Home

Casa XPS User's Manual - The Molecular Materials Research Center

1. Gaussian Lorentzian Product Form 2 E m f F GL x F E m 5 x E 2 1 4m Gaussian Lorentzian Sum Form x E SGL x F E m opf am2 P ms m gt F 1440 2 57 A List of Line Shapes Exponential Asymmetric Blend based upon Voigt type line shapes Given either of the above Gaussian Lorentzian symmetric line shapes an asymmetric profile is obtained from a blend function as follows Y x GL x 1 GL x x T x Where exp 2 xSE T x k F E otherwise 1 Alternative Asymmetric Line Shapes An asymmetric line shape due to Ulrik Gelius Uppsala Sweden offers a class of profiles by modifying the Voigt function via an ad hoc adjustment The profile is given by Line Shapes and Backgrounds A x a b F E GL x F E m w a b AW x F E m G F E x lt E GL x F E m x gt E Where AW x a F E o Ps and 7 0 3 w a b b 0 7 CEN The parameters a and b determine the shape of the asymmet ric portion of the curve Doniach Sunjic a theoretically based asymmetric line shape Doniach and Sunjic 6 performed an analysis for both pho toemission and X ray line shapes both of which result in an underlying profile given by the expression below The for mula includes an asymmetry parameter a that characterizes the asymmetry for a particular metal like material F is relat ed to the FWHM and the position E is again related to but 58 not equal to the po
2. Tagged Regions Analytical Applications tween the components based upon tags nor do any of the annotation options involving quantification tables unless the Use Tag Field checkbox is ticked The tag mechanism is however always used when the Combined button on the Re port Spec property page from the Quantification Parameters dialog window is selected No other quantification reporting options use tags Tagging for different line shapes Although the primary reason for using tagged reports is to reference results from peak fits to survey spectra it is possi ble to use the same mechanism to remove some of the ambi guities from comparisons between intensities derived from peak fits that employ different line shapes Any line shape involves a functional form that may extend beyond the ac quisition regions to which it is applied especially when asymmetric peaks are involved The tag mechanism allows the peak fitted results to be referenced to the underlying in tegration region and so intensities for the chemical states are calculated based upon the data rather than the implementa tion of the synthetic components Figure 81 shows a set of quantification tables These tables derive from components and integration regions taken from the two high resolution scans below the survey spectrum Note that the Al 2p doublet is fitted using a Doniach Sunjic asymmetric line shape while the O 1s spectrum is modeled using symmet
3. dragging the central divider the splitter bar in side the CHAPTER 3 frame allocating a larger or smaller area to the graphical dis play the spectrum or the block structure as appropriate Data held within a VAMAS ISO file appears in rows and columns in the Block window each row corresponds to data acquired under the same experimental variable value while the columns correspond to spectra regions resulting from measurements on the same chemical species or transition energy range The mouse is used to select a set of one or more blocks from the right hand side the Browser and these data blocks are displayed by selecting one of the two leftmost toolbar buttons shown in Figure 15 IDEL T Figure 15 Spectrum Display Options The first icon will display a scrolled list of spectra or re gions one per tile The second of the two allows many spectra to be overlaid in the same tile In both cases the x and y scale values are determined by the maximum range values for the spectra or regions requested The corresponding but 25 Understanding the Data tons on the right of the toolbar provide display and overlay within the already set scale values If there are no points in the range then a message indicating No Data Points in Window appears this facility enables the overlay of de tailed regions onto a wide scan for example Selecting the ISO 14976 Blocks Each ISO file requires at least one E
4. note that the original data are not affected by this correc tion only the quantification The manual entry type in box in general should only be used when automatic correc tion is not available The Update button applies a new man ual value or the automatic setting after any change in the status of the checkbox Note that the width of the region list may be altered in the usual fashion by dragging on the edge of the divider in the header bar Components see Tile Display on page 27 and also see Line Shapes and Backgrounds on page 55 This tab defines the parameters for each synthetic compo nent for a particular region and also controls peak fitting Note that the Cross Section is the Tougaard 3 parameter constant set consisting of four comma separated values see U 4 Tougaard short form U 4 on page 67 Offset provides a numerical adjust for spectrum overlay in display after background subtraction the value entered is a percntage of the largest peak to be displayed and may only be positive Positive values result in movement of the region downwards offset towards the baseline Max Height and Min Height are derived from the region under consideration signal values above below back ground and used e g with AES signals or other deriva tives in determining peak to peak concentration See page 11 for notes on other button use 139 The Report Spec tab enables customised
5. Other information drawn on a tile can be changed through the Display property page The font and text for the title of a tile may be adjusted The optional header information font may be changed similarly A range of optional display items can be toggled on off allowing the spectrum to appear for example with or without quantification regions present Data Display and Browser 540 538 536 534 532 530 528 526 Binding Energy eV Figure 17 3 D plot Colours The Colours property page provides the means to set the colours used throughout the system Spectra backgrounds to spectra synthetic components regions and residual plots may have the colours adjusted Fill colours used to display 3 D plots are also changed through the options on this page Selecting a button on the Colours property page brings up a dialog window that allows the existing colours to be viewed and new values set In the case of spectra sixteen colours may be chosen These colours are displayed in the Custom Color section of the dialog window and changed using the Define Custom Colors expanded form of the di alog window To change a colour within the Custom Color set first click on the colour you wish to change then select a new colour either from the default palette or using the colour values on the expanded section of the window On pressing the button labelled Add to Custom Colors the 28 Fonts colour squ
6. 180 C Is 1 kinetic energy eV 1191 6 0 05 1 counts per channel d pulse counting 0 5 1 400E 9 42 2250 179 59 end of experiment 144 2 Multiple block depth profile composed of XPS wide scan spectra Note the line numbers for the experiment header and block 1 have been deliberately left the same as in example 1 so that differences stand out clearly Extra lines variables and values in this example are denoted with a sign Item Description format identifier institution identifier instrument model i d operator i d experiment i d comment lines experiment mode scan mode regions experiment variables experiment variable label experiment variable units parameter exclusion entries manual items future experiment items future block entries blocks block i d sample i d year 4 digits month day hour minute second hours GMT lines in block comment comment lines used internally Line 1 2 3 4 5 6 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 ISO 14976 Format Item in DATA2 vms file VAMAS Surface Chemical Analysis Standard Data Transfer Format 1988 May 4 Not Specified AXIS 165 SAC kratos VISION User export home kratos data 2043it1 dset 0 NORM REGULAR 1 1 Etch Time reoocococoos 17 OSnIn a a 2 OSnIn
7. A te 3 EEE d 110 s 2 t g s R e 1 08 a ad EP A nE i 9 Ape D 3 eo ES WRS S aa e te bg gt 2 e ee uae pee a nares aes RAD SA Be 64 4 t t ge t tal eet Ps rs s TAO Shap ce oF fe veo on H ue e ZN ot s i 0 85 T x x eM p 0 80 o 50 100 150 200 250 300 350 400 450 Figure 68 Peak areas from one end point ground from the information in the simulated noise In Figure 68 the background is computed using only one point at each end of the integration region while Figure 68 is the equivalent experiment but in this case an average of 21 points defines the background limits It is clear from the two scatter plots that the intensity calcu lation using just one data point at each end of the integration region is much more sensitive to noise The introduction of more data channels when determining the background im proves the situation 103 Monte Carlo Methods Uncertainties in Intensity Calculations Peak Area determined from Integration Region Intensities Determined by Peak Fitting Twenty One End Points Defines Region Limits 1 20 Two peaks of equal width and area but separated by leV 145 represent an intractable problem for a peak fitting algorithm if the intensity of the individual peaks is required and no fur 1o ther information is available that can be used to provide con 105 2 straints for the peak fitting par
