar expansion will be performed The distance is understood as the distance of the centroids of the two orbitals Default O When the distance between two orbitals distance of centroids is higher than the absolute value of distance a u the multipole cor rection of the exchange dispersion
57. are printed in the beginning of each local calculation For 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 key DOMSEL A typical domain output here for water looks like this 23 LOCAL CORRELATION TREATMENTS 165 Orbital domains Orb Atom Charge Crit 2 1 1 O1 LD 0 00 3 H2 0 84 1 00 3al 1 ol 2 02 1 00 4 1 1 01 1 96 1 00 Ouse 1 01 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 Improper domains could result from improperly localized orbitals or forgotten NOSYM directive This does not only negatively affect performance and memory requirements but can also lead to bogus results Poor localization is sometimes an intrinsic problem in particular for strongly conjugated systems In rare cases it might also happen that the localization procedure does not converge The default for the selection criterion DOMSEL is 0 98 This works usually well for small basis sets like cc p
58. are selected idleig 0 functions corresponding to the smallest diagonal elements of projected orbital matrix are eliminated idleig 1 Functions corresponding to the largest coefficients in the 23 LOCAL CORRELATION TREATMENTS 175 DELCMIN value eigenvectors of are deleted default Since degenerate eigenvec tors can arbitrarily mix the selection may not be unique and depend on the diagonalization method see DELBAS Only effective with DELEIG 1 Only basis functions with coeffi cients larger than value in the eigenvectors of small eigenvalues can be deleted default 0 1 Parameters for multipole treatment of exchange operators MULTMETHOD option DSTMLT level Used internally by the MULTP card don t mess with it Determines the expansion level of the multipole expansion of distant pairs e g 1 means dipole approximation 2 quadrupole approxima tion and so on Default is O MULTP card 3 Parameters for energy partitioning IEPART value EPART cutoff Miscellaneous options SKIPDIST skipdist ASYDOM jiterm LOCS ING locsing MAXANG Ilmax 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 larges
59. basis sets e g ECP 1 OMWB e The Hay ECPs and corresponding basis sets ECP 1 and ECP2 e Some of the Karslruhe basis sets SV TZV and for some elements SVP TZVP TZVPP TZVPPP e The Binning Curtiss sets for Ga Kr BINNING SV BINNING SVP BINNING VTZ and BINNING VTZP e Most of the Pople basis sets using their standard names e g 6 31G 6 311 G D 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 h20 cc pVTZ basis 1A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input Hl OE H2 0 r H1 theta basis VTZ luse VTZ basis hf closed shel1l scf is entirely equivalent to h20 cc pVTZ basis IA title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input HL 04 43 H2 0 r H1 theta basis spdf o vtz c spd h vtz c At 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 func
60. be set see section f ftcflag adds a token to the specifiers for the Fortran preprocessor ftc largefilesl nolargefiles controls whether large file gt 2Gb support is wanted This option is not relevant or used on all architectures For Linux PC it should be specified only if the kernel and system libraries also support large files p31 p4l athlon specifically identifies a particular hardware in order to force appropri ate run time libraries where possible These options are supported only on Linux systems If any of these options is given the MOLPRO executable will A INSTALLATION OF MOLPRO 277 be named molpro_p3 exe molpro_p4 exe ormolpro_athlon exe in the mpp case e g molpro_p3_tcgmsg It is possible to install different platform variants simultaneously in the same MOLPRO tree see section A 3 4 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 or athlon 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 configure for each case and specify the appropriate libraries when configure p
61. but without using lo cal blocking 1 e full matrices are used This is the most expensive method and only for testing LOCAL 3 Fully local calculation This is the fastest method for lo cal calculations with no weak pairs LOCAL 4 Fully local calculation Default This is the fastest method for large molecules with many weak pairs and requires minimum memory SAVE record Allows the domain information to be saved on record name ifil for later restart using START This can be used to freeze the domains as function of geometry Note that the domain information is automati cally stored if a SAVE directive is given see above and in this case the record given on the SAVE card will overwrite any record given as SAVE option START record Retrieves domain information previously saved using SAVE Note that the domain information is automatically restored if a START directive is given see above and in this case the record given on the START card will overwrite any record given as START option PIPEK option If this option is given and option gt 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 23 LOCAL CORRELATION TREATMENTS 173 in the orbital record cf ORBITAL and LOCALIZE directives In the latter case the most recent localized orbitals are used Setting option 1 switches the local
62. calculations see COORD keyword section 32 2 1 32 2 16 Hessian starting guess from a frequency calculation HSTART HSTART irec ifil This option allows you to use a cartesian hessian matrix computed in a FREQUENCIES calcu lation or using the cpmcscf hessian option in multi as a starting guess to the Hessian in a geometry optimization The Hessian is transformed automatically into the coordinate system the optimization is performed in This option might be very useful in transition state opti mization i e by using a cheap analytical SCF Hessian as a starting guess to a higher level 32 GEOMETRY OPTIMIZATION 237 optimization See example hcn_mp2_ts com below Note that dummy atoms are ignored and therefore the coordinates of dummy atoms should not depend on variables The starting hessian is read from record irec on file fil compare the SAVE subcommand of FREQUENCIES section 33 f If no irec ifil is given the last FREQ record found on file ifil 2 will be used Note that the HSTART option has higher priority than the NUMHES card If a hessian is found using the HSTART command the NUMHES command will be ignored Further note that if the hessian is read from file 1 it is assumed that it has been calculated at the current geometry and the hessian update is disabled On the other hand if the hessian is found on file 2 or 3 it is assumed that it has only been computed in the first iteration and the hessian update is per
63. 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 summary of internal configurations N N 1 and N 2 electron print internal configurations N N 1 N 2 print summary of reference configurations 18 THE CI PROGRAM REF 1 PSPACE HEL HSS SPQ TEST 0 TEST 1 TEST 2 CPU ALL 18 7 Examples 148 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 xxx Single reference CISD and CEPA 1 for water r 0 957 angstrom theta 104 6 degree geomet ry 0 Hl 7 07 43 H2 0 r H1 theta z matrix geometry input examples h20_cepal com hf wf 10 1 TOTAL SCF ENERGY 76 02680642 CEGC 3 dal coro lim 10 1 TOTAL C
64. directives 17 7 1 Matrix elements over one electron operators EXPEC oper oper2 OPeTn TRAN oper oper2 0pern Calculate expectation values and transition matrix elements for the given one electron 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 17 7 2 Matrix elements over two electron operators EXPEC2 0per 0per OP Tn TRANZ2 oper oper2 Opern 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 Ly TRAN2 LXZ calculates matrix elements for 4 LyL LzLy TRAN2 LXX LYY LZZ calculates matrix elements for L L and TA The matrix ele ments for the sum Z are also printed 17 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 17 THE MCSCF PROGRAM MULTI 129 17 8 1 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
65. during compilation Typically 30Mb is needed for the finally installed program 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 STMPDIR STMPDIR2 STMPDIR3 These variables should be set before the program is installed preferably in profile or cshrc since at some stages the installation procedures will check for them cf section A 3 6 6 If the program is to be built for parallel execution the Global Arrays toolkit version 3 1 or later is needed This is available from http www emsl pnl gov 2080 docs global ga html and should be installed prior to compiling 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 normally recommended to use MPI where other possibilities exist For more information see section A 3 3 A INSTALLATION OF MOLPRO 274 7 The source distribution of MOLPRO which consists of a base compressed tar archive with a file name of the form molpro 2002 6
66. eA wR Gear eg ell Gels Sou em Maine ae a a ant hls g g 27 4 PROGRAM CONTROL 29 4 1 Starting a J b CAF lo ee ee ee Ok Ble a a a 29 4 2 Endingajob 2 2 ee ee 29 4 3 Restarting a job RESTART o o e 29 4 4 Including secondary input files INCLUDE o 30 4 5 Allocating dynamic memory MEMORY o o 30 4 6 DO loops DO ENDDO e 30 4 6 1 Examples for do l00pS o o o 31 CONTENTS 4 7 Branching IF ELSEIF ENDIF 0 0 0 0 0000 eee ee ee Al TF statements ooo wk we we ee ea a 4 7 2 GOTO commands 2 0 00 ee tatas 4 7 3 Labels LABEL 1 e o 4 8 Procedures PROC ENDPROC 1 ee e 4 9 Textcards TEXT e 4 10 Checking the program status STATUS o 4 11 Global Thresholds GTHRESH o e 4 12 Global Print Options GPRINT NOGPRINT o 4 13 One electron operators and expectation values GEXPEC 4 13 1 Example for computing expectation values 4 13 2 Example for computing relativistic corrections SAA tan dts oo at Bt aid AN O2 DERE O Bs 4 263 amp 2 es ee eth es EE a SR Rw Ee ee a Behe eae red ee eee Gee eS ee hw A SA DATA 24 ba te ae ad oe ee baad Ban awa te be od ae ad eas 5 5 Assigning punch files PUNCH lt 2 20 20 0000 00000 eee eee 5 6 MOLPRO s
67. ea INDEX 288 HCTH93 96 ECP LDA 108 effective core potential 78 LSDAC 101 ELSEIF BI LSDC 101 ENDDO 30 LTA 97 ENDIF BI LYP ENDZ 66 MK00 97 EOM 159 MKOOB 98 EOMPAR 160 P86 98 EOMPRINT 160 PBE 99 ERASE 40 PBEO Examples PBEX 9 allene_opt_bmat com 247 PW86 100 allene_optmp2 com 248 Pw91 100 allene_optscf com 246 PW91C 100 ar2_rel com 38 82 192 Pw91X 101 auh_ecp_1lib com 80 Pw92C 101 bh_mrci_sigma_delta com 148 S 102 butane_opt_transition com 249 TH1 102 caffeine_opt_diis com 248 TH2 103 cn_sa_casscf com 134 TH3 103 cndft com 109 TH4 104 cu_ecp_explicit com THGFC 105 field com 190 THGFCFO 105 form_freq com THGFCO 106 2 com 20 THGEL 104 2f merge com VS99 106 20_c2v_cs_start com 86 VSXC 106 2o_cassef com 133 VWN 107 20_ccsd com 158 Density matrices 20_ccsdt_vtz com 21 DF LMP2 167 20_cepal com 148 DF MP2 155 20_diffden_molden com 69 DFT 90 20_direct com 63 20_dma conm 20_field com Difference gradients 132 Diabatization 201 DIIS 131 157 20_forces com 225 DIP 190 20_freqdft com 26 DIP 190 20_gexpec1 com i86 dipole field 190 20_gexpec2 com 37 DIRECT 55 88 20_manymethods com 28 B1 distributed multipole analysis 1
68. 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 OF FSET card which must follow the orbital directive 34 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 ORBITAL 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 34 ORBITAL MERGING 258 34 7 Symmetric orthonormalization ORTH ORTH n1 N2 N8 Symmetrically orthonormalizes the first n vectors in each symmetry i These vectors must be supplied before by ORBITAL and MOVE or ADD directives 34 8 Schmidt orthonormalization SCHMIDT SCHMIDT n1 n2 ng Schmidt orthonormalizes the first n vectors in each symmetry i These vectors must be supplied before by ORBITAL and MOVE or ADD directives 34 9 Rotating orbitals ROTATE ROTATE iorb1 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
69. easily installed If a myrinet network is available myrinet should be chosen This requires in addition 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 A INSTALLATION OF MOLPRO 275 the machine architecture It is then possible to install different versions simultaneously in the same MOLPRO tree see section 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 prompts for the licence key obtainable as described above The key may also be given using the k option on the command line or given through the environment variable SMOLPRO_KEY 5 configure asks whether you wish to use system BLAS subroutine libraries MOLPRO has its own optimized Fortran version of these libraries and this can safely be used On most machines however it will be advantageous to use a system tuned 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 prompt
70. 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 recomputed 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 a
71. for redundancy check using DOMSEL 1 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 DELETG 1 Parameter for multipole treatment of exchange operators DSTMLT 0 3 expansion level for distant pairs Parameters for energy partitioning IEPART 0 If nonzero do energy partitioning EPART 3 0 cutoff parameter for determining individual monomers 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 MAXANG MAXL 99 restriction for Boughton Pulay domain selection CHGMIN 0 01 minimum Mulliken charge for BP domain selection CHGMINH 0 05 minimum Mulliken charge of H atoms for BP domain selection CHGMAX 0 40 If charge larger than this value atom is always included MAXBP MAXBP 0 determines how to rank atoms in Boughton Pulay domain selection MULLIKEN LOCMUL 0 determines how to rank atoms for domains PIPEKAO LOCAO 0 activates AO localization criterion NONORM 2 determines whether projected functions are normalized LMP2ALGO MP2ALGO 1 if nonzero use low order scaling method in LMP22 iterations OLDDEF 0 allows to revert to older defaults Thresholds THRPIP 1 d 12 Threshold for Pipek Meze
72. for table save the table in file h2o tab title for table sort table examples h20_pes_cesdt com 3 INTRODUCTORY EXAMPLES 28 Results for H20 basis VDZ R1 R2 THETA SCF CCSD CCSD T 1 6 1 6 100 0 FIr 99757338 76 20140563 76 20403920 Lo 1 6 100 0 76 00908379 76 21474489 76 21747582 Vi Le 100 0 76 02060127 76 22812261 76 23095473 2 0 LQ 110 0 76 01128923 76 22745359 76 23081968 2 0 2 0 110 0 76 00369171 76 22185092 716422931212 You can use also use DO loops to repeat your input for different methods h20 benchmark method hf fci ci cepa 0 cepa 1 cepa 2 cepa 3 mp2 mp3 mp4 aci ccsd becd qci t ccsd t bccd t casscf mrci acpf basis dz Double zeta basis set geometry o0 h1 0 r h2 0 r h1 theta Z matrix for geometry r 1 ang theta 104 Geometry parameters do i 1 method Loop over all requested methods Smethod i call program e i energy save energy for this method ames h20_manymethods com enddo esc e 1 scf energy efci e 2 lfci energy table method e sat for lprint a table with results Title for table title Results for H20 basis Sbasis R r Ang Theta theta degr This calculation produces the following table Results for H20 basis DZ R 1 Ang Theta 104 degr METHOD E E ESCE E EFCI HF H15 399891339 00000000 13712077 FCI 76 13609416 13712077 00000000 CI 76 12844693 12947355 00764722 CEPA 0 76 134
73. grid dimension 15 3F12 6 number of grid points n3 step vector for third grid dimension 15 4F12 6 atomic number charge and coordinates one such record for each atom 6E13 5 n X m 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 3n3 containing the gradient with the three cartesian components contiguous For the laplacian there is a further record of length n3 25 7 GOPENMOL calculate grids for visualization in gOpenMol GOPENMOL filename iflag n 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 25 PROPERTIES AND EXPECTATION VALUES 195 crd A CHARMm CRD format 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 symm
74. h0 ao isa load h0 h0 load h0 DS load qmzz oper qmzz load quadrupole moment qmzz add h01 h0 field qmzz add quadrupole moment to h0 same result as above with second moments save h01 1210 1 h0 Isave h0 hf Ido scf with modified h0 quad field ladd quadrupole field to h0 hf Ido scf with modified h0 same result as above with matrop field zz field xx 0 5 field vy 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 35 24 Exercise SCF program Write a closed shell SCF program for H20 using 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 Fo
75. in degree for angles 32 GEOMETRY OPTIMIZATION 228 32 GEOMETRY OPTIMIZATION It is possible to invoke the geometry optimization program in two ways Automatic geom etry optimization using the OPTG command or geometry optimization with an explicit input sequence using OPT Normally geometry optimizations should be done using the OP TG com mand This calls OPT and all other necessary programs automatically 32 1 Geometry optimization step OPT The OPT program reads the geometry definitions and geometry parameters energies and gradi ents of the present and previous points from a geometry record It is necessary that the gradients have been computed using the FORCE command see above before calling OPT The program then predicts a new optimum geometry i e takes one optimization step by default optimizing all variable parameters on which the geometry depends and writes this back to the geometry record The optimization is performed in a space of scaled parameters the scaling being such that the initial hessian matrix has unit diagonal elements The variable OPTCONV is set to the length of the step taken in scaled parameter space and can be tested after the OPT step using the IF command to decide whether to return for another geometry For subcommands of OPT see OPTG OPT can also be used for automatic geometry optimization using a sequence of input commands In this case one can specify the first input command needed for computing th
76. input is needed if the MPn card directly follows the corresponding HF SCF Otherwise occupancies and orbitals can be specified as in the CI program The resulting energies are stored in variables as explained in section 6 7 20 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 4 13 For an example see section 21 6 1 The density matrix can be saved using DM record ifil See also sections and 20 2 Density fitting MP2 DF MP2 RI MP2 Density fitting MP2 RI MP2 can be performed with standard density or and Poisson fitting basis sets The present implementation works only without symmetry The input is as follows DF MP 2 Optionally a card DF IT can follow on which the following options can be specified BASIS_MP 2 string Fitting basis sets e g JKFIT default for standard density fitting or DENSITY POISSON for mixed density Poisson fit ting These basis sets must have been defined in a previous BASTS block 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 Scre
77. 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 18 THE CI PROGRAM 144 18 4 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 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 0 then destroy files icfil igfil at end IREST if nonzero restart NATORB if nonzero natural orbitals are calculated and printed The number of printed external orbitals per symmetry is min natorb 2 WFN
78. 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 17 4 2 Selecting configurations SELECT refl 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 refl rec1 file recI gt 2000 The configurations 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 17 THE MCSCF PROGRAM MULTI 120 ref2 rec2 file rec2 gt 2000 Additional configurations are read from the spec ified record If rec2 is nega
79. 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 If nonzero use low order scaling method in LMP2 iterations This may require more CPU time in calculations for smaller molecules 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 Note that GTHRESH is not an input command of the local program and must be given before the METHOD card 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 S lt thresh one function is deleted The default is
80. 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 W directory This determines the destination of permanent wavefunction dump files used for storing information like orbitals or Cl vectors etc These files are essential for restarting a job As explained for the integral files above per manent wavefunction files will be copied to directory after completion of the job The default for directory is SHOME wfu k key where key is the licence key obtainable as described in section A 1 m G The default local memory and GA memory should be checked to be appro priate for the hardware environment A INSTALLATION OF MOLPRO 279 n N The number of processors or their identity can be specified explicitly in the configuration file but very often it is neither desirable nor necessary to do so Where possible the molpro program extracts a reasonable 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 com mand line or rely on molpro s attempts to get it from the batch system A 3 7 Tuning MOLPRO can be tuned for a particular system by
81. keys 1 means print also the entry element Specify chemical element Ifomitted all elements are searched key Specify record key If omitted all keys are searched type Specify entry type i e s p Ifomitted all types are searched format One of text default molpro MOLPRO input format table tabular or html 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 scriptmolpro_basis presents a graphical and forms based interface for performing searches It may be installed locally but is also normally available at http www molpro net current molpro_basis 10 3 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 10 BASIS INPUT 73 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 e 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 e The older segmented Dunning Hay double zeta sets for the first row DZ and DZP e The Roos ANO basis sets ROOS e The Stuttgart ECPs and associated
82. 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 4 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 lt 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 label ends with a colon 4 7 3 Labels LABEL LABEL This is a dummy command sometimes useful in conjunction with GOTO 4 PROGRAM CONTROL 33 4 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 statem
83. matrix DENSITY 187 25 2 3 Linear molecules LINEAR GENERAD 187 25 2 4 Maximum rank of multipoles LIMIT 188 25 2 5 Omitting nuclear contributions NONUCLEAR 188 Peer or eee 188 eat ik al 188 25 2 8 Notes andreferences e 188 A ee aes oe Dee oy ee ee ee ee Ada iS 188 AES RT ees RLM er tech cdg oe al dG ta ls dae A 189 25 3 1 Calling the population analysis program POP 189 25 3 2 Defining the density matrix DENSITY 189 25 3 3 Populations of basis functions INDIVIDUAL 189 whe a Seed oe ee eke ane Heke Bo ah ee ad Rae 4 189 25 4 Finite field calculations o ee 189 25 4 1 Dipole fields DIP o o ee 190 25 4 2 Quadrupole fields QUAD o o o 190 25 4 3 General fields FIELD 0 0 o eee dtes 190 siap E e e las e a we 190 25 5 Relativistic corrections a a s ss ss soss dpo toa daa aa aa E o a 192 a fe ake od Ri ee Ge Ee a hk 192 dd A SO we ew A 193 25 6 1 DENSITY source of density o o 193 25 6 2 ORBITAL source of orbitals 193 CONTENTS XV ka Bh ei ah Oe eb A bab at 193 25 6 4 BRAGG spatial extent of grid 2 2 2 ee 194 25 6 5 ORIGIN centroid of grid 2 2 2 2 2 2 0 00000 194 fe Bi a aw hs ce BoE ae ee A 194 AS 194 26 DIABATIC ORBITALS 196 27 NO
84. 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 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 O 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 6 VARIABLES 46 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 equal 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 a new value THETA 2 4 are unchanged THE
85. 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 molproircand molproi rc 2 1 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 ol output outfile specifies a different output file 2 GENERAL PROGRAM STRUCTURE 2 x executableexecutable specifies an alternative MOLPRO executable file dl 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 whereby an existing output file is saved backup switches it on again vl verbos causes the procedure to echo debugging information noverbose selects quiet operation default cho procedures causes the contents of the default procedure files to be echoed at run time noecho procedures selects quiet operation default f
86. mies is remembered by the program across restarts via the MOLPRO variable DUMMYATOMS Dummies can be reset to their original charges using a DUMMY card with no entries 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 center definitions 9 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 9 8 2 Example interaction energy of OH Ar 10 BASIS INPUT 0H 2Sig Ar linear memory 2 m geometry q1 o ql ro h ql rh ar ql rar o theta h 0 roh 1 8 rar 7 5 theta 0 ro roh 16 17 rh roh 1 17 basis avdz 71 dummy center in center of mass o 180 Igeometry of OH Igeometry of Ar OH bond length distance of Ar from center of mass langle OH Ar distance of O from center of mass Idistance of H from center of mass basis set text calculation for complex eht occ 8 3 3 wf 27 1 1 recsd t e_ohar energy text cp calculation for OH dummy ar rhf oceo 3 1 1 wf 9 1 1 recsd t e_oh energy RHEF for total system CCSD T for total system save energy in variable e_ohar make Ar a dummy center
87. 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 ORBITAL 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 preceeding 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 which specific for CASPT2 3 and are described below 19 2 Coupling MRCI and MRPT2 For particularly difficult cases with strong intruder problems or in which second order pertur bation theory fai
88. nitort 8 14 7 7 Direct SCF DIRECT options If this card is present the calculation is done in direct mode See section 8 for options Normally it is recommended to use the global GD IRECT command to request the direct mode See section 8 for details 14 THE SCF PROGRAM 89 Table 8 Miscellaneous options for the SCF program Option description SHIFTA shifta Level shift for o spin orbitals see SHIFT card SHIFTB shiftb Level shift for B spin orbitals see SHIFT card ACCURA accu Convergence threshold see ACCU card ENERGY thrden Energy convergence threshold default depends on accu UNOMIN unomin Minimum occpation number for UNO CAS default 0 02 UNOMAX unomax Maximum occupation number for UNO CAS default 1 98 POTFAC potfac Scale factor for potential energy in first iteration default 1 0 IPTYP iptyp Interpolation type see IPOL card IPNIT pnit First iteration for DIIS interpolation see IPOL card IPSTEP ipstep Iteration increment for DIS interpolation see IPOL card MAXD 1 S maxdis NITORD nitord NITCL nitcl NITORT nitort MAXI T maxit NDRU ndru NPRT nprt JACOB I jacobi Max number of Fock matrices used in DIIS interpolation default 10 Parameter of reordering orbitals see SHIFT card Parameter for fock matrix see SHIFT card Parameter for orbital orthonormalization see ORTH card Maximum number of iterations see MAXIT card Number of virtual orbitals printed s
89. obtainable from http www netlib org atlas The easi est and safest is to use a pre built library and we found that atlas3 2 1_Linux_ATHL works very well on current hardware The appropriate linker options to provide are L blasdir lcblas 1f77blas latlas An even faster BLAS library 1ibathlonb1las a is available on the MOLPRO web page see below Intel PIV Atlas library atlas3 3 0_Linux_P4SSE2 tgz otherwise as for Athlon above Alternatively the Intel mk1 libraries can be used in conjunction with the Intel fortran compiler INSTALLATION OF MOLPRO 276 For the cases where copyright rules permit these libraries as well as BLAS libraries for other systems such as HP PA RISC 32 and 64 bit and IA64 Intel mk1 can be obtained from http www molpro net blaslibs Specification of these libraries can be simplified by placing any relevant downloaded libraries in the directory blaslibs configure searches 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 command line option see below Any directory structure in the web copy of these libraries should be preserved in the local copy The simplest way to ensure all this is to fetch complete set of libraries using wget cut dirs 1 nH np r http www molpro net blaslibs configure prompts for the destination directory INSTBIN for final installation of
90. 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 23 LOCAL CORRELATION TREATMENTS 174 DELBAS ibaso deleted if the norm of the projected function is below THRCOR de fault 0 1 This parameter determines the method for eliminating redundant functions of pair domains ibaso 0 The space of normalized eigenvectors of S which corre spond to small eigenvalues is eliminated default if no gradients are computed ibaso gt 0 individual basis functions are eliminated The value of ibaso affects details of the method to determine redundant functions ibaso 1 Redundant functions eliminated from pair domains using Jacobi method for diagonalization of overlap matrices This is the de fault if properties or gradients are computed ibaso 2 Redundant functions are eliminated from pair domains us ing Householder method for diagonalization of overlap matrices ibaso 3 Redundant functions are eliminated from orbital and pair domains using Jacobi method for diagonalization of overlap matri ces ibaso 4 Redundant functions are eliminated from orbital and pair domains using Householder method for diagonalization of overlap matrices The diagonalization method has only an effect for DELEIG 1 if degenerate eigenvalues are present ibaso gt 2 has only an effect if NONORM 0 Parameters for selection of redundant func
91. on the WF card must form a block which comes directly after the WF card The cards can be in any order however 17 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 2S singlet 0 doublet 1 trip let 2 etc Note that these values take sensible defaults if any or all are not specified see section 2 13 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 17 3 2 Defining the number of 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 17 3 3 Specifying weights in state averaged calculations WEIGHT w 1 w 2 w nstate 17 THE MCSCF PROGRAM MULTI 119 w i is th
92. options Iglobal thresholds optional gdirect options Iglobal direct optional gexpec opnames global definition of one electron operators basis basisname basis specification If not present cc pVDZ is used varl value var2 value Isetting variables for geometry and or wavefunction definitions geometry geometry specification program program or procedure name ras lend 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 2 4 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 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 In
93. points on energy surfaces and reaction path following The stan dard algorithms are based on the rational function approach and the geometry DIIS approach 32 GEOMETRY OPTIMIZATION 229 Also available is the quadratic steepest descent following method of Sun and Ruedenberg see J Sun and K Ruedenberg 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 see F Eckert P Pulay and H J Werner J Comp Chem 18 1473 1997 The OPTG must directly follow the input for the wavefunction used in the geometry optimiz ation It will call FORCE OPT INT and as needed HF RHF MCSCF CI CCSD etc For each of these programs the input file is automatically repositioned to the last corresponding input before the OPTG card so any input for RHF MCSCF CI CCSD etc can be used and will be correctly processed It is essential however that the most recently optimized orbitals are used in the wavefunction for which the geometry is optimized Any input needed for OPTG must directly follow the OPTG card The gradients are computed analytically for HF DFT MP2 QCISD or MCSCF wavefunctions otherwise the gradients are computed by finite differ ences see OPTG NUMERICAL Davidson corrected energies or excited state energies can be optimized using the VARIABLE and STATE subdirective Various options in particular convergence criteria c
94. 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 2 GENERAL PROGRAM STRUCTURE 11 Table 2 Numbering of the irreducible representations in Do Dor No Name Function 1 Ag RY 2 B3u Xx 3 Boy y 4 Big xy 5 Biu 6 Bog XZ 7 B3g YZ 8 Au XYZ Table 3 Numbering of the irreducible representations in the four dimensional groups Cry Can D2 No Name Function Name Function Name Function 1 Aj S Z Ag S XY A S 2 B X XZ Ay Z B3 X YZ 3 B2 y yz Bu x y B2 Yy XZ 4 A2 xy By XZ YZ B xy Table 4 Numbering of the irreducible representations in the two dimensional groups Es C2 Ci No Name Function Name Function Name Function 1 1 A S X Y Xy A S Z Xy Ag 8 XY XZ YZ 2 A Z XZ YZ Bo x y x2 yZ u X Y Z 2 GENERAL PROGRAM STRUCTURE 12 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 6 3 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 calcul
95. rewound otherwise it is appended 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 rows according to increasing values of the given columns The columns are sorted in the order they are specified Specify print options TABLE HEADING TITLE WARNING FORMAT SORT The default is print for the first three and noprint for the last three Disable print for given keys Don t write data to the punch file data are written by default Specify start and end indices of the variables to be printed Print also linear regression and quadratic fits of the data columns P LOT CMD 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 p
96. second derivatives for SCF MCSCF Efficiency improvements for open shell coupled cluster Further parallelization Open shell MP2 LMP2 and LCCSD Some of these features will be included in the base version whereas others will be available only as optional modules The above list is for information only and no representation is made that any of the above will 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 iv References All publications resulting from use of this program must acknowledge the following MOLPRO a package of ab initio programs designed by H J Werner and P J Knowles version 2002 1 R D Amos A Bernhardsson A Berning P Celani D L Cooper M J O Deegan A J Dobbyn F Eckert C Hampel G Hetzer P J Knowles T Korona R Lindh A W Lloyd S J McNicholas F R Manby W Meyer M E Mura A Nicklass P Palmieri R Pitzer G Rauhut M Schiitz U Schumann H Stoll A J Stone R Tarroni T Thorsteinsson and H J Werner 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
97. 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 syml 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 21 THE CLOSED SHELL CCSD PROGRAM 160 21 9 1 Parameters for EOM CCSD EOMPAR Normally no further input is needed However some defaults can be changed using the EOMPAR directive EOMPAR key 1 valuel key 2 value 2 where the following keywords ke y are possible MAXDAV nv Maximum value of expansion vectors per state in Davidson procedure default 10 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 Nu
98. 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 On Intel Linux the Portland pgf90 com piler version 3 2 or higher is recommended For using this compiler it is required that the environment variable PGI is set and points to the root directory where the compiler is in stalled e g usr local pgi MOLPRO has also been tested with the Intel Compiler ifc Version 5 0 1 In order to use this compiler the environment variable A32ROOT must be set and point to the appropriate directory normally opt intel compiler50 ia32 On Alpha Linux the Compaq compiler is rec ommended The directory in which the compiler is located must be in your PATH 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 weet utility for batch mode http transfers although not needed for installation 1s essential for any subsequent application of patches that implement bug fixes 4 About 200Mb disk space strongly system dependent more with large blocksize file sys tems and where binary files are large
99. the CI program BRUECKNER parameters can be modified using the BRUECKNER directive 21 THE CLOSED SHELL CCSD PROGRAM 157 The Brueckner orbitals and approximate density matrix can be saved on a MOLPRO 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 21 3 1 The BRUECKNER directive BRUECKNER orbbrk ibrstr ibrueck brsfak This directive allows the modification 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 ib