8. RSE Relative Sensitivity Factor for the data within a region The integrated data are divided by this value e Start Start point for the integration interval e End End point for the integration interval e Background Tougaard Shirley Linear None e Average WidthNumber of channels used to fix end points for the backgrounds 42 Regions Armed with the values described in the above table a region is created by either pressing the Create button on the Re gion property page or by clicking the column buttons on the scrolled list of regions The start and end points for a region are taken from the zoom parameters currently in effect when the region is created Before pressing a create button zoom into the spectral fea ture until the display shows the range which characterises the intensity in question The new region created at this stage will have start and end energies defined by the display After creation the limits can be adjusted under mouse con trol To perform this operation it is necessary that the Quan tification dialog window is visible and the active page corresponds to the Region property page Position the cur sor near an end point for a region then hold down the left mouse button A box will appear showing the extent of the region The end under the cursor is altered when the mouse is moved whilst continuing to hold down the mouse button Should the same procedure be followed but instead of ini tially pos
9. The current set of active element markers can be view on the Periodic Table property page also located on the ele ment library dialog window The element markers may be activated and deactivated via the Periodic Table as well as the scrolled element list Step 4 Annotate peaks using element markers Once all the spectral lines are identified and markers have been placed on the displayed spectrum the peaks can be an notated using the names stored in the element library This step not only creates a labelled spectrum but these annota tion peak labels are used to create a set of quantification re gions at the press of a button The annotation dialog window includes a property page en titled Peak Labels Figure 8 All the peak markers active on a spectrum are listed on the Peak Labels property page and may be selected using the mouse and the control key to add to the selection When the Apply button is pressed those la bels selected in the list of names on the Peak Labels property page become annotation labels on the spectrum The peak labels are positioned on the spectrum according to the loca tion of the nearest peak so it is important that the spectrum is calibrated before this procedure is performed Step 5 Create quantification regions The Quantification Parameters dialog window is available from the top toolbar button indicated in Figure 9 Integration regions are created and adjusted via the Quantification Pa rameters dia
10. The file is a series of single item lines each terminated by a single character return The first line is the version number Each subsequent entry in the file has items that are included one per line as follows 38 Library File Structure Element Transition Label Name Mass Daltons Energy Type B E or K E Energy eV F W H M Line shape e g GL 30 9 Relative Sensitivity Factor 10 Ionisation source X ray anode Photoelectric transitions should be entered with item 5 en ergy type set to BE this allows the same energy to be used for different X ray anodes Auger lines always appear at the same kinetic energy independent of the excitation source and these values should be specified with energy type KE Note that the ionisation source is also specified so that source specific information can be offered for spectra that match the acquisition characteristics found in the library line shapes are specified using the abbreviations listed in Line Shapes and Backgrounds on page 55 The beginning of the default CasaXPS library file is shown below Ril Oe E aS 0 H 1s H ls 0 000000e 000 BE 1 400000e 001 1 000000e 000 Element Library GL 30 2 000000e 004 Mg 0e 001 oO Oe 003 An entry allows the species and transition to be specified but also permits a name other than these two labels to be also en tered For example
11. as appropriate It is intended that these examples should pro vide a greater insight into some of the powerful features of the system and an indication of a possible analytical ap proach which may be applied effectively to certain prob lems It is not suggested that these are the only aproaches possible or available even within CasaXPS but merely that these indicate known good methodology Organic Polymers and Curve Fitting The availability of data from well characterized samples such as those offered by Beamson and Briggs owe much to the popularity of XPS as a tool for understanding the chemistry of polymers A typical C 1s envelope Figure 18 Beamson G and Briggs D The XPS of Polymers Database CD ROM Surface Spectra Ltd 2000 CHAPTER 12 74 includes structure that offers chemical information about a sample but without some initial starting point it is difficult to construct an appropriate model for the data enve lope The high resolution spectrum in Figure 74 derives from poly acrylic acid PAA reacted with inorganic material or par tially reacted in an acid base reaction If Gaussian Lorentz ian GL line shapes are added in an arbitrary way the curve fit yields little information about the sample other than to say that it deviates from the published data for PAA and therefore demonstrates the presence of additional chemistry at the surface Figure 75 shows a synthetic model for this data e
12. ky Golay polynomial This is appropriate since noise inter ference would cause problems to an algorithm used for analytical functions and so the implied smoothing operation involved in Savitzky Golay differentiation would explicitly be required with other techniques The parameters used in differentiation are therefore identi cal to those for smoothing the data To get a feel for what was actually done to the spectrum during the differentiation procedure a smooth operation with the same parameters applied to the original spectrum can provide an insight into the shape of the data that produced the derivative Integration A more novel use of the Savitzky Golay polynomial is in calculating the integral for a spectral range Integration by any means has implied smoothing involved since it is essen tially an averaging process However when noise is in volved there is little benefit in using sophisticated Newton Processing Cotes or other quadrature methods as these generally in volve forcing a function to take on the data values at the cor responding nodes The virtue of putting a quadratic through three points containing noise is somewhat doubtful Inte grating the data using a least square fit of a quadratic to more than three points seems better however from a practical per spective if there is a significant difference between these two operations then there s trouble somewhere To summarise integration is perform
13. spectra exhibit deviations from idealized profiles due to a range of instrumental and physical effects e The response function of the electron analyzer which may be asymmetric e The profile of the x ray line shape predicted to be asym metric for non monochromatic lines from metal anodes 6 e Intrinsic life time broadening of the core level hole state usually assumed to be Lorentzian in nature e Phonon broadening e Differential surface charging of the sample In addition to these instrumental considerations the shape of a synthetic peak is also influenced by the choice of back ground algorithm used to remove so called extrinsic elec CHAPTER 8 trons from the data Figure 31 and Figure 32 show examples zl w Hem Pu FWHM L ik ima ae HES Oe Ee TANF 2 MAT BE DT Ee e ed m B i Lig Late Figure 31 Au 4f doublet fitted using an offset linear back ground The line shape used in the fit is a Doniach Sunjic form con voluted with a Gaussian that results in an asymmetry index of 0 49 for each peak 55 of peak fits where the difference between the two outcomes lies in the choice of background Doniach Sunjic profiles have been used to model the recorded doublet where it has become necessary to adjust the asymmetry parameter as well as the width of the Gaussian used to broaden the under lying Doniach Sunjic shape The most obvious difference 5 ao my Hine P Fou Lh kn H deo MITE
14. 2 1999 12 22 17 40 2 0 3 Lens Mode Magnetic Resolution Pass energy 80 Anode Al 180 W 145 by Kratos Vision system CasaXPS adds history here technique experiment variable value source source energy source strength source width x mu source width y mu source polar angle source azimuth analyser mode analyser resolution characteristic magnification of analyser analyser work function target sample bias analysis width x mu analysis width y mu analyser axis polar angle analyser azimuth species transition state charge of analysed particle abscissa label abscissa units abscissa start abscissa increment corresponding variables corresponding variable label corresponding variable units d none signal mode signal collection time channel scans for this block signal time correction sample normal tilt 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 Step meV 300 0 Dwell ms 136 Acquired On Tue Dec 22 17 40 02 1998 XPS 0 Al 1486 6 180 1E 37 1E 37 1E 37 1E 37 FAT 80 1E 37 1E 37 0 1E 37 1E 37 1E 37 1E 37 Wide none 1 kinetic energy eV 926 7 0 3 2 Intensity d Transmission d pulse counting 0 136 1 0 1E 37 146 sample normal azimuth 59 J1E 37 sample rotation angle
15. 2p Profile Factors Eigenvalue RMS RE RSD IE IND 1000 Chi sq Calc Chi sq Expected Al 2p 103 7476 855 1 90305E 13 0 0 0 2 03675E 25 0 The chi square indicates that the data matrix can be repro duced to within experimental error using two abstract fac tors This is a result that is consistent with the physical nature of the sample It is also interesting from a mathemat ical standpoint to note that using all the abstract factors to reproduce the data matrix returns a chi square of zero al lowing for round off errors in the computation This should always be the case and provides an easy check to see that the calculation has been performed correctly All the statistics except the Indicator Function point to two abstract factors being sufficient to span the factor space for the data matrix It is worth examining the data set using a subset of the spec tra and Target Testing the spectra not used in the PCA This allows anomalies to be identified such as spikes in the data Selecting a representative subset of spectra for the PCA then target testing the remainder is particularly useful for large sets of data Table 9 Target Test Report for a Subset of Al 2p Data Set Target AET REP RET SPOIL Al 2p 3 Al 2p 8 Al 2p 23 20 93032 10 38374 18 17295 1 750135 0 39033 0 04947 Al 2p 28 23 83028 11 05117 21 11288 1 910467 0 41839 0 017257 Al
16. Data 90 PCA and CasaXPS c6 0 cack winged bots oe aad 94 Viewing the Data in Factor Space 96 Monte Carlo Methods Uncertainties in Intensity Calculations 97 Monte Carlo Data Sets 0 4 98 Brror Matix aa sete k oie epee ee 98 Contents Monte Carlo A Simple Example 99 Quantification a na aa ce eae ee eee os 102 Monte Carlo End Point Determination 102 Integration Region Limits 102 Intensities Determined by Peak Fitting 104 SUMAN wach Screen taad teow RE es 106 Analytical Applications e08 107 Organic Polymers and Curve Fitting 107 Quantification using Tagged Regions 111 Introduction 2 i462 eces ey aeecee reg ew hes 111 Tagging regions in a survey scan 112 Tagging for different line shapes 113 Trend Analysis for Metal Oxidation 114 Tntrod cti gn A 4 ee RU i BU UN ae 114 PCA for Aluminium Oxidation Sequence 116 Synthetic Model for an Oxidation Sequence 117 Peak Fit Using Doniach Sunjic Line Shapes 118 Adjusting a Custom Quantification Report 120 Using Different File Formats 123 Kratos DS800 Binary Files 123 VG Belipse Filesi sens wa i en t eee 124 VGX900 Ron Unwin 00 125 Dayta System Files Bristol IAC system 127 Kratos Vision 1 x 2 x ASCII files 128 SSI M Probe Files 4 02 38 4 fen shy pea
17. Definitions and Formulae Glossary of terms Block ISO A sub unit of an Experiment file q v which consists of a header a set of parameters that only apply to that block followed by a series of ordinate values which may represent a curve e g a depth profile a spectrum or a map Experiment ISO A complete ISO 14976 file which defines the context of and data acquired during a surface chemical anal ysis determination Consists of a header containing of a series of parameters which apply to the measurement procedure as a whole followed by one or more blocks q v of data VAMAS the Versailles project on Advanced Materials And Standards Formulae Peak shapes Gaussian Lorentzian Product Form 2 x exp 41n2 1 m F GL x F E m 2 r me Gaussian Lorentzian Sum Form m 2 pagan F x E SGL x F E m o an ms F 149 Doniach Sunjic cos 32 1 a atan DS x amp F E 1 a 2 Appendix 4 References and other Resources World wide web links CasaXPS UK Surface Analysis Forum Kratos Analytical Physical Electronics Thermo VG American Vacuum Society AVS ECASIA QSA Queen Mary Westfield College Institute of Physics UK Royal Society of Chemistry UK American Chemical Society ACS American Institute of Physics AIP IUVSTA Materials Research Society USA F x E http www casaxps com http www uksaf org http ww
18. Excel where the data can be formatted for printing either as a table or in graphical form Report Files and Excel When a report is generated the data are displayed in a scrolled list This represents an alternative view for a VA MAS file document If a report view is the active MDI frame indicated by the frame title bar colour then the main menu only offers File and Window menus The options on the File menu under these circumstances are restricted to Save As this allows a File Dialog to be used to specify a text file to receive the report Once a report has been saved to disk Excel can open the text file simply by selected the file name via the File Open op tion on the Excel main menu A Wizard for loading text files into Excel guides the way through the available options On completion the columns of data appear in a spreadsheet for mat Excel provides many tools for presenting spreadsheets in both tabulated forms as well as graphically These far ex ceed the sophistication that could be implemented in CasaXPS and so the use of a spreadsheet for presenting the results seems most appropriate 54 Line Shapes and Backgrounds A range of physically possible line profiles in core level XPS is possible and simple Gaussian or Lorentzian func tions are only rarely entirely adequate In the case of metal samples it has been shown that asymmetric profiles should be expected on theoretical grounds 6 however recorded
19. ISO 14976 corresponding to the fields used by XPS normally stored in ISO format The VAMAS appellation often used loosely and interchange ably with ISO 14976 arose because the early work on stand ard data transmission format was sponsored and encouraged as part of VAMAS the Versailles project on Advanced Ma terials And Standards from the early 1980 s see for exam ple VAMAS Surface Chemical Analysis Standard Data Transfer Format with Skeleton Decoding Programs W A Dench L B Hazell M P Seah and the VAMAS Communi 1 http www vamas org TWA 02 is the Surface Chemical Analysis work area with secretariat in the USA ISO 14976 VAMAS Files Introduction ty Surf Interface Anal 13 63 1988 The present ISO standard is closely related to the VAMAS original but not quite the same It is not intended to mean here that every variation of ISO XPS format is supported within CasaXPS there are many allowed variations even though the basic skeleton structure is well defined and invariant While every effort has been made to support a range of XPS formats VAMAS flavours as interpreted by particular systems it has not been possible as yet to accommodate the entire set of combinations and permutations that are allowed by the com plete file specification Many of the processing techniques rely on equally spaced abscissa For this reason irregular ISO files are not support ed That is to say
20. Mg 2p 104 The first step towards creating a model for each of these transitions is to identify a representative spectrum for each column of the browser Click the column header for C 1s All the carbon spectra then become the browser selection Pressing the toolbar button for displaying the blocks one per tile the spectra are displayed in the left hand scrolled view Ep bakk Figure 30 Spectrum Display Toolbar Choose a carbon spectrum that shows a well formed C 1s en velope The limits for the energy range and hence the back ground to the spectrum needs to be well defined A good idea of where the carbon peaks are located is essential for es tablishing a realistic background shape 51 Before proceeding further ensure that your element library is loaded See Element Library on page 38 Another important consideration is intensity calibration Be fore quantifying the raw data it may be necessary to apply a transmission adjustment See Processing on page 34 for a discussion of the issues involved with intensity calibration The next decision determines what values are used in the re gion In the event that only the total counts for carbon is re quired no synthetic peaks are necessary and the name plus the R S F values will be the key information used by the quantification report On the other hand if the chemical states within the carbon envelope are to be profiled then the R S F within the re
21. Some of these ideas can be tried out on the known peak structures for PMMA PVA PVC and PIB Quantification First Derivatnre Intensity d Second Derivative T T TTT T TTF rfr a L T TTT YT T T 296 294 292 290 288 286 284 282 280 Binding Energy eV Figure 28 PMMA peak structure with first and second de rivatives PMMA has four peaks two well resolved and two that merge together to from a broad structure towards the lower binding energies Spotting the shoulder associated with the third and fourth peaks and determining the locations for these poorly resolved peaks is critical for constructing a physically meaningful model The presence of a background also interferes with identifying the true structure although the data in Figure 27 has been prepared without this compli cation Differentiating the artificial data once and then twice can help to see the underlying peak structure see Figure 28 The shoulder of the third peak in the PMMA envelope can be seen in the second derivative where a kink in both deriv atives highlights the shoulder in the data 50 Quantification by Example The procedure for quantifying a set of spectra will be ex plained using an ISO 14976 file This file includes measure ments for four transitions taken after etch cycles have been used to change the state of the sample The problem is to cre ate suitable models for each of the transitions using a com bination of regi
22. a carbon line corresponds to a chemical state arising from PVC is identified by entering C for the species Is for the transition and C ls PVCa for the name It is the name that appears as the first item in the scrolled list of transitions Although the expected energies for X ray induced transi tions are universal constants quantification data will be in practice specific to an instrument An element library can be customised through the energy ordered scrolled list on the property page labelled Element Table Right clicking on an item name will produce a dialog window that allows the items associated with the indicated line to be updated New entries can be created using this mechanism but for 39 large numbers of new entries a text editor with cut and paste can be the best method of modifying the element library Items which are specific to an instrument or instrument con figuration include the relative sensitivity factors R S F full width half maximum F W H M and the line shape In strument manufacturers will supply some of this informa tion but individual experience will determine appropriate values for items such as the F W H M and line shapes These parameters clearly vary from analysis to analysis and specific figures are present here only to provide sensible initial values for the creation of synthetic components val ues that are sometimes useful Loading an Element Library The property page labell