100. three row elements You may also put in individual matrix elements of the hessian value sets starting value for hessian matrix element between parameters paraml param2 If param2 is omitted it defaults to paraml diagonal element If the Model Hessian is disabled the initial hessian is diagonal with values 1hartree bohr 2 for all lengths 1 hartree radian 2 for all angles This is usually quite reasonable except for cases such as dihedral angles A reasonable strategy for complicated cases is to perform an optimization with a small basis set at the SCF level with PRINT HESSIAN in order to obtain an approximate starting hessian These values are set before processing the START record see above This option is obsolete if the Model Hessian is used default unless heavy elements are present In transition state searches the hessian matrix is evaluated numerically see NUMHES section 82 2 15 Alternatively the cartesian hessian matrix evaluated in a previous frequency calcu lation see FREQUENCIES section 33 can be used with the HSTART command see section B2 2 16 It is also possible to use the numerical hessian or the hessian from a frequency calcu lation in minimizations Note that numerical hessians cannot be computed when dummy atoms holding basis functions are present 32 2 6 Transition state saddle point optimization ROOT ROOT root Specifies the eigenvector of the hessian to be followed root 1 specifies a minimi
101. threshold for generation of 2 external integrals Defaults THR_D2EXT THREST_DTRAF THR_DTRAF THREST default 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 58 THRINT_D2EXT THRPROD_D2EXT THR_D3EXT THREST_D3EXT THRINT_D3EXT THRPROD_D3EXT THR_D4EXT THREST_D4EXT THRINT_D4EXT THRPROD_D4EXT THR_DCCSD THREST_DCCSD THRINT_DCCSD THRPROD_DCCSD THRMAX_DCCSD Integral threshold for generation of 2 external integrals Defaults THR_D2EXT THRINT_DTRAF THR_DTRAF THRINT default Product threshold for generation of 2 external integrals Defaults THR_D2EXT THRPROD_DTRAF THR_DTRAF THRPROD default 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 gene
102. 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 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 9 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 rath
103. used The specification of isym is optional it can also be defined using the SYMMETRY key dentype Density type This can be one of CHARGE charge density SP IN UHF spin density TRANSITION transition density matrix The default is CHARGE symmetry Specifies a particular state symmetry Alternatively the state symme try can be specified using STATE see above spin Spin quantum number i e O for singlet 1 2 for doublet 1 for triplet etc Alternatively MS2 can be used ms2 2Ms i e O for singlet 1 for doublet 2 for triplet etc Alternatively SPIN can be used nelec Number of electrons iset Set number of orbitals The orbital sets are numbered in the order they are stored 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 Use SCF orbitals ORBITAL 2140 2 Us 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 1 2 GENERAL PROGRAM STRUCTURE 14 2 17 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 low
104. 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 two integrals of a triple differ DKEXT 3 use in core algorithm and integral triples if all inte grals of a triple differ 1f 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 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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 61 THRMAX_DKEXT Initial value for THREST_DKEXT in CI and CCSD calcula tions If nonzero it will also be used for
105. vectors 28 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 B and LA 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 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 17 5 8 Secondly the DDR procedure can be used to find the transformation amon
106. world wide web Those en titled to the code should obtain it from http www molpro net distrib 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 are 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 The whole key needs to be inserted in its entirety as directed below The web pages provide a facility to have this key delivered by email to a registered licensee 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 two different RPMs one contains the program and the other documen tation and either may be installed independently of the other At present these RPMs are not relocatable and will install under usr local Once the program has been installed using for example rpm Uhv molpro 2002 6 1386 rpm some post installation configuration is necessary and this is accomplished by changing or adding options in the script file that runs the program usually usr local bin molpro Most importantly a valid licence token must be given in a k option Other configuration options as descri
107. 0 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 HOME w fu this is the default molpro h2 com h2 com contains A Y file 2 h2 wf punch h2 pun basis vdz examples geometry angstrom hl h2 h1 74 h2 com hf 3 As before but the file h2 wf is sent to the directory tmp wfu molpro W tmp wfu h2 com 3 2 Simple SCF calculations The first example does an SCF calculation for H20 using all possible defaults x x h20 A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input HIFOJ examples H2 0 r H1 theta h20_scf com hf closed shell scf In the above example the default basis set VDZ is used We can modify the default basis using a BASIS directive h20 cc pVTZ basis 1A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input nea al examples Bap Orey iiy theta h2o_scf_vtz com basis VTZ luse VTZ basis a hf closed shell scf 3 3 Geometry optimizations Now we can also do a geometry optimization simply by adding the card OPTG Ex AZO A title r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input HL OE H2 0 r H1 theta examples basis 6 31g luse Pople basis set h2o_scfopt_631g com hf closed shell scf optg Ido scf geometry optimization 3
108. 0 52 52 53 53 53 54 55 63 CONTENTS 9 7 Redefining and printing atomic masses o o o 9 8 Dummy centres 9 8 1 Counterpoise calculations 2 2 0 o o o 9 8 2 Example interaction energy of OH Ar 10_ BASIS INPUT 10 1 Cartesian and spherical harmonic basis functions 10 2 The basis set library 10 3 Default basis sets 10 4 Default basis sets for individual atoms 10 5 Primitive set definition 10 6 Contracted set definiti0onsS e 10 7 Examples 11 EFFECTIVE CORE POTENTIAL 11 1 Input from ECP library 2 ee ee 11 2 Explicit input for ECPs 2 2 0 0 0 o o e 11 3 Example for explicit ECP input o o e 11 4 Example for ECP input from library o 12 CORE POLARIZATION POTENTIALS 12 1 Input options 12 2 Example for ECP CPP 2 0000000000 2 eee 13 RELATIVISTIC CORRECTIONS 13 0 1 Example for computing relativistic corrections 14 THE SCF PROGRAM ent SAE eee ne Gel Be ok en at ae NO 14 1 1 Defining the number of occupied orbitals in each symmetry De bk a ea ey oa ed he eA PtGhon Sada e be ae at oae bos at dhe Bb hated wae wed ee 14 3 Starting orbitals TRESEN TEE E Ee ee ee aa fad Udo A ee ee VETA 14 5 Using additional point group symmetry o o o o
109. 0 SPIN ORBIT COUPLING 219 bra2ms 2 x Ms value of the bra wavefunction ket2ms 2 x Ms value of the ket wavefunction lsop Cartesian component of the Spin orbit Hamiltonian This can be one of LSX LSY or LSZ in 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 Cl vectors using the MRCI program see chapter 118 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 tha
110. 016 15 1 21 P86 J P Perdew Phys Rev B 33 8822 1986 VWN with gradient correction e C rjo ap 3 1 6 a ES Pa Pp a 8 p e A 0y 1 h8 Ba Me ERA AA 9 21 8 1 0 gt A q ki li mi n1 K pet where ae Ig h q k2 l2 m2 n2 w q kz3 l3 m3 n3 q A p c d A in 5 el Dic q arctan gen pod in eh 2 on aretan 2 c d V4d 2 X i c d ci d p 0 007390075 Vo C r p 6 PENE 1 2 C 2 0 002568 ar Br2 C r 0 001667 r T Er br2 10000873 z 0 11 98 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 15 THE DENSITY FUNCTIONAL PROGRAM a 0 023266 B 0 000007389 E 8 723 0 472 k 0 0310907 0 01554535 1 6m 1 0 10498 0 325 0 0047584 m 3 72744 7 06042 1 13107 and n 12 9352 18 0578 13 0045 15 1 22 PBE PBE PW91C PBEX J P Perdew K Burke and M Ernzerhof Phys Rev Lett 77 3865 1996 15 1 23 PBEO PBEO 0 75PBEX PW91C 0 25 Exact Exchange C Adamo and V Barone J Chem Phys 110 6158 1999 15 1 24 PBEX PBE Exchange Functional J P Perdew K Burke and M Ernzerhof Phys Rev Lett 77 3865 1996 1 E n 3 MO T where ws F S 1 R R t E R 0 804 u n 3 and 5 0 066725 99 111 112 113 114 115
111. 0675 3 22494 1 68698 0 0235810 15 1 35 THGFL D J Tozer N C Handy and W H Green Chem Phys Lett 273 183 1997 Density dependent first row exchange correlation functional for closed shell systems K Y or i l where n 4 Ri P PB t 7 6 4 3 3 2 5 3 and 1 06141 0 898203 1 34439 0 302369 190 191 192 193 194 195 196 197 198 199 15 THE DENSITY FUNCTIONAL PROGRAM 105 15 1 36 THGFC D J Tozer N C Handy and W H Green Chem Phys Lett 273 183 1997 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 K O R X 200 i 1 where n 12 201 Ri P P 202 Vi 2 Vi 2 Ooa ga Opg Xi PB gt on 11 11 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 F 3 2 5 3 2 204 v 0 0 0 0 1 1 1 1 2 2 2 2 205 and o 0 864448 0 565130 1 27306 0 309681 0 287658 0 588767 0 252700 0 0223563 0 0140131 0 0826608 0 0556080 0 00936227 206 15 1 37 THGFCFO D J Tozer N C Handy and W H Green Chem Phys Lett 273 183 1997 Density and gradient dependent first row exchange correlation functional The closed and open shell parts are fitted to training sets of closed and open shell systems independently K Y 0 R S X Y 207 i l where n 20 208 Ri
112. 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 23 LOCAL CORRELATION TREATMENTS 177 THRCOR thresh 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 card 23 7 Additional options available on the ATTENUATE card The defaults reported for the following keys are likely to change in the future Most important options DECAY 0 SHORTMLT level LONGMLT level This is the decay parameter that determines the splitting of the Cou lomb operator in the split approach Larger values of put more weight to the long range part of the operator which means that the multipole correction will have more difficulties to converge but the transformation of the short range part will be faster Default 0 20 Determines the expansion level of monopolar multipole expansions in the context of the split Coulomb operator approach Default 15 Determines the expansion level of bipolar multipole expansions in the context of the split Coulomb operator
113. 18 3 Additional reference symmetries REF sym 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 IT 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 WE 15 1715 define wavefunction symmetry 1 REF 2 define additional reference symmetry 2 and for the doublet A state Wey Loy ey 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 WE 15 2914 define wavefunction symmetry 2 REF 1 define additional reference symmetry 1 REF 3 define additional reference symmetry 3 or WF 15 3 1 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 18 3 1 Selecting configurations SELECT refl ref2 refthr refstat mxshrf This card is used to sp
114. 3070930 1 2888938190 0 8907763360 H 3 4903070930 1 2888938190 0 8907763360 H 2 6289534570 2 5638654230 0 0000000000 H 4 1360211370 1 5529079440 0 0000000000 H 3 6817059520 0 0685850980 0 8931597470 H 3 6817059520 0 0685850980 0 8931597470 hf optg coord bmat method dii S Optimization in natural internal coordinates Optimization method Geometry DIIS 32 GEOMETRY OPTIMIZATION Results ITER ENERGY OLD ENERGY NEW 1 667 68596573 667 72289939 2 667 72289939 667 73501326 3 667 73501326 667 73561597 4 667 73561597 667 73564985 5 667 73564985 667 73565369 6 667 73565369 667 73565399 32 3 5 Transition State of Bicyclo 1 1 0 butane ring opening DE 03693366 01211387 00060271 00003388 00000384 00000029 Bicyclo 1 1 0 butane Transition State memory 1 m D OO Oo OO GRADMAX 08872281 04342539 01213581 00342477 00098604 00029103 Define Active Variables Geometry Specification Z Matrix Transition State search Use Quadratic St basis 3 21G L1 1 495 ang L2 1 418 ang L3 1 463 ang L4 1 093 ang L5 1 111 ang L6 1 098 ang L7 1 097 ang L8 1 110 ang L9 1 106 ang A1 92 1 degree A2 62 1 degree A3 136 0 degree A4 123 5 degree A5 122 4 degree A6 124 7 degree A7 126 7 degree A8 117 9 degree D1 120 4 degree D2 4 4 degree D3 108 8 degree D4 107 5 degree D5
115. 35 7 0146 28 382 35 033 20 428 76 B 0 56258 0 0171 1 3064 1 0575 0 8854 77 C 1 09025 0 7992 5 5721 5 8676 3 0454 78 and A 0 006 0 2 0 004 79 15 THE DENSITY FUNCTIONAL PROGRAM 97 15 1 17 LTA Local t Approximation J P Perdew and Y Wang J Chem Phys 111 911 1999 LSDA exchange functional with density represented as a function of T K 5Y En 80 where 3a 4 5 E a c on 81 and ERT T i m 15 1 18 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 A Z K 4 POPP ABa Papp 47 78 18 2p2 3 5 11 PsOss alon 8220p3 5 2 8 18 0 OLY PaO 2 o pps eps 5 2 8 18 ds 7 T 2p 3 p os gt 83 where EE e pez e 84 c dZ a 85 pl B 0 04918 86 A 0 132 87 c 0 2533 88 d 0 349 89 3 2 3 and d NT z 1 91 15 1 19 MK00 Exchange Functional for Accurate Virtual Orbital Energies F R Manby and P J Knowles J Chem Phys 112 7002 2000 3m K 92 S Ts 05 4 15 THE DENSITY FUNCTIONAL PROGRAM 15 1 20 MKOOB Exchange Functional for Accurate Virtual Orbital Energies F R Manby and P J Knowles J Chem Phys 112 7002 2000 MKOO with gradient correction of the form of B8 8X gy pps x T Vs 4 1 6BXs ea where B 0 0
116. 35023712 0 0 0 0 0 80039692 37236272 0 0 0 0 0 80039692 The frequencies intensities and further details can be found in the output file 3 9 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 H20 potential geometry 1x o Aly 07 11 h2 0 r2 1 h1 theta i basis vdz angles 100 104 110 distances 1 6 1 7 1 8 1 9 2 0 i 0 do ith 1 angles do irl 1 distances do ir2 1 irl i i 1 r1 i distances irl r2 i distances ir2 theta 1 angles ith AE escf i energy ecsd t eccsd 1 energc eccsat i energy enddo enddo enddo table r1 r2 theta escf eccsd eccsdt head rl r2 theta scf ccsd ccsd t save h2o tab title Results for H20 basis Sbasis SOT 3 1 2 This produces the following table luse cs symmetry z matrix define basis set list of angles list of distances linitialize a counter loop over all angles H1 0 H2 loop over distances for O H1 loop over O H2 distances rl ge r2 lincrement counter Isave rl for this geometry Isave r2 for this geometry save theta for this geometry Ido SCF calculation save scf energy for this geometry Ido CCSD T calculation save CCSD energy Isave CCSD T energy lend of do loop ith lend of do loop irl lend of do loop ir2 produce a table with results modify column headers
117. 4 3 General fields FIELD FIELD operl facl oper2 fac2 FIELD operl facl oper2 fac2 Adds one electron operators operl oper2 with the corresponding factors facl fac2 to the one electron hamiltonian The available operators are 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 25 4 4 Examples The first examples shows various possibilities to add perturbations to the one electron hamilto nian 25 PROPERTIES AND EXPECTATION VALUES 191 H20 finite fields memory 4 m R THETA 0 96488518 ANG 101 90140469 geomet ry H1 O H1 R H2 0 R H1 THETA hf wf 10 1 Iscf without field f 0 05 dip f ladd dipole z field to h0 hf Ido scf with modified h0 field dmz f ladd dipole z field to HO same result as previous exampl hf Ido scf with modified h0 quad f add quadrupole qmzz field to h0 les hf Ido scf with modified h0 eas fais field com field qmzz f add quadrupole qmzz field to h0 same result as previous exampl hf Ido scf with modified h0 field zz f xx 0 5 f yy 0 5 f hf field zz f field field hf field hf The se fields ladd general field same result as quad a
118. 4 3 Projected excited state calculations PROJECT record nprojc Initiate or continue a projected excited state calculation with information stored on record If nprojc gt O 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 O 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 states use Clip fproject 3000 3 1 Clp 7 project 3000 3 ci project 3000 3 18 4 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 18 4 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
119. 625 0 000768156 0 03 10377 0 0720326 0 0446562 0 266802 1 50822 1 94515 0 679078 15 1 33 TH3 D J Tozer and N C Handy Mol Phys 94 70 1998 179 180 181 182 183 184 Density and gradient dependent first and second row exchange correlation functional of the form TH2 but with t 7 6 4 3 3 2 5 3 2 3 2 5 3 25 3 1 2 5 3 11 2 7 6 4 3 3 2 5 3 8 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 v 0 0 0 0 1 1 1 1 2 2 2 0 0 0 0 0 0 0 0 185 186 187 15 THE DENSITY FUNCTIONAL PROGRAM w 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 and o 0 142542 0 783603 0 188875 0 0426830 0 304953 0 430407 0 0997699 0 00355789 0 0344374 0 0192108 0 00230906 0 0235189 0 0331157 0 0121316 0 441190 2 27167 4 03051 2 28074 0 0360204 15 1 34 TH4 D J Tozer and N C Handy Mol Phys 94 70 1998 104 188 189 Density an gradient dependent first and second row exchange correlation functional of the form TH2 but with t 7 6 4 3 3 2 5 3 5 3 2 5 3 5 3 2 2 5 3 4 2 7 6 4 3 3 2 5 3 5 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 v 0 0 0 0 1 1 1 1 2 2 2 0 0 0 0 0 0 0 0 w 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 and 0 0677353 1 06763 0 0419018 0 0226313 0 222478 0 283432 0 0165089 0 0167204 0 0332362 0 0162254 0 000984119 0 0376713 0 0653419 0 0222835 0 375782 1 9
120. 792 9 9 9 1 1828200 4074500 0500000 5 6 1560908 6297641 5474485 3 11 1293530 8727459 8338853 2 16 1227807 1282973 3306094 Zs 2s Lis 4 Zo 2s 6 3 247536 293352 353062 2 1818752 4155787 4419262 7 1498119 08903231 9015764 12 1232020 9341224 3625506 17 1279904 1591890 8192878 Enter geometry optimization Reaction path following First step is along the transition vector Print optimization history 92 249729 92 92 302671 92 92 354060 92 3 1760568 1 4753427 50 1407672 47 8 T3915 17 1 7427947 36 6753140 32 13 1 1236534 1 9767135 Le 19 9713780 14 18 1355902 dye 2 1884619 2 0 4100429 0 The minimum reached is the HNC molecule 1385201 STSL 1164836 254961 314152 354083 4 1711937 5158853 8796949 9 4681651 4 0569670 2222411 19 1374274 1868721 0400977 250 examples hcn_isomerization com 92 260937 92 322784 92 354084 5 1636706 5757858 9463232 10 1 1324402 8259215 1961654 5 11212118 2 0879631 5176033 20 1 1371834 2 1873023 0160280 92 269447 92 337223 First step in the opposite direction 32 OPTIMIZATION HISTORY ENE RGIES Positions 92 92 92 Le L1 L2 Al L1 L2 Al L1 L2 Al L1 L2 Al L1 L2 Al 246043 273233 331223 339713 923 92 92 6 T2 110 153
121. 84 2 degree D6 109 3 degree D7 106 1 degree geometry Cl c2 1 DI C3 212 1 Al c4 1 13 A2 3 D1 H5 1 14 A3 3 D2 H6 2 L5 1 A4 4 D3 H7 3 L6 A5 1 D4 H8 3 L7 A6 1 D5 H9 4 L8 1 A7 2 D6 H10 4 L9 1 A8 2 D7 int rhf optg root 2 method qsd Results ITER ENERGY OLD ENERGY NEW pest Descent Method GRADMAX 249 examples butane_opt_transition co1 32 GEOMETRY OPTIMIZATION 1 153 88760305 153 90344614 2 153 90344614 153 90480163 3 153 90480163 153 90492606 4 153 90492606 153 90492110 5 153 90492110 153 90493986 6 153 90493986 153 90494020 7 153 90494020 153 90494021 015843 001355 000124 0 000004 000018 000000 000000 09 0 05684518 49 0 01484734 43 0 00659821 97 0 00428601 76 0 00329871 34 0 00030504 01 0 00015487 32 3 6 Reaction path of the HCN HNC isomerization Xx kxk memory 1 m basis 3 21G 11 1 18282 ang 12 1 40745 ang al 55 05 degree geomet ry x C N 1 11 H 2 12 1 a1 int rhf optg method qsdpath option idir 1 print history HCN lt gt NHC Isomerization Reaction Path Starting geometry is transition state Cs Symmetry Results The minimum reached is the HCN molecule OPTIMIZATION HISTORY ENERGIES Positions L1 L2 Al L1 L2 Al L1 L2 Al L1 L2 Al Reaction path following using option idir 1 Results 92 246043 92 276316 92 343620 92 923 S92 246064 284467 349
122. 87 DM Zo _mrei_ vez com 21 20_optmp2 com 25 h h h h h h h h h h h h h h h h h h h h h h h h h h h h DMA 187 20_optmp2_runccsdt com 26 DO 30 20_pes_ccsdt com 27 BI DO loops 30 20_pop com 189 DONT 127 20_proce com 23 DUMMY 70 20_property com 186 Dummy centres Q X 66 20_put_molden com 69 DUMP 109 184 20_sc com 20 20_scf_vtz com 20 73 ECP 20_scf_vtz_explicit com 73 library 78 20_scfopt_6319 com 20 INDEX 20_sto3gstartl com 85 20_sto3gstart2 com 85 20_table com 21 20_vqz_fp com 78 20_vqz_fp_explicit com 20_xyzinput com 67 20p_mrci_trans com 148 25_diab com 125 196 2s_diab1 com 202 2s_diab2 com 204 cn_isomerization com 250 cn_mp2_ts com 237 fdimer_cpcopt 1 com 239 fdimer_cpcopt2 com 241 _ecp com 221 f_nacme com 199 lih2 _S0T0 com 24 lih2_D0D1 com 244 matrop com 268 matropfield com 269 n2_rasscf com n2f2_ccsd com n n n ee Ey ey ey o o o ey iy yy E Ju Rai WI Go ooj BY O 3 2 a2_ecp_cpp c o_mergel com TIES o_merge2 com oh_macros com oh_runccsdt oh_runmrcil aa meg 3 3 oh_runmrci2 oh_runmrci3 Q O 3 Q O 5 MENEE oh_runmrci4 a ohar_bsse com 70 pf5_freq com 254 s_so com 220 EXCHANGE 109 EXPEC 36 87 123 143 EXPEC2 128 Expectation val
123. 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 6 8 Displaying variables Variables or the results of expressions can be displayed in the output using SHOW and TABLE 6 8 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 gt 6 more than one line is needed but in any case a new line is started after n elements SHOW n 10 10 4 prints E in the format given with newline forced after n elements 7 TABLES AND PLOTTING 53 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 u
124. 90643 13593304 00118773 CEPA 1 76 13304720 13407381 00304696 CEPA 2 76 13431548 13534209 00177868 CEPA 3 76 13179688 13282349 00429728 MP2 76 12767140 12869801 00842276 MP3 76 12839400 12942062 007 70015 MP4 76 13487266 13589927 00122149 QCI 76 13461684 13564345 00147732 CCSD 76 13431854 13534515 00177561 BCCD 76 13410586 13513247 00198830 QCI T 76 13555640 13658301 00053776 CCSD T 76 13546225 13648886 00063191 BCCD T 76 13546100 13648762 00063315 CASSCF 76 05876129 05978790 07733286 MRCI 10 133118309 13414496 00297580 ACPF 76 13463018 13565679 00146398 One can do even more fancy things like for instance using macros stored as string variables See example oh_macros com for a demonstration 4 PROGRAM CONTROL 29 4 PROGRAM CONTROL 4 1 Starting a job The first card of each input should be xxx text where text 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 4 2 Ending a job The end of the input is signalled by either an end of file 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
125. AT 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 MAXIT see maxiter card MAXITI see maxiter card MAXDAV see maxdav card MAXVI see maxdav card NOSING see nosing card NOPAIR see nopair card MXSHRF see select card IKCPS 0 In CIKEXT only K CP is calculated this option taken when and only when no singles IKCPS 1 only K CP is calculated Implies that modified coupling co efficients are used IKCPS 2 K CP and K CP are calculated Default is IKCPS 2 except when single reference configuration when IKCPS 1 IOPTGM Option for density matrix routines IOPTGM 0 all quantities in density matrix routines are recalculated for each intermediate symmetry max CPU min core 18 THE CI PROGRAM 145 IOPTGM 1 quantities precalculated and stored on disk max I O min core IOPTGM 2 quantities precalculated and kept in core min CPU max core IOPTOR If nonzero calculate intermediate orbitals for each pair Might improve convergence in some cases in particular if localized orbitals are used 18 4 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 disc sector length INTREL number of integer
126. BITAL 2120 2 OVE 1 1 0 4 ROT 3 1 4 1 45 ROT 5 1 6 1 453 PRINT 1 ORTH 6 2 2 save 2150 2 WaltA oce 6 2 2 wi 15 2 1 wf 15 3 1 start 2150 2 260 IN atom c2v symmetry 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 c2v symmetry read move move move move move move move move move read move orbitals of N atom ls orbital to output vector 1 1 2s orbital to output vector 3 1 2pz orbital to output vector 5 1 2px orbital to output vector 2 examples 2py orbital to output vector 1 3 no_mergel com virtual orbitals of symmetry virtual orbitals of symmetry virtual orbitals of symmetry virtual orbitals of symmetry orbitals of O atom all oxygen orbitals into place NDNNRRPR UO rotate linear rotate linear 2s orbitals to make bonding and antibonding combinations 2pz orbitals to make bonding and antibonding combinations set print option symmetrically orthonormalize the valence orbitals the resulting orbitals are printed save merged orbitals to record 2150 2 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 calc
127. CH 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 32 2 14 Numerical gradients NUMERICAL NUMERICAL active step activez step2 With this option 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
128. D J Tozer and N C Handy J Chem Phys 109 6264 1998 The original HCTH functional with parameters optimized on a set of 93 training systems K Pa Pp Pa 0 Pg 0 Ao Ain d A1 Aan d 2 43n d 21 Aan d A1 Y e ps 0 Bo Bin xs A2 Bon Xi A2 Bary x5 A2 Ban x5 A2 FA 4 3 2 2 0 2 3 0 2 4 2 7 Ar Ps Co Cin x A3 C2n X5 A3 C3n xs 43 Can 25 23 65 where d 2 93 2 66 u 7 n 0 u 0 67 A 0 72997 3 35287 11 543 8 08564 4 47857 68 B 0 222601 0 0338622 0 012517 0 802496 1 55396 69 C 1 0932 0 744056 5 5992 6 78549 4 49357 70 A 0 006 0 2 0 004 71 and B is the correlation energy per particle of the Local Spin Density Approximation PW92C 15 1 15 HCTH120 A D Boese N L Doltsinis N C Handy and M Sprik J Chem Phys 112 1670 2000 HTCH optimized on a set of 120 training systems extending the set of HCTH93 to include Anionic atoms and molecules 2 row anions and H bonded dimers A 0 51473 6 9298 24 707 23 110 11 323 72 B 0 48951 0 2607 0 4329 1 9925 2 4853 73 C 1 09163 0 7472 5 0783 4 1075 1 1717 74 and A 0 006 0 2 0 004 75 15 1 16 HCTH147 A D Boese N L Doltsinis N C Handy and M Sprik J Chem Phys 112 1670 2000 HTCH optimized on a further extended set of 147 training systems A 0 542
129. DECI Here is an example of advanced use of macros and string variables 6 VARIABLES 45 test for parser text This fancy input demonstrates how string variables and macros can be used text basis vdz Idefine basis set geometry 0 H 0 r define geometry z matrix text methods method rhf 2 casscf 2 mrci text active spaces spaces 3 1 1 3 4 2 2 3 5 2 2 text symmetries symset 1 2 1 2 3 1 2 text weight factors for state averaged casscf werghts Ly oly dh Ba Sy iy 0 az EJ text scf occupation scfocc 3 2 1 text bond distance examples r 1 85 oh_macros com hf do i 1 method loop over methods occ S S spaces i set active space for this run set symmetry S symset i set symmetries for this run weight S S weights i set weights for this run Smethod i now run method e 1 Senergy save energies in strings dipol i1 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 6 5 Indexed Variables Vectors Variables 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
130. DKEXT in MP3 and MP4 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 thresholds THRINT_DKEXT and THRPROD_DKEXT are ob tained by multiplying their input or default values by THRMAX_DKEXT THREST_l 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 DISKSIZE 0 THRDISK 1 d 3 MEM _ DFOCK BUF_DFOCK l1 SWAP_DFOCK SWAP DMP 2 DTRAF 1 PAGE_DTRAF PAGE 1 SCREEN_DTRAF SCREEN MAXSHLO1_DTRAF NSHLO1 32 MINSHLO1_DTRAF 0 MAXSHLO2_DTRAF NSHLO2 16 MINSHLO2_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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 62 Table 7 Default thresholds and alias names for direct calculations Parameter A
131. E 33 2 Thermodynamical properties THERMO o o 33 3 Examples e sag rr a E e A 34 ORBITAL MERGING 34 1 Defining the input orbitals ORBITAD o o 34 2 Moving orbitals to the output set MOVE o o 34 3 Adding orbitals to the output set ADD o o 34 4 Defining extra symmetries EXTRA o e 34 5 Defining offsets in the output set OFESET o 34 6 Projecting orbitals PROJECT 2 o o a 34 7 Symmetric orthonormalization ORTH o o 34 8 Schmidt orthonormalization SCHMIDT a oaa 34 9 Rotating orbitals ROTATE 2 o o e e o 34 10Initialization of a new output set INIT o o 34 11Saving the merged orbitalS o o 34 12Printing options PRINT o o e o Se ok ae rta boa gia ghere Gece a atte aug ete a dee ere A BAAS GB ios te 8 ie a ea g Bb le we etd e eee eb ie ee ee ah pe Ane Une Pe ee eas Ce RA eae ne ee eee 35 MATRIX OPERATIONS 35 1 Calling the matrix facility MATROP o o o e 35 2 Loading matrices LOAD 2 ee ieee Monee a aaa bra tl fale a ad eee eed Goa Laia apena eo eee eee eer ee ae OOE 35 3 Saving matrices SAVE 2 ee 35 4 Adding matrices ADD lt oaa o 35 5 Trace of a matrix or the product of two matrices
132. El Azhary G Rauhut P Pulay and H J Werner J Chem Phys 108 5185 1998 QCISD and LQCISD gradients G Rauhut and H J Werner Phys Chem Chem Phys 3 4853 2001 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 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 Schiitz G Hetzer and H J Werner J Chem Phys 111 5691 1999 G Hetzer M Schiitz
133. F 0 is the recommended algorithm for very large molecules cf linear scaling LMP2 chapter 23 DTRAF 2 alternative algorithm to generate the exchange operators directly in projected basis Usually this algorithm 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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 60 THREST_LMP2 THROQ1_LMP2 THRO2_LMP2 THRAO_ATTEN 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 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
134. F 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 submitted for publication are available in MOLPRO In both cases a high spin RHF refer ence wavefunction is used 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 alternative 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 The 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 defi
135. H Stoll and H J Werner J Chem Phys 113 9443 2000 Local MP4 SDTQ CCSD T QCISD T C Hampel and H J Werner J Chem Phys 104 6286 1996 M Schiitz and H J Werner J Chem Phys 114 661 2001 M Schiitz and H J Werner Chem Phys Lett 318 370 2000 M Schiitz J Chem Phys 113 9986 2000 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 Thorsteins son and D L Cooper in Quantum Systems in Chemistry and Physics Volume 1 Basic prob lems and models systems eds A Hern ndez 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 Hartke and H J Werner J Chem Phys 111 4523 1999 vi CONTENTS vii Contents
136. HLO2_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 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 General threshold for generation of 2 external integrals If given this is used as a default for all D2EXT thresholds described be low Prescreening
137. I PROGRAM 139 Specifies an orbital configuration to be included in the reference function 1 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 18 3 4 Defining state numbers STATE nstate nroot l 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 18 3 5 Defining reference state numbers REFSTATE nstatrnrootr 1 nrootr 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 explicit
138. I SD ENERGY 76 22994348 ci occ 3 1 1 core 1 wf 10 1 cepa 1 TOTAL CEPA 1 ENERGY 76 23799334 Valence multireference CI for X and A states of H20 gthresh energy 1 d 8 r 0 957 angstrom theta 104 6 degree geomet ry 0 z matrix geometry input Hi Oy ee H2 0 r H1 theta hf wf 10 1 TOTAL SCF ENERGY 76 02680642 multi occ 4 1 2 closed 2 core 1 wf 9 2 1 wf 9 1 1 tran ly IMCSCF ENERGY 75 66755631 IMCSCF ENERGY 75 56605896 examples ci occ 4 1 2 closed 2 core 1 wf 9 2 1 save 7300 1 h2op_mrci_trans com TOTAL 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 Transition moment lt 1 3 LY 1 1 gt 0 962004881 a u 18 THE CI PROGRAM 149 BH singlet Sigma and Delta states r 2 1 geomet ry b h b r hE ocer wt 6 13 multi occ 3 1 1 core 1 wf 6 1 state 3 lquant 0 2 0 wf 6 4 lquant 2 tran 1z expec2 1z1z examples Sigma states nergies 25 20509620 24 94085861 bh_mrci_sigma_delta cor ci oce 3 1 1 core 1 wt 6 1 state 2 1 3 Delta states nergies 24 98625171 ci octa 3 1 1 coro 1j wf 06 l statos 1 2 Delta state xy component ci ocd 3 1 17 core Le wt 6 4 19 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 150 19 MULTIREFERENCE RAYLE
139. IGH SCHRODINGER 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 RS2C RS3 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 normally be used if third order is not required note that RS3C is not available 19 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 RS2 and RS3 programs use the same configuration 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 contracte
140. IR 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 i 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 following algorithm i e METHOD SRSTEEP In this algorithm s1max determines the frequency of the recalculation of the numerical hessian If the total step size of the last steps exceeds slmax the hessian will be recalculated other wise it will be updated By default simax is two times the maximum step size of the optimization step steplength see STEP section 32 2 9 If you are using METHOD QSD the SLMAX option is obsolete and the NUMHES command see above should be used instead 32 2 18 Optimizing energy variables VARIABLE VARIABLE name 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 1 holds last reference energy for state i ENERGD 1 holds last Davidson corrected energy for state
141. In Specifies a simple list of orthogonalization pairs Orbital i is made orthogonal to i 3 to i4 etc GROUP label i in 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 different groups can be used together or in combination with orbital numbers in ORTH or PAIRS ij in specifies the list of orbitals in the group Thus the combination GROUP A 1 2 GROUP B 3 4 ORTH A B will orthogonalize the pairs of orbitals 1 3 1 4 2 3 and 2 4 STRONG 29 THE VB PROGRAM CASVB 215 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 29 11 Wavefunction analysis 29 11 1 Spin correlation analysis NO SCORR With this option expectation values of the spin operators Sy 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 29 11 2 Printing weights of the valence bond structures For fu
142. M2 Mg ni 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 occ see CISD OCI and CCSD 18 2 4 Defining the orbitals ORB1IT 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 2 16 The default is the set of orbitals from the last SCF MCSCF or CI calculation 18 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 is the number of electrons sym is the number of the irreducible representation spin defines the spin symmetry spin 2S 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 18 THE CI PROGRAM 137
143. 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 17 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 27 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 matrix 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 refere
144. MOLPRO User s Manual Version 2002 6 H J Werner Institut fiir Theoretische Chemie Universit t Stuttgart Pfaffenwaldring 55 D 70569 Stuttgart Federal Republic of Germany P J Knowles School of Chemical Sciences University of Birmingham Edgbaston Birmingham B15 2TT United Kingdom February 2003 Copyright 2003 University of Birmingham 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 programs with
145. Multi step optimization A loop over two or more optimization steps may be specified using ALTERN WNiter 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 29 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 29 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 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 label The alternative format SYMELM label sign key 1 key 2
146. N ADIABATIC COUPLING MATRIX ELEMENTS 198 NR 198 201 207 29 1 Structure Of the input o 207 A a id e NE d 208 IN 208 29 3 Other wavefunction dIrectiVes e e ee 208 Daca wk ea a a a eke k 208 Seite ase an do Buea ol 208 in ware Eee eae Pied A oe eres S 209 pio ate we esi 209 coe oS eal ee aa a ne oO ee ee ee A 209 ee es go ae acy ie dee Bae Ant a de ae ee ee 210 bd Ro Oe Doe A Oe AG ede BE Ae es 210 29 7 2 Guess for structure coefficients 2 020000 210 29 7 3 Read orbitals or structure coefficients 4 210 CREA eras hg BAe Ds bead he de a at E 211 29 9 Optimization control oa ee 211 La eho hake Oe EAE Ge Ee EE a 211 29 9 2 Number of iterations 2 2 2 20 2 0 eee ee ee 211 arene eres 211 ite ye ela eae ae doe aaa ee area 211 E a ee ee 212 he ae A eae aed a de a Bo ee ae 212 oh tee a ole bd Chek 212 iwi Gull ERRE 212 29 10 2 The IRREPS keyword o o e e 212 29 10 3 The COEFFS keyword o o o e e 213 29 10 4 The TRANS keyword 0 o o 213 e e E dd me ee cee oe 213 29 10 6 The SYMPROJ keyword 2 0 0 2 0 000200000 213 ee et Sa me eB a dae 214 aseos 214 pel 214 e eE E dl es 214 Ea A eye Ree eee ew a 215 da a a aac Roe a A 215 ios ee 215 asus 215 29 12Controlling the amount of output 2 2 ee ee 216 29 US Service Mode aa ia ah ae ww ee AR a ee we ww ww 216