23. all those that have already been constructed Sometimes a Trend Analysis for Metal Oxidation Analytical Applications Figure 82 Oxidation Sequence data set may have a dominant trend and therefore the ab stract factors apparently have chemical meaning in their own right The temptation to interpret these abstract factors as chemically meaningful quantities should be avoided nor should the number of significant factors be slavishly used to imply the number of synthetic component that are required to model a data envelope However PCA can produce some interesting insights into the data and with the appropriate care provides supporting evidence for a synthetic model A trend analysis for the oxidation of a piece of aluminium 115 provides a good example of how PCA can assist in the inter pretation of a data set The essential problem is to construct a synthetic model for the Al 2p or Al 2s photoelectric lines measured periodically as an Aluminium sample oxidises in a vacuum chamber The relative sizes for the individual syn thetic lines provide a quantitative measure for the trends within the data The problem lies in knowing what chemical states are present and therefore what features the model must accommodate PCA for Aluminium Oxidation Sequence The spectra shown in Figure 82 were recorded on a Scienta ESCA300 XPS spectrometer at RUSTI and represent a trend observed as a clean Aluminium sample created by argon ion etchi
24. and may need changing in view of other in put however such a model is only possible when peak fitting routines offer mechanisms for fixing parameters with re spect to one another The role played by the pure PAA syn thetic model is that of a foundation shape from which differences in the unknown polymer can be assessed These additional peaks may still require further interpretation but with the aid of chemical knowledge and supporting evidence a meaningful model can emerge from seemingly intractable data Quantification using Tagged Regions Introduction Routine measurements using XPS often involve acquiring a wide scan spectrum to determine the general composition of a sample The survey spectrum is then used to select narrow energy regions where detailed structures are present These narrow scan spectra are typically needed when overlapping peaks are responsible for the data envelope such as is often seen in the case of C 1s spectra Synthetic line shapes must be used to extract the chemical state information in the data but intensities determined from these models are not always comparable with intensities determined from integration wide scan regions The most notable case is when asym metric line shapes are used to model the data Moreover the transmission characteristics of an instrument require that each operating mode must be characterized by a transmis 111 sion function and although it is possible to correct the
25. changing a file name and or saving to a different directory Save Picture brings up a save file browser enabling the displayed graphic to be saved as a Windows Enhanced MetaFile EMF Print opens the standard Windows print dialog box for the system default printer Print Setup invokes the standard Windows print set up box for the system default printer recent files the system indicates the four most recently opened files a push down list Clicking on one of these opens a new Experiment Frame containing the file Exit quits CasaXPS entirely 131 Command Summary View Toolbar toggles sets on or off visibility of the Toolbar Options Bar toggles visibility of the Options bar Status Bar toggles visibility of the Status bar Toolbar Options Bar Window New Window launches a fresh blank processing win dow a tile within the CasaXPS work area Cascade overlaps a number of processing windows Main menubar tidily so that the top and left edges are visible behind the front one Tile fills the work area uniformly with the open windows so that all their contents are visible The usefulness of this command depends on the screen area available but it works best with a even number and is seldom effective with more than four tiles on an average 800x600 pixel screen Arrange Icons tidies any minimised processing win dows docking them in rows along the lower edge of the work a
26. e g if it is required to set the FWHM value to constraint results in a peak fit shown in Figure 76 Two ad 1 1 the constraint interval should be entered with the value ditional peaks appear in the model and further more three of 1 1 1 1 Actually it is sufficient to set the parameter value out the peaks look like they may have something to do with pure side the constraint range currently defined for the parameter Organic Polymers and Curve Fitting 108 Analytical Applications Spectra contained in the Beamson and Briggs polymer data base offer the opportunity to examine more complex poly mer data in the context of known synthetic models Figure 77 is an example of such data where a set of three line shapes has been used to model the clean PAA C 1s data en velope The important feature is that the stoichiometry and chemical shifts for the C 1s lines are incorporated into the model and this information is then transferable to other pol ymer spectra The peaks in Figure 77 are linked in area but only the position of peak C 1s b and C 1s a are con strained by an offset Repos Commies tats Bate Pepe fo Ty Cia tie Figure 76 Same C 1s envelope as Figure 75 but the synthetic peaks are all constrained to have FWHM equal to 1 1 eV Organic Polymers and Curve Fitting To link a component parameter the constraints must be adjusted as follows Each synthetic component defined on the Quanti
27. either a periodic table in terface or an energy ordered scrolled list Information se Graph Annotation lected via these mechanisms can be used to annotate the spectra or create quantification objects such as regions or synthetic peak shapes Graph Annotation Although Casa XPS provides Microsoft enhanced metafiles for displaying spectra via Word and Excel XPS specific an notation and other important information can be added to the display through the Casa XPS options These include e Tabulated results and summaries extracted from the quantification regions and components e Peak labels derived from the element library e Basic text overlays Annotation may be positioned on the graphs either with re spect to the axes or the data This allows annotation to be po sitioned at a fixed location relative to the axes independent of zoom state or to follow the data as a peak is expanded to fill the display area Processing Understanding the data envelope the peak shape for any transition can sometimes be assisted by smoothing differ entiating or integrating the data Techniques for providing these alternative views of the spectra are made available to the users of CasaXPS through the processing option The spectra can be calibrated with respect to energy and intensi ty The latter requires the transmission characteristics of the Introduction instrument to be included in the ISO 14976 file as one of the correspondi
28. existing or ganisation or equipment CasaXPS and the User s Manual are copyright 2001 Casa Software Ltd Contents Introduction 5 iiins eave etna kein ak eee hed 5 Ge tting Started ve 08 ea what aes eee ee 6 Installing CasaXPS 22 0teei awiae ee eden 6 Starting Casa XPS util bt pate ee 6 Terminology 0 e ee eee eee eee 6 A Quick Tour of CasaXPS 04 7 Step 1 Load the experiment file 8 Step 2 Select a spectrum for processing 8 Step 3 Identify peaks 9 Step 4 Annotate peaks using element markers 11 Step 5 Create quantification regions 11 Step 6 Adjust region start and end points 12 Step 7 Print the results 14 Data Display and Browser Windows 15 Element Library o1 4 08e saci Hees Geek oes 16 Graph Annotation 008 17 Processing 4 su brakes iae cunconi n ia tae LRes 17 Q uantificat on O05 athena SA we ea es ee ee 17 ISO 14976 VAMAS Files 17 ISO 14976 File Format ceeeeee 19 Partially Encoded Format Versions 19 File Structure 2 2 4 0 ekegadee bd coed ees 20 Experiment header ys esewe dpe dg news wel 21 Data Blocks venida eels oath IO ie at 21 Binding vs Kinetic Energy 22 Experimental Variable 22 Transmission Correction and Quantification 22 Transmission Functions 4 23 Q
29. from RSD namely IE Umbedded Error and IND Indicator Function given by l n 2 IE RsD And RSD IND a o IE and IND are statistics that should decrease as the number of primary abstract factors is increased Once all the primary factors have been included these statistics should begin to increase since at this point factors from the noise subspace start to interfere with the accuracy of the data de scription This minimum is therefore an indicator of the di mensionality of the data subspace Chi square Bartlett proposed using the chi square criterion for situations similar to XPS data where the standard deviation varies from one data channel to the next The procedure involves reproducing the data matrix using the abstract factors Each abstract factor is progressively in cluded in a linear combination in the order defined by the Principal Component Analysis size of the eigenvalues and weighted by the co ordinates of the corresponding eigenvectors The chi square value for a set of n abstract factors is computed using di a Xi Ey ewe i lj 1 oi where ad is an element of the data matrix d is the corre sponding approximation to the data point constructed from the first n abstract factors with the largest eigenvalues The standard deviation for XPS data o is the square root of d The expected value for each n is given by Xr expected r n c n A comparison betwee
30. in the Toolbar If you are using an emulator package on a processor other than an Intel Pentium ensure that you obey the emulator s instructions for installation of applications packages that you understand the limitations if any e g mouse button availability of the system and that you have sufficient space and processing power for the convenient use of CasaXPS Starting CasaXPS Simply click on the desktop short cut if as recommended you installed one or double click on the programme icon in Windows Explorer Ensure that you have stored your Introduction ISO 14976 data files in an easily accessible directory or have the demo data on the installation CD available Ensure also that you have an appropriate element library available one needs to be installed before the first use of Library system see Element Library on page 38 a simple ge neric version is available on the installation disk but check with your instrument operator which is the most appropriate file for the data acquired by that particular instrument At some point you may wish to install and use the on line help files see note on your CD inlay and you should have a def directory prepared if you wish to store synthetic components and other items for future use Terminology CasaXPS component windows appear on screen as shown in Figure 1 where the annotation defines their names as used throughout this manual I
31. in the active tile 8 Scroll through the carbon spectra and check that sensible regions have been defined and that the backgrounds are indeed representative of the data Repeat steps 6 through 8 until satisfactory results are obtained Once the above sequence of steps have been performed for Cs O 1s and Mg 2p it is then time to generate syn thetic components for the Al 2p data Figure 25 shows a representative spectrum taken from a sample for which alu minium metal and aluminium oxides were present To create a model for an aluminium envelope the first step is to create a region In this case the region is best named Re Al 2p and a value of zero should be entered for the R S F value By entering zero for the R S F value any quantification report will not include the intensity calculated from the region in the percentage concentration values Two pairs of components named Al 2p Metal and Al 2p Oxide describes the data seen in Figure 25 These corre spond to two sets of doublets for aluminium This knowl edge can be introduced into the model via the constraints for the component parameters The relative intensity and spac ing of the components of a spin orbit split doublet are usu 52 ally well known and inclusion of this information results in a chemically more meaningful model and also one in which the optimisation routines will work more effectively Armed with the component model for one spectrum fro
32. infinite extent vs defined region The grey area introduced by modelling data with functional forms that blur the boundaries between peak and background are also present when line shapes are extracted from experimental data as well as issues associated with ar bitrarily truncating the model data Good statistics for a model do not necessarily translate into good quantification results for practical reasons Peak Fit Using Doniach Sunjic Line Shapes The preceding discussion attempts to highlight the dangers of modelling spectroscopic data using synthetic line shapes Nevertheless trends within data such as the set shown in Fig ure 85 must be assessed by some means and an approxima tion chosen In the case of the Al 2s peaks the Doniach 118 Figure 85 Al 2s metal data fitting using a Doniach Sunjic line shape Sunjic line shape provides the best form for modelling the observed data Given these doubts it is therefore wise to test the results against other techniques in order to build confi dence in the conclusions In this case PCA is used to cross check Figure 85 shows the initial spectrum in the data set fitted us ing a single Doniach Sunjic line shape Note how the back ground approximation is a simple linear form offset from the spectrum The model from the first spectrum in the sequence is used as the basis for modelling the remaining spectra An additional Trend Analysis for Metal Oxidation Analytical Application
33. only the Re Al 2p region needs to be edited Move the cursor over the name field and then press the right hand mouse button The dialog window for editing the name formula field appears and the current val ues for these two items are entered into the text edit fields Change the name from Re Al 2p to Al 2p then edit the formula to read Al 2p Metal Al 2p Oxide Note that the list above the name formula table contains all the names de fined for the report On the edit dialog window press the button labelled OK observe that the name formula for aluminium has changed to the desired values Table 5 Table 5 Name Formula List after editing Name Formula C ls C ls O Is O Is Al 2p Al 2p Metal Al 2p Oxide Mg 2p Mg 2p The custom report is generated by pressing the Apply but ton on the Report Spec page Nine columns of data are dis played in a scrolled list view The first column is the experimental variable and is followed by eight columns cor responding to the names formulae previously prepared The first four of these eight columns list the raw areas that were specified via the formulae and the second set of four col Quantification umns represents the percentage concentrations for the same items The values within this report may be written to file in an AS CII format Each column of data is TAB separated This al lows the text file to be read into a spreadsheet program such as
34. ordinate val ues of t to load the corresponding abstract factors is com pared to the original target vector A target vector that belongs to the subspace spanned by the primary abstract fac tors should result in a predicted vector that is identical to the initial target vector Errors in the original data matrix and similar errors in the measured target vector mean that the predicted and target vector differ from each other as well as from the pure target vector x x but without error Esti mates for these differences allow a comparison to be made between the predicted and target vector and a decision as to which targets to include in the target combination step The apparent error in the test vector AET measures the dif ference between the test and predicted vectors in a root mean square RMS sense Similarly two other RMS quantities es timate the real error in the target vector RET and the real error in the predicted vector REP These error estimates form the basis for the SPOIL function defined to be approx imately equal to the ratio RET REP Principal Component Analysis by Example The first example illustrating the characteristics of PCA uses a set of artificial data 84 Target Factor Analysis Three sets of spectra prepared from synthetic components are used in the PCA The structure of the artificial data de rives from Carbon 1s states within three compounds name ly PMMA PVA and PVC Figure 53 The proportion
35. output of quanti Mau Domei ia ia Bent Som fied values with the ability to apportion and aggregate areas and set up mathematical relationshipships as well as pro duce standard reports The Data Editor provides access to individual channel counts to enable for example spike re moval Note that this is a potentially dangerous procedure which can modify original data Processing see Processing on page 34 The Smoothing Integration and Differentiation tabs have similar controls and use related procedures Zero ad just in Integration provides for a shift of the minimum or dinate value after integration to the axis zero PCA see Principal Component Analysis on page 80 is an entirely Processing Dialog windows Command Summary different and specialised procedure The two Calibration procedures are straightforward see Energy Calibration on page 35 Intensity Calibration on page 36 as is the Calcu lator The Test Data tab provides not only standardised files for benchmarks but also access to the Monte Carlo pro cedures see Monte Carlo Methods Uncertainties in Inten sity Calculations on page 97 Processing History provides a complete or selective un do function as well as a record of applied methods Dieavades kag Panis PTA Cole Pecenieg Mace Covet iterate Satna Reset removes all processing from the selcted block and re turns it to the ori
36. previously empty window If the tick box labeled Data only is ticked then the spec tra will be copied using the results of any processing otherwise an exact image of the original data will be transferred to the new file An empty frame represents an empty VAMAS file and by copying spectra into an empty frame one creates a new uni form experiment within which it is possible to compare the spectra but in addition the transferred data may subsequent 71 The Calculator ly be saved to disk so that the results of any data manipula tion performed on the spectra are preserved For example the difference of two spectra can be copied to a new file by performing the subtraction and then copying the VAMAS block altered by this operation using the Data Only check box option Any charge correction or other processing can be preserved using this method The Calculator The spectrum calculator is accessible from the Processing Dialog window An important use of the calculator is that of constructing dif ference spectra where the intention is to use the processed data as part of a quantification procedure There is a funda mental requirement for these operations to maintain the in tegrity of the data and therefore a limited set of calculation options is available to support this end The principal limita tion is that the number of acquisition channels must be iden tical in each spectrum used in the calculation The data may