147. NG 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 All other matrices can be saved in triangular or square form to plain records using the TRIANG 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 PARI TY 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 35 MATRIX OPERATIONS 265 If NATURAL orbitals are generat
148. O 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 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 2 19 2 1 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 line 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
149. OCC CICL OSED CLOSED 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 records from which configurations can be selected as CISELECT only used if CISELCT is not present defines occupancy restrictions as RESTRICT only used if CIRESTRICT is not present number of occupied orbitals in each symmetry as CIOCC only used if CIOCC is not present number of closed shell orbitals in each symmetry as CICLOSED only used if CICLOSED is not present 6 VARIABLES 52 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 or DFTNAME ifun name of ifun th component of density functional DFTFAC ifun factor multiplying i fun th component of density functional DFTEXFAC 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 6 3 SET NELEC
150. PECIAL pecial Variables SPIN 10 n wn nN INDEX SPINBASTS 209 START ES 121043 LOH 216 E STATE 118 139 STATUS B4 STEP 130 234 String variables 43 STRONG 214 STRUC 210 Summary of keywords 14 SYM 87 112 SYMELM 212 symmetry WE card 10 additional MCSCE 127 SCF Integral program SYMPROJ 213 System variables 43 TABLE 53 Tables 53 TEST 130 THERMO 253 HR 108 110 HRESH 115 131 145 156 TRAN 128 TRAN2 128 TRANH 141 TRANS 143 213 TRNINT 131 TRUST 235 a H HF 83 HF SCF 83 Ks 90 KS SCF 90 NCOMPRESS 64 PDATE 234 VARIABLE variables Indexed Introduction 9 Setting Special String System VB 207 VB VBDUMP VBWEIGHTS 215 Vector operations S GGG G G 292 vibrational frequencies VORONOT wavefunction definition WEIGHT wF 10 83 118 135 T83 WRITE 216 XYZ 67 63 Z matrix ZMAT gal
151. Ps To Tp pa ar g 2cf A d 1 S 4zs 4cr p ads dire f Mx z 0 A x z a 3 x z 0 A x z 0 1 0 x z gt cp 3 3r 5 p 0 98 0 3271 0 7035 F x 2 p q C d e f 0 218 219 220 221 222 223 224 225 226 227 15 THE DENSITY FUNCTIONAL PROGRAM 107 q 0 003557 0 03229 0 007695 228 r 0 00625 0 02942 0 05153 229 t 0 00002354 0 002134 0 00003394 230 u 0 0001283 0 005452 0 001269 231 y 0 0003575 0 01578 0 001296 232 a 0 001867 0 005151 0 00305 233 and B is the correlation energy per particle of the Local Spin Density Approximation PW92C VS99 is aliased to VSXC 15 1 40 VWN Vosko Wilk Nusair 1980 local correlation energy S H Vosko L Wilk and M Nusair Can J Phys 58 1200 1980 K pe 234 where 3 1 6 a 236 p e A 0y 1 237 9 4 3 9 4 3 9 1 lt 1 2 4 A A 1 239 9 213 1j a a A q k1 11 m1 n1 240 q k2 l2 m2 n2 241 a q k3 13 m3 n3 242 x2 C O c d q A p c d A in lt a Ded arctan ae x p c 2 Q c d XG cd n H 2 ah arctan ged 243 Q c d V4d e 244 X i c d i ci d 245 k 0 0310907 0 01554535 1 6n 246 l 0 10498 0 325 0 0047584 247 m 3 72744 7 06042 1 13107 248 and n 12 9352 18 0578 13 0045 249 15 THE DENSITY FUNCTIONAL
152. RANS 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 sets 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 US 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 30 6 Examples 30 6 1 SO calculation for the S atom using the BP operator 30 xK kK SPIN ORBIT COUPLING 221 SO calculation for the S atom geometry s basis spd s vtz luse uncontracted basis INES OCOS la Es LO e Irhf for 3P state multi casscf WE 16 4 23wt 16 6 23 wt 16 7 2 wi 16 1 07 state 3 11D and 1S states wf 16 4 0 wf 16 6 0 wf 16 7 0 3P states ci wf 16 1 0 save 3010 1 state 3 save casscf wavefunctions using mrci noexc ci wf 16 4 0 save 3040 1 noexc
153. RE 13 DENSITY RECORD record TYPE 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 pseudocanonical 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 LOCAL orbitals Therefore in most cases the orbital type needs not to be specified state 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
154. Radial integration grid RADIAL o 15 4 3 Angular integration grid ANGULAR 15 4 4 Atom partitioning of integration grid VORONOTI 15 4 5 Grid caching SAVE NOSAVE 02 00002 eee 15 4 6 Grid symmetry SYM NOSYM o o o e 15 4 7 Grid printing PRINT 2 2 2 0 0 000000 02 o 16 ORBITAL LOCALIZATION 16 1 Defining the input orbitals ORBITAD o o 16 2 Saving the localized orbitals SAVE o o 16 3 Choosing the localization method METHOD o 16 4 Delocalization of orbitals DELOCAL 16 5 Localizing AOs LOCAO 16 6 Selecting the orbital space 16 6 1 Defining the occupied space OCC o o o o o ooo o 16 6 2 Defining the core orbitals CORE o 16 6 3 Defining groups of orbitals GROUP OFFDIAG 16 6 4 Localization between groups OFEDIAG o REA TR Rd We ac 16 7 1 No reordering NOORDER o e e e 16 7 2 Defining reference orbitals REFORB o o 16 7 3 Selecting the fock matrix FOCK 02 000004 16 7 4 Selecting a density matrix DENSITY 00 16 8 Localization thresholds THRESH 2 2 ee 16 9 Printing options PRINT 17 THE MCSCF PROGRAM MULTI 17 1 Structure of the input 17 2 Defining the orbital subspaces 2 2 e
155. SCF 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 TYPE orbtype STATE state SP IN spin MS2 ms2 SAVE 0rbsav 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 on the ORBITAL directive see section 17 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 2 16 orbl orb2 is a pair of orbit
156. SD using MRCI code Bring ci cepa 1 lcepa 1 using MRCI code h20 so mp2 Second order Moeller Plesset a mp3 Second and third order MP mp4 Second third and fourth order MP4 SDTQ mp4 notripl MP4 SDQ cisd CISD using special closed shell code ccsd t coupled cluster CCSD T aci t Iquadratic configuration interaction QCISD T Brueckner CCD T calculation becd t 21 6 2 Single reference correlation treatments for NoF2 FER N2E2 CIS G EOM ETRY C2h rnn 1 223 ang rnf 1 398 ang alpha 114 5 geometry N1 N2 N1 F1 N1 F2 N2 basis vtz method hf cisd do i 1 method Smethod i e i energy enddo table method e title Results for n2f2 define N N distance define N F distance Idefine FNN angle rnn rnf N2 alpha rnf N1 alpha F1 180 cc pVTZ basis set ecsd t qcisd t bccd t lall methods to use 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 examples n2f2_ccsd com basis Sbasis This calculation produces the following table Results for n2f2 METHOD CISD 308 BCCD T 308 CCSD T 308 QCISD T 308 basis VTZ E E ESCE 4634948 0 78283137 6251173 0 94445391 6257931 0 94512967 6274755 0 94681207 21 THE CLOSED SHELL CCSD PROGRAM 159 21 7 Saving the density matrix DM record ifil The effective first order densi
157. SPECIFICATION AND INTEGRATION 70 9 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 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 immediately follow each other If there is any other input between two sets of mass cards all mass definitions but not t ype from the first one are overwritten by the last one Note that specifying different isotope masses for symmetry related atoms lowers the symmetry of the system if the molecular centre of mass is 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 9 8 Dummy centres DUMMY atomI atom2 Sets nuclear charges on atoms 1 2 etc to zero for doing counterpoise calculations for example atoml atom2 can be atom numbers or tag names Note that the current setting of dum
158. 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 6 4 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 ECORR ENERGY ESCF define a macro HF Ido SCF calculation ESCF ENERGY store SCF energy in variable ESCF ULTI Ido CASSCF DEMC SECORR store CASSCF correlation energy in variable DEMC RCI Ido MRCI DECI SECORR store MRCI correlation energy in variable
159. TA 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 MRCI 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 6 6 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 Rcopies RtoS 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 denotes elements R 2 toR R but R 2 denotes a single element of R VARIABLES 47 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 element
160. TML version in the directory molpro2002 6 doc manual top level file is manual htm1 It is generally recom mended that the documentation is unpacked in the installation source tree so that the documen tation can be copied to its final destination as specified in the CONFIG file generated by the configure command To install the documentation and interactive basis set tool issue make install inthe doc directory Numerous example input files are included in the manual and can alternatively be seen in the directory molpro2002 6 examples B RECENT CHANGES 283 B Recent Changes B 1 New features of MOLPRO2002 6 Relative to version 2002 1 there are the following changes and additions B 2 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 6 and 6 3 The total charge of the molecule can be specified in a variable CHARGE or on the WF card see section 2 14 Improved numerical geometry optimziation using symmetrical displacement coordinates see sections and 32 Improved numerical frequency calculations using the symmetry AUTO option see section 33 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 vers
161. TRACE 35 6 Setting variables SETH o ee 35 7 Multiplying matrices MULT o o e e o 35 8 Transforming operators TRAN o o e e ee 35 9 Transforming density matrices into the MO basis DMO 35 10Diagonalizing a matrix DIAG 2 0 00000202 ee 35 11 Generating natural orbitals NATORB o e ee ee 35 12Forming an outer product of two vectors OPRD o 35 13 Forming a closed shell density matrix DENSA o 35 14Computing a fock matrix FOCK 0 0 00000000 eee eee xvii 246 247 248 248 249 250 252 253 253 254 256 256 256 256 257 257 257 258 258 258 258 258 258 259 259 259 CONTENTS 35 15Computing a coulomb operator COUL 35 16Computing an exchange operator EXCH 35 17Printing matrices PRINT 35 18Printing diagonal elements of a matrix PRID 35 19Printing orbitals PRIO 35 20 Assigning matrix elements to a variable ELEM 35 21 Reading a matrix from the input file READ 35 22 Writing a matrix to an ASCII file WRITE o o 35 23Examples 35 24Exercise SCF prograM A Installation of MOLPRO A l Obtaining the distribution materials A 2 Installation of pre built binaries A 3 Installation from source files A 3 1 Overvi
162. VDZ For larger basis sets like cc pWTZ we recommend to use a slightly larger value of 0 985 to ensure that enough atoms are included in each domain There are some other options which affect the domain selection CHGMIN value determines the minimum allowed Mulliken charge for an atom ex cept H in a domain i e atoms with a smaller absolute charge are not included even if the DOMSEL criterion is not fulfilled default 0 01 CHGMINH value as CHGMIN but used for H atoms default 0 03 CHGMAX value If Mulliken charge is larger than this value the atom is included inde pendent of any ranking MAXBP maxbp If maxbp 1 the atoms are ranked according to their contribution to the Boughton Pulay overlap If maxbp 0 default the atoms are ranked according 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 MULLIKEN option Determines method to determine atomic charges option 0 default squares of diagonal elements of S2C are used option 1 Mulliken gross charges It appears that the first choice works better with diffuse basis sets MERGEDOM number If this option is given all orbital domains containing number or more atoms in common are merged number 1 is treated as number 2 default 0 This is particularly useful for geometry optimizations of conjugated or aromatic systems like e g benzene In the latter case MERGEDOM 1 causes the g
163. Y DMZ Overlap Kinetic energy Potential Delta function v4 Darwin term of relativistic correction Mass veclocity term of relativistic correction Total relativistic correction Dipole moments XX YY ZZ XY XZ XY Second moments XXX XXY XXZ XYY XYZ XZZ YYY YYZ YZZ ZZZ Third moments OMXX QMYY QMZZ EFX EFY EFZ FGXX FGYY FGZZ D DX D DY D DZ LSX LSY LSZ ine EX LY L EXTER LYLY LED QMXY QMXZ QMXY Quadrupole moments Electric field FGXY FGXZ FGXY Electric field gradients Velocity One electron spin orbit Total angular momentum squared L Electronic angular momentum LXLY LXLZ LYLZ Two electron angular momentum 6 VARIABLES 50 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 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 sym
164. _DTRAF default THRO1_LMP2 THRO1 THR_LMP 2 THRPROD_DTRAF default THRO2_LMP2 THRO2 THR_LMP2 THRINT_DTRAF default THRAO_ATTEN THRATTEN THREST_LMP THR_DKEXT THRKEXT THREST_DKEXT THRAO_DKEXT THR_DKEXT THREST THRINT_DKEXT THRSO_DKEXT THR_DKEXT THRINT THRPROD_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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 63 8 1 Example for integral direct calculations memory 2 m method hf mp2 ccsd qci bccd multi mrci acpf rs3 some methods basis vdz basis geometry o0 h1l o0 r h2 0 r hl theta geometry gdirect direct option r 1 ang theta 104 bond length and angle do i 1 method loop over met kgamples Smethod i Irun method i h2o_direct com e 1 energy save results in variables dip 1 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 76 23369819 0 76875001 ACPF 76 23820180 0 76872802 RS3 76 23549448 0 75869972 9 GEOMETRY SPECIFICATION AND INTEGRATION 64 9 GEOMETRY SPECIFICATION AND INTEGRATION Before sta
165. _save reset HF HF distance to current value text HF1 CP MONOMER dummy f2 h2 second hf is now dummy hf scf for first monomer mp2 mp2 for first monomer ehfl energy save mp2 energy in variable forces compute mp2 gradient for first monomer add 1 subtract from previous gradient text HF2 CP MONOMER dummy f1 h1 first hf is now dummy hf Iscf for second monomer mp2 mp2 for second monomer ehf2 energy save mp2 energy in variable forces compute mp2 gradient for second monomer add 1 subtract from previous gradient dummy lreset dummies text DIMER CALCULATION Lis Iscf for dimer mp2 mp2 for dimer edimer energy Foarrcas save mp2 energy in variable lLoomnite mn aradient for dimor examples hfdimer_cpcoptl com 32 GEOMETRY OPTIMIZATION 241 In the next example the monomer structures are kept fixed and the interaction energy is opti mized 32 GEOMETRY OPTIMIZATION 242 HF dimer mp2 CP optimization with fixed monomers basis avdz maxit 20 Imax number of iterations text OPTIMIZED VALUES OF GEOMETRY VARIABLES RFF 5 31431160 Rl 1 75768738 R2 1 75298524 HETA1 7 03780227 HETA2 111 25930975 geomet ry x noorient noorient must be specified since gradients are added fl 2 EL BEF hl f1 1 75768738 f2 thetal h2 2 1 75298524 fl theta2 hl 180 text CALCULATION AT LARGE SEPARATION rff_save rff save current rff distance rff 1000 dim
166. a ee 143 18 4 14 Transition moment calculations 00 0000 08 143 E r aes a Ag Wee ees eas ae 143 18 4 16 Natural orbitals 2 ee 143 ooh A hey ee a se ee Go es R 144 db teh a BS oe al hh a a ae 145 18 5 Miscellaneous thresholds 2 a aa 145 A NN 146 wot Gah a a aca ra ad aa Oe aS E 148 19 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY I50 A ti ire AS RAE we 150 es A el be ee ee 151 ee ERAS ee ee 151 a aa a 152 AAA as Ok Ge a ee ee 152 Dok a Gace Wark Mo aa ae daw doe ad wae A 152 19 7 Options for CASPT2 and CASPT3 0 02002000004 152 20 M LLER PLESSET PERTURBATION THEORY 155 20 1 Expectation values for MP2 o e 155 20 2 Density fitting MP2 DF MP2 RI MP2 000 155 156 A A eat ed et 2 em at a 156 21 2 Quadratic configuration interaction QC o o 156 rere 156 oh ee Be a le Bes O Ca 157 ee eee 157 21 5 The DTTS directive o ea a cae aR Oa a 157 AAA AAA A le ee am a es ee 158 veer eer 158 A 158 e Bt O e de ake E we 159 21 8 Natural Orbitals e 159 A Geek E E ce eek ae 159 21 9 1 Parameters for EOM CCSD EOMPAR o o 160 21 9 2 Print options for EOM CCSD EOMPRINTD 160 22 OPEN SHELL COUPLED CLUSTER THEORIES 162 23 LOCAL CORRELATION TREATMENTS 163 E SI ee ee hee ee 163 23 2 Getting started ss foe we eb eS ewe ew ea a a 163 239 DOME IME oe ew Boke a
167. a ee eee Bea le ee ep ea aed 164 23 31 AWAY Sol i edo os A ed e e dk EL 164 23 3 2 Linear scaling MP2 o ooo e 166 23 4 Density fitting LMP2 DF LMP2 RI LMP2 167 23 4 1 Intermolecular interactions eae 167 23 4 2 Gradients and frequency calculations 168 CONTENTS xiv 234 3 Basis SO S o scc 04 bas kk E ee A EE 169 23 5 Furthercommands s cc ee i ee pea ce E a E E E e EO E 170 ea bah ae hee ee aa 170 23 5 2 Restarting a calculation START o o 170 breast oe O ari iaa a 170 A 170 Sell 171 A aca ita a AA A e e E 172 23 7 Additional options available on the ATTENUATE card 177 183 A A ea ee er de eee te a hs eS 183 LA lh de Rd A A a ES a dw 183 24 3 Frozen core orbitals 2 a a o 183 24 4 Defining the state symmetry o o o 183 A tee ee Be Poe Ge Pe ee ee Aen 184 ra bakhd ban Bo e Pe acd Bae hae de 184 25 PROPERTIES AND EXPECTATION VALUES 185 a A A A e ae dh Mee Y 185 25 1 1 Calling the property program PROPERTY 185 25 1 2 Expectation values DENSITY o o 185 25 1 3 Orbital analysis ORBITAL o ee eee 185 aaa 185 SP oe a a Moe aaa So ae MIA a de cd 186 BOR He um ee eign cae ett Ee ee BRE ee in a ee eo tes 186 fe nee ao A a oo 187 25 2 1 Calling the DMA program DMA o 187 25 2 2 Specifying the density
168. actor A value lower than 1 means the damping function will be squeezed while a value higher than 1 will cause it to be stretched A value of O disables the scaling as well as 1 A value of 0 9 is recommended when using a Taylor expansion Default 0 Prints a list of uncoupled pair energies before the MP2 iterations Can be used as a convenient diagnostic when getting totally implau sible correlation energies or no convergence or UNREASONABLE NORM messages 23 LOCAL CORRELATION TREATMENTS 181 Table 9 Summary of Local multp options and their default values Parameter Alias Default value Meaning General Parameters LOCAL 4 determines which program to use SAVE SAVDOM 0 specifies record for saving domain info START RESTDOM 0 specifies record for reading domain info PIPEK LOCORB 0 activates or deactivates PM localization SAVORB SAVLOC 0 specifies record for saving local orbitals DOMONLY 0 if set only domains are made if 2 only orbital domains are made Parameters to select weak and distant pairs WEAKPAIR WEAKP 1 criterion for selection of weak pairs DISTPAIR DISTP 0 8 criterion for selection of distant pairs VERYDIST VERYD 0 15 criterion for selection of very distant pairs Parameters to define domains DOMSEL CHGFRAC 0 98 selection criterion for orbital domains DELCOR IDLCOR 2 delete projected core AOs up to certain shell DELBAS IBASO 0 determines how to remove redundancies Parameters
169. ad 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 Cl optimization icstep microiteration increment between Cl optimizations 17 THE MCSCF PROGRAM MULTI 131 INTOP T maxito maxitc maxrep nitrep iuprod Special parameters for internal optimization scheme For experts only NONLINEAR itmaxr ipri drmax drdamp gfak1 gfak2 gfak3 irdamp ntexp 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 THRDIV thrdiv THRDOU 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 1 d 7 DIIS disvar augvarmaxdis maxaug idsci igwgt igvec idstrt idstep Special parameters for DIIS conver
170. al 97 97 98 E a A oe ee AA A ee 98 E Er e ee 99 pao 99 it a 99 A E oe al Sh ae ta Se ine 8 WR Mi 100 15 1 26 PW91 PW91 PW9IX PW9IC 2 ee 100 15 1 27 PW91C Perdew Wang 1991 GGA Correlation Functional 100 15 1 28 PW91X Perdew Wang 1991 GGA Exchange Functional 101 ae MOL 15 1 30 S Slater Dirac Exchange Energy o o 102 SALSA a do e o Oe A ee ee 102 15 L32 THO Pe e o ss ee ee e a 103 ID TSS THO ace a a eM we a a a a ee we ae 103 VST S4 THAT Pik al oe eh oe ad Ee he ee eae ea 104 15 1 35 THGEF L pee ee eA ba trie ci d ewe eee ee 104 I5 L36 THGEC oe oh edb SE io g he p Ta ai ioa oe eo bh 105 15 137 THGECEO og ks ae ea ea ha ee RAO ee Oe eee we 105 15 1 36 THGECO 00 coca ae gained PES be Made oe eee eae 106 E 106 oe 107 DAO aA eee ee a ee ee E a 108 obs Gk a Ma eet le es aden ws oe dogma a ae A aU ON ak ae 108 15 2 1 Density source DENSITY ODENSTTY 108 15 2 2 Thresholds THB ee 108 15 2 3 Exact exchange computation EXCHANGE 109 15 2 4 Exchange correlation potential POTENTIAL 109 15 2 5 Grid blocking factor BLOCK o o e 109 15 2 6 Dump integrand values DUMP o 109 iit A aid ok AA O 2s 109 15 4 Numerical integration grid control GRID o o 110 15 4 1 Target quadrature accuracy THR o o 110 CONTENTS 15 4 2
171. als 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 7367 1988 Below we present an example for the first two excited states of H2S which have B and A2 symmetry in Cz and A symmetry in Cs We first perform a reference calculation in Cz sym metry and then determine the diabatic orbitals for displaced geometries in Cs symmetry Each subse
172. als 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 11 2 Explicit input for ECPs For each of the pseudopotentials the following information has to be provided e acard of the form ECP atom Ncore lmax des where More 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 J is the corresponding number of terms in the SO part Z pen Ncore Umax 1 I max EN Vas Vinae ED VI Vins PH 2 AMAS 1 0 l 1 the semi local terms with angular momentum projectors P are supplemented by a local term for l lmax e a number of cards specifying V the first giving the expansion length n in max max Nmax mj 2 yr Vimax e de cir 5 en j l and the following n ones giving the parameters in the form max m1 V1 C13M2 V2 C25 a number of cards specifying the scalar relativistic semi local terms in the order l 0 1 lmax 1 For each of these terms a card with the expansion length n in at l yr l m 2 Y r Vi Vina cyr i et j l has to be given and immediat
173. an 90 Those parts which are machine depen dent are maintained through the use of a supplied preprocessor which allows easy intercon version 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 lil systems A fuller description of the hardware and operating systems of these machines can be found at http www molpro net machines html 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 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 Future enhancements presently under development include Local coupled cluster theory LCCSD as described in J Chem Phys 104 1996 6286 and J Chem Phys 114 661 2001 with perturbative treatment of triple excitations as described in Chem Phys Letters 318 370 2000 and J Chem Phys 113 9986 2000 e Enhancements to the efficiency of the DFT integration Analytical energy gradients for CCSD LCCSD and CAS PT2 Analytical
174. an be specified on the OP TG command OPTG keyl value key2 value where key can be MAXIT to set the maximum number of optimization cycles The default is 50 GRAD sets the required accuracy of the optimized gradient The default is 3 1074 STEP to set the convergence threshold for the geometry optimization step if value gt 1 the threshold is set to 10 The default is 3 1074 ENERGY sets the required accuracy of the optimized energy The default is 1 1076 GAUSSIAN Use Gaussian convergency criteria SRMS sets for Gaussian convergency criterion the required accuracy of the RMS of the optimization step The default is 0 0012 GRMS sets for Gaussian convergency criterion the required accuracy of the RMS of the gradient The default is 3 1074 BAKER Use Baker s convergency criteria see J Baker J Comp Chem 14 1085 1993 NUMERICAL Force the use of numerical gradients The standard MOLPRO convergency criterion requires the maximum component of the gradient to be less then 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 a
175. and FCI 24 1 Defining the orbitals ORB1IT 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 24 2 Occupied orbitals OCC N1 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 24 3 Frozen core orbitals CORE A M2 Ng35 ni 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 24 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 24 THE FULL CI PROGRAM 184 24 5 Printing options PRINT code value 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
176. 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 29 7 1 Orbital guess ORB i 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 17 5 4 Use of one of the three latter keywords is recommended 29 7 2 Guess for structure coefficients STRUC C1 C2 CNVB Specifies a starting guess for the NVB 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 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 29 7 3 Read orbitals or structure coefficients The READ keyword can take one of the following forms READ ORB iorb1 TO iorb2 AS jorb1 TOjorb2 FROM record READ STRUC istrucl TO istruc2
177. ansform density to natural MO could also be done usi print diagonal elements occupation numbers transform D_ao to canonical MO basis Same as above sim imultiply d_can by 1 diagonalizes density D_can transforms canonical orbitals to natural orbitals prints new natural orbitals make natural orbitals using MCSCF density D_ao directly prints new natural orbitals should be the same as abov form mcscf scf difference density make natural orbitals for difference density write difference density to ASCII file denfile Istore natural orbitals for difference density in dump r 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 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 35 MATRIX OPERATIONS 270 memory 2 m R 0 96488518 ANG THETA 101 90140469 geomet ry H1 Os AL Eg H2 0 R H1 THETA hf wf 10 1 field 0 05 define field strength matrop load h0 h0 load one electron hamiltonian load xx oper XX load second moments load yy oper yy load zz oper Zz 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 h0 hf Ido scf with modified
178. ar relativistic effects 1 Use the Douglas Kroll relativistic one electron integrals 2 Compute a perturbational correction using the Cowan Griffin operator see section 4 13 3 Use relativistic effective core potentials see section 11 For all electron calculations the prefered way is to use the Douglas Kroll Hamiltonian It is simply activated by setting DKROLL 1 somewhere in the input before the first energy calculation 13 0 1 Example for computing relativistic corrections ar2 geometry arl ar2 arl1 r r 2 5 ang bt expec rel darwin massv e_nrel energy show massv darwin erel dkroll 1 ht e_dk energy show massv darwin erel show e_dk e_nrel geometry definition bond distance non relativisitic scf calculation compute relativistic correction using Cowan Griffin operator save non relativistic energy in variable enrel Ishow individual contribution and their sum examples luse douglas kroll one electron integrals ar2_rel com lrelativistic scf calculation save relativistic scf energy in variable e_dk show mass velocity and darwin contributions and their sum Ishow relativistic correction using Douglas Kroll 14 THE SCF PROGRAM 83 14 THE SCF 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 calls the spin unrestricted Hartree Fock program In contrast to ol
179. are printed 35 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 35 MATRIX OPERATIONS 268 35 21 Reading a matrix from the input file READ READ name type subtype symmetry values Reads a square matrix symmetry 1 from input The values can be in free format but their total number must be correct Comment lines starting with or are skipped The matrix can be read from another ASCII file by including this into the input using the INCLUDE command see section 2 2 type is a string which can be used to assign a matrix type If appropriate this 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 35 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 p
180. art bond angle examples h20_optmp2_runcesdt co loptimize energy at mp2 level Ido single point MP4 at optimized mp2 geometry Ido single point ccsd t calculation At the end of the output the following summary of results is printed RESULTS FOR BASIS VTZ METHOD STATE S MP2 D 1 0 0 76 MP2 1 0 0 76 MP3 1 0 0 76 MP 4 SDQ Ll 0 0 76 MP 4 SDTO dd 0 0 16 MP2 1 0 0 ANO CCSD 1 0 0 76 CCSD T 1 0 0 76 CCSD T 1 0 0 76 The MP2 energy appears repeatedl calculations 3 8 4 DFT frequency calculation ENERGY 31865774 31865774 32273584 32484084 33305159 31865774 32454712 33221602 33240959 y since it is computed in the MP2 MP4 and CCSD T The following input performs a DFT B3LYP geometry optimization and frequency calculation for water 3 INTRODUCTORY EXAMPLES 27 geometry 0 Z matrix for water H1 O R H2 0 R H1 THETA R 0 96 Ang Istart bond distance Theta 104 start bond angle examples basis 6 31g functional b3lyp freqdft The results are RESULTS FOR BASIS 6 31G ETHOD STATE S FREQ KS B3LYP Tradl 0 0 LO ZPE Lal 0 0 0 E ZPE dol 0 0 STG HTOTAL 1 1 0 0 76 GTOTAL TI 0 0 76 Pople basis set ldefine fucntional run frequency calculation h2o_freqdft com optional b3lyp is default ENERGY DIPX DIPY DIPZ 38101813 0 0 0 0 0 80039692 02697203 0 0 0 0 0 80039692 35404610 0 0 0 0 0 80039692
181. ases 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 is 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 5 2 DELETE DELETE filel file2 Deletes the specified files file refers to the logical MOLPRO file numbers as specified on the FILE card 5 3 ERASE ERASE filel file2 Erases the specified files file refers to the logical MOLPRO file numbers as specified on the FILE card 5 FILE HANDLING 41 5 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 fil 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 nam ifill nam2 ifil2 If nam2 0 nam2 nam 1 If nam 0 all records are copied fro
182. ates distant pairs i e no pairs are treated approximately Default is O MULTP card 8 VERYDIST distance If all atoms of orbital domain i are separated by at least distance a u from any atom of orbital domain j pair ij is neglected Set ting distance 0 default eliminates very distant pairs i e no pairs are neglected Reasonable values for distance would be 12 15 a u Default is O MULTP card 15 Parameters to define domains DOMSE L value Threshold for selecting the atoms contributing to orbital domains us ing the method of Boughton and Pulay The default is value 0 98 value 1 0 would include all atoms into each orbital domain The cri terion is somewhat basis dependent The larger the basis the fewer functions will be selected with a given threshold The default value usually works well for double zeta basis sets For larger basis sets e g cc pVTZ it is recommended to use value 0 985 In most cases the domain selection is uncritical for saturated molecules However for delocalized systems it is recommended to check the printed or bital domains In cases of doubt compare the domains you get with a smaller basis e g cc pVDZ See also the MAXANG option below DELCOR nshell 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
183. ation 256 orbital spaces 12 Orbitals 12 orbitals closed CL 136 MCSCE 117 closed shell 12 core 12 CL 135 FCI 183 MCSCF frozen MCSCF internal 12 EL 135 occupied 12 CL 135 FCI 183 MCSCE 117 ORBPERM 21T ORBPRINT 88 129 ORBREL ORTH 88 214 258 ORTHCON 214 PAIR 140 PAIRS 140 214 Parallel 3 PARAM 145 Plotting 68 POP 189 population analysis 189 POTENTIAL 109 PRINT 112 115 129 146 184 186 216 PROC B3 Procedures 17 33 program structure 1 PROJECT 141 257 properties 185 CI 143 MCSCE 128 PROPERTY 185 pseudopotential P SPACE 120 142 PUNCH MI 291 PUT 68 QCI 156 oct 135 QUAD 190 QUAD 190 quadrupole field RADIAL 110 RADIUS 188 reaction path 232 237 READ 210 READPUN 7 records 8 REF 137 References REF STATE 139 REL 192 Relativistic corrections 192 RESTART 9 29 RESTRICT 119 138 RHF RHF SCF RI LMP2 167 RI MP2 155 RKS 90 RKS SCF 90 ROOT 233 ROTATE 86 122 258 RS2 150 RS2 150 RS2C 150 RS3 150 RS3 150 Running MOLPRO SADDLE 211 SAMC 224 save BA 112 113 122 42 209 233 SCALE 223 SCF 83 SCHMIDT 258 SCORR 215 SELECT 119 137 SERVICE 216 SET 42 HIFT 87 141 HOW 52 sorted integrals S
184. ation but then not in subsequent 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 2 15 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 m1 m2 mMg CORE C0 C02 COg CLOSED cl clo cls where m is the number of occupied orbitals including core and closed co the number of core orbitals and cl is the number of closed shell orbitals including the core orbitals in the 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 Note that the OCC CORE 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 orbitals which are occupied in any of the reference CSFs In the MCSCF core orbitals are doubly occupied in all CSFs and frozen not optimized while closed denotes all doubly occupied orbitals frozen pl
185. ation count is smaller than nitcl only the closed shell part of the Fock matrix is used default nitcl 0 14 THE SCF PROGRAM 88 14 7 2 Maximum number of iterations MAXIT maxit sets the maximum number of iterations to maxit The default is maxit 30 14 7 3 Convergence threshold ACCU accu The convergence threshold is set to 10 accu This applies to the square sum of the density matrix element changes The default is accu 10 14 7 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 14 7 5 Interpolation IPOL iptyp 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 14 7 6 Reorthonormalization of the orbitals ORTH nitort In the RHF program the orbitals are reorthonormalized after every nitort iterations The default is
186. ation performs state averaged CASSCF and subsequent MRCI calculations for the ground and first excited state of the OH radical A full valence active space is used FER OH geomet ry 0 H 0 1 83 set symmetry 2 3 1 spin 1 2 Pix 2Piy runmrci SA CASSCF and MRCI The following table is printed at the end of the output RESULTS METHOD STATE S ENERGY DIPX DIPY CASSCF 1 32 0 5 75 41331789 0 0 0 0 CASSCF S 0 5 75 41331789 0 0 0 0 CASSCF Ler 0 05 75 24125256 0 0 0 0 RCI 1 2 0 5 75 55518444 0 0 0 0 RCI D 1 2 0 5 75 56014871 0 0 0 0 RCI P 1 2 DUES 75 55853208 0 0 0 0 RCI 1 3 0 5 75 55518444 0 0 0 0 RCI D Lie 0 5 75 56014871 0 0 0 0 RCI P Le 0 5 75 55853208 0 0 0 0 RCI Lgr 0 5 75 39442202 0 0 0 0 RCI D ToT 0 5 75 40040680 0 0 0 0 RCI P 1 1 0 5 75 39846312 0 0 0 0 This calculation performs MRCI calculations for both I enough to recognize that they are degenerate However eliminate this drawback and 2 Sigma states examples oh_runmrcil com DIPZ 67158730 67158730 69975340 66457191 66457191 66457191 66457191 66457191 66457191 70484623 70484623 70484623 QO OG OG OOG GOGO GOG O and IL The procedure is not clever one can easily modify the input to 3 INTRODUCTORY EXAMPLES 25 RO OT geomet ry 0 H 0 1 83 set symmetry 2 3 1 spin 1 2Pix 2Piy and 2Sigma states specifying spin 1 is optional runcas SA CASSCF for all three s
187. ations of basis functions INDIVIDUAL INDIVIDUAL 25 3 4 Example h20 population analysis geometry o0 h1 0 r h2 0 r h1 theta Z matrix geometry input r 1 ang bond length theta 104 bond angle basis 6 311g misa hf Ido scf calculation P e a h20_pop com pop Mulliken population analysis using mcsc ensity individual give occupations of individual basis functions If specified the Mulliken populations of each individual basis function are printed 25 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 25 PROPERTIES AND EXPECTATION VALUES 190 25 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 25 4 2 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 25
188. ative this orbital will be occupied with negative spin only allowed in UHF 14 2 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 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 2 7 The default for file is 2 143 Starting orbitals The START directive can be used to specify the initial orbitals 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 All files are searched Second Try to find orbitals from a previous SCF or MCSCF calculation All files are searched 14 THE SCF PROGRAM 85 Third If no orbitals are found the starting orbitals are generated usin
189. ator is diagonalized or of localizing the orbitals 17 5 1 Defining the starting guess START record options record dump record containing starting orbitals As usual record has the form irec ifil where irec is the record number e g 2140 and ifil 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 17 THE MCSCF PROGRAM MULTI 122 that frozen core orbitals should always be taken from an SCF or MCSCF calculation at the present geometry and must be specified separately on the CORE 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 CORE card in a previous MCSCF calculation
190. atrix 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 otherwise 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 analytical sec ond derivatives are only possible for closed shell Hartree Fock HF and MCSCF wavefunctions without symmetry It is not yet possible to calculate IR intensities analytically Note that due to technical reasons the analytical MCSCFE second derivatives have to be computed in the MCSCF program us inge g multi cpmcscf hess see MULTI before they can be used in FREQUENCIES If analytical 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 calcula
191. aveco DENSIT 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 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 matrix and the natural orbitals are saved See also DM or NATORB cards 18 THE CI PROGRAM 143 18 4 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 18 4 13 One electron properties EXPEC
192. avefunction 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 2 7 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 5000 MCSCF gradient information 5100 CP MCSCF gradient information 5200 MP2 gradient information 5300 Frequencies 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 re
193. b ltitle for table sort table examples h20_pes_cesdt com The next example shows how to loop over many methods h20 benchmark method hf fci ci cepa 0 cepa 1 cepa 2 cepa 3 mp2 mp3 mp4 X qcei ccsd bccd qci t ccsd t bccd t casscf mrci acpf basis dz geometry o0 h1l o0 r h2 0 r hl theta r 1 ang theta 104 do i 1 method method 1 e 1 energy enddo esc e 1 efci e 2 table method e ser EGI Title for table title Results for H20 basis Sbasis R r Ang F ELSE F END 4 7 Branching Double zeta basis set Z matrix for geometry Geometry parameters Loop over all requested methods call program save energy for this method examples h20_manymethods com Iscf energy lfci energy print a table with results Theta Stheta degr IF blocks and IF ELSEIF blocks can be constructed as in FORTRAN 4 7 1 IF statements IF blocks have the same form as in Fortran 4 PROGRAM CONTROL 32 IF logical expression THEN Statements ENDIF If only one statement is needed the one line form IF logical expression statement can be used except if statement is a procedure name ELSE and 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
194. 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 DISKSIZE Max disk size in MB allowed in semi direct SCF calculations 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 57 THRDISK MEM_DFOCK SWAP_DFOCK Only integrals larger than this threshold are stored on disk in semi direct SCF calculations Integral buffer size for semi direct SCF If this is larger than DISKSIZE the calculation is done semi direct incore no in tegrals are written to disk If MEM_DFOCK is negative de fault a default buffer size is used If MEM_DFOCK is zero all available memory is used Enables or disables label swapping in fock matrix calculation test purpose only General options for direct integral transformation DTRAF PAGE_DTRAF SCREEN_DTRAF MAXSHLO1_DTRAF MINSHLO1_DTRAF MI MAXSHLO2_DTRAF MAXCEN_DTRAF NS
195. bed in section A 3 6 may also be specified 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 installation 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 op tions performs housekeeping functions and starts the one or more processes that do computation 2 The molpro exe executable which is the main back end For par allel 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 The molpro rc file which contains default options for molpro cf section A 3 6 A INSTALLATION OF MOLPRO 273 4 The molproi rc file which contains MOLPRO script procedures 5 Machine ready basis set and other utility libraries Validation A suite of self checking test jobs is run to provide assurance that 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
196. bital 2130 2 move 1 1 2 1 1 1 move 3 1 0 4 4 1 orbital 2100 2 move 1 1 0 4 save 2131 2 chtsroce 4 1 bestart 2131 2 orbital 2132 2 merge orbital 213 hs move 3 13 Os 1 ete gt move 1 1 0 save 2141 2 0 multisoces b 1 ls7start 2141 2 34 13 2 NO 259 luse C2v symmetry rhf for f atom use C2v symmetry Iscf for h2 Imcscf for h2 linear geometry for F H2 examples h2f_merge com rhf orbitals for F atom move orbitals 1 1 2 1 move all remaining starting at 4 1 hf orbitals for H2 imove these to fr positions save merged orbitals rhf 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 Imcscf orbitals for H2 imove 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 C gt as the molecular calculation 34 ORBITAL MERGING NO merge r 2 1 geometry x y n ERE occ IL Ls wf 7 4 3 orbital 2110 2 geomet ry x y 0o rhe oce 3 1 17 wf 8 4 2 orbital 2120 2 geometry n 0 n r ERGE ORBITAL 2110 2 OVE 1 1 1 1 OVE 2 1 2 1 3 1 OVE peel oe ly eed OVE 1 2 1 2 OVE lasy Lvs OVE 4 1 7 1 OVE 2 2 3 2 OVE 2 pp oe OVE 1 4 OR
197. blocks of the exchange operators is suppressed If a positive value is given the multipole correction of the ionic exchange blocks is also suppressed A zero value disables the suppression of multipole corrections Default 0 Options for least squares fit generation of interaction coefficients 23 LOCAL CORRELATION TREATMENTS 178 F ITMLTP option Specifies how the coefficients for the multipole expansion of long range integrals are calculated option 0 Taylor expansion option 1 Least squares fit Default 1 F 1DGRID value Sets the number of quadrature points used to generate integrals that arise in the one dimensional fit i e for the monopolar multipole ex pansion Default 50 F 2DGRIDR value Sets the number of quadrature points used along the r radial coordi nate when generating integrals for the two dimensional fit i e for the bipolar multipole expansion Default 50 F 2DGRIDP value Sets the number of quadrature points used along the p angular coor dinate for the two dimensional integrals Default 20 F1DBORDER value If greater than 0 sets the upper bound of integration in bohr and selects the Gauss Legendre quadrature for the one dimensional inte grals If 0 selects Gauss Laguerre quadrature Default 0 F 2DBORDER value The same for the two dimensional integrals F 1DGAMMA y Sets the negative exponent of the weight function for the one dimensional fit Smaller values are better for more diffuse densities Shou
198. bove Ido scf with modified h0 same as before with separate field commands t Xx 0 5 f ry VVr 0 5 f Ido scf with modified h0 lremove field Iscf without field cond example shows how to compute dipole moments and polarizabilities using finite 25 PROPERTIES AND EXPECTATION VALUES 192 H20 finite field calculations r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix input HL yO ee 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 Smethod m calculate energy e k energy save energy examples enddo h20_field com enddo k 0 n method do m 1 method k k 1 energ m e k dipmz m e k n e 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 Ipolarizability enddo table method energ dipmz dpolz title results for H20 r R theta theta basis Sbasis 25 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
199. 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 lt 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 6 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 6 2 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 section 6 7 1 which can be used for further processin
200. cal 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 Schiitz G Rauhut and H J Werner J Phys Chem 102 5197 1998 The energy partitioning algorithm is activated either by supplying the global ENEPART card ENEPART epart iepart 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 SOD H11 H12 2 02 H21 H22 energy partitioning relative to centre groups intramolecular correlation 43752663 xchange dispersion 00000037 dispersion energy A 00022425 ionic contributions 00007637 The centre groups correspond to the individual monomers determined for epart 3 In the present example t
201. cent MCSCF calculation or otherwise no orbitals are frozen If the CORE card is given as CORE record file then the orbitals corresponding to atomic inner shells are taken i e 1s for Li Ne 1s2s2p for Na Ar etc A CORE card without any specification resets the number of frozen core orbitals to zero 17 2 3 Closed shell orbitals CLOSED A1 M2 Mg ni 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 CSFs In contrast to the core orbitals see CORE these orbitals are fully optimized Please note that the program output sometimes says CORE when it means CLOSED and FROZEN when it means CORE historical reasons 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 CORE 17 THE MCSCF PROGRAM MULTI 118 17 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 this card has no effect 17 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
202. cf vector for 2Py state ci wf 7 5 1 noexc save 5200 2 save casscf vector for 2Pz state ci wf 7 2 1 save 6000 2 mrci for 2Px state ci wf 7 3 1 save 6100 2 mrci for 2Py state ci wf 7 5 1 save 6200 2 mrci for 2Pz state m LER OCG 1 2 25 7 2 casscf with larger active space Why Peed Wey lp op LEE Ed average 2P states ci wf 7 2 1 noexc save 5010 2 ci wf 7 3 1l noexc save 5110 2 ci wf 7 5 l noexc save 5210 2 ci wf 7 2 1 save 6010 2 ci wf 7 3 1 save 6110 2 ci wf 7 5 1 save 6210 2 casscf esy ly Lo Lye t lsmat ecp 5000 2 5100 2 5200 2 Ido spin orbit calculations casscf Cet pep en ye lsmat ecp 5010 2 5110 2 5210 2 tex ci tex DEE Gt Hf Hf ci MECT y OCC by LLnL lsmat ecp 6000 2 6100 2 6200 2 MECL OCC rly 2273 2 lsmat ecp 6010 2 6110 2 6210 2 tex El tex DA E Gh I i ct 4 a F 222 lokal term s terme 0 p terms with wei d terms with wei f terms with wei ECP SO for examples i_ecp com ECP SO for p ter ECP SO for d ter f ter 31 ENERGY GRADIENTS 223 31 ENERGY GRADIENTS 31 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 MCSCF In the MCSCF case the wavefunction must either be fully optimized
203. cient IPROCS 3 Only singles with one or two holes in the closed shells are in ternally contracted in RS2 using a projection operator TPROCI 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 THINT 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 IDF IA 1 corresponds to Dyall s CAS A met
204. cies as in transition state optimiza tions 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 displacement_type can be one of the following affects only numerical gradients 32 GEOMETRY OPTIMIZATION 231 SYM Use symmetric displacement coordinates default This is the only recom mended 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 recommended If option is set to NOROT the cartesian coordinates are not transformed to minimze rotations 32 2 2 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 32 2 3 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 32 2 4 Selecting the optimization method METHOD METHOD key key defines the optimization method For minimization the following options are valid fo
205. 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 4 PROGRAM CONTROL 37 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 QM 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 4 13 1 Example for computing expectation values The following job computes dipole and quadrupole moments for H20 h20 properties geometry o0 h1l o0 r h2 0 r h1 theta Z matrix geo
206. cord 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 using 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 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 2 GENERAL PROGRAM STRUCTURE 9 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 2
207. ction 32 2 17 SRSTEEP Old Version of OSDPATH In transition state searches and reaction path following Z Matrix coordinates should be used Although it is also possible to use cartesian or BMAT coordinates the computational effort is usually much larger since the hessian matrix has to be calculated numerically in all 3 x N possi ble degrees of freedom 32 2 5 Approximating hessian matrix elements 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 The Model Hessian is parameterized for the elements up to the third row and is used by default unless the molecule contains atoms from higher rows Alternatively the model Hessian of Schlegel can be used HESSIAN key value paraml param2 where key can be MODEL Use Lindh s Model Hessian in optimization default 32 GEOMETRY OPTIMIZATION 233 MODEL SCHLEGEL Use Schlegel 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 For minimizations the Model Hessian provides very good approximations the Hessian matrix improving convergence rapidly so it was chosen as default At present it is implemented for the first
208. ction 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 OPT 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 32 GEOMETRY OPTIMIZATION 240 HF dimer mp2 CP optimization basis avdz maxit 20 Imax number of iterations text OPTIMIZED VALUES OF GEOMETRY VARIABLES RFF 5 31431160 R1l 1 75768738 R2 1 75298524 HETA1 7 03780227 HETA2 111 25930975 geomet ry x noorient noorient must be specified since gradients are added f1 EZ EL Ef hl f1 rl f2 thetal h2 gt EZ 22 fl theta2 hl 180 do iter 1 maxit loptimization loop text CALCULATION AT LARGE SEPARATION rff_save rff Isave current rff distance rff 1000 calculation at large separation text HF1 MONOMER dummy 2 h2 second hf is now dummy hf scf for first monomer mp2 mp2 for first monomer ehflinf energy save mp2 energy in variable forces compute mp2 gradient for first monomer text HF2 MONOMER dummy f1 h1 first hf is now dummy hf Iscf 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 from previous gradient rff rff
209. ctive 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 17 8 7 The default record name both here and in MULTI is 1900 1 29 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 29 THE VB PROGRAM CASVB 210 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 START 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 29 7 Specifying a guess GUESS key 1 key 2 3 The GUESS keyword initiates the input of a guess for the valence bond orbitals
210. cture BASIS If a basis is not specified at all for any unique atom group then the program assumes a default For further details including respecifying the default to be used see the specification of the BASIS subcommand below 10 5 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 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 a type atom key scale nprim Load basis named key from the library with angular symmetry type S P D F G H or I This basis is added from all atoms with number i atom on the A cards see above If scale is present all exponents are scaled by scale 2 If nprim is specified the first nprim exponents only are taken from the library If nprim is negative the last nprim 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 de
211. cu 1 0d 7 record 5100 1 CPMCSCF GRAD 1 1 spin 0 5 accu 1 0d 7 record 5101 1 CPMCSCF GRAD 2 1 spin 0 5 accu 1 0d 7 record 5102 1 examples orbprint 5 lih2_DOD1 com NATORB 6666 ci 0 Force SAMC 5100 1 CONICAL 6100 1 Force SAMC 5101 1 CONICAL 6100 1 Force SAMC 5102 1 CONICAL 6100 1 opt STEP 0 15 0 2 2 COORD bmat UPDATE bfgs 5 if optconv 1t 1 d 4 exit enddo This second example optimizes the CI SO TO in LiH2 ground state is Singlet excited state is Triplet 32 GEOMETRY OPTIMIZATION 246 ee AS QO CI memory 3 M basis sto 3g geometry nosym Li hl Li r h2 Li r hl theta r 3 31510281 theta 30 57744006 maxstep 40 do i 1 maxstep If I eq 1 then int cart pri 2 hf wf 4 1 0 else int cart end if multi occ 7 core 0 closed 0 wf 4 1 0 state 1 wf 4 1 2 state 1 CPMCSCF GRAD 1 1 spin 0 accu 1 0d 7 record 5101 1 CPMCSCF GRAD 1 1 spin 1 accu 1 0d 7 record 5100 1 orbprint 5 NATORB 6666 ci 0 examples lih2 _SOTO com Force SAMC 5101 1 CONICAL 6100 1 NODC Force SAMC 5100 1 CONICAL 6100 1 NODC opt gradient 1 d 5 step 1 d 4 ISTEP 0 15 0 2 2 COORD bmat UPDATE bfgs 5 if optconv 1t 1 d 4 exit enddo 32 3 Examples 32 3 1 Allene Z matrix 32 GEOMETRY OPTIMIZATION 247 xxx Allene geometry optimization using Z Matrix memory 1 m basis sto 3g rcc 1 32 ang rch 1 08 ang acc 120 degree Geometr
212. d 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 H OF 0 in the basis of the configuration state functions 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 19 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 151
213. d can be followed by CON cards which define state specific P space configurations 17 4 5 Projection to specific A states in linear molecules Since MOLPRO can only use Abelian point groups e g C2 instead of C for linear molecules A 2_ 2 States as well as E 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 LQUANT option can be used to specify which states are desired LQUANT lam 1 lam 2 lam nstate lam i is the A quantum number of state i i e O for 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 17 5 Restoring and saving the orbitals and CI vectors MULTI normally requires a starting orbital guess In this section 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 oper
214. d energies are stored in variables as explained in section 6 7 As well as the energy the T diagnostic T J Lee and P R Taylor Int J Quant Chem S23 1989 199 is printed and stored in the variable T1 DIAG for later analysis 21 1 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 For further information on triples corrections see under RCCSD 21 2 Quadratic configuration interaction QCI OCT or OCISD performs quadratic configuration interaction QCISD Using the OCT T or OCISD 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 DUS directives see section 21 5 21 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 vanish 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
215. d 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 ES vee 27 87 MIXTOT 0 00 T526 27 88 MIXCI 0 00 19 11 35 83 MIXTOT 0 00 15 36 28 73 28 QUASI DIABATIZATION 205 h2s Diabatization and NACME calculation memory 3 m gprint orbitals civector geometry x noorient Ss Hs ely 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 always be used for diabatization This basis is too small for real application Referenc geometry Displaced geometries Samll displacements for finite difference NACME calculation Orbital dumprecord at reference geometry IMRCI record at reference geometry IMRCI record at displaced geometries C2v geometry hf roc 9 2 wh 18 2 E orbitalsp2L002 mult
216. der 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 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 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 di rective allows to modify the level shift in the RHF program and EXPEC to calculate expectation values of one electron operators see section 4 13 14 1 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
217. directory specifies the directory where the run time file 1 will be placed overriding di rectory 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 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 GENERAL PROGRAM STRUCTURE 3 2 1 2 Running MOLPRO on parallel computers MOLPRO will run on distributed memory multiprocessor systems including workstation clus ters under the control of the Global Arrays parallel toolkit There are also some parts of the 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 multiu
218. 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 18 2 Specifying the wavefunction 18 2 1 Occupied orbitals OCC nN1 N2 Ng ni 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 18 2 2 Frozen core orbitals CORE A1 M2 Ng35 18 THE CI PROGRAM 136 ni 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 18 2 3 Closed shell orbitals CLOSED A1
219. duct threshold for screening in first half transformation SPARSE value If value is on zero use sparse algorithm in second half transformation default LOCFIT value If value 1 use united orbital fitting domains default If value 2 use united pair fit domains RDOMAUX value Radius for extending the pair fitting domains Should be nonzero if LOCFIT 2 KSCEEN value If value 1 use Schwarz screening linear scaling algorithm MINBLK value Minimum AO blocking size MAXBLK value Maximum AO blocking size MAXF I T value Maximum block size for fitting functions MAXBATCH value Blocking size for first half transformation RI MP2 is an alias for the command DF MP 2 At present expectation values and gradients cannot be computed with DF MP2 23 4 1 Intermolecular interactions 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 cards or the SAVE command see section 23 5 1 If the asymptotic energy is not needed it is sufficient to do this initial calculation using option DOMONLY 1 These domains should then be reused in the subsequent calculations at all other intermolecular distances by using the START record option or the START command see section 23 5 2 Only in this way the basis set superposition error is minimized and normally negligible of course this does not
220. e 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 19 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 LSHIF T shift DIRECT RS2C Gn SHIFT shift DIRECT 19 7 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 19 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 153 OPTION codel value code2 value Of relevance for the CASPT2 3 program are the following options TPROCS 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 ineffi
221. e print a table with results title Results for H20 basis S basis title for the table The job produces the same table as before If you 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 3 8 Using default Procedures Using the default procedures described in section the above inputs can be simplified even more 3 8 1 RCCSD T for different states The following calculation performs RCCSD T calculations for the ground and first excited state of the OH radical examples h20_proce com 3 INTRODUCTORY EXAMPLES RO OH geomet ry 0 H 0 1 83 runccsdt set symmetry 2 spin 1 runccsdt X 2 Pi state epi energy set symmetry 1 spin 1 runccsdt A 2 Sigma state esig energy de esig epi toev Excitation energy in eV This produces the following results RESULTS METHOD STATE S ENERGY HF SCF L 2 0 5 75 39004124 RCCSD Le 0 5 75 55736436 RCCSD T L 2 0 5 75 55912676 RCCSD T ES 0 5 HT 9096S 7 HF SCF L 1 0 3 75 22787407 RCCSD L 1 0 9 75 39738157 RCCSD T L 1 0 5 75 39914839 RCCSD T L 1 0 5 7939913991 SETTING DE 4 35323484 3 8 2 SA CASSCF and MRCI 24 examples oh_runccsdt com The following calcul
222. e 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 17 9 2 Difference gradients 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 17 9 3 Non adiabatic coupling matrix elements for SA MCSCF For computing non adiabatic coupling matrix elements analytically use CPMCSCF NACM state state2 ACCU thresh RECORD record where s
223. e 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 18 3 5 the state averaged fock operator is made for all reference states and the WEIGHT card refers to the corresponding states 19 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 n RS2C n RS3 n where n can take the values 1 to 4 Instead of the value n one can also specify G1 G2 etc 19 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 245 215 1995 The shift can be specified on the RS2 or RS2C card RS2 Gn LSHIFT shift RS2C Gn SHIFT shift Typical choices for the shift is are 0 1 0 3 Only two figures after the decimal point ar
224. e 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 en vironment variables in shell form for example STMPDIR or tmp SUSER these will be expanded at run time If multiple scratch file systems are avail able it is advantageous to present a list of directories of which there is one in each file system 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 of bin molpro rc the environment variables STMPDIR TMPDIR2 STMPDIR3 are used to construct the list of scratch directories for the d option Thus these environment variables should at make time be filled with the names of directories on each available scratch file system cf section A 3 3 I directory 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 direc tory 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 no copying will take place On some main frames the scratch directory is erased automatically after a
225. e ee ee 17 2 1 Occupied orbitals 17 2 2 Frozen core orbitals 17 2 3 Closed shell orbitals 17 2 4 Freezing orbitals 17 3 Defining the optimized StateS ee ee 17 3 1 Defining the state symmetry oo o 17 3 2 Defining the number of states in the present symmetry Fah tain bite avs Maes Ga wee Acne ge ae Gee mie Beem E ahaa ae ia ioe es ha be aon eo Bee Bees ing og ee AE ad amp Bceaw 17 4 4 Selecting the primary configuration set depen Anais e ei A a a a Se ea e Mek ed sah et He a S 17 5 5 Natural orbitals 17 5 6 Pseudo canonical orbitals 17 5 7 Localized orbitals 17 5 8 Diabatic orbitals Xi 110 111 112 112 112 112 113 113 113 113 113 113 114 114 114 114 114 114 115 115 115 115 115 115 CONTENTS xii ehh a ie de de 126 17 6 1 Selecting the Cl method o ooo 126 17 6 2 Selecting the orbital optimization method 127 E ia ee OA oe fe Po ek ws 127 ee ee 127 it tay da Fes a ks Bt hed 128 Wigs pon HS fn ee wh RO aL ao aay he 128 N 128 ada pedos 128 LE aaa e a A de ek o 128 ao aio sana MIES ae dde erase 129 ads Gea ae dc e e O de re 129 17 8 3 Maximum number of iterations o 130 ek a eee oo dl A ee lado Got eo 130 aiaa a ue Beare eee ee Ee gee 130 paa 131 po Ue oa e Bd a 131
226. e energy and gradient using the STARTCMD option OPT STARTCMD command Similar to the calculation of numerical gradients see above command must be found in the input before the OPT card i e the sequence of input cards starting with command and ending with OPT defines one optimization step For example in order to optimize the geometry at the ccsd t level using numerical gradients the following input could be used hf loptimize orbitals ccsd t compute ccsd t geometry forces numerical startcmd hf compute numerical ccsd t gradients opt startcmd hf optimize geometry The convergence criteria are the same as explained below for the OPTG procedure The con vergence thresholds can be modified using further options on the OPT card exactly in the same way as explained below for OPTG For example the threshold for the gradient can be changed using OPT STARTCMD command GRAD 1 d 4 Further subcommands for OPT are possible which are the same as for OPTG described in the next section 32 2 Automatic geometry optimization OPTG The OPTG command is used to perform automatic geometry optimizations for all kinds of wave functions The coordinates to be optimized can be chosen using the COORD directive see sec tion optgeo coord Various optimization methods can be selected as described in section 32 2 4 MOLPRO allows minimization i e search for equilibrium geometries transition state optimiz ation i e search for saddle
227. e neglected This is a prerequisite to obtain linear scaling for large molecules Using LOCAL without choosing appropriate settings manually will result in O N scaling Be sure to read the applicable parts of the next section before starting your own calculations 23 3 Doing it right 23 3 1 Always Turn off symmetry Otherwise you won t get appropriately localized orbitals local orbitals will tend to be symmetry equivalent instead of symmetry adapted Symmetry is in principle OK only if all atoms are symmetry unique This allows the treatment of planar molecules in C symmetry when using the LOCAL directive But note that the multipole program does not support symmetry at all so choose always C symmetry with the MULTP directive To turn off symmetry specify nosymas the first line of your geometry input e g geomet ry nosym Ol H1 01 roh H42 01 roh h1 hoh Use NOORIENT Under certain circumstances it may happen that the domains and correlation energies are not rotationally invariant We therefore recommend to use the NOORIENT option in the geometry input to avoid unintended rotations of the molecule when the geometry changes This is particularly important for geometry optimizations and for calculations of interaction energies see section 23 4 1 Check your orbital domains Local correlation methods are less black box than the canonical ones It is therefore recommended always to check the orbital domains which
228. e 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 4 Facilities 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
229. e 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 16 after the 23 LOCAL CORRELATION TREATMENTS 169 HF card Indicative of insufficient SCF accuracy are small positive energy changes near the end of the geometry optimization 23 4 3 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 23 LOCAL CORRELATION TREATMENTS 170 23 5 Further commands 23 5 1 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 command is given the domain information as well as the amplitudes are saved for MPn the amplitudes are not saved If just the domain informati
230. e value is specified this is used for all states 6 VARIABLES SPIN MCSTATE STATE WEIGHT LQUANT MCSELECT SELECT MCRESTRICT RESTRICT CONFIG MCOC C OCC MCCL OSED CLOSED MCCO RE CORE MCSTART COREORB MCORB MCSAVE 51 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 E 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 MCCORE only used if MCCORE is not present record of starting orbitals record of frozen core orbitals record for saving optimized orbitals 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 CISELECT SELECT CIRESTRICT RESTRICT CIOC C
231. e 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 17 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 17 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 orb orb orb are included in the 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
232. ecessary 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 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 TYPE orbtype STATE state SP IN spin MS2 ms2 SAVE 0rbsav 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 r
233. ecify 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 WF or REF RESTRICT optional SELECT optional CON optional 18 THE CI PROGRAM refl recl file ref2 rec2 file refthr refstat mxshrf 138 rec gt 2000 The configurations are read in from the specified record See section 17 5 4Jabout 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 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 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 t
234. ecommended the new orbitals are stored in the default dump record 2140 2 or the one given on the ORBITAL directive see section 17 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 2 16 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 17 THE MCSCF PROGRAM MULTI 125 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 orbitals 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
235. ecord 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 9 4 By default imaginary and low frequency modes are not stored By specifying DUMPALL rather than DUMP all modes are written out By default all computed frequencies including low and imaginary ones are printed The fol lowing options can be used to modify the print level 33 VIBRATIONAL FREQUENCIES FREQUENCIES 253 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 belonging to ro tations and translations and their normal modes default PRINT LOW 1 suppresses the print PRINT IMAG print imaginary vibrational frequencies and their normal modes default PRINT IMAG 1 suppresses the print Imaginary frequencies 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 cm 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 t
236. ect 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 2 2 use DKEXT to compute exchange operators in DMP 2 one integral evaluation This is only useful in local DMP 2 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_DMP 2 THRPROD_DTRAF THR_DTRAF THRPROD default Specific options for direct local MP2 LMP 2 DTRAF THR_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_DTRA
237. ed 35 MATRIX OPERATIONS 264 35 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 35 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 4 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 35 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 35 3 Saving matrices SAVE SAVE name record type At present type can be DENSITY ORBITALS FOCK HO ORBEN OPER TRIA
238. ed and saved in a dump record the occupation numbers are automatically stored as well This is convenient for later use e g in MOLDEN 35 4 Adding matrices ADD ADD result facl matl 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 fac 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 35 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 35 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 35 7 Multiplying matrices MULT MULT result matl mat2 facl fac2 calculates result fac2 result facl matl mat2 The strings result matl mat2 are the internal names of the matrices If fac is n
239. ed 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 This 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 Q1 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 A second atom needed to define the angle po p p2 The same rules hold for the specification as for pj Internuclear angle 0 po p p2 This angle is given in degrees and must be in the range 0 lt a lt 1800 A third atom needed to define the dihedral angle B po p1 p2 p3 Only applies 1f J 0 see below 9 GEOMETRY SPECIFICATION AND INTEGRATION 67 B Dihedral angle B po p p2 p3 in degree This angle is de f
240. ed in this way see section 6 7 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 4 PROGRAM CONTROL 34 ENDIF ULTI CI ENDPROC This procedure can be used for a calculation of a vertical ionization potential of H20 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 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 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 1 e variables are commonly known to all procedures and all variables defined in procedures will be 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 4 9 T
241. ed 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 i8 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 For Linux PCs we recommend the following BLAS libraries Intel PIII lsblaspiil 2f_03 00 a inthe ASCI library which can be ob tained from http www cs utk edu ghenry distrib To use this library link it to a file name that the linker can understand for example libblas a in s 1sblaspiil 2f_03 00 a libblas a and when configure prompts you for the library type L blasdir lblas where blasdir is the absolute path of the directory holding the BLAS library AMD Athlon Atlas library
242. ee ORBPRINT card Test print parameter If nonzero use Jacobi diagonalization 14 7 8 Options The OPTION directive can be used to set various miscellaneous options including those de scribed before OPTION option value option3 value gt optionz valuez The following table list the available options 15 THE DENSITY FUNCTIONAL PROGRAM 90 15 THE DENSITY FUNCTIONAL PROGRAM Density functional theory calculations may be performed using one of the following commands DFT calculate functional of a previously computed density KS or KS SCF calls the closed shell self consistent Kohn Sham program RKS or RKS SCF calls the spin restricted open shell Kohn Sham program UKS or UKS SCF calls the spin unrestricted open shell Kohn Sham program Each of these commands may be qualified with the key names of the functional s which are to be used command keyl key2 key3 If no functional keyname is given the contents of the MOLPRO vector variable DF is interpreted as a list of functionals If DF is empty DF TNAME is examined otherwise the default is LDA see below Following this command may appear commands specifying options for the density functional code followed in the Kohn Sham case by further SCF options exactly as for the Hartree Fock programs 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 va
243. ely following n cards with the corresponding parameters in the form MM analogously a number of cards specifying the coefficients of the radial potentials AV of the SO part of Vps 11 3 Example for explicit ECP input 11 EFFECTIVE CORE POTENTIALS 80 2 E 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 geomet ry cu basis ECP 1 10 3 ECP input 1 NO LOCAL POTENTIAL Zee POSS 2 S POTENTIAL 2 30 22 355 770158 2 13 19 70 865357 2 P POTENTIAL 2 33 13 233 891976 2 13 22 53 947299 2 D POTENTIAL 2 38 42 31 272165 2 13 26 2 741104 examples 8s7p6d 6s5p3d BASIS SET cu_ecp_explicit com s 1 27 69632 13 50535 8 815355 2 380805 952616 112662 040486 01 6 Lady 2311327 3 0968 LIL 5D4 0810 p 1 93 504327 16 285464 5 994236 2 536875 897934 131729 030878 c 1 2 022829 1 009513 C 3 4 24645 792024 d 1 41 225006 12 34325 4 20192 1 379825 383453 1 c 1 4 044694 212106 453423 533465 end rts el energy Lhe occ 4 1 1 1 1 1 43 Closed 4 1 1 1 4 7b we bo pl e2 energy de e2 el1 toev Delta E 0 075 eV 11 4 Example for ECP input from library AUH CCSD T binding energy of the AuH molecule at r exp using the scalar relativistic 19 valence electron pseudopotential of the Stutt
244. en 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 cards 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 cards are determined automatically as usual 23 5 4 Correlating subsets of electrons ATOMLIST 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 only at the SCF level The ATOMLIST directive allows the specification of a subset of atoms 23 LOCAL CORRELATION TREATMENTS 171 ATOML1IST atoml atom2 The program will then correlate only electrons in orbitals whose domains are exclusively cov ered by the given atoms Electrons in a bonding orbital from one of the given atoms to one which is not part of the list are not correlated This 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 23 5 5 Energy partitioning for molecular cluster calculations ENEPART The local character of occupied and virtual orbitals in the lo