37. scrolled list may be adjusted by select ter the return key is pressed A Quick Tour of CasaXPS 12 Although manual input is achieved through the Regions property page an easy way to check the start and end points for each integration region is to use the Zoom List toolbar buttons and adjust the integration limits under mouse con trol Provided the Regions Property Page is the visible page on the Quantification Parameters dialog window the inte Introduction gration regions on the active spectrum will be marker by vertical bars The presence of these bars indicates that the limits for the regions can be adjusted under mouse control If the cursor is dragged starting at the position of one of these vertical bars the corresponding region end point will be ad justed when the mouse button is released Zoom Out button Figure 10 Cycle through the integration regions using the Zoom Reset button fol lowed by the Zoom Out button The integration regions on a survey spectrum are typically too narrow to be adjusted in this way without the use of a A Quick Tour of CasaXPS zoomed display To facilitate the use of the mouse CasaXPS allows all the current integration regions to be entered onto the Zoom List and then using the Zoom Out toolbar button or Ctrl right click of the mouse the current set of integra tion regions can be zoomed into and adjusted Press the Zoom Reset toolbar button indicated in Figure 10 The spec
38. square and its expected values do seem to point to a 3 dimensional subspace The crossover between the two quantities suggests the need for three ab stract factors when approximating the data matrix using the results of PCA Principal Component Analysis and Real Data XPS depth profiles generate sets of spectra that are idea for examination via PCA The spectra are produced by repeat edly performing etch cycles followed by measuring the count rate over an identical energy region The resulting data set therefore varies in chemical composition with respect to Principal Component Analysis etch time and the common acquisition conditions provide data in a form that is well suited to PCA An aluminium foil when profiled using a Kratos Analytical Axis ULTRA provides a good example of a data set that can be analysed using some of the features on offer in CasaXPS These data are not chemically or structurally interesting but do show how trends can be identified and anomalies isolat ed 82 80 78 76 74 72 70 Binding Energy eV Figure 58 Al 2p Depth profile Figure 58 shows a plot of the Al 2p energy region profiled against etch time The data envelopes change in shape as the surface oxide layer is removed by the etch cycles to reveal the homogeneous bulk aluminium metal It should also be noted from Figure 59 that the data contains an imperfection One of the energy scans includes data ac quired during an instrumenta
39. synthetic components used to model the O 1s region with the intensities from the integration regions for the C ls and the Al 2p regions Table 14 has been calculated from a combination of the Peak O s intensities and the previously used Al 2p and C Is regions The real value comes when the O Is region say has an interference peak which requires excluding from the final results by fitting multiple peaks to the data but only selecting a subset of these peaks for the purposes of the quantification report intensities Peak Al 2p C ls Peak O Is Al 2p Cls CPSeV CPSeV CPSeV O 1s 17024 9 11360 8 2860 9 54 4856 36 3585 9 15586 Adjusting a Custom Quantification Report 122 Using Different File Formats In order to further enhance the cross platform approach in herent in ISO 14976 CasaXPS provides several file conver sion routines for the native file format of systems which either do not have any VAMAS output capability or which do not implement the specification fully brief descriptions of procedures for these conversions are given below Kratos DS800 Binary Files Kratos DS800 binary files can be directly converted through CasaXPS The first step to creating anew VAMAS file is to enable the Convert toolbar button The Convert option is disabled un less the selected sub frame window representing a file con tains no VAMAS regions If no empty sub frames a
40. the asymmetry function a background approxi mation must be introduced to facilitate the optimisation procedures that ultimately determine the peak parameters 117 The Al 2s line is an interesting one to study since the data envelopes should result from only two photoelectric lines namely the metal and the oxide Al 2s resonance features al though a small plasmon structure from the Al 2p line should also appear at the same BE as the Al 2s lines This is in con trast to the Al 2p data envelope since in this case the data en velope is constructed from doublet structures that therefore complicate the nature of the model The data set under study offers a near pure Al 2s Metal spectrum from which the asymmetry can be assessed and then applied to the subse quent spectra in the oxidation sequence The theoretical line shape for a metal such as the aluminium has been shown to have an asymmetric line shape given by Doniach and Sunjic however from a practical perspective the Doniach Sunjic line shape presents a number of prob lems The principal problem is the question of area under the curve and how it relates to the intensity of the photoelectric line The Doniach Sunjic line provides a very good fit for the observed data provided the background does not attempt to model the background contribution from the Al 2s line it self The functional form for the Doniach Sunjic line typical ly extends beyond the data region and therefore integrate
41. the distribution but are not present when the same experiment is performed but with an improved background specification Figure 73 Monte Carlo Simulation for Two Peaks of Same Intensity and Width But Separated by 1eV Intensity Peak 2 Intensity Peak 1 Figure 73 Individual intensities Linear background with twenty one end points Monte Carlo Methods Uncertainties in Intensity Calculations Summary Monte Carlo simulation provides a valuable tool for under standing the uncertainties associated with data reduction for AES XPS spectra The results presented here are available in other forms 5 161 07 however the visual feedback of fered by Monte Carlo simulations provides an insight for analysts who feel less comfortable with a mathematical de scription of the same concepts By using Monte Carlo simu lation an understanding of the errors involved in extracting information from AES XPS spectra is hopefully opened to a wider audience 16 P J Cumpson and M P Seah Random Uncertainties in AES and XPS Peak Energies Areas and Quantification NPL Report DMM A 26 May 1991 Surf Interface Anal 18 345 1992 and 18 361 1992 17 S Evans Surf Interface Anal 18 323 1992 106 Analytical Applications The following sections provide examples of the treatment of particular analytical problems in greater depth emphasising the special tools and procedures available within CasaXPS
42. tries to illustrate the nature of the problem Consider a set of three spectra each spectrum has three ac quisition channels s1 4 3 6 s2 2 3 2 s3 2 0 4 The data matrix is given by 422 D 330 624 These three vectors belong to a 3 dimensional space how ever they do not span 3 dimensional space for the following reason If a linear combination of the vectors s1 s2 and s3 is used to construct a new vector v then v always lies in a plane a 2 dimensional sub space of 3 dimensional space The fact that v lies in a plane is a consequence of the following relationships between the three spectra s3 s1 s2 SO v asl bs2 cs3 asl bs2 c sl s2 a c sl b c s2 Thus two principal components exist for the set of three spectra The analysis of the data matrix in the above simple example Theory of Principal Component Analysis Principal Component Analysis has been performed by observation Unfortunately real spec tra are not so simple and spotting the linear relationships be tween the columns of the data matrix requires a more sophisticated approach PCA also known as Eigenanalysis provides a method for identifying the underlying spectra that form the building blocks for the entire set of spectra The data matrix is trans formed into a new set of r dimensional vectors These new vectors span the same subspace as the original columns of the data matrix however they are now characterise
43. ure 35 For this reason it is essential to use the Doniach Sun jic line shape in situations where the asymmetry parameters are close to the same value so that calculated areas can in some sense be compared Modifications to the Doniach Sunjic function The advantages of the Doniach Sunjic profile are lost when synthetic models are required for quantification purposes It is necessary to introduce ad hoc cutoff behaviour before fi nite intensities can be reported and once obtained the peak areas can only be used in a quantification reports when rela tive sensitivity factors are know for the specific cutoff crite rion adopted Empirical observations suggest that pure Doniach Sunjic profiles do not always yield good fits for electrons with higher kinetic energy KE when monochromatic X ray sources are employed For example Al 2p doublets record ed using a monochromatic X ray source appears less Lorentzian in nature to the higher KE side of the peak than the Doniach Sunjic shape will allow Figure 36 however the same line measured using an Aluminium anode does ex hibit the predicted but somewhat broader shape The latter is thought to be due to the double influence of the Doniach Sunjic shape from the photoemission process as well as the X ray line profile while monochromatic X rays have lost the Doniach Sunjic distribution as a consequence of the fil 62 tering process M zla Hira Pre Pa L Area kiia ao GOS DL a
44. use XPS for surface characterisation In contrast to the special purpose control systems that form part of commercial XPS instruments CasaXPS offers a compact portable efficient and user friendly processing system to anyone with an IBM compatible Pentium PC running Microsoft Windows 95 or later or suitable emula tor It incorporates much of in many cases more than the processing functionality of the instrument linked packages without recourse to unfamiliar operating systems or hard ware or proprietary file formats It is designed from the out set on the basis of the ISO 14976 Surface Chemical Analysis Standard Data Transfer Format and so by design has a uni versal cross platform approach independent of anything other than a reasonable adherence to the ISO standard Spec tra collected in the standard format may be selected viewed and processed in a simple yet powerful way and the results of CasaXPS data reduction presented in a variety of graph ical and tabular formats are available for incorporation into CHAPTER 1 and use as data by other popular Microsoft packages such as Word or Excel Additional spectrum input filters for other commercial file formats are also available supple menting the ISO standard and broadening its usefulness Casa XPS has been written entirely in native C using the Microsoft Developers Studio programming environment It employs Microsoft Foundation Class MFC libraries t