245. eneration of full T domains i e the domains for all three 2 orbitals comprise all carbon basis functions Note that the merged domains are generated after the above print of orbital domains and information about merged domains is printed separately See section 23 4 2 for further discussion of geometry op timizations 23 LOCAL CORRELATION TREATMENTS 166 These options can be disabled by setting their values to zero 23 3 2 Linear scaling MP2 Linear scaling of the CPU time as well as memory and disk requirements with molecular size for a fixed basis set can be achieved for very extended systems if the calculation is performed in integral direct manner i e the 2 electron integrals are never stored on disk 3 This requires the DIRECT module In order to achieve low order scaling the MULTP directive must be used Normally the program uses appropriate defaults and no further options must be set Thus the typical input structure looks as follows memory 64 m specify enough memory if you try a very large calculation file 2 name wfu save orbitals and other info for later restart gdirect lenable integral direct mod basis basis specification geometry nosym noorient geometry specification hf scf calculation this often takes most of the time since at present this is not linearly scaling locali pipek lOrbital localization this card is optional If not given Pipek Mezey localization is done automatically within
246. enever 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 THR le 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 SAVE SYM 15 4 1 Target quadrature accuracy THR THR acc accr acca 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 specifying accr and acca respectively 15 4 2 Radial integration grid RADIAL
247. ening product threshold for first half transformation SPARSE value If Non zero use sparse algorithm in second half transforma tion default RI MP2 is an alias for the command DF MP 2 At present expectation values and gradients cannot be computed with DF MP2 21 THE CLOSED SHELL CCSD PROGRAM 156 21 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 NOS INGL directive The code also allows 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 18 The only special input directives for this code are BRUECKNER and DIIS as described below The convergence thresholds can be modified using THRESH ENERGY thrden COEF F 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 4 11 The compute
248. ents 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 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 defin
249. ep the domains fixed during geom etry optimizations frequency calculations or whenever smooth potential energy functions are required In optimizations with very large geometry changes it may be useful to determine new domains at the optimized geometry and repeat the geometry optimization with these domains Particular care must be taken in optimizations of highly symmetric aromatic systems like e g benzene In De symmetry the localization of the tr 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 large changes of the localized orbitals If now the domains are kept fixed using the SAVE and START options 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 23 3 1 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 7 orbitals Note that this problem does not occur if the symmetry of the aromatic system is lowered by a substituent Finally we note that the LMP2 gradients are quit
250. er calculation at large separation text HF1 MONOMER dummy 2 h2 second hf is now dummy Aes scf for first monomer mp2 mp2 for first monomer ehflinf energy save mp2 energy in variable text HF2 MONOMER dummy f1 h1 first hf is now dummy hts scf for second monomer mp2 mp2 for second monomer ehf2inf energy save mp2 energy in variable rff rff_save reset HF HF distance to initial value do iter 1 maxit loptimization loop text HF1 CP MONOMER dummy f2 h2 second hf is now dummy hf scf for first monomer mp2 mp2 for first monomer examples ehfl energy Isave mp2 energy in variable hfdimer_cpcopt2 com forces compute mp2 gradient for first monomer scale 1 multiply gradient by 1 text HF2 CP MONOMER dummy f1 h1 HE mp2 ehf2 energy forces add 1 dummy text hf mp2 edimer energy forces add 1 DIMER CALCULATION first hf is now dummy Iscf for second monomer mp2 for second monomer save mp2 energy in variable compute mp2 gradient for second monomer Isubtract from previous gradient lreset dummies Iscf for dimer mp2 for dimer save mp2 energy in variable compute mp2 gradient for dimer add to previous gradient 32 GEOMETRY OPTIMIZATION 243 32 2 20 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 t
251. er than the atomic symbol 9 GEOMETRY SPECIFICATION AND INTEGRATION 68 geomt yp xyz geomet ry 3 number of atoms This is an example of geometry input for water with an XYZ file O 0 0000000000 0 0000000000 0 1302052882 H 1 4891244004 0 0000000000 1 0332262019 H 1 4891244004 0 0000000000 1 0332262019 hf examples h20_xyzinput com 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 9 4 9 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 Xx 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 9 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
252. er values might be useful to reduce the CPU time The format of the DIRECT directive is DIRECT keyl valuel key2 value 2 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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 56 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 product 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
253. ert 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 ERS first data line always INCLUDE include other input file FILE definition of named files TEXT prints text TITLE defines a title for the run CON specifies orbital configurations a 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 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 However commas are required in certain cases to make the meaning unique For instance typing 3 4 evaluates to one field with value 7 but 3 4 is the input for two fields with values 3 and 4 We recommend the general use of commas in order to avoid unexpected result
254. escribed 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 17 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 17 THE MCSCF PROGRAM MULTI 128 17 6 5 Saving the density matrix The first order density matrix in AO basis is written automatically to the dump record specified on the ORBITAL card default 2140 2 Ifno 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 the SA density as well the individual state densities are saved See section 2 16 for information about how to recover any of these densities for use in later programs 17 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
255. eted by local correlation methods J Chem Phys 110 7210 1999 23 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 LMP4 LOCI SD T LCCSD t or LCISD directives Further options can be given on the same input card or on subsequent LOCAL or MULTP cards Alternatively one can also give the LOCAL or the MULTP directive after the normal MP2 MP3 MP4 OCI SD T CCSD T or CISD directives Thus the two input cards METHOD LOCAL key1 value key2 value2 are equivalent to LMETHOD key1 value key2 value2 23 LOCAL CORRELATION TREATMENTS 164 where METHOD is one of MP2 MP3 MP4 OCI SD T CCSD T or CISD Similarly METHOD MULTP key1l value key2 value2 is equivalent to LMETHOD MULTP key1 value key2 value2 The full set of options is described in section and summarized in Table 9 The LOCAL and MULTP directives only differ in the defaults that they assume for the input keys The LOCAL directive requests a traditional local correlation calculation where all pairs of occupied orbitals that are correlated by MP2 are treated equal regardless of their distance The MULTP directive turns on additional approximations that depend on the distance between the orbitals The distant pairs are treated by a multipole approximation as described in Ref 2 and very distant pairs ar
256. etry pl1t If ORBITAL is given then for each orbital num ber symmetry specified the file filename_orbital_number symmetry plt is produced and contains the orbital grid in gOpenMol internal format e The default is not to produce any orbitals or densities and so only the atomic coordinates are dumped e The default is to use unformatted binary files and this should not normally be changed e 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 26 DIABATIC ORBITALS 196 26 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 MC
257. eversions The following options can be given cache directory c d 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 U U Username for web server password pp 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 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 A INSTALLATION OF MOLPRO 282 patcher r xx yy 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 needed A 3 11 Installation of documentation The documentation is usually supplied as a compressed tar file with a name of the form molpro doc 2002 6 tar gz This file unpacks to a top level directory molpro2002 6 this may either coincide or not with the top level directory contain ing source code or not according to taste The postscript and PDF user s manual is found in the directory molpro2002 6 doc with the H
258. ew A 3 2 Prerequisites A 3 3 Configuration A 3 4 Configuration of multiple executables in the same MOLPRO tree A 3 5 Compilation and linking A 3 6 Adjusting the default environment for MOLPRO A 3 7 Tuning A 3 8 Testing A 3 9 Installing the program for production A 3 10 Getting and applying patches A 3 11 Installation of documentation B Recent Changes B 1 New features of MOLPRO2002 6 B 2 New features of MOLPRO2002 B 3 Features that were new in MOLPRO2000 B 4 Facilities that were new in MOLPRO98 xviii 267 267 267 267 267 267 268 268 268 270 272 272 272 272 272 273 274 277 277 278 279 279 279 280 282 283 283 283 284 285 287 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 PR
259. ex pansion only neglect monopole 2 pole interactions where p is the requested multipole level option 3 Use monopole integrals everywhere translation and ex pansion but neglect all monopole 2 pole dipole 2 pole quadrupole 2P lpole interactions and so on This is entirely consistent but re duces the effective multipole level by 1 as compared to the other op tions Default 1 Multipole operators MAXMLTPL option Defines the highest level of multipole operators that are created Has to be greater or equal than each of DSTMLT LONGMLT and SHORTMLT is otherwise overwritten by the default which is max DSTMLT LONGMLT SHORTMLT Greater values are a useless waste of CPU unless you save the operators for later reuse MULTPAGE option option 0 Suppress paging of multipole operators during multipole expansion After their creation all operators are read from disk into memory Will crash if not enough memory is available option 1 Read operators from disk when needed Small perfor mance impact Default 1 Essentially obsolete keys for Taylor expansions TRUNCATE option Determines if the simple or exhaustive truncation truncation scheme will be used for multipole expansions Exhaustive truncation means that unlike a classical multipole expansion all interactions of mul tipoles up to the highest order are taken into account e g in an ex pansion of level two the exhaustive scheme will include quadrupole quadru
260. ext 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 4 10 Checking the program status STATUS STATUS nr prog1 prog6e This command reads the status information from file nr and checks the status of the spec ified program steps prog to prog may be HF RHF UHF MCSCF CI MULTI 4 PROGRAM CONTROL 35 FORCES If none of these is specified the status of the last step is checked If one of prog to progs is CRASH or STOP the program will either crash or stop if status was not o k STOP is default If CLEAR is specified a bad status is cleared so there will be no crash at subsequent status checks Examples STATUS 1 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 useful to avoid that the next program in a chain is executed STATUS 2 MULTI CI STOP will check the status of the most previous MULTI and CI steps which had allocated file 2 and stop if something was wrong STATUS 1 RHF CLEAR will clear status flag for last RHF No action even if RHF did not converge Note that the status variables are recovered in a restart 4 11 Global Thresholds GTHRESH A number of global thresholds can be set using the GTHRESH command outside the individual programs t
261. f batches NUMBATCH 0 manually set number of batches BATCHDIAM 35 maximal diameter of batches BATCHALGO 2 algorithm to determine batches WEIGHTPREV 0 5 parameter for algorithm BATCHALGO 1 RANSEED 1 initialize random number generator for simulated annealing Further numerical stability options CUTOFF 15 orbital cutoff MONOPOLE 1 if and how to treat monopole integrals Multipole operators MAXMLTPL auto manually set level of multipole operators to create MULTPAGE 1 turn on paging of multipole operators during multipole expansion Essentially obsolete keys for Taylor expansion TRUNCATE 0 truncation pattern of multipole expansion DAMP 0 damping function for orbitals SCALEDAMP 0 scaling factor for the damping function Stuff for debugging PATREN 0 print a list of uncoupled pair energies 24 THE FULL CI PROGRAM 183 24 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 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 comm
262. f the batch centres Default 2 WEIGHTPREV value Affects the one dimensional path laid through the molecule for BATCHALGO 1 Smaller values mean a more systematic directed order of atoms from 23 LOCAL CORRELATION TREATMENTS 179 one end to the other Larger values mean that the distance from one atom in the path to the next will become smaller Has to be between 0 and 1 Default 0 5 RANSEED value Negative values initialize the random number generator for the sim ulated annealing algorithms Positive numbers suppress initialisation of the random number generator Default 1 Further numerical stability options CUTOFF distance Applies a simple cutoff to orbitals before the transformation of the multipole operators Orbital coefficients belonging to AOs that are more than distance a u away from the orbital centre will be deleted distance 0 means don t use a cutoff Default 15 MONOPOLE option Specifies how to treat monopole integrals Monopole integrals should be zero due to the orthogonality between occupied and virtual orbitals Therefore they are usually not included in the calculation However this does not hold exactly when an orbital cutoff is applied Including monopole integrals in the calculation might therefore improve the nu merical stability option 0 Neglect monopole integrals option 1 Use monopole integrals in the translation but neglect them later on option 2 Use monopole integrals everywhere translation and
263. fault contractions i e such cards should be followed only by C without any other specifications for contractions b 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 cray 13 each following card 20 exponents cray 15 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 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 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 Such behaviour can be forced at any stage by issuing the command INT and this is the recommended way of ensuring that a new basis set is adopted note however that INT will cause atomic orbital integrals to be e
264. 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 SP If style is XYZ an XYZ file will be written see also section 9 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 9 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 9 GEOMETRY SPECIFICATION AND INTEGRATION 69 The interface to the gOpenMol program offers an alternative visualization possibility and is described in section The example below generates all the information required to plot the molecular orbitals of water and to visualize the normal modes of vibration x x H20 geometry angstrom o h o roh h o roh h theta roh 1 0 theta 104 0 rhf optg frequencies examples h20_p
265. finite radius of convergence The method is available by replacing the LOCAL or MULTP card by the ATTENUATE card ATTENUATE key1 value ke y 2 value2 It does everything the MULTP card does i e distant pairs are still treated by ordinary multipole expansion plus it will enable the split Coulomb operator treatment of weak and strong pairs and select reasonable defaults See section 23 6 for details If you don t want distant pairs to be treated by ordinary multipole expansion simply specify DISTPAIR 0 on the ATTENUATE card Note that this method will only work in the context of integral direct calculations 23 6 Options Various options can be specified using key value pairs qualifying the LOCAL or MULTP com mand 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 correspond to the variable name used in the program Table 9 summarizes the keys aliases and default values In the following the parameters are described in more detail General Parameters 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
266. for the total CASSCF wavefunction as well as the orthogonal complement to Yyg The default for the number of configurations requested Neonf is 10 If Noonr 1 all configurations are included 29 12 Controlling the amount of output PRINT q io 3 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 1 Standard level of output 2 Extra output The areas for which output can be controlled are iy Print of input parameters wavefunction definitions etc in Print of information associated with symmetry constraints 13 General convergence progress la Progress of the 2nd order optimization procedure is Print of converged solution and analysis i6 Progress of variational optimization i7 Usage of record numbers on file 2 For all the default output level is 1 If is gt 2 VB orbitals will be printed in the AO basis provided that the definition of MOs is available 29 13 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 symmetr
267. formed in subsequent iterations HCN lt gt NHC Isomerization Transition State Optimization and Frequencies memory 1 m 11 1 18268242 ang 12 1 40745082 ang al 55 05153416 degree basis 3 21G geomet ry nosymm Nek pled H 2 12 1 al1 int hf frequencies analytical save 5300 2 print low print imag mp2 optg root 2 method rf step 1 0 0 3 10 hstart 5300 2 frequencies print low print imag HF SCF amples Vibrational frequencies for HF SCF analytical Hes Stan f i cn_mp2_ts com Save Hessian in record 5300 2 Print low vibrational frequencies Print imaginary vibrational frequency MP2 Optimize Geometry Transition State Search Rational Function Optimizer Allow large stepsize Use SCF Hessian from FREQ record 5300 2 as starting guess Vibrational frequencies for MP2 numerical Hessian Print low vibrational frequencies Print imaginary vibrational frequency 32 2 17 Reaction Path Following options OPTION key param where 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 32 GEOMETRY OPTIMIZATION 238 the transition state This is the default If ID IR 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 mini mum If using a larger ID
268. future Note that you have to install the local module in order to use local correlation methods If you want linear scaling you also need the direct module References General local CCSD 1 C Hampel and H J Werner Local Treatment of electron correlation in coupled cluster CCSD theory J Chem Phys 104 6286 1996 Multipole treatment of distant pairs 2 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 3 M Schiitz 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 LMP2 Gradients and geometry optimization 4 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 LMP2 vibrational frequencies 5 G Rauhut A El Azhary F Eckert U Schumann and H J Werner Impact of Local Approx imations on MP2 Vibrational Frequencies Spectrochimica Acta 55 651 1999 Intermolecular interactions and the BSSE problem 6 M Schiitz G Rauhut and H J Werner Local Treatment of Electron Correlation in Molecu lar Clusters Structures and Stabilities of H20 n n 2 4 J Phys Chem 102 5997 1998 7 N Runeberg M Schiitz and H J Werner The aurophilic attraction as interpr
269. g 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 6 3 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 6 1 Setting variables A variable can be defined using variable expression unit unit is an optional string which can be used to associate a unit to a variable A variable definition is recognized by the equals sign in the first field of the input card For instance R 1 ANG THETA 100 DEGREE defines the variables R and THETA but THRESH ENERGY 1 d 8 GRADIENT 1 d 5 6 VARIABLES 43 does not define variables here ENERGY and GRADIENT are keywords to be recognized by the program 6 2 String variables String variables can be set as other variables in the form variable string If string contains blanks or other special characters like it must be given in quotes Instead of string also another string variable can be used e g METHOD P ROGRAM where PROGRAM is a string variable set by the program see section special variables The same name must not be
270. g approximate atomic densities or eigenvectors of h see below Since these defaults are usually appropriate the START card is not required in most cases 14 3 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 2s 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 not 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 14 3 2 below Example r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input fi Opes H2 0 r H1 theta basis STO 3G first basis set examples hf Iscf using STO 3G basis h20_sto3gstartl com basis 6 311G Isecond basis set hf scf 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 T
271. g 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 28 QUASI DIABATIZATION 202 SMAT The first nstate x nstate elements contain the state overlap matrix bra index rans fastest UMAT The first nstate x nstate elements contain the 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 Cl vectors only without orbital correction are stored in the variables SMATCI UMATCI HDIACI a
272. gart Koeln group gprint basis orbitals geomet ry 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 ht rcesd t core 1 1 1 1 el energy geometry h rhf e2 energy rAuH 1 524 ang molecular calculation geomet ry au h au rAuH hf ccsd t core 2 1 1 e3 energy de e3 e2 el toev binding energy 3 11 eV examples auh_ecp_lib com 12 CORE POLARIZATION POTENTIALS 81 12 CORE POLARIZATION POTENTIALS 12 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 ncentres e CPPADD factor e CPPSETLfcpp CPB INIT lt ncenters gt abs lt ncenters gt further cards will be read in the following format lt atomtype gt lt ntype gt lt Qq gt lt Oy 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 lt 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 po
273. gence acceleration For experts only 17 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 17 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 29 5 The default value for trnint is 1900 1 17 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 17 THE MCSCF PROGRAM MULTI 132 17 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 sam
274. 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 33 VIBRATIONAL FREQUENCIES FREQUENCIES 254 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 standard Temperature and Pressure T 298 150 K p 1 atm Subcommands of THERMO are PRINT THERMO additional information such as atomic masses partition functions and ther modynamical function in calories is printed to the output SCALE factor in calculating the thermodynamical properties use vibrational frequencies scaled with factor in order to take account of systematic errors of the wave function e g using SCF wavefunctions factor 0 89 is reasonable TEMP tmin tmax tstep calculate the thermodynamical properties at different temperatures start ing with tmin K up to tmax K in steps of tstep K PRESSURE p calculate the thermodynamical properties at a given pressure of p atm 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 33 3 Examples formaldehyde fregency calculation memory 8 m basis vdz gthresh energy 1 d 8
275. he 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 18 3 2 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 orbo 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 they 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 18 3 3 Explicitly specifying reference configurations CON N1 12 N3 N4 18 THE C
276. he energy De fault 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 de fault frequencies are saved on record 5300 2 START irec ifil Restart numerical frequency calculation from record irec on file fil 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 hes sian default COORD 3N Don t use symmetry unique displacements not recommended using finite differences 33 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 82 2 18 Note that numerical hessians cannot be computed when dummy atoms holding basis functions are present 33 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 pointgroup has to be the Schoenflies Symbol e g C3v for ammonia linear molecules have to be C v or D
277. he 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 r 1 85 theta 104 set geometry parameters geomet ry 0 z matrix geometry input Hil 04 43 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 h20_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 14 THE SCF PROGRAM 86 14 3 2 Starting with previous orbitals START RECORD record file specifications reads previously optimized orbitals from record record on file file 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
278. he first letter G is optional but should be used to avoid confusion with program specific THRESH cards The syntax is GTHRESH key value 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 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 ENERGY Convergence threshold for energy default 1 d 6 GRADIENT Convergence threshold for orbital gradient in MCSCF default 1 d 2 STEP Convergence threshold for step length in MCSCF orbital optimization default 1 d 3 ORBITAL Convergence threshold for orbital optimization in the SCF program default 1 d 5 CIVEC Convergence threshold for CI coefficients in MCSCF and reference vector in CI default 1 d 5 COEFF Convergence threshold for coefficients in CI and CCSD default 1 d 4 PRINTCI Threshold for printing CI coefficients default 0 05 PUNCHCI Threshold for punching CI coefficients default 99 no punch 4 PROGRAM CONTROL 36 4 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
279. he 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 Cl vectors alone 27 NON ADIABATIC COUPLING MATRIX ELEMENTS 200 1if non adiabatic coupling memory 1 m basis f avdz li vdz E L007 109 10 00 LLS by 1 25 01 dr 0 01 geometry li f 1li rlif rlif 3 ht oce 4 1 L 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
280. he updated hessian matrix Note that its diagonal elements 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 de bugging Several print options can be specified with one PRINT command 32 2 21 Conical Intersection optimization CONICAL To optimize a Conical Intersection CI 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 NOTE Previous versions required an explicit Gradient Difference calculation which is now replaced point 2 by a second gradient calculation This can be done by adding three different CPMCSCF cards in the MULTI input CPMCSCF NACM S Sj ACCU 1 0d 7 record record1 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 infos should be stored Parameter SPIN is half of the value in the WF card used to define the electronic state One must remember to 1 specify always three different record file in the CPMCSCEF cards 11 eval
281. heir default value of zero ORBITAL Orbital truncation threshold DENSITY Density truncation threshold FOCK Fock matrix truncation threshold 15 THE DENSITY FUNCTIONAL PROGRAM 109 15 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 i e the exchange is assumed to be contained in the functional and only the Coulomb interaction is calculated explicitly FACTOR facl fac2 Provide a factor for each functional specified The functionals will be combined accordingly By default all factors are one 15 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 15 2 5 Grid blocking factor BLOCK BLOCK 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 15 2 6 Dump integrand values DUMP DUMP file status Write o
282. hod for the case that CASSCF references 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 NOREF 1 Default Interactions between reference configurations and singles are omitted NOREF 0 Interactions between reference configurations and singles are included This causes a relaxation of the reference coefficients but may lead to intruder state problems 19 MULTIREFERENCE RAYLEIGH SCHR DINGER PERTURBATION THEORY 154 IMP 3 2 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 20 M LLER PLESSET PERTURBATION THEORY 155 20 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
283. i ENERGP 1 holds last Pople corrected energy for state i ENERGC holds CCSD QCI BCCD energy in CCSD T QCI T BCCD T calcu lations 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 32 2 19 Optimizing counterpose corrected energies Geometry optimization of counterpoise corrected energies is possible by performing for the total system as well as for each individual fragment separate FORCE calculations The gradients and 32 GEOMETRY OPTIMIZATION 239 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 intera
284. i occ 9 2 closed 4 1 wf 18 2 state 2 natorb reforbl noextra LOCO 9 2 closed 4 1 wf 18 2 0 state 2 orbital reforbl save refci Text Displaced geometries do i 1 r data truncate savci 1 reforb reforbl do j 1 3 r2 r i dr j multi oce 9 2 closed 4 1 wf 18 2 0 state 2 start reforb orbital 3140 2 3 diab reforb noextra reforb 3141 2 cis ocd 9 2 closed 4 Lo wf 18 2 0 state 2 orbital diabatic save savci eadia energy if j eq 1 then el 1 energy 1 e2 1 energy 2 end if ci trans savci savci dm 7000 2 3 ci trans savcitj refci dm 7100 2 3 ci trans savcitj savcitl am 7200 2 4 IB1 and 1A2 states Save reference orbitals on reforbl Dont use extra symmetries IMRCT at referenc 11B1 and 1A2 states Use orbitals from previous CASSCF Save MRCI wavefunction geometry Loop over different r values Itruncate dumpfile after referenc Loop over small displacements for NACME Set current r2 Wavefunction definition Starting orbitals Dumprecord for orbitals Generate diabatic orbitals relative to reference geometry Dont use extra symmetries lUse orbitals for j 1 as reference for j 2 3 lUse diabatic orbitals Save MRCI for displaced geometries examples Save adiabatic energies for use in ddr g 3 h2s_diab2 com Save adiabatic energies for table printing Compute transition densities at R2 DR j Save transition den
285. 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 L L 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 ACPF as described in Theor Ch
286. im 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 M ller 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 11 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 MP2 and QCISD 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 and MRCI wavefunctions Spin orbit coupling as described in Mol Phys 98 1823 2000 e Some two electron transition properties for MCSCF wavefunctions e g L etc Population analysis Orbital localization
287. 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 energies 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 30 SPIN ORBIT COUPLING 220 30 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 30 5 2 Options for spin orbit calculations Some options can be set using the OPTION directive in any order OPTIONS WIGNER value HLSTRANS value MATE L 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 HLST
288. ined as the angle between the planes defined by po p1 p2 and p1 p2 p3 180 lt B lt 180 Only applies if J 0 see below J If this is specified and nonzero the new position is specified 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 p 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 D2 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
289. ing may be long IO print debugging information in I O routines 17 8 2 Convergence thresholds Convergence thresholds can be modified using ACCURACY GRADIENT conv STEP sconv ENERGY econv where 17 THE MCSCF PROGRAM MULTI 130 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 4 11 Normally the above default values are appropriate 17 8 3 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 increasing the limit In most cases the choice of active orbitals or of the optimized states is not appropriate see introduction of MULTI 17 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 17 8 5 Special optimization parameters STEP radius trust1 tfac1 trust2 tfac2 Special parameters for augmented hessian method For experts only GOPER igop Use G operator technique in microiterations Default If igop It 0 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 gr
290. ing chapters 2 GENERAL PROGRAM STRUCTURE 15 MEMORY UNCH ILE START NCLUDE ASIS Q fol H fa fa ro ea ROME TRY FU AR gt E H kal HRESH GTHRESH IRECT GDIRECT EXPEC GEXPEC NDDO 7HRAREAER g la mM a j x le o H H zi H H le oQ A y Da H E LSEIF ie Z Oo HI try LABEL U Q D O H H D O jw ELETE ERA Z n eg ATROP wa JQ aj a Cl fa w IH 52 ARTESIAN PHERICAL USER I indicates start of a new calculation 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 endof 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
291. ing else in the input apart from this commands may come in any order 29 THE VB PROGRAM CASVB 208 29 2 Defining the CASSCF wavefunction CASVB is interfaced with the determinant part of MULTI 1 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 17 5 4 The three latter are recommended 29 2 1 The VBDUMP directive VBDUMP vbdump It is advisable to restore the wavefunction definitions using VBDUMP cards here and in the CASSCF calculation see Section 17 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 automatically based on the latest MCSCF calculation however STATE and WEIGHT information will not be restored in such a case 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 29 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 CORE Frozen core orbitals WE Wavefunc
292. input errors B RECENT CHANGES 286 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 records for orbitals density matrices or operators 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
293. ion 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 14 6 Expectation values EXPEC oper oper2 0pern Calculates expectation values for one electron operators oper oper2 Opern See section 4 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 4 13 for details 14 7 Miscellaneous options All commands described in this section are optional Appropriate default values are normally used 14 7 1 Level shifts SHIFT shifta shiftb nitord nitcl 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 If the iter
294. ion This means that integral direct procedures as described in M Schiitz 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 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 licence The linear scaling LCCSD T methods as de scribed in M Schiitz and H J Werner J Chem Phys 114 661 2001 M Schiitz and H J Werner Chem Phys Lett 318 370 2000 M Schiitz 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 B RECENT CHANGES 284 4 10 B 3 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 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 procedure