45. 1 10861690000 266 7944 615 8053 194 7347 7602 534 29055 89 1800 C 1s 2 684568100 13 0963 29 86145 13 35445 466 5852 172 5176 1592 C 1s 3 1433862 0 01064209 0 02964045 0 01623474 0 604907 1 0 000190564 1386 C 1s 4 1 230771 0 000391784 0 002107629 0 001332982 0 05854525 9 86356E 08 1182 C 1s 5 0 0037 12633 0 000385433 0 001279202 0 000904532 0 05116807 1 20367E 07 980 C 1s 6 0 000545838 0 000230365 0 001168993 0 000905498 0 07306205 4 86336E 08 780 C 1s 7 0 000473059 0 000240069 0 001018602 0 000852223 0 113178 5 18512E 08 582 C 1s 8 0 000306354 0 000155331 0 00089 1207 0 000797119 0 2228016 2 20338E 08 386 C 1s 9 0 000200465 0 00012725 0 00076887 0 0007294 14 0 7688698 1 91008E 08 192 C 1s 10 0 000118823 4 68061E 13 0 0 0 6 13259E 23 0 It is interesting to see how the eigenvalues change with re spect to the three data sets Figure 55 and Figure 56 The same spectra varied in different ways results in slightly dif ferent orientations for the principal component axes and hence different eigenvalues The PCA statistics IE and IND have implied a dimensional ity other than three Table 6 The clue to the correct dimen sionality of the data lies in the relative size of the eigenvalues The fourth eigenvalue is in two cases better than five orders of magnitude smaller than the third eigen value This statement has been made with the benefit of a good understanding of what is present in the data In real sit uations su
46. 2 4762 1 08333 It is very useful to try to fit the artificial peaks without any constraints with respect to the other peaks and then gradual ly include the relationships shown in the above table The chi square value for the curve fit should go to zero This is the case since the artificial data does not contain noise Or dinarily in the presence of noise the chi square value should be about equal to the number of degrees of freedom also shown on the Components property page 47 Regions 296 294 292 290 288 286 284 282 280 Binding Energy eV Figure 26 PIB artificial peak structure A good exercise is to attempt to reproduce the peak envelope for PIB What should be observed is how difficult it is to produce a good fit for PIB The three peaks are very hard to identify unless additional information is given to the fitting procedures If the offsets are first provided then the fit im proves but the exact match does not materialise until both the correct intensity ratios are supplied as well as the relative positions Another useful feature of fitting this artificial data is that the strengths of the two fitting algorithms on offer can be seen PIB is a stern test for the Marquardt method The uncertainty as to which of the many combinations of similar peak pa rameters presents a plateau in the parameter space that fails to give any good direction towards the optimum values On the other hand the Simp
47. 2p 33 19 83736 11 47927 16 17861 1 409376 0 42997 0 065749 Al 2p 38 19 8507 12 01348 15 80274 1 315418 0 44268 0 10609 Al 2p 43 19 9069 12 46508 15 52116 1 245171 0 4531 0 133366 Al 2p 48 57 16561 12 70691 55 73546 4 386233 0 45854 0 14688 Al 2p 53 15 37333 13 18052 7 912861 0 600345 0 46791 0 174614 Al 2p 58 21 39836 13 30379 16 76004 1 259795 0 46901 0 184805 Al 2p 63 19 92528 13 5238 14 63296 1 082016 0 47386 0 195062 Al 2p 68 27 13522 13 78354 24 06775 1 746122 0 48087 0 203826 93 Principal Component Analysis Target Factor Analysis Table 9 Target Test Report for a Subset of Al 2p Data Set Target AET REP SPOIL Al 2p 3 Al 2p 8 Al 2p 73 19 10189 13 88023 13 12332 0 945469 0 48192 0 210646 Al 2p 78 20 9575 13 98145 15 61204 1 116625 0 48264 0 218455 Al 2p 83 19 03813 14 15492 12 7314 0 899433 0 48483 0 229382 Al 2p 88 18 38591 14 11378 11 78317 0 83487 0 48374 0 228046 The SPOIL function and AET statistics Table 9 show that Al 2p 48 differs in some respect from the other spectra in the list tested The spectrum in question corresponds to the trace displaying the spikes seen in Figure 58 Also another spec trum that could be looked at is Al 2p 68 The AET value is high compared to the other spectra Such spectra may high light interfaces where either new chemical states appear e1 ther directly from features in the data or indirectly through changes in the background due features ou
48. 31 page found on the same dialog window 80 104 FWHM Area 70 3 37292 1793731 8 45 52 4 25922 1183253 1 39 04 60 4 08363 150498 4 15 44 1 7 r T T phd 1000 800 600 400 200 0 Binding Energy eV Figure 20 Annotation using region information Moving Annotation and the History Mechanism Annotation either tables or text can be moved using the list of annotation entries currently defined for the active spec trum The spectrum display will be updated with the annota tion now at the new location Position for annotation items may be referenced to the display tile frame or the data itself for convenient placement e g when multiple blocks are dis played in overlay and labels may be orientated vertically or horizontally Positioning targets small squares with central dots may be set with the mouse anywhere within the display tile and disappear when the annotation window is dismissed Other attributes for the annotation can be adjusted using the history list Select the line describing the annotation for Moving Annotation and the History Mechanism Graph Annotation which the change is required Then adjust the settings using the options found below the history list Press Apply to see the result of the changes Text lines may have the orientation adjusted in addition to altering the contents via the text edit box Fonts and colour are reselectable through the history list and the edit options on the A
49. 41 These markers serve only as a guide to the presence or absence of a feature in the spectrum a peak maximum which you may want to identify for example by using the Element Library routines as described on page 9 Markers may be set and cleared independently of Peak Labels use the Clear Elements button for the Labels TF Graph Annotation Text Annotation Peak labels are merely multiple instances of text annotation Individual lines of text can be added to the display using the Text property page A text edit box allows the desired text to be entered and any attribute such as vertical horizontal font or colour may be chosen before pressing the Apply button 33 Processing An XPS spectrum represents an envelope that is derived from many unresolved peaks The relative intensity of these peaks may vary as a function of the kinetic energy imparted to the electrons being recorded as well as the characteristics of the electron optics lens mode pass energy In addition the position of the peaks may differ from the expected value due to a combination of instrument energy calibration and charging effects on the sample CasaXPS therefore provides a number of processing options that assist in understanding the data envelope and allow adjustments to the spectrum to enable compensation for these analysis variables The processing options are available from the Options main menu item or from the main toolbar button sho
50. 60 1E 37 additional params 61 0 ordinate values 62 878 min y 63 427 maxy 64 15744 first data point 65 2967 first transmission value 1 second data point 66 2954 second transmission value 1 875 total data amp transmission points last data point 950 458 last transmission value 951 1 block i d 952 OSnInaa 6 sample i d 953 OSnIn 6 year 4 digits 954 1999 month 955 12 day 956 22 hour 957 17 minute 958 43 second 959 27 hours GMT 960 0 lines in block comment 961 3 Lens Mode Magnetic Resolution Pass energy 80 Anode Al 180 W Step meV 300 0 Dwell ms 136 Sweeps 1 Acquired On Tue Dec 22 17 43 27 1998 technique 965 XPS experiment variable value 30 source 967 Al source energy 968 1486 6 source strength 969 180 and so on as above for this block block 2 last data point 1884 504 last transmission value 1885 1 block i d 952 OSnInaa 11 147 sample i d 953 OSnIn 11 and so on as above for this block block 3 and also for another 114 blocks until last data point 15894 504 last transmission value 15895 1 terminator 15896 end of experiment Appendix 2 ISO 14976 and the World Wide Web MIME type The ISO 14976 file format is designed for simple unambiguous communication of information derived from surface chemical analysis experiments The World Wide Web is one of the most important communicatio
51. 622169 0 5048702 0 3569971 20 19481 1 247441 980 C 1s 6 226 9936 0 01656361 0 19049 14 0 147554 11 90571 0 002773735 780 C 1s 7 29 13143 0 002688086 0 00847462 0 007090376 0 9416245 5 0833E 05 582 C 1s 8 0 03943089 0 000339289 0 003105158 0 002777337 0 7762894 3 05649E 07 386 C 1s 9 0 003409306 0 0002063 14 0 001523904 0 001445703 1 523904 9 22657E 08 192 C 1s 10 0 000466779 1 26006E 12 0 0 0 7A912E 22 0 Real data includes noise The effect of noise on a PCA cal culation can be see from Figure 57 together with the report in Table 7 The data in the file clstestl vms has been used together with a pseudorandom number generator to simulate noise that would typically be found in XPS data The conse quence of including a random element in the data is that the eigenvalues increase in size and lead to further uncertainty with regard to which eigenvalues belong to the set of prima ry abstract factors Note that the abstract factors in Figure 57 are plotted in the reverse order to the ones in Figure 55 and Figure 56 89 Target Factor Analysis Abstract Factors After Introduction of Noise Eigenvalue 5095 0 946795 2 715206 6 872e08 1 088e10 300 298 296 294 292 290 288 286 284 282 280 Binding Energy eV Figure 57 clstestl vms data plus artificial noise Fortunately the chi square statistic becomes more meaning ful when noise is introduced into the problem A comparison between the computed chi