295. it might also influence the accuracy of the SCF if given before the HF card On the other hand requesting a more accurate energy will tighten the thresholds Tighter thresholds will also be chosen automatically if the AO overlap matrix has very small eigenval ues which can happen for large and diffuse basis sets The prescreening thresholds for LMP2 can also be changed using the specific options THRAO_LMP2 THRQ1_LMP2 and THRO2_1MP2 on the DIRECT card see section 8 but this is only recommended for experienced users 23 LOCAL CORRELATION TREATMENTS 167 23 4 Density fitting LMP2 DF LMP2 RI LMP2 Density fitting LMP2 can be performed with standard density or and Poisson fitting basis sets The present implementation works only without symmetry The input is as follows DF LMP2 Imp2 options Optionally a card DF IT can follow on which the following options can be specified appropriate default values are available BASIS_MP 2 string Fitting basis sets e g JKF IT default for standard density fitting or DENSITY POISSON for mixed density Poisson fitting These basis sets must have been defined in a previous BASTS block THROV value Screening threshold for 2 index integrals of fitting basis THRAO value Screening threshold for 3 index integrals in the AO basis THRMO value Screening threshold for half transformed 3 index integrals THRSW value Threshold for Schwarz screening THRPROD value Pro
296. ization off If option gt 1 the localized orbitals are printed Note Boys localization can only be performed using the LOCALIZE command The program will use the Boys or bitals 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 on record name ifil for later use e g plotting The two orbital 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 DOMONLY value 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 Parameters for selection of weak and distant pairs WEAKPAIR distance If all atoms of orbital domain i are separated by at least distance a u from any atom of orbital domain j pair ij is treated by MP2 The default is distance 1 which means that all pairs for which i and j have no atom in common are treated as weak pairs Setting distance 0 eliminates weak pairs i e all pairs are fully included in the calculation This option has no effect for local MP2 calculations DISTPAIR distance If all atoms of orbital domain i are separated by at least distance a u from any atom of orbital domain j pair ij is treated approx imately by MP2 provided the multipole approximation is activated Setting distance 0 elimin
297. 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 SDO cal culations The threshold will be reduced to THREST once a certain accuracy has been reached see VARRED or latest af ter MAXRED iterations In CI and CCSD calculations also the initial thresholds THRINT_DKEXT and THRPROD_DKEXT are influenced by this value For a description see THRMAX_DKEXT If THRMAX 0 the final thresholds will be used from the begin ning in all methods SCREEN Enables or disables prescreening SCREEN gt 0 full screening enabled SCREEN lt 0 THRPROD is unused No density screening in direct SCF SCREEN lt 1 THRINT is 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
298. larizability and lt cutoff gt is the exponential parameter of the cutoff function When lt ncenters 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 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 polarization integrals are effective at the moment 12 2 Example for ECP CPP 13 RELATIVISTIC CORRECTIONS 82 Na2 Potential curve of the Na2 molecule using 1l ve ECP CPP gprint basis orbitals rvec 2 9 3 0 3 1 3 2 3 31 do i 1 rvec rNa2 rvec i geomet ry na na na rNa2 basis ecp na ecplO0sdf s na even 8 3 5 p na even 6 3 2 d na 12 03 cpp init 1 na ly 9047771027 bt ehf 1 energy Ccisd core eci 1 energy enddo table rvec ehf eci ang ecp input basis input examples na2_ecp_cpp com CPP input 13 RELATIVISTIC CORRECTIONS There are three ways in MOLPROto take into account scal
299. late all pairs which can be formed from orbitals orbl isy1 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 18 3 7 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 18 4 Options 18 4 1 Coupled Electron Pair Approximation CEPA ncepa Instead of diagonalizing the hamiltonian perform CEPA calculation CEPA type ncepa This is currently available only for single configuration reference functions 18 4 2 Coupled Pair Functional ACPF AQCC CPF ncpf gacpfi gacpfe ACPF ncpf gacpfi gacpfe AQCC nepf gacpfi gacpfe Instead of diagonalizing the hamiltonian perform CPF calculation ncpf 2 not yet imple mented ACPF calculation ncpf 0 or AQCC calculation ncpf 1 For ACPF and AQCC 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 18 THE CI PROGRAM 141 18
300. ld be positive leading to a negative exponent Default 1 7 F 2DGAMMA y The same for the two dimensional fit WEIGHT3D option Selects what type of weight function is used for the fits option 0 Flat option 1 Spatial Remember to change F 1DGAMMA F 2DGAMMA accordingly when us ing a flat weight function y 1 0 is then a reasonable value Default 1 Options for determination of batches NUMBATCH value If 0 selects automatic determination of the number of batches If gt 0 provides a manual override for this number Default 0 BATCHDIAM value Maximal acceptable diameter of batches Used to automatically de termine the number of batches If zero disables batches i e only one batch will be created Default 35 BATCHALGO option Selects the algorithm used to determine the batches option 0 Manually set batch centres Three arrays with name BATCHX BATCHY BATCHZ have to be set in the input before the MP 2 card that contain the x y and z coordinates of the desired batch centres option 1 Simple algorithm Determines a path through the molecules and distributes the batch centres along this path option 2 A robust simulated annealing algorithm Will only use atom positions as batch centres might therefore fail for strongly sep arated dimers and similar systems where there is no atom near the optimal batch centre position option 3 4 A less robust simulated annealing algorithm that directly tries to optimize the batches instead o
301. lias 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 82 THREST_DSCF THRDSCF lt 1 d 10 depending on accuracy and basis set THRMAX_DSCF THRDSCF_MAX THRMAX THR_DTRAF 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 THRDMP 2 THR_DTRAF THREST_DMP2 THRAO_DMP2 THR_DMP 2 THREST_DTRAF default THRINT_DMP2 THRSO_DMP2 THR_DMP 2 THRINT_DTRAF default THRPROD_DMP2 THRP_DMP2 THR_DMP 2 THRPROD_DTRAF default THR_LMP2 THRLMP 2 THR_DTRAF THREST_LMP2 THRAO_LMP2 THR_LMP 2 THREST
302. localized in order to ensure invariance of subsequent electron correlation treatments This behaviour can be modified using the OCC and CORE direc tives 16 6 1 Defining the occupied space OCC OCC 01 02 defines the highest orbital o in each symmetry i to be localized 16 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 o are localized in symmetry i 16 6 3 Defining groups of orbitals GROUP OF FDIAG GROUP orb1 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 16 6 4 Localization between groups OF FDIAG OFFDIAG If this card is present localize between groups instead of within groups 16 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 BOYS This requires a Fock operator from the preceding energy calculation For localization of
303. lotfile can be modified using the FILE or PLOTFILE directive 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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 55 8 INTEGRAL DIRECT CALCULATIONS GDIRECT References Direct methods general M Schiitz R Lindh and H J Werner Mol Phys 96 719 1999 Linear scaling LMP2 M Schiitz 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 break 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 no
304. ls to predict reliable results a new method that couples MRCI and CASPT2 has been developed This variant is invoked using the CIPT2 directive 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 19 3 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 1 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 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 presented in Chem Phys Lett 288 299 1998 19 MULTIREFERENCE RAYLEIGH SCHRODINGER PERTURBATION THEORY 152 In the case that several states ar
305. lt 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 axis 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 Ci X Y Cry XY Z Can XZ YZ Dy 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 Dap for linear molecules See section for more details of symmetry groups and ordering of the irreducible represen tations Also see section for more information about automatic generation of symmetry planes 9 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
306. ly 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 18 THE CI PROGRAM 140 18 3 6 Specifying correlation of orbital pairs PATR iorbl isyl iorb2 isy2 np 1s 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 iorb1 isy iorb2 isy np Corre
307. m 2 x 2 rotations between individual orbitals Other orbital manipulations can be performed using the LOCALI program see section 16 or the MATROP program section B5 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 data 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 34 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 34 2 Moving orbitals to the output set MOVE MOVE orb1 sym1 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
308. m file ifil to file ifil2 5 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 5 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 1f 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 6 VARIABLES 42 LENBUF length of integral buffer file 1 NTR length of integral records must be multiple of 3 track LTR disk sector length assumed in CI default 1 is reasonable NCACHE machine
309. m is solved in a basis of Slater determinants unless a CONFIG card is given Some procedures may be disabled using the DONT directive 17 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 17 THE MCSCF PROGRAM MULTI 127 17 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 ending with an END card ITERATIONS DO method 1 iter1 TO iter2 DONT method2 iter3 TO iter4 END 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 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 17 6 3 Disabling the optimization In addition to the ITERATIONS directive d
310. may also be used 29 10 2 The IRREPS keyword IRREPS ij i2 3 29 THE VB PROGRAM CASVB 213 The list 7 i2 specifies which irreducible representations as defined in the CASSCF wave function are antisymmetric with respect to the label operation If an irreducible representation is not otherwise specified it is assumed to be symmetric under the symmetry operation 29 10 3 The COEFFS keyword COEFF S 1y ly 3 The list i1 i2 specifies which individual CASSCF MOs are antisymmetric with respect to the label 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 29 10 4 The TRANS keyword TRANS din Ur C1 1p C12 gt Cage nes Specifies a general ngim X Naim transformation involving the MOs ij in Specified by the c coefficients This may be useful for systems with a two or three dimensional irreducible representation or if localized orbitals define the CASSCF wavefunction Note that the specified transformation must always be orthogonal 29 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 SYMELM may be specified accordi
311. mber 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 ltmax Maximum number of iterations in EOM CCSD default 20 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 DIFOCK difo If set to 0 the program uses an approximate diagonal of H for look ing for the initial configurations corresponding to the nmax maxex lowest diagonal elements and for the vector update in the Davidson procedure If set to 1 the diagonal of the Fock matrix is used in stead If set to 2 the update procedures from CCSD program are used DIFOCK 1 and 2 should give exactly the same results Default for non local EOM CCSD even based on local CCSD is 0 For the local EOM CCSD DIFOCK is always set to 2 EOMLOCAL eoml If set to 0 non local calculation default If set to 1 it simulates the local EOM CCSD using the standard EOM program EOMLOCAL 1 is a test option and shouldn t be used for the time being INIMAX is used only if INISINGL and INIDOUBL are both zero All keywords can be abbreviated by at least four characters 21 9 2 Print options for EOM CCSD EOMPRINT The following print options are for testing purposes a
312. metries 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 RootinSym 1 2 3 1 NN WN WO 6 7 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 CHARGE Total charge of the molecule can be given instead of nelec NELEC number of electrons MCSYM METRY wavefunction symmetry This can be an array for state averaged cal culations SYMMETRY as MCSYMM only used if MCSYMM is not present MCSPIN 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 on
313. metry CI for 2 states wavefunction saved to record 6002 2 Compute overlap and transition density lt R R DR gt Save transition density to record 8100 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 backward difference store result in variable nacmelm 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 end of loop over differend bond distances nacmeav nacmelp nacmelm 0 5 average the two results forward and backward differences table r nacmelp nacmelm nacmeav nacme2 print a table with results title Non adiabatic couplings for LiF title for table 28 QUASI DIABATIZATION 201 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 10 5 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 019223339 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
314. metry 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 optimizing 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 P SPACE 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 II A etc states 17 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 CORE 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 17 THE MCSCF PROGRAM MULTI 117 f configuration specification for that state symmetry SELECT CON RESTRICT g definition of the p
315. metry default record 700 except if the optimization was aborted and no other calculations have been performed since then The information from the previous calculation is to be used to construct an approximate hessian for a starting guess first and last specify the first and last geometry blocks respectively to be used from record The default is to use all blocks It is not necessary that the method used for the previous optimization is identical to the present one This can be useful if several optimizations for different methods follow each other For example the hessian from an SCF optimization can be used as starting guess for a subsequent MCSCF optimization However it is required in such a case that the SAVE and START records are different 32 2 9 Setting a maximum step size STEP STEP steplength drmax damax drmax1 damax1 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 drmax 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 damax default 10 degrees 32 2 10 Number of point used in hessian update UPDATE UPDATE type nstep This option chooses the
316. metry input r 1 ang bond length theta 104 bond angle gexpec dm sm qm compute dipole and quarupole moments methods hf multi ci Ido hf casscf mrci do i 1 methods loop over methods Smet hods i Irun energy calculation e 1 energy dip 1 dmz save dipole moment in variable dip oe examples quadxx 1 qmxx save quadrupole momemts 2 h20_gexpec2 com quadyy 1 qmy y quadzz i qmzz smxx 1 xx save second momemts smyy i yy smzz i zz enddo table methods dip smxx smyy SMZZ print table of first and second moments table methods e quadxx quadyy quadzz print table of quadrupole moments This Job produces the following tables ETHODS DIP SMXX SMYY SMZZ HF 0 82747571 5 30079792 3 01408114 4 20611391 ULTI 0 76285513 5 29145148 e LL7LT397 4 25941000 CI 0 76868508 5 32191822 3 15540500 4 28542917 ETHODS E QUADXX QUADYY QUADZZ HF 76 02145798 1 69070039 1 73937477 0 04867438 ULTI 76 07843443 1 60318949 1 65831677 0 05512728 CI 76 23369821 1 60150114 1 64826869 0 04676756 4 PROGRAM CONTROL 38 4 13 2 Example for computing relativistic corrections ar2 geometry arl ar2 arl1 r r 2 5 ang 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 Igeometry definition bond distance Inon relativisitic scf calculation compute relativistic correction using Cowan Griffin o
317. modified by specifying RSTEP value ASTEP value on the numerical card Note that step sizes which are too large may lead to optimization failure For each active variable two additional energy calculations are necessary for each geometry optimization step so this may be expensive Symmetrical displacement coordinates are normally used to minimize the number of energy calculations see COORD keyword section 32 2 1 For optimization of special energies see VARIABLE section 32 2 18 32 2 15 Numerical Hessian NUMHES NUMHES mstep type icalc thresh 32 GEOMETRY OPTIMIZATION 236 This option allows you to calculate numerical second derivatives of the energy by finite differ ences If you use analytical gradients these are differentiated once whereby it is possible to use forward differences needs one additional gradient calculation for each coordinate or cen tral differences more accurate needs two additional gradient calculations for each coordinate For transition state optimizations it is usually sufficient to use forward differences If you use numerical gradients the energy is differentiated twice In this case only central differences are possible mstep defines the number of optimization steps after which the numerical hessian is recalculated mstep 1 Don t calculate numerical hessian default for minimization mstep 0 Calculate numerical hessian only once at the start of the optimization de fault for transition state sea
318. n fregdft Performs DFT frequency calculation freqmp2 Performs MP2 frequency calculation freqcas Performs CASSCF frequency calculation The procedure does not support state averaged calculations Note in most cases the SCF calculation is not performed if starting orbitals are found Several procedures may be specified after each other 2 19 MOLPRO 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 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 furture 3 INTRODUCTORY EXAMPLES This section explains some very simple calculations in order to help the new user to understand how easy things can be 3 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 lt lt basis vdz geometry angstrom h1 h2 h1 74 hf 1 3 INTRODUCTORY EXAMPLES 2
319. name_libname rc As described above the different executables can then be chosen on a specific machine by setting the environment variable MOLPRO_RCFILE to molpro_procname_libname rc Note that if MOLPRO_RCFILE is not set molpro rc will be used by default which will correspond to the last molpro_ 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 A INSTALLATION OF MOLPRO 278 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 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 discussion of options d dirl dir2 wher
320. nce 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 orb 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 Cl 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 stored in the same record In that case and also when there are just two states 27 NON ADIABATIC COUPLING MATRIX ELEMENTS 199 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 states where state is of the form istate isym the symmetries of both states must be t
321. nd 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 28 QUASI DIABATIZATION 203 h2s Diabatization memory 3 m gprint orbitals civector geometry x noorient Inoorient should always be used for diabatization S Alis EL h2 s r2 h1 theta basis avdz This basis is too small for real application r1 2 5 Reference geometry theta 92 r 2 50 2 55 2 60 IDisplaced geometries reforb 2140 2 Orbital dumprecord at reference geometry refci 6000 2 IMRCI record at reference geometry savci 6100 2 IMRCI record at displaced geometries text compute wavefunction at reference geometry C2v r2 r1 hes oe O 2 wt LS 2 43 orbital 2100 2 multi occ 9 2 closed 4 1 wf 18 2 state 2 11B1 and 1A2 states natorb reforb Save reference orbitals on reforb noextra Dont use extra symmetries LOCA 9 2 closed 4 L MRCI at reference geometry wf 18 2 0 state 2 11B1 and 1A2 states orbital reforb Use orbitals from previous CASSCF save refci Save MRCI wavefunction Text Displaced geometries do i 1 r Loop over different r values data truncate savci 1 Itruncate dumpfile after referenc r2 r 1 Set current r2 multi occ 9 2 closed 4 1 wf 18 2 0 state 2 Wavefunction definition start reforb Starting orbitals N examples orbital 3140 2 Dum
322. nd for looking for the convergence prob lems 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 21 THE CLOSED SHELL CCSD PROGRAM 161 DIAGONAL PSPACE HEFF RESIDUUM ipr 2 also print warning information about complex eigenvalues ipr 3 also print hamiltonian and overlap matrix in trial space 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 long 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 residuum 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 22 OPEN SHELL COUPLED CLUSTER THEORIES 162 22 OPEN SHELL COUPLED CLUSTER THEORIES Spin unrestricted RH
323. nd gradient dependent first row exchange correlation functional n K y O R S X Y i l where n 21 Ri Pat Pp _ 204 Si p Vi 2 Vi 2 Dan Oph i 2p4vi 3 E Y es A ie p8 3 t 7 6 4 3 3 2 5 3 4 3 3 2 5 3 3 2 5 3 1 2 3 2 5 3 2 7 6 4 3 3 2 5 3 1 102 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 15 THE DENSITY FUNCTIONAL PROGRAM u A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 v 0 0 0 0 1 1 1 1 2 2 2 2 0 0 0 0 0 0 0 0 0 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 and o 0 728255 0 331699 1 02946 0 235703 0 0876221 0 140854 0 0336982 0 0353615 0 00497930 0 0645900 0 0461795 0 00757191 0 00242717 0 0428140 0 0744891 0 0386577 0 352519 2 19805 3 72927 1 94441 0 128877 15 1 32 TH2 D J Tozer and N C Handy J Chem Phys 102 3162 1998 103 175 176 177 178 Density and gradient dependent first row exchange correlation functional of the form TH1 but with n 19 t ae 4 3 3 2 5 3 3 2 5 3 1 A 1 5 3 132 5 3 e TGA 7 6 4 3 3 2 5 3 u 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 v 0 0 0 0 0 1 1 1 1 2 2 2 0 0 0 0 0 0 0 w 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 and o 0 678831 1 75821 1 27676 1 60789 0 365610 0 181327 0 146973 0 147141 0 0716917 0 0407167 0 0214
324. nd 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 The defaults for the convergence parameters can also be changed by using a global GTHRESH directive 1 e GTHRESH OPTSTEP step OPTGRAD grad ENERGY energy 32 GEOMETRY OPTIMIZATION 230 32 2 1 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 displacement_type option This option chooses the optimization space and the coordinate system in which the optimization 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 vari ables on which the z matrix depends can be chosen using the ACTIVE or INACTIVE subdirec tives If the z matrix depends on no variables or xyz input is used all 3N cartesian coordinates are optimized opt_coord determines the coordinates in which the optization takes place By default local normal coordinates are used Optionally cartesian coordinates or natural internal coordinates can be used displacement_type specifies how numerical gradie
325. ned 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 We note that the present implementation performs the partial integral transformation using an algorithm which is not optimal in its memory use and therefore needs quite a large memory if extensive basis sets are used Future improvements will remove this bottleneck 23 LOCAL CORRELATION TREATMENTS 163 23 LOCAL CORRELATION TREATMENTS 23 1 Introduction The local correlation program of MOLPRO can currently perform closed shell MP2 D MP3 D MP4 SDQ CISD QCISD and CCSD calculations So far only MP2 is publicly available Linear scaling versions of the other methods are still experimental and will be released later Perturbative energy corrections for triple excitations are under development and will also be available in the near
326. ng to ORBREL i i2 labell label2 Orbital i is related to orbital i2 by the sequence of operations defined by the label specifications defined previously using SYMELM The operators operate right to left Note that 7 and i2 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 29 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 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 averaged 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 29 THE VB PROGRAM CASVB 214 29 10 7 Freezing orbitals in the optimization FIXORB 1 i2 This command freezes the orbitals specified in the list i z to that of the starting guess Alternatively the special keywords ALL or NONE ma
327. nt nvirt virtual orbitals in each symmetry By de fault the orbitals are not printed unless the ORBPRINT option see section 17 8 1 is present or the global GPRINT ORBITALS see section 4 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 2 16 for information about dump records and how specific orbital sets can be requested in a later calculation 17 5 6 Pseudo canonical orbitals CANORB record options or 17 THE MCSCF PROGRAM MULTI 124 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 17 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 LOCALI 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 17 5 8 Diabatic orbitals In order to construct diabatic states it is n
328. nto MO basis NATORB Computes natural orbitals DIAG Diagonalizes 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 from input SET Assigns a value to a variable See the following subsections for explanations 35 1 Calling the matrix facility MATROP The program is called by the input card MATROP without further specifications MATROP 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 35 MATRIX OPERATIONS 263 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 35 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 LOAD command has slightly different opti
329. nts and hessians are computed This defaults to symmetric displacement coordinates and should normally not be modified These defaults can be modified using the COORD directive opt_space can be one of the following ZMAT Optimize all variables on which the z matrix depends default 3N Optimze all 3N cartesian coordinates Z Matrix input coordinates will be destroyed on this entry opt_coord can be one of the following NORMAL Optimization in local normal coordinates This is default if the Model Hes sian 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 plus additional information about this optimization is written to the specified file These coordinates resemble in part the valence coordinates used by vibrational spectroscopists and have the advantage of decreasing coupling between different modes This often increases the speed of convergence The use of this option is highly recom mended especially in minimization of large organic molecules with rings Nevertheless you should keep in mind that these coordinates are constructed automatically and there exist exotic bond structures which might not be treated properly e g weakly bonded spe
330. ny configuration parameters which you deem necessary For further information see any comments in the CONFIG 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 hard ware Note that if i4 is used large files gt 2Gb are supported on most systems but even then the sizes of MOLPRO records are restricted to 16 Gb since the internal addressing in MOLPRO uses 32 bit integers If i8 is used the record and file sizes are effectively unlimited 2 In the case of building for parallel execution the option mpp must be given on the command line 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 it can also be built with a GA library made with the LAPI target configure prompts for the type default t cgmsg and then for the directory holding the associated libraries Normally tcgmsg is recommended which is most efficient on most systems and also most
331. o manipulate orbitals In UHF only the B spin orbitals are rotated 14 THE SCF PROGRAM 87 14 5 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 C2 instead of C the 82 52 orbitals appear in symmetry 1 4 In other cases a linear geometry may occur as a special case of calculations in Cs symmetry and then one component of the Tr orbitals occurs in sym metry 1 A The program is able to detect such hidden extra symmetries by blockings in the one electron hamiltonian h 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 extra symmetries are ordered with increasing l 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 EM iyi 11272 If necessary the program will reorder the orbitals in each iteration to force this occupat
332. o that of other programs However cartesian 10 BASIS INPUT 72 functions can be requested using the CARTESIAN 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 10 2 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 1libmol and 1ib 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 1ilbmol 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 is libmol pprint eelement k key t type format where the parameters are print Output level 0 means list matching
333. 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 15 4 5 Grid caching SAVE NOSAVE NOSAVE disables the disk caching of the grid i e forces the recalculation of the grid each time it is needed SAVE forces the use of a grid cache where possible 15 4 6 Grid symmetry SYM NOSYM NOSYM switches off the use of symmetry in generating the integration grid whereas SYM forces the use of any point group symmetry 15 4 7 Grid printing PRINT PRINT 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 16 ORBITAL LOCALIZATION 113 16 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 method can be either BOYS or PIPEK By default the valence orbitals from the last energy calculation a
334. ometry separately in different records and variables 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 6 Variables can also be used for passing input parameters to the program This is useful for procedures which are described in Section 4 8 2 12 Multiple passes through the input It is 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 2 GENERAL PROGRAM STRUCTURE 10 Table 1 The symmetry generators for the point groups Generators Point group null card C i e no point group symmetry X orYorZ Cs XY C2 XYZ Ci X Y Cry XY Z Cah XZ YZ Do X Y Z Dan 2 13 Symmetry MOLPRO can use Abelian point group symmetry only For molecules with degenerate symme try an Abelian subgroup must be used e g Cz or Dap 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 specif
335. on 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 25 2 9 Examples The following input calculates SCF multipole moments for water 25 PROPERTIES AND EXPECTATION VALUES 189 h20 distributed multipole analysis geometry o0 h1 0 r h2 0 r h1 theta Z matrix geometry input r 1 ang bond length theta 104 bond angle basis 6 311g examples hf Ido scf calculation h20_dma com dma limit 4 results for total multipoles are 25 3 Mulliken population analysis 25 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 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 25 3 2 Defining the density matrix DENSITY DENS1TY 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 25 3 3 Popul
336. on should be stored the SAVE option on the LOCAL card can also be used cf section 23 6 23 5 2 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 direc tive If the START command 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 cal culation 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 LOCAL card can be used cf section 23 6 23 5 3 Defining orbital domains DOMAIN Normally 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 The selection criterion can be modified by the DOMSEL key see section 23 6 It is also possible to define the domains by hand using the DOMAIN directive DOMAIN orbital atoml atom2 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 centred at the giv
337. ons 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 35 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 0RB 35 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 35 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 35 2 4 Loading S 2 LOAD name SMH loads S where S is the overlap matrix in the AO basis The keyword SMH needs not to be given if name SMH 35 2 5 Loading the one electron hamiltonian LOAD name HO LOAD name H01 loads the one electron hamiltonian in the AO basis HO1 differs from HO by the addition of perturbations if present see sections 25 4 1 25 4 2 The keyword HO H01 needs not to be given if name H0 H01 The nuclear energy associated to HO or H01 is internally stor
338. oper oper2 oper3 5 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 18 4 14 Transition moment calculations TRANS readc1 readc2 oper oper2 opers 5 Instead of performing an energy calculation only calculate transition matrix elements between wavefunctions saved on records readcl 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 18 4 15 Saving the density matrix DM record ifil idip 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 18 4 16 Natural orbitals NATORB RECORD record ifil P RINT nprint CORE natcor 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
339. operties are not to be evaluated 25 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 Imax if a value is specified otherwise the value of Imax specified by the LIMIT 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 25 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 25 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 Aldert
340. orb ci save 3500 2 multi vb optg 30 SPIN ORBIT COUPLING 218 30 SPIN ORBIT COUPLING 30 1 Introduction Spin orbit matrix elements and eigenstates can be computed using either the Breit Pauli BP op 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 em 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 4 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 eigenval
341. orbitals equivalent to angle 90 These vectors must be supplied before by ORBITAL and MOVE or ADD directives 34 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 34 11 Saving the merged orbitals SAVE 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 34 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 34 ORBITAL MERGING 34 13 Examples 34 13 1 HF This example merges the orbitals of Hz and F xxx example for merg print orbitals basis rh2 1 4 rhf 300 basis vdz geometry x y F text F that 9 1 lpoce 3 le lrerbital 2 30 2 text H2 geomet ry x Vy H1 H2 H1 rh2 Hfrsorbital 2100 2 multii oce Z orbital 2101 2 text FH2 geometry F H1 F rhf H2 H1 rh2 F 180 merge or
342. 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 17 5 5 To print additional information at the end of the calculation use PRINT keyl key2 Printing is switched on for key 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 configuration 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 Warn
343. orbitals 1 1 and 2 1 The orbitals are read from record 2100 2 11 EFFECTIVE CORE POTENTIALS 78 10 7 Examples This shows the use of default basis sets for HO H20 basis V0Z f p R 0 95 ANG THETA 104 DEGREE geometry 0 H1 0 R H2 0 R H1 THETA hf Ido closed shell SCF examples h2o0_vqz_fp com This is equivalent to the explicit input form H20 R 0 95 ANG THETA 104 DEGREE geometry 0 H1 0 R H2 0 R H1 THETA basis spdf 0o vqz c sp h vqz c hf Ido closed shell SCF examples h2o0_vqz_fp_explicit com 11 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 which defines a pseudopotential for an atom specified either by a chemical symbol 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 11 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 pro
344. ored 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 25 2 1 Calling the DMA program DMA DMA This command initializes the DMA program 25 2 2 Specifying the density matrix DENSITY DENS1TY 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 25 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 25 PROPERTIES AND EXPECTATION VALUES 188 25 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 25 2 5 Omitting nuclear contributions NONUCLEAR NONUCLEAR The nuclear contributions to pr
345. orresponding to small eigenvalues of the overlap matrix op tion DELBAS 0 If the redundancies are exact i e if the overlap matrix for a domain has zero eigenvalues it is in principle irrelevant which function is deleted In practice however the selection sometimes influences the numerical stability On the other hand if the overlap matrix has very small but nonzero eigenvalues the computed energy slightly depends on which basis functions are eliminated We tried very hard to make the selection algorithm as robust as possible but pitfalls in certain cases cannot be fully excluded Problems with the redundancy elimination normally occur only for very small molecules 2 or 3 atoms with very small ba sis sets If difficulties are encountered it is recommended to use PRINT DOMAINS to obtain more detailed information about domains and redundant functions The default behaviour can be changed using the DELBAS DELEIG DELSHL TYPECHECK and THRLOC options Whenever possible the domains should be made rotationally invariant which can be achieved by eliminating shells of basis functions see DELSHL option Sometimes it may be necessary to modify the threshold THRLOC to obtain the desired result In order to avoid problems when rotational invariance is not strictly fulfilled we recommend always to use of the NOORIENT option in the geometry input for optimizations or frequency calculations The SAVE and START options should always be used to ke
346. osed or open shell calls internally contracted multireference perturbation theory calls closed shell MP2 program calls closed shell MP3 program calls closed shell MP4 program calls closed shell CISD program calls closed shell coupled cluster program calls closed shell Brueckner CCD program calls closed shell quadratic configuration interaction program calls spin unrestricted open shell coupled cluster program calls spin restricted open shell coupled cluster program calls determinant based full CI program calls orbital localization program calls orbital manipulation program Properties and wavefunction analysis 2 GENERAL PROGRAM STRUCTURE 17 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 PLOT calls orbital and density plot program IGLO PIGLO NMR call magnetic property programs Gradients and geometry optimization FORCES calls gradient program OPT calls geometry optimization 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 2 18 Default procedures For convenience of use MOLPRO provides a numbe
347. ot 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 35 MATRIX OPERATIONS 266 35 8 Transforming operators TRAN TRAN result Op C calculates result 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 35 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 35 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 print gt 1 also the eigenvectors are printed 35 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 35 12 Forming an outer product of two vectors OPRD OPRD result matrix orb1 orb2 factor Takes the column vectors v
348. owles 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 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 sym
349. p record for orbitals h2s diabl diab reforb Generate diabatic orbitals relative to referenc Secret ey noextra Dont use extra symmetries cirocc 9 2 closed 4 1 wf 18 2 0 state 2 11B1 and 1A2 states orbital diabatic lUse diabatic orbitals save savci ISave MRCI for displaced geometries el 1 energy 1 Save adiabatic energies e2 1 energy 2 ci trans savci savci Compute transition densities at R2 dm 7000 2 Save transition densities on this record ci trans savci refci Compute transition densities between R2 and R1 dm 7100 2 Save transition densities on this record ddr density 7000 2 7100 2 Densities for lt R2 R2 gt and lt R2 R1 gt orbital 3140 2 2140 2 lOrbitals for lt R2 R2 gt and lt R2 R1 gt nergy el i e2 i Adiabatic energies mixing 1 2 2 2 Compute mixing angle and diabatic energies mixci i mixangci 1 Mixing angle obtained from ci vectors only hideji 4 sahdai acd 1 iDiahatio energqiec ohtained from ci wectoare only 28 QUASI DIABATIZATION 204 This calculation produces the following results Diabatic energies for H2S obtained from ClI 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 co
350. pe mass 70 ITERATIONS 127 Keywords 14 KS 90 KS SCF 90 LABEL B2 LATTICE 69 LDA 108 LIBMOL 78 libmol 72 library 78 LIMIT 188 LINEAR 187 LINESEARCH 235 LOCAL 124 Local correlation 163 LOCALI 113 Localization space 114 LOCAO 113 LOCORB 124 loops 9 LQUANT 121 Macros in string variables 44 MASS Mass velocity 192 Matrix operations 262 MATROP 262 MAXDAV 142 MAXITER 88 130 41 217 290 MCSCF 6 MCSCF MEMORY 30 Memory allocation 9 MERGE 256 METHOD 231 MOLDEN 68 molpro 1 Molpro help 19 Molpro2000 Molpro2002 Molpro2002 6 283 Molpro98 285 molpro_basis MOVE 256 MP2 155 MP3 155 MP4 155 MPP 3 MPP systems B Mulliken analysis 189 MULTI 116 MULTI NACM 132 225 NACME 132 198 ATORB 123 143 159 NELEC 10 NOCASPROS 211 NOEXC 140 NOEXTRA 127 NOGPRINT 36 Non adiabatic coupling NONLINEAR 131 NONUCLEAR 188 NOORDER 115 NOPAIR 140 NOSAVE 112 NOS INGLE 140 Nos YM 112 NOSYMPROJ 213 NUMERICAL 225 235 Numerical gradients 225 NUMHES 235 c 12 5 14 LE SEEE 583 FFDIAG 114 FFSET 257 PEN 84 Z Q OOCOO0OC0 CO INDEX ORB 210 ORBIT 183 ORBITAL 2 84 113 115 122 136 T85 orbital localization 113 orbital manipul
351. per 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 ECP 1 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 ECP 1 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 number 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 11 EFFECTIVE CORE POTENTIALS 79 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 pseudopotenti
352. perator save non relativistic energy in variable enrel show individual contribution and their sum examples use douglas kroll one electron integrals ar2_rel com lrelativistic scf calculation save relativistic scf energy in variable e_dk show mass velocity and darwin contributions and their sum Ishow 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 4 PROGRAM CONTROL 39 Table 5 One electron operators and their components Generic Parity Components Description name OV 1 Overlap EKIN 1 Kinetic energy POT 1 potential energy DELT 1 delta function DEL4 1 At DARW 1 one electron Darwin term i e DELT 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 ZZ XY XZ YZ second moments TM 1 XXX XXY XXZ XYY XYZ XZZ YYY YYZ YZZ ZZZ third moments MLTPn 1 all unique cartesian products of order n multipole moments OM 1 OMXX OMY Y OMZZ OMXY OMXZ OMYZ quadrupole moments and R OMRR 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 electric field gradients DMS 1 DMSXX DMSYX DMS ZX DMSXY DMSYY
353. pole terms while the simple scheme won t Valid options are option 0 Always use simple truncation 23 LOCAL CORRELATION TREATMENTS 180 DAMP order option 1 Use simple truncation for bipolar expansions but exhaus tive truncation for monopolar expansions This significantly improves convergence when a Taylor expansion is used but also accelerates the onset of divergent behaviour for large expansion levels and should therefore always be used in connection with a damping function for multipole operators option 2 Use simple truncation for monopolar expansions but ex haustive truncation for bipolar expansions including distant pair treat ment option 3 Always use exhaustive truncation The latter two options are not useful and only included for the sake of completeness Default 0 Specifies the form of a damping function that is applied to orbitals before the transformation of the multipole operators order is actu ally the order of a Taylor series that is part of the function and that mimics the behaviour of the monopolar multipole expansion Only multiples of 4 are reasonable values Higher values mean a harder damping function i e less damping at short distances but more at long distances A value of 0 disables the damping A value of 8 is recommended when using a Taylor expansion Default 0 SCALEDAMP scalefactor The damping function can additionally be scaled by the value of Stuff for debugging PATREN option scalef
354. presentation 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 93 343 366 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 Prog Theor Chem Phys in press All publications resulting from use of this program must acknowledge some or all of the above For an up to date bibliography see http rs2 ch liv ac uk dlc CASVB html 29 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 17 10 b Definition of the 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 everyth
355. 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 18 THE CI PROGRAM U n II U 1 ll o FU 1 II Q00000000000 U 1 I N Q Q 1 H TOT D GSS 0 GSS 1 DPQ EPO HPO DPI DSS DSI LOG CC 0 CC 1 DEN 1 DEN 2 DEN 3 DEN 4 GAM 1 GAM 2 GAM 3 PAIRS 0 PAIRS 1 CORE 0 CORE 1 REF 0 147 print paging information for CIGPQ print matrices GP at exit of CIGPQ print paging 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 Q print coupling coefficients B P Q print coupling coefficients y P Q print coupling coefficients for pair internal interactions not yet used 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
356. 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 012 ONact 18 assumed 29 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 29 5 Recovering CASSCF CI vector and VB wavefunction The appropriate MOLPRO records may be specified explicitly using the START directive note however that use of the vbdump mechanism described in section 29 2 1 is preferable whenever possible 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 17 5 4 A default value for ci is determined from the most recent vbdump record s 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 CASVB calculation If the VB wavefunction was previously saved in the AO basis the orbitals will be projected onto the present a
357. quent 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 reforb 2141 2 26 DIABATIC ORBITALS 197 H2S diabatic A states basis VDZ luse cc pVDZ basis set geometry 1x 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 Iglobal 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 orbital 2140 2 Isave orbitals to 2140 2 examples 2140 2 gt reforb 0 h2s_diab com 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 Iset 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 Isave new orbitals to record diab reforb compute diabatic orbitals using reference orbitals stored on record reforb reforb 2141 2 set variable reforb to the new orbitals enddo 27 NON ADIABATIC COUPLING MATRIX ELEMENTS 198 27 NON ADIABATIC COUPLING
358. r 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 default 5 and the interpolation type i e the subspace in which the GDIIS interpolation is made METHOD DIIS number type type may be GRAD interpolation using the gradients default working good for rigid molecules STEP interpolation using Quasi Newton steps which could be advantageous in dealing with very floppy molecules ENER inter polation using energies which is an intermediate between the above two QSD Quadratic steepest descent method of Sun and Ruedenberg SRMIN Old version of OSD For transition state searches invoked with the ROOT option see section 32 2 6 key can be 32 GEOMETRY OPTIMIZATION 232 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 espe cially 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 conve
359. r of default procedures for standard ap plications Each procedure performs a full calculation and prints a summary of the computed energies at the end of the output It is possible to call several procedures one after the other in the same job The procedures are provided in the file molproi rc which is automatically included at the begin ning of each input when the unix molpro command is used Inclusion of the procedures can be avoided using themolpro floption The user can also define his own procedures in his molproi rc file For details see section 4 8 2 18 1 Procedures for energy calculations The wavefunction symmetry and spin can be modified using the variables SYMMETRY and SP IN respectively The number of electrons can be modified using the variable NELEC SCF calculations are only done if no orbitals are available or if the symmetry or spin changed since the last SCF optimization In CASPT and MRCI calculations the CASSCF is only done if the last optimized orbitals are not of MCSCF type runscf Performs SCF calculation rundft Performs DFT calculation The functional can be specified using el ther the FUNCTIONAL or DFTNAME variable default B3LYP runmp2 Performs SCF and MP2 calculations runmp3 Performs SCF MP2 and MP3 calculations runmp4 Performs SCF and MP2 MP4 calculations 2 GENERAL PROGRAM STRUCTURE 18 runmp4sdq Same as runmp4 but without triples runccsd Performs SCF and CCSD calculations U
360. r the following geometry and basis set geometry o0 h1 0 r h2 0 r h1 theta Z matrix geometry input r 1 ang bond length theta 104 bond angle basis vdz basis set thresh 1 d 8 convergence threshold the result could look as follows 35 MATRIX OPERATIONS SCF has converged in 24 iterations E H 3 El 0000000 000000000000000000o0owz Io ol ZJ O R 00 ia i si Y Ae WwW Y JJ J aa a al 0214 0214 0214 0214 0214 0214 0214 0214 0214 AANA ONBPWNHRrF DMO WMAAIADADH PWN FP N NM NW Nr OW N Ww Lhe eg I OE VL Al s A Tee r E Te ie lle al bar Sia ee a Re ee ES t AE a ESS a E E E j AAAA A eA A O o a O O O AY AV N A 0214 It does not converge terribly fast 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 S 0050 00 190 O10 070 OS OO OS 0 E O DIP 17407361 06209922 10199751 79658706 43669203 17616098 05644998 63401784 91637513 76319435 86107911 80513445 83990621 81956198 83202128 82464809 82912805 82646089 82807428 82711046 82769196 82734386 32755355 82742787 271 A INSTALLATION OF MOLPRO 272 A Installation of MOLPRO A 1 Obtaining the distribution materials MOLPRO is distributed to licensees on a self service basis using the
361. ration 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 generalized 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 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 59 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 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 2 0 use fully dir
362. rches mstep n Calculate numerical hessian after each n optimization steps This is useful for difficult transition state optimizations e g 1f the eigenvalue structure of the hessian changes during the optimization type defines the finite differences to be used type 0 Use forward differences default type 1 Use the more accurate central differences icalc defines in which way the Hessian matrix shall be recalculated icalc 0 Static regeneration Recalculate complete Hessian matrix numerically after each mstep optimization steps default icalc 1 Partial regeneration Recalculate selected Hessian matrix elements if the rel ative deviation of this element before and after update see UPDATE section is larger than thresh If thresh is not specified a default value of thresh 0 05 1 e a maximum deviation of 5 is used icalc 2 Dynamic regeneration 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 thresh 0 5 a u is used Note that the static regeneration of the complete Hessian matrix after mstep iterations is not disabled if the partial icalc 1 or dynamical icalc 2 regeneration is used i e if you want to use only the partial or dynamical regeneration you should set mstep to zero If cartesian coordinates are used the molecular symmetry can be used to shorten the number of gradient and or energy
363. re 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 16 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 16 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 16 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 16 4 Delocalization of orbitals DELOCAL DELOCAL If this card is present the orbitals are delocalized 16 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 16 ORBITAL LOCALIZATION 114 16 6 Selecting the orbital space By default only the valence orbitals are
364. rge 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 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 don
365. rgence than the RF or DIIS method The hessian recalculation safeguard may be turned off using the METHOD OSD NOHESS input card SRTRANS Old version of OSD For reaction path following the input key is OSDPATH Quadratic Steepest Descent reaction path following This methods deter mines reaction paths intrinsic reaction coordinates IRCs by following the exact steepest descent lines of subsequent quadratic approximations 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 steepest descent lines is exceeded the hessian is recalculated numerically see OPTION section 32 2 17 For details see J Sun and K Ruedenberg J Chem Phys 99 5269 1993 It is also possible to recalculate the hessian after each m steps using the NUMHES m command see section 32 2 15 If the hessian matrix is recalculated in every optimiza tion 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 se
366. rgy calculation which produced the input orbitals 16 7 4 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 2 16 16 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 16 7 16 9 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 17 THE MCSCF PROGRAM MULTI 116 17 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 Kn
367. riable DF TF UNS the total is in DFTFUN and the corresponding individual functional names in DF TNAME Energy gradients are available for self consistent Kohn Sham calculations 15 1 Density Functionals In the following Pa and pg are the o and spin densities the total spin density is p The gradients of the density enter through Cua VPa VPa pp VPp VPp Sap OBa VPu Vos O Cua Opp 20 9f 1 V Oaa y OBB Pa P Va V Pa Vp V Pp V Va p 3 Additionally the kinetic energy density for a set of Kohn Sham orbitals generating the density can be introduced through a B Ta Y Ivo tp Vo T Ta HT 4 All of the available functionals are of the general form F Ps Ps Oss O55 Osz Ts Ts Vs Us park Ps Ps Oss Oss Oss Ts Ts Us Us 5 where 5 is the conjugate spin to s 15 THE DENSITY FUNCTIONAL PROGRAM 91 15 1 1 B86 Xa y A D Becke J Chem Phys 84 4524 1986 Divergence free semiempirical gradient corrected exchange energy functional 4 3 K z 6 E 1 YXs where i 3 3 Ea 7 o 5 Se B 0 0076 8 and y 0 004 9 15 1 2 B86MGC Xa y with Modified Gradient Correction A D Becke J Chem Phys 85 7184 1986 B86 with modified gradient correction for large density gradients 274 3 4 3 BXSPs K CPs 10 2 1 Aay2 4 where te 3 3 E 11 a 5 ve B 0 00375 12 and A 0 007 13 15 1 3 B86R Xa y Re optimised A D Becke J Chem Phys 107
368. rimary 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 states of different symmetry 17 2 Defining the orbital subspaces 17 2 1 Occupied orbitals OCC n1 n2 Ng ni specifies numbers of occupied orbitals including CORE and CLOSED in irreducible repre sentation 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 17 2 2 Frozen core orbitals CORE N1 N2 record file ni is the number of frozen core orbitals in irrep number i These orbitals are doubly occupied in all configurations and not optimized 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 n It should always be given if the orbital guess is from a 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 CORE card is omitted then the numbers of core orbitals are taken from the most re
369. ring make install ie make DEFAULT 1 install then a symbolic link is made to INSTBIN name Furthermore If the file INSTBIN molpro does not already exist or if the variable DEFAULT is set to molpro during make install thenasymbolic link is made from INSTBIN name to 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 When the program has been verified and or installed the command make clean can be used to remove compilation logs make 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 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
370. rmally 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 i 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 most 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 larg
371. rmation is stored An exception is the un usual case that several different CPMCSCF calculations have been formed in a previous MCSCF calculation In this case the SAMC directive must be used to select the desired record 31 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 B2 2 19 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 together are evaluated at exactly the same nuclear geometry otherwise wrong results could result due to unintended rotations of the system 31 1 2 Scaling gradients SCALE SCALE factor 31 ENERGY GRADIENTS 224 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 32 2 19 31 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
372. ro 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 4 5 Allocating dynamic memory MEMORY MEMORY n scale Sets the limit on dynamic memory to n 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 4 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 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 2 23 330 Oy Dr Oy 6 0 ngstr m The same could be achieved as follows RVEC 1 0 1 2 1 4 1 6 1 872 0 2 5 3 0 4 0 5 0 6 0 ANG DO I 1 RVEC R RVEC 1 ENDDO Up to 20 DO loops may be nested Each DO must end with its own ENDDO Jump
373. rompts for them These library paths are 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 molpro_procname exe If the ARCH option is not given the last one configured will be generated In addition a file molpro_ 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 molpro_procname rc input More conveniently one can set the Unix environment variable MOLPRO_RCFILE to molpro_ procname xc and then simply use molpro without an option The recommended mechanism is to set the environment variable MOLPRO_RCFILE 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 p3 p 4 and or athlon and then link using make MPPLIB libname where libname can be tcgmsg mpi or myrinet The ARCH and MPPLIB options can be combined e g make MPPLIB libname ARCH procname and this will generate the executable molpro_procname_libname exe and the default file molpro_proc
374. roperties will have been significantly in error There was a programming error in the transformation of gradients from cartesian to inter nal coordinates which in some cases resulted in slow convergence of geometry optimiza tions The error is now fixed Vibrational frequencies formerly by default used average 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 multireference 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 B RECENT CHANGES 285 8 Linear scaling CPU and memory LMP2 as described by 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 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 file card Also if there are unjustified messages coming up in very la
375. roperty 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 computed directly in the SCF and MCSCF programs but in this case no orbital contributions are printed program 25 PROPERTIES AND EXPECTATION VALUES 187 h2o properties geometry o0 h1l o0 r h2 0 r hl1 theta Z matrix geometry input r 1 ang bond length theta 104 bond angle gexpec dm qm Iglobal request of dipole and quadrupole moments hf Ido scf calculation examples multi state 2 do full valence CASSCF h2o_gexpecl com natorb state 1 1 compute natural orbitals for state 1 1 natorb state 2 1 compute natural orbitals for state 2 1 25 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 st
376. rrection 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 DIAB 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 ClI vectors R El E2 H11CI H22 I H2 TEI 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 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 an
377. rther details regarding the calculation of weights in CASVB see T Thorsteinsson and D L Cooper J Math Chem 23 105 26 1998 VBWEIGHTS key1 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 29 11 3 Printing weights of the CASSCF wavefunction in the VB basis For further details regarding the calculation of weights in CASVB see T Thorsteinsson and D L Cooper J Math Chem 23 105 26 1998 CIWEIGHTS keyl key2 Neonfl 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 29 THE VB PROGRAM CASVB 216 given
378. rting 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 It is 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 BASTS blocks respectively 9 1 Sorted integrals By default two electron integrals are evaluated once and stored on disk This behaviour may be overridden by using the input command gdirect see section 8 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
379. rueck Iteration increment between orbital updates default 1 brsfak Scaling factor for singles in orbital updates default 1 21 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 21 5 The DIIS directive DITIS itedis incdis maxdis itydis This directive allows to modify the DUS 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 21 THE CLOSED SHELL CCSD PROGRAM 158 21 6 Examples 21 6 1 Single reference correlation treatments for H O seh h20 test memory 1 m lallocate 1 MW dynamic memory geometry o0 h1 0 r h2 0 r h1 theta Z matrix geometry input basis vtz cc pVTZ basis set r 1 ang bond length theta 104 bond angle hf Ido scf calculation text examples for single reference correlation treatments ci CI
380. 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 Nn2 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 machines html If after due efforts to fix problems of a local origin the problem cannot be resolved the developers of MOLPRO would appreci ate receiving a report There i
381. s 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 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 2 GENERAL PROGRAM STRUCTURE 5 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 2 GENERAL PROGRAM STRUCTURE 6 2 3 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 wavefunction file optional gprint options Iglobal print options optional gthresh
382. s 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 A INSTALLATION OF MOLPRO 280 make MPPLIB libname ARCH procname install into the same INSTBIN and INSTLIB directories but note that the INSTLIB directories must be distinct for i4 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 if the variable DEFAULT is set du
383. s can be selected using specifications as explained in section 2 16 Note that the density matrices are stored in the same record as the orbitals 25 1 3 Orbital analysis ORBITAL ORBITAL record file specifications If this card is present the orbitals are read from record record file and an 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 25 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 4 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 centre 0 you need not read in coordinates XYZ cartesian coordinates of the point only if centre 0 25 PROPERTIES AND factor EXPECTATION VALUES 186 the operator is multiplied by this factor The default is factor 1 except for REL In this cases proper factors for relativi
384. s 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 supercedes 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 Le 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 p
385. s 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 4 PROGRAM CONTROL 4 6 1 Examples for do loops 31 The first example shows how to compute a potential energy surface for water H20 potential geomet ry x o h1l o r1 i h2 0 r2 1 h1 theta i basis vdz angles 100 104 110 distances 1 6 1 7 1 8 1 9 2 0 i 0 do ith 1 angles do irl 1 distances do ir2 1 irl i i 1 r1 i distances irl r2 i distances ir2 theta 1 angles ith HE escf i energy ccsd t eccsd i energc eccsdt i energy enddo enddo enddo table r1 r2 theta escf eccsd eccsdt head rl r2 theta scf ccsd ccsd t save h2o tab title Results for H20 basis S basis SOLEIL Z luse cs symmetry z matrix define basis set list of angles list of distances linitialize a counter loop over all angles H1 0 H2 loop over distances for O H1 loop over O H2 distances r1l ge r2 lincrement counter Isave rl for this geometry Isave r2 for this geometry save theta for this geometry Ido SCF calculation Isave scf energy for this geometry Ido CCSD T calculation save CCSD energy Isave CCSD T energy lend of do loop ith lend of do loop irl lend of do loop ir2 produce a table with results modify column headers for table save the table in file h2o ta
386. s of R from Angstrom 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 and A 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 i 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 st ring variables 6 7 Special variables 6 7 1 Variables set 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 KELVIN 1 d0 3 157733d5 HARTREE JOULE 1 d0 2625 500d0 HARTREE CAL 1 d0 627 5096d0 HARTREE
387. s per REAL 8 word IVECT 0 scalar machine IVECT 1 vector machine MINVEC call MXMB in coupling coefficient routines if vector length larger than this value IBANK number of memory banks for vector machines If IBANK gt 1 vector strides which are multiples of IBANK are avoided where appropriate LTRACK number of REAL 8 words per track or block for file alloca tion LTR determines how matrices are stored on disc If LTR LSEG all matrices start at sector 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 VO if much paging is necessary but reduce I O if everything fits in core NCPUS Maximum number of CPUs to be used in multitasking 18 5 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 18 THE CI PROGRAM 18 6 Print options 146 print CI coefficients which are larger than this value omit two electron integrals which are smaller than this value
388. s the maximum number of iterations in the second order optimizations Default is Niter 50 29 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 29 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 29 THE VB PROGRAM CASVB 212 29 9 5 Defining several optimizations More than one optimization may be performed in the same CASVB deck by the use of OPTIM keywords OPTIM 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 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 29 9 6
389. s to make bonding and antibonding linear combinations ROT 4 1 6 1 45 rotate 2pz orbitals to make bonding and antibonding linear combinations PRINT 1 set print option ORTH 6 2 2 symmetrically orthonormalize the valence orbitals the resulting orbitals are printed PROJ 2100 2 Project valence orbitals out of scf orbitals of the molecule and add virtual orbital set SAVE 2150 2 save merged orbitals to record 2150 on file 2 dummy remove dummies multi occ 6 2 2 perform full valence casscf for NO wf 15 2 1 2Pi state wf 15 3 1 2Pi state start 2150 2 start with merged orbitals 35 MATRIX OPERATIONS 262 35 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 i
390. sed 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 25 PROPERTIES AND EXPECTATION VALUES 194 25 6 4 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 25 6 5 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 25 6 6 Format of cube file The formatted cube file contains the following records A job title A brief description of the file contents 15 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
391. sed to show several variables more easily SHOW qm dm shows all variables whose names begin with QM 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 6 9 Clearing variables Variables can be deleted using CLEAR namel name 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 7 TABLES AND PLOTTING 7 1 Tables Variables can be printed in Table form using the command TABLE varl var2 The values of each variable 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 i THETA i ECI 1 for each 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
392. sent 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 CONFIG card see section 17 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 17 THE MCSCF PROGRAM MULTI 123 17 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 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
393. ser environment 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 systems 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
394. ses RCCSD in open shell cases runccsdt Performs SCF and CCSD T calculations Uses RCCSD T in open shell cases runuccsd Performs SCF and CCSD calculations Uses UCCSD in open shell cases runuccsdt Performs SCF and CCSD T calculations Uses UCCSD T in open shell cases runbccd Performs SCF and BCCD calculations closed shell only runbecdt Performs SCF and BCCD T calculations closed shell only runqcisd Performs SCF and QCISD calculations closed shell only runqcisdt Performs SCF and QCISD T calculations closed shell only runcas Performs SCF and CASSCF calculation The wavefunctions for state averaged CASSCF calculations can be defined using the variables SYMMETRY SPIN STATE WEIGHT and NELEC caspt2 Performs SCF CASSCF and CASPT2 calculations In case of state averaged CASSCF reference functions see runcas the CASPT2 is performed for each state separately caspt3 As CASPT2 but also computes third order energy runcaspt2 Same as caspt2 runcaspt3 Same as caspt3 runmrci Performs SCF CASSCF and MRCI calculations In case of state averaged CASSCF reference functions see runcas the MRCI is performed for each state symmetry separately Several states in the same symmetry are treated simultaneously Transition moments are automatically computed between all states However these are not printed in the summary at the end of the output and must be extracted from the output or punch file runacpf Performs
395. sities on this record Compute transition densities between R2 DR j Save transition densities on this record Compute transition densities between R and R2 DR j Lasa Franeition deneitias on thie record and R1 28 QUASI DIABATIZATION 206 The calculation produces the following table Mixing angles and non adiabatic coupling matrix elements for H2S R MIXCI MIXTOT DCHI NACMECI 2 009 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 NACMECT and by differentiating the mixing angle DCHD are in close agreement 29 THE VB PROGRAM CASVB 207 29 THE VB PROGRAM CASVB CASVB is a general program for valence bond calculations written by T Thorsteinsson and D L Cooper 1996 2000 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 re
396. 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 2 13 14 1 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 WF card 14 THE SCF PROGRAM 84 14 1 2 Specifying closed shell orbitals CLOSED A1 M2 Mg 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 an OPEN card 14 1 3 Specifying open shell orbitals OPEN orb sym orbz symz2 0rby syMn 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 if 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 neg
397. start 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 symmetry 1 Orbitals which have been moved 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 34 5 Defining offsets in the output set OF FSET OFFSET iof iof2 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
398. stic corrections are used unless factor is given The two commas before factor are needed to preserve compatibility with Molpro 96 25 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 printed 25 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 h20 properties geometry o0 h1 0 r h2 0 r h1 theta r 1 ang theta 104 hf property orbital density dm am multi state 2 natorb state 1 1 natorb state 2 1 property orbital state 1 1 density state 1 1 dm qm property orbital state 2 1 density state 2 1 dm qm Z matrix geometry input bond length bond angle Ido scf calculation call property program lread scf orbitals lread scf density matrix compute dipole moments and print orbital contributions compute quadrupole moments and print orbital contributi Ido full valence CASSCF compute natural orbitals for state 1 1 compute natural orbitals for state 2 1 examples call property program h20_property com 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 contributi call p
399. 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 Koppel 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 Below we present an example for the first two excited states of H2S which have B and A2 symmetry in Cz and A symmetry in Cs We first perform a reference calculation in Cz 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 reforb 2141 2 17 THE MCSCF PROGRAM MULTI 126 H2S diabatic A states basis VDZ luse cc pVDZ basis set geometry 1x luse Cs symmetry planeyz fix orientation of the molecule noorient
400. t a formatted file will be written otherwise unformatted fortran i o will be used 11 12 3 specify the number of grid points in each of three dimensions If not specified sensible defaults are chosen 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 25 6 1 DENSITY source of density DENSITY density source GRADIENT density source LAPLACIAN density source Compute the density and optionally its gradient and laplacian lt density source gt 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 25 6 2 ORBITAL source of orbitals ORBITAL orbital list orbital source lt orbital list gt 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 lt orbital source gt 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 25 7 25 6 3 AXIS direction of grid axes AXIS x y Z x y z specify the unnormali
401. t 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 are computed all values are stored in the respective variable arrays with the bra states running fastest 30 4 Calculation and diagonalization of the entire SO matrix HLSMAT type record1 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 30 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
402. t 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 zero 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 Enables the use of asymmetric domains for distant pairs The asym metric domain approximation supplements the multipole approxima tion 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 improved 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
403. t 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 UHF NAT 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 orCCSD 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 7 Ger 2 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 calculations 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
404. tar gz together possibly with one or more module archives with file names of the form molpro module 2002 6 tar gz The modules contain code which is not generally distributed or features which are not always required to install the code An example of the former is the program developers kit module deve lop an example of the latter is the documentation module doc 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 al1 2002 6 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 molpro2002 6 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 CONFIG This file contains machine dependent parameters such as com piler options Normally CONFIG will not need changing but you should at the least examine it and change a
405. tate 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 17 THE MCSCF PROGRAM MULTI 133 17 10 Optimizing valence 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 29 Energy based opti mization of the VB parameters is the default and the output level for the main CASVB iterations is reduced to 1 17 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
406. tates examples set symmetry 2 1 remove Piy oh_runmrci2 com runmrci MRCI for 2Pix and 2Sigma This produces RESULTS METHOD STATE S ENERGY DIPX DIPY DIPZ CASSCF T2 0 5 75 41331789 0 0 0 0 0 67158730 CASSCF 1 3 0 5 75 41331789 0 0 0 0 0 67158730 CASSCF Lal DUES 75 24125256 0 0 0 0 0 69975340 MRCI 182 0 5 75 55518444 0 0 0 0 0 66457191 MRCI D 1 2 0 5 75 56014871 0 0 0 0 0 66457191 MRCI P Lei 0 5 75 55853208 0 0 0 0 0 66457191 MRCI 1 51 0 5 75 39442202 0 0 0 0 0 70484623 MRCI D IsI 0 5 75 40040680 0 0 0 0 0 70484623 MRCI P TeL 05 75 39846312 0 0 0 0 0 70484623 You may want to extend the active space to include the 27 orbitals This can be achieved by setting the variable OCC OH geomet ry 0 H 0 1 83 Geometry definition set symmetry 2 3 1 spin 1 2Pix 2Piy and 2Sigma states occ 4 2 2 4 sigma and 2 pi occupied les runcas SA CASSCF for all three states a ace oh_runmrci3 com set symmetry 2 1 remove Piy runmrci MRCI for 2Pix and 2Sigma For accurate calculations of the electronic transition moment also the 16 orbitals contribute significantly These are in symmetry 1 6 2_ 2 and 4 6 In order to force the 5a orbital to become the 5 we must use the SYM directive in the SCF calculation Since it is not possible to insert this into the procedure we must write the SCF input explicitly OH basis avqz Use aug cc pVQZ basis geomet ry 0 H 0 1 83
407. te symmetry n2_xasscf com restrict Y Li LL Restriction to singles and doubles restrtidt 1 l 33 Dd bli Take out singles ee natorb ci print Print configurations Print natural orbitals and CI coeffs 18 THE CI PROGRAM 135 18 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 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 18 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 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
408. ternally TRUE is stored 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 2 GENERAL PROGRAM STRUCTURE expr LE expr expr NE expr expr expr expr expr expr expr expr Less Equal Not Equal Exponentiation Exponentiation Parenthesis no effect Change sign Keep sign no effect 2 5 Intrinsic functions Expressions may contain the following intrinsic functions ABS expr MAX expr expr MIN expr expr EXP expr LOG expr LOG10 expr SORT expr NINT expr INT expr SIN expr COS expr TAN expr ASIN expr ACOS expr ATAN expr COSH expr SINH expr TANH expr MOD exprl expr2 Absolute value Largest value of arbitrary number of numbers or expressions Smallest value of arbitrary number of numbers of expressions Exponential Natural Logarithm Common Logarithm Square Root Next nearest integer Truncate to integer Sine Cosine Tangent Arcsine Arccosine Arctangent Hyperbolic cosine Hyperbolic sine Hyperbolic tangent Remainder
409. the LMP2 program mp2 multp Local MP2 with multipole approximation for distant pairs Notice that for small systems a local MP2 with these default options might take more time than a conventional MP2 There are two reasons for this firstly the default options in the LMP2 case are chosen so that the memory and disk requirements are minimized This requires the evalu ation of each unique integral twice in the transformation Secondly the LMP2 equations have to be solved iteratively which takes some additional time It is possible to speed up local MP2 calculations for small or medium size molecules using the following option on the GDIRECT card GDIRECT PAGE 1 In this case each unique integral is computed only once but intermediate storage of the half transformed integrals on disk is required The disk requirements scale cubically in this case and therefore this option cannot be used for very large cases The CPU time depends sensitively on the prescreening thresholds used in the transformation Their choice in turn depends on the required accuracy The default thresholds 1 d 7 normally ensure that absolute numerical errors in the energy are around 1 uH For many applications lower accuracy is sufficient and a change of the default values can achieved be requesting a less accurate energy GTHRESH ENERGY 1 d 5 This card must be given before the MP2 directive and will then increase the prescreening thresh olds to 1 d 6 note that
410. the ORBITAL card is not needed because the location of the or bitals is stored in the MCSCF dump record 31 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 31 1 5 State averaged MCSCF gradients SAMC Normally no further input is required for computing gradients for state averaged MCSCF Note however that a CPMCSCF GRAD state directive is required in the SA MCSCE calcu lation see CPMCSCF The gradients are then computed automatically for the 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 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 CPMCSCF is passed to the FORCE program in a certain records default 5101 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 direc
411. tion card STATE Number of states for this wavefunction symmetry WEIGHT Weights of states For the exact definition of these cards see sections 17 2 and 17 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 29 4 Defining the valence bond wavefunction 29 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 29 THE VB PROGRAM CASVB 209 configurations are defined in terms of the active orbitals only and may be specified using one or more CON cards note that the RESTRICT and SELECT keywords are not used in CASVB CON N1 12 N3 N4 5 The configurations can be specified by occupation numbers as in section 13 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
412. tion of the hessian the symmetry of the molecule may be low ered Giving SYMM AUTO the program uses the maximum possible symmetry of the molecular wavefunction in each energy gradient calculation and this option therefore minimizes the com putational effort With SYMM NO no symmetry is used during the frequency calculation de fault For single 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 18 3 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 C 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 r
413. tions are possible as follows P examples h20_scf_vtz com examples h20_scf_vtz_explicit con 10 BASIS INPUT 74 The maximum angular momentum in the basis set can be reduced using syntax such as BASIS VQZ D which would omit the f and g functions that would normally be present in the VQZ basis set BASIS VQZ D P would specify additionally a maximum angular momentum of 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 3s2pl1d 2s generates a 6 31G sized basis set from the Roos ANO compilation 10 4 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
414. tions if DELBAS gt 0 DELSHL idlshl TYPECHECK typechk DELE 1 G idleig This parameter determines if whole shells of basis functions i e all p functions for a given exponent at one atom should be simultane ously eliminated This may be useful in order to guarantee rotational invariance in geometry optimizations and frequency calculations idlshl 1 eliminate as many functions of a shell simultaneously as possible but never more than determined by small eigenvalue of the overlap matrix default idlshl 2 as idlshl 1 but also eliminate functions with identical norm simultaneously idlshl 3 eliminate all functions of a shell simultaneously even if a larger number of functions is deleted than determined by small eigen values of the overlap matrix This must be used with care since very poor energies may sometimes result idlshl 4 as idlshl 3 but also eliminate functions with identical norm simultaneously idlshl gt 4 as idlshl 4 but equivalent functions centred at all sym metry equivalent atoms are considered to form a shell not recom mended If nonzero activates basis function type restrictions in redundancy check For a given atom only basis functions corresponding to occu pied atomic orbitals are allowed to be deleted For instance on first row atoms at most two s functions and one p shell will be deleted No functions are deleted from hydrogen or He atoms This option determines how redundant basis functions
415. tive 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 cpmcescf nacm 1 1 2 1 save 5101 1 Ido cpmcscf for coupling of states 1 1 2 1 31 ENERGY GRADIENTS 225 cpmcescf nacm 1 1 3 1 save 5102 1 Ido cpmcscf for coupling of states 1 1 3 1 cpmescf nacm 2 1 3 1 save 5103 1 Ido 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 force samc 5103 1 compute NACME for states 2 1 3 1 See also test job 11f_nacme test 31 1 6 Non adiabatic coupling matrix elements NACM see SAMC 31 1 7 Difference gradients for SA MCSCF DEMC see SAMC 31 1 8 Example Calculate SCF Gradients for Water alpha 104 degree set geometry parameters r 1 ang geomet ry 0 define z matrix HL 0 ay H2 0 xr H1 alpha basis vdz basis set hf rdo scf examples forces compute gradient for SCF h2o_forces com mp2 mp2 calculation forces mp2 gradients multi casscf calculation forces casscf gradient 31 2 Numerical gradients It is possible to compute gradients by finite differences using FORCE NUMERICAL STARTCMD command where command is the first command in the input needed for the current energy calculation The command must be found in the input before the FORCE card 1 e
416. tive all records between rec and abs rec2 are read All configurations found in this way are merged 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 17 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 n n2 etc are the occupation numbers of the active orbitals 0 1 or 2 For example for OCC 5 2 2 CLOSED 2 1 1 n is the occupation of orbital 3 1 number sym n2 is the occupation of orbital 4 1 n3 of 5 1 ng of 2 2 and ns of 2 3 Any number of CON cards may follow each other Example for the BH molecule OCC Le Lol four sigma one pi orbitals are occupied CORE 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 2sigma 2 lpi_x
417. to avoid confusion with program specific PRINT cards The syntax is GPRINT key1 valuel key2 value2 NOGPRINT key1 key2 Normally value can be omitted but values gt O 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 Print basis information DISTANCE Print bond distances default ANGLES Print bond angle information default If gt O dihedral angles are also printed ORBITAL Print orbitals in SCF and MCSCF CIVECTOR Print CI vector in MCSCF PAIRS Print pair listin CI CCSD CS Print information for singles in CI CCSD CP Print information for pairs in CI CCSD REF Print reference CSFs and their coefficients in CI PSPACE Print p space configurations MICRO Print microiterations in MCSCF and CI CPU Print detailed CPU information IO Print detailed 1 O information VARIABLE Print variables each time they are set or changed default 4 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 the GEXPEC card expectation values are
418. 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 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 2 2 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 The input is read sequentially by 2 GENERAL PROGRAM STRUCTURE 4 a controlling program when the controlling program calls a program module this module con tinues to read the input file until it finds an unknown keyword After the module has performed its function control is returned to the controller 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 You may type your input upper or lower case The input processor will conv
419. 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 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 OF FSET 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 34 3 Adding orbitals to the output set ADD ADD orb1 syml orb2 sym2 orb3 sym3 iofffac istart iend This adds orbitals orbl symI 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 34 ORBITAL MERGING 257 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 34 4 Defining extra symmetries EXTRA EXTRA exsym orb1 syml orb2 sym2 orb3 sym3 ioff fac i
420. to run pat cher before running configure with make C utilities patcher bootstrap A INSTALLATION OF MOLPRO 281 To use the patcher program in the top level directory issue the command patcher apply revert list cache directory user password url local verbose no action patchl patch2 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 l 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 specify your username and password through command line options or else the program will prompt for them They are then remembered in the file CONFIG in the cache directory In case of problems first consult the file pat cher 1log which contains the output from indi vidual patch applications and r
421. ty matrix is computed an stored in record record on file ifil This currently works for closed shell MP2 and QCISD See also NATORB 21 8 Natural orbitals NATORB RECORD Jrecord ifil PRINT nprint CORE natcor l 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 natcor 1 the core orbitals are frozen by excluding them from the natural orbital transformation 21 9 Excited states using linear response CCSD LR EOM CCSD Excitation energies can be computed using linear response LR theory also called equation of motion EOM approach Accurate results can only be expected for singly excited states The states to be computed are
422. uate the CPMCSCF for USG always as last 111 skip the DC evaluation 1f the Cl involves states with different spin eg a Singlet Triplet crossing because the vector would be identically zero Three sets of FORCE cards only two for Singlet Triplet CD follow the MULTI input They will be like FORCE SAMC record n file CONICAL record4 file NODC 32 GEOMETRY OPTIMIZATION 244 where record file is one of the records containing CPMCSCF info again FORCE card that evaluates the USG must be the last one and record4 file points to a free record used for internal storage by the CONICAL code record4 file must be the same in all the CONICAL cards WARNING The present implementation works properly only if file in the CONICAL cards The optional keyword NODC is used in case of different spins eg S T crossing when DC is not evaluated and therefore not used The actual optimization is performed by repeatedly calling OPT inside a DO LOOP cycle until the variable OPTCONV is below a predefined threshold The example below optimizes the CI DO D1 in LiH ground and excited states are both Doublets 32 GEOMETRY OPTIMIZATION 245 ee AS QO CI memory 3 M basis sto 3g geometry nosym Li hl Li r h2 Li r hl theta r 3 31510281 theta 30 57744006 maxstep 20 do i 1 maxstep If I eq 1 then int cart pri 2 hf wf 4 1 0 else int cart end if multi occ 7 core 0 closed 0 wf 5 1 1 state 2 CPMCSCF NACM 1 1 2 1 ac
423. ue other than 1 allows a different amp to be used m is not necessary for this radial scheme 252 r 0Q 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 r a 2 1 x log 253 log 2 x og 75 y with using standard Gauss Chebyshev quadrature of the second kind for the x space integration m is not necessary for this radial scheme no ni 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 accr is used unless either is zero in which case it is ignored 15 4 3 Angular integration grid ANGULAR ANGULAR method acca crowd LMI Nip ee i un 7 max max max max LMAX O Fp A 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 as defined by C W Murray N C Handy and G J Laming Mol Phys 78 1993 997 Each type of grid specifies a family of which the vario
424. ues 36 Expressions 6 EXTRA 257 FACTOR 109 FC1 183 FIELD 190 FIELD 190 FILE 40 Files 7 F IXORB 214 FIXSTRUC 214 FOCK 115 142 FORCE 223 FREEZE 118 Q o 3 289 FREQUENCIES 252 frequencies 252 energy variables 253 FULL 215 Full CI G1 152 Gaussian 68 GDIRECT 55 GENERAL 187 GEOMETRY 65 Geometry files 69 Molpro 92 style 68 Writing CRD files 68 Writing Gaussian input 68 Writing MOLDEN input 68 Writing XMol files XYZ input 67 Z matrix 66 geometry geometry optimization 228 automatic 229 conical intersection 243 convergence criteria 229 counterpoise correction 238 DIIS method energy variables 238 quadratic steepest descent method 229 rational function method 228 231 saddle point transition state GEXPEC 36 GOPENMOL 194 GOTO B2 GPARAM 41 GPRINT 36 gradients 223 GRADTYP 223 GR1D 110 ROUP GTHRESH 35 GUESS 210 Help 19 HESSIAN 232 hessian 232 model 232 numerical 235 restart 236 HF HF SCE Hints 1 Q INDEX HSTART 236 IF F blocks INACTIVE INCLUDE ndexed Variables INDIVIDUAL INIT 258 input format B input structure 6 Integral direct integrals INTOPT Intrinsic functions intrinsic reaction coordinate 232 237 Introductory examples IPOL IPRINT IRC TRREPS 212 Isoto
425. ues by precomputed values which are passed to the spin orbit program by the MOLPRO variable HLSDIAG 30 2 Calculation of SO integrals The one and two electron spin orbit integrals over the BP Hamiltonian can be precomputed and stored on disk using the command LSINT X Y 2 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 1 1 Of course in ECP LS calculations the LSINT card is not needed 30 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 record 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 3
426. ular 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 10 6 Contracted set definitions a C first last c1 c2 cnycn l General specification of a contracted function first last defines the range of primitives to be contracted c c2 are the 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 default 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 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 2C 1 12 2100 2 1 1 generates two contractions using the first 12 coefficients from
427. ulation 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 orbitals 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 34 ORBITAL MERGING 261 NO merge geometry n 0 n r r 2 1 rhif oCo 271 Irhf for NO wf 15 2 1 12Pi state orbital 2100 2 save orbitals to record 2100 on file 2 dummy O oxygen is dummy ENE NOS dls rhf nitrogen WEI PA 14S state orbital 2110 2 save orbitals to record 2110 on file 2 dummy n Initrogen is dummy cht occ 3 1 1 rhf for oxygen wf 8 4 2 13P state orbital 2120 2 save orbitals to record 2120 on file 2 ERGE call merge program ORBITAL 2110 2 read orbitals of N atom OVE 1 1 1 1 move input vector 1 1 to output vector 1 1 OVEjZLi3 dl 33a move input vectors 2 1 3 1 to output vectors 3 1 and 4 1 OVE Pls 27152 move input vector 1 2 to output vector 1 2 examples OVE 1 3 1 3 move input vector 1 3 to output vector 1 3 no_merge2 com ORBITAL 2120 2 read orbitals of O atom OVE 1 1 3 1 move input vectors 1 1 to 3 1 to output vectors Hei They a S65 1 OVE 1 2 1 2 move input vector 1 2 to output vector 2 2 OVE 1 37163 move input vector 1 3 to output vector 2 3 ROT 3 1 5 1 45 rotate 2s orbital
428. unch file If status NEW ERASE or em REWIND a new file is written otherwise as existing file is ap pended 35 23 Examples The following example shows various uses of the MATROP commands 35 MATRIX OPERATIONS h20 matrop examples geometry o0 h1 0 r h2 0 r h1 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 1 elem d21 Dscf 2 1 1 1 elem d12 Dscf 1 1 2 1 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 Dscf natorb C_diff diffden write diffden denfil save C_diff 2500 2 269 Z matrix geometry input bond length bond angle Ido 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 transform s into MO basis same as above print result should be unit matrix f 7 mples Itrace of S_MO number of basis functions ples matrop com form trace DS number of electrons form SC S Cnat tr
429. update type to be used and limits the number of points used for the hessian update to nstep In minimizations type may be BFGS Use BFGS update of hessian default IBFGS Use BFGS update of the inverse hessian CGRD Use Conjugate Gradient update see also CUT TRUST NONE Don t do any update 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 On input nstep steps will be used for update ignoring any symmetry constraints In transition state optimizations type may be 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 32 GEOMETRY OPTIMIZATION 235 32 2 11 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 Augmented Hessian procedure This refers to the scaled parameter space default 0 5 The initial step size is stepmx see STEP card 32 2 12 Setting a cut parameter CUT CUT threshold Specifies a threshold for orthonormalization used in conjugate gradient update of hessian de fault 1 d 3 see also UPDATE 32 2 13 Line searching LINESEAR
430. us 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 l 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 15 THE DENSITY FUNCTIONAL PROGRAM 112 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 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 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 15 4 4 Atom partitioning of integration grid VORONOT VORONOI m Controls Becke Voronoi partitioning of space The algorithm
431. us 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 2 16 Selecting orbitals and density matrices ORBITAL DENSITY As outlined in section 2 7 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 ORBITAL and DENSITY directives is as follows ORBITAL RECORD record TYPE orbtype STATE state SYM METRY symmetry SPIN spin MS2 ms2 N ELEC nelec SET iset 2 GENERAL PROGRAM STRUCTU
432. used an error would result since the ENDDO or ENDIF cards would not be found 4 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 O0 lt r lt 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 4 PROGRAM CONTROL 30 4 4 Including 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 nonze
433. used for a string variable and a real or logical variable 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 METHOD MCSCF R 1 5 TEXT method results for R SR 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 Q Note that the quotes are necessary in these cases 6 3 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 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 6 VARIABLES SCFSPIN Same as SPIN but only for HF MCSPIN
434. ut 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 15 3 Examples The following shows the use of both non self consistent and self consistent DFT 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 examples endft com 15 THE DENSITY FUNCTIONAL PROGRAM 110 15 4 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 data for the DFT or KS calculations that will use the grid 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 dirty wh
435. ut_molden com 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 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 gprint orbitals geometry y planexz 0 H1 0 r h2 0 r hl alpha r 1 8 alpha 104 int hf wf 10 1 orbital 2100 2 multi wf 10 1 orbital 2140 2 matrop load dscf density 2100 2 load scf density i examples load dmcscf density 2140 2 load mcscf density h20 diffd 1d add ddiff dmcscf 1 dscf compute dmcscf dscf P ACEN ONS COU natorb neworbl dscf natorb neworb2 dmcscf natorb neworbs ddiff save neworbs 2110 2 save ddiff 2110 2 put molden h2o_ddens molden orb 2110 2 9 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 9 6 Lattice of point charges LAT X y Z q Immediately following the Z matrix specification may appear one or more LAT cards defin ing a lattice of point charges x y z are the cartesian coordinates and q the charge of the point Note that internal coordinates may not be used and that only the symmetry unique lattice points should be defined 9 GEOMETRY
436. vailable in one of the next releases INDEX Index comments in input 4 kkk comma 4 4 29 end of input record 4 ACCURACY 88 129 ACPF 140 ACTIVE 231 ADD 188 223 238 256 ALTERN 212 ANGULAR 111 AOINT 64 aocc 140 arrays 9 Atomic mass 70 BASIS 72 74 basis cartesian spherical harmonic basis set 71 contraction even tempered primitive BCCD 156 BLOCK BMAT 230 BRUECKNER 157 CANONICAL 124 CANORB 123 CASPROJ 21 CASSCF 116 207 CASVB 207 CCSD 156 CCSD 135 156 CCSD T CEPA 140 CHARGE 12 CL 135 c1 135 CI PRO 135 CISD 157 CI5SD 135 CIWEIGHTS 215 CLEAR 53 CLEARALL 53 LOSED 12 84 117 136 COEFF S 213 Q 287 COMPRESS CON 120 CONF 1G 126 CONICAL 243 COORD 230 coordinates 230 B matrix 230 cartesian 230 natural internal Z Matrix 230 copr 130 core AEAEE Cowan Griffin 192 CPF 140 CPMCSCE 132 CPP 81 CRD 68 CRIT PI CUBE 193 CUT 235 Darwin 192 DATA 9 41 DDR 198 Default procedures 17 DELETE 40 188 DELOCAL 113 DELSTRUC 214 DEMC 225 DENSITY 12 108 1 15 185 187 189 Density functionals B86 B86MGC 91 B86R B88 92 B88C 91 B88X B95 93 B97 93 B97R BR BRUEG BW p4 CS CS1 95 CS2 95 G96 HCTH120 96 HCTH147 96
437. valuated and this overhead should of course not be incurred unnecessarily in non direct calculations c type atom EVEN nprim ratio centre 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 10 BASIS INPUT TT 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 logc n 1 2 i logr4 5 n 1 2 i logd T ag Example 1 SP 1 VTZ C SP 1 EVEN 1 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 1 C SP 1 EVEN 2 2 5 generates the generally contracted s p triple zeta basis sets for atom 1 Two energy optimized d functions 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 type atom EVENR nprim aa ap bb bp Generates an even tempered set of nprim functions according to the reg
438. 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 24 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 25 PROPERTIES AND EXPECTATION VALUES 185 25 PROPERTIES AND EXPECTATION VALUES 25 1 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 4 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 25 1 1 Calling the property program PROPERTY PROPERTY invokes the property program 25 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 state
439. 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 2002 6 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 it will also if necessary revert and reapply patches Note that if you need to run pat cher before compiling MOLPRO you can build it from the top level directory with make C utilities patcher or if you need
440. wo 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 supermolecular calculation i e by subtracting the two corresponding monomer correlation energies from the intramolecular correlation component of the complex given in the output 23 5 6 Split Coulomb operator treatment of weak and strong pairs ATTENUATE This method is still in development and has to be considered experimental We document it here because we hate undocumented features but please do yourself a favour and don t use it for the time being If you get in trouble with it we won t help you The method relies on the partitioning of the Coulomb operator into a rapidly decaying short range part containing the singularity and a smooth long range part The integrals over both parts 23 LOCAL CORRELATION TREATMENTS 172 of the Coulomb operator are then treated separately The short range integrals are obtained by transformation of the short range integrals in the AO basis which is faster than the conventional transformation as more efficient screening is possible The long range integrals are treated by a multipole expansion In contrast to conventional multipole expansions this expansion has an in
441. written as BASIS DEF VTZ H VDZ BASIS DEFAULT VTZ l use cc pVTZ basis as default H VDZ use cc pVDZ for H atoms D H 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 HO7 use Huzinaga 7s for Hydrogen C 1 4 contract first four s functions PH 10 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 SPD O VTZ luse VTZ basis for all oxygen atoms 10 BASIS INPUT 76 SPD 1 VTZ luse VTZ basis for atom group 1 Instead of the BASIS END block one can also use the stru
442. y Cl Cial ree Oly Cl ree p02 LEG hl cl ich h2 cl rch WS CS POR h4 c3 rch Z matrix input examples e2 45 allene_optscf com c1 180 q1 0 c2 acc ql 0 e2 ace hl 180 ELA A190 G2 ace hz 90 hf optg default optimization using model hessian Results ITER ENERGY OLD ENERGY NEW DE GRADMAX 1 114 41781970 114 42128294 0 00346324 0 10090139 2 114 42128294 114 42170099 0 00041806 0 02723890 3 114 42170099 114 42171728 0 00001629 0 00670578 4 114 42171728 114 42171910 0 00000182 0 00220625 5 114 42171910 114 42171910 0 00000000 0 00000895 32 3 2 Allene in natural internal coordinates xxx Allene geometry optimization using natural internal coordinates memory 1 m basis sto 3g rcc 1 32 ang rch 1 08 ang acc 120 degree Geometry nosym el Corak rec Olye rec ay CZy FOS hi cl rch h2 c1 rch h3 3 ECH h4 c3 rch hf optg coord bmat Z matrix input examples c2 45 allene_opt_bmat com c1 180 q1 0 c2 acc ql 0 tac hl 180 eZ ace nl 90 o2 acc h2 90 default optimization using model hessian luse natural internal coordinates Results ITER ENERGY OLD ENERGY NEW DE GRADMAX 1 114 41781970 114 42134430 0 00352460 0 06107869 2 114 42134430 114 42168324 0 00033894 0 00464120 3 114 42168324 114 42171906 0 00003582 0 00181807 4 114 42171906 114 42171910 0 00000004 0 00020759 32
443. y be used These orbitals are eliminated from the optimization procedure but will still be normalized and symmetry adapted according to any ORBREL keywords given 29 10 85 Freezing structure coefficients in the optimization FIXSTRUC iy in 3 Freezes the coefficients for structures ij i2 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 29 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 29 10 10 Orthogonality constraints ORTHCON key I key 2 5 The ORTHCON keyword initiates the input of orthogonality constraints between pairs of valence bond orbitals The sub keywords key i can be 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 1 i2 3 Specifies a list of orbitals to be orthogonalized All overlaps between pairs of orbitals in the list are set to zero PAIRS y
444. y 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 23 LOCAL CORRELATION TREATMENTS 182 Table 10 Summary of at tenuate options and their default values Parameter Default value Meaning Most important options DECAY 0 20 split parameter SHORTMLT 15 level p of monopolar multipole expansion LONGMLT 13 level p of bipolar multipole expansion Specifying which integrals to treat by which multipole expansion type RMAIN 1 when to switch from monopolar to four block treatment RIONIC 0 when to switch from monopolar to bipolar treatment of ionic blocks SUPPRESS 0 when to suppress cross excited blocks Options for least squares fit generation of interaction coefficients FITMLTP 1 use least squares fit instead of Taylor F1DGRID 50 no of quadrature points for 1D fit F2DGRIDR 50 no of quadrature points for 2D fit r F2DGRIDP 20 no of quadrature points for 2D fit F1DBORDER 0 end of integration interval for 1D fit F2DBORDER 0 end of integration interval for 2D fit r F1DGAMMA 1 7 negative exponent of weight function for 1D fit F2DGAMMA 1 7 negative exponent of weight function for 2D fit WEIGHT3D 1 use spacial instead of flat weight function Options for determination o
445. y orbital basis are written to channel iabs iwrite The default action is 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 idim4 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 29 THE VB PROGRAM CASVB 217 29 14 Examples RAS AZ Al singlet state geometry angstrom c hl e 1 117 h2 c 1 117 h1 102 4 int hf multi occ 4 1 2 closed 1 6 in 6 CASSCF natorb ci save 3500 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 A Rao 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 s 2 13 36 2 013 0 4538 1233 k 1 2 0 032828 0 231204 geometry li h 1li r int hf wf 4 1 multi oce 407070 closed 0 0 0 0 nat
446. y 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 Cz and C27 The possibilities in this case are shown 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 The irreducible representations of each group are numbered 1 to 8 Their ordering is important and given in Tables P H 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 2 14 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 WF nelec irrep spin or alternatively WF NELEC nelec SYM METRY irrepl 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
447. ype is optional and only affects the numerical calculation of the gradient for 3N coordinates By default SYM is used SYM Use symmetrical displacements This yields the minimum number of displace ments and always preserves the symmetry of the wavefunction This is the 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 recommened either 31 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 31 2 3 Stepsizes 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 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 31 ENERGY GRADIENTS 227 STEP varname value where the value must be in atomic units for distances and
448. ystem parameters GPARAM o 6 VARIABLES ha e eee rare A A e ere be ree ee grees ede ee ee Ge Bah be ean Y ee ee aoe gee ae aa aS ia ar ee Sear R Mod dud Sa dak Bh aed eter ety 6 5 Indexed Variables Vectors gt e sr 0 nein Sg oe thee ee Oe ae ee eee eee eo Reema eee ee He Gale Reed nea oe ARS 6 7 1 Variables set by the program 000 eer eer eee Cig har ae ot nb ae ER ee ae Ge 6 8 1 The SHOWcommand 000 0000084 6 9 Clearing varlables o o ee ee 7_ TABLES AND PLOTTING 7 METALES iae bea a ted WS He aed a we we BLA A 172 Plotting iee a ve a eb ea eee ee aoe a ES bas 8 INTEGRAL DIRECT CALCULATIONS GDIRECT 8 1 Example for integral direct calculations 9 GEOMETRY SPECIFICATION AND INTEGRATION 91 Sorted integrals urinarias bee See be Oe eS ae Berka 9 2 Symmetry specification 2 o ee 9 3 Geometry specifications 2 ee 9 3 1 Z matrixinput o 0 3 2 AO e si Gee oh ee we bate eae Be be ee a e E 9 3 3 MOLPRO92 input 2 ee 9 4 Writing Gaussian XMol or MOLDEN input PUT 9 4 1 Visualization of results using Molden 9 5 Geometry Files os lt a ce mai a eioi a k o e 9 6 Lattice of point charges a oaoa ee viii 31 31 32 32 33 34 34 35 36 36 37 38 40 40 40 40 41 41 41 42 42 43 43 44 45 46 47 47 5
449. zation default root 2 specifies a transition state 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 particular a good starting geometry and a good approximation to the hessian is needed The latter is achieved by evaluating the hessian numerically see NUMHES section 32 2 15 or using a precomputed cartesian hessian with the HSTART command see section 32 2 16 32 2 7 Saving optimization information SAVE SAVE record Specifies a record on which the geometry definitions parameters energies and gradients for the present optimization are stored By default this is record 700 and it is overwritten in each new optimization If it is intended to use a START directive in a subsequent optimization a different number should be given e g 710 720 The geometry record is saved on all permanent files 32 GEOMETRY OPTIMIZATION 234 32 2 8 Restarting a geometry optimization START START record first last Specifies a record from which the geometry definitions parameters energies and gradients of a previous optimization are read This record must have been written with a SAVE card in a previous optimization It should not be the geo

MOLPRO

Contents

Download Pdf Manuals

Related Search

Related Contents