52. Comment Edit Block Information Block Block E h Set Experiment variable linear Edit Species Transition i Edit Source Analyser parameters For Display Rs 1044 135 Processing Dialog windows Page Layout see Tiles of Spectra on page 26 Ja ofa a l3 7 u jn E ow uw wn le Vide bagaj i ym Efe Coleen ras ms Irra Fa hima of Poradi Par ate ra wm om Tirirgra Ce a _ Sixteen predefined layouts are provided corresponding to arranging spectra in symmetrical rows up to the maximum 16 allowed All predefined formats may be changed as re quired for example to provide a project specific layout set the three controls in each layout functioning as radio but tons allowing selection of only one alternative or combina tion for each case Tile Display Parameters see Tile Display on page 27 The six tabs give access to tile individual spectra param eters in a straightforward way Settings may be applied to in dividual tiles or all tiles in an Experiment via Global tab X Axis Independent variable provides a radio button choice be tween BE and KE for x axis display and a means of label ling the axis as desired the type in boxes and setting the Processing Dialog windows Command Summary typeface Font esis Via Gaonaip ali cim iai F Tae Min and max enable setting the span of the display which may be greater
53. FHA m 6 ti oh ra Lo 55 2S Ja 10000 Ed Bo GN SANG ee W Erig Beer hY ai Beige Fee ht Figure 89 Synthetic peaks for the O 1s spectrum are now named differently from the integration region 121 problem lies with the same name being used for more than one quantification item The integration region for the O 1s spectrum has the same name as both the synthetic compo nents and therefore the three quantities associated with the O Is spectrum are summed to provide the intensity used in the custom report This was not the intention To prevent this type of ambiguity it is best to use a different name for the integration region from that used for the synthetic components Figure 89 shows the same data where the synthetic peaks have been as signed a common name which is different from the integra tion region A Custom report generated from this newly named set of quantification items is shown in Table 13 and can be seen to be identical to the report in Table 11 Table 14 Custom report using the Synthetic Peaks for the O 1s intensity but the integration regions for the C 1s and Al 2p Analytical Applications Table 13 Custom report after the synthetic components have been renamed Ols Al2p Cis Gpsev Cesev Cese viS SAAP Es 2 17050 5 11360 8 2860 9 54 523 36 3287 9 14836 The value of the Custom report lies in the ability to combine intensities such as those from the
54. Figure 65 and Figure 66 By projecting horizontal or vertical lines onto the parameter axes in such a way that the appropriate number of points lie between the projection lines for a given confidence limit the uncertainty for a parameter can be assessed in the context of others Again the uncertainty for a parameter should be viewed in the context of all other correlated distributions and yet the ellipsoid in 6 dimensional parameter space is dif ficult to quantify the procedure based on scatter plots should be seen as merely a step in the right direction rather than arriving at the ultimate destination In the case of the peak area shown in both Figure 65 and Figure 66 the esti mate taken from the diagonal element of the error matrix would seem to be reasonable Both scatter plots show that about 65 of the points lie inside projection lines positioned at about 0 95 and 1 05 This interval represents about 5 of 114 2 CPSeV see Figure 63 and is not too different from the uncertainty taken from the error matrix 6 2 CSPeV Monte Carlo A Simple Example Monte Carlo Methods Uncertainties in Intensity Calculations Eemia Pci Fo Peab Arena Arpe Me Pegh T Mima Maii Fiai T Figure 65 Scatter plot showing the anti correlation between the peak area parameter distributions All the parameter distributions are reported relative to the in itial parameters used to define the data envelope As a result the relative parameter distri
55. HATO Duna a e iien CasaXPS User s Manual for Version 2 0 October 2001 Product design Casa Software Ltd Documentation Acolyte Science Manual version 1 2 CasaXPS and the User s Manual are copyright 2001 Casa Software Ltd All rights reserved Information in this document is subject to change without notice and does not represent a commitment on the part of the copyright holder The software described in this document is furnished under a license agreement No part of this publication may be copied reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photo reproduction recording or otherwise without the prior written consent of Casa Software Ltd Acknowledgements CasaXPS wishes to thank all those connected with the production of this manual and the software system it describes In particular thanks are due to those who provided experimental data results and interpretations which form the bulk of the examples here Notable among these contributors are Morgan Alexander Graham Beamson David Briggs Peter Cumpson Kevin Harrison Len Hazel Fran Jones Simon Page Roy Paynter and Martin Seah All third party Trade Marks are acknowledged and used without prejudice Any references to company names in sample output and tem plates are for demonstration purposes only and are not intended to imply any endorsement or refer to any particular
56. PCA without background subtraction shows the smoothest trend in Figure 87 and the limited input assump tions suggests this curve is the best description of the trend within the data set although not necessarily the best descrip Adjusting a Custom Quantification Report Analytical Applications tion of the chemical processes involved The trend analysis based on PCA is modelling changes in both the background and the intensity of the photoelectric lines but does show that these adjustments throughout the data set are well mod elled by two underlying shapes in the form of abstract fac tors Adjusting a Custom Quantification Report The standard quantification reports treat each quantification item as a separate entity and therefore a percentage concen tration is reported for each item used to quantify a sample Custom reports on the other hand sum the intensities for any quantification items defined with the same name It is therefore important when using the Custom report option to label synthetic peaks with a different name from the region used to define the background for those peaks Consider the following example Figure 88 shows three high resolution spectra used to quan tify a sample The O 1s region is fitted using two synthetic components whilst the C 1s and the Al 2p spectra are quan tified using integration regions only Each spectrum has an integration region defined and the names for these regions are O 1s Al 2p a
57. R OS DEOL emos sia f kr mE DIEI DSO semia ate aj r f l al h r it 7 i 1 ij 4 i E a lj L j I i i t a j 1 f J t l 1 T t j w F k Fi P i F ta 4 e Erp Poa i is Dikt Bry fee Figure 32 Au 4f doublet fitted using an iterated Shirley back ground The line shape used in the fit is a Doniach Sunjic form con voluted with a Gaussian that results in an asymmetry index of 0 008 for each peak between the two peak fits is the reduction in the amount of asymmetry required in the line shape when a Shirley 5 5 Shirley D A Phys Rev 55 4709 1972 Line Shapes and Backgrounds background is used This is best measured by the asymmetry index 6 which changes from 0 49 in the case of a linear background to 0 008 when a Shirley background is em ployed Although theoretically based the Doniach Sunjic profile suffers from ill defined areas The integral of the Doniach Sunjic function is infinite for non zero values of the asym metry parameter 7 and therefore any direct use of this line shape requires a somewhat arbitrary use of cut offs to allow finite peak areas to be reported Changing the shape of the peaks through adjustments to the asymmetry parameter and the width of the broadening function has altered the relative proportion of the spin orbit split peaks in the Au 4f doublet Figure 31 and Figure 32 The background choice has al tered the repor
58. Shirley background SGL p K b0 b1 Gaussian Lorentzian sum formula modi fied by a Shirley type background prescribed by Castle et al A linear polynomial determined from bO and b1 adjusts the step in the Shirley background Further adjustments to the basic shapes The basic shapes that result from the various functional forms can be further modified by numerically convoluting the profile with a Gaussian For example GL 100 50 will convolute a pure Lorentzian with a Gaussian characterized by a width of 50 The value for the width is the number of digital nodes used to describe the Gaussian and is therefore an arbitrary unit characteristic of the numerical form for the Gaussian N B Any line shape that requires a digital convolution will cause slower performance than a shape that does not involve this procedure 60 Asymmetric Line Shapes The exponential tail described by Sherwood in the book Practical Surface Analysis edited by Briggs and Seah at tempts to approximate asymmetric line shapes commonly found in photoelectric peaks Several researchers have of fered alternative deformations of the symmetric Voigt pro file however none but Doniach Sunjic backs their approaches with a theoretical basis Nevertheless practical surface analysis requires practical solutions and the need for a rigid line shape model that matches the observed profiles makes ad hoc forms acceptable tools Philosophically theo retically based solutions ar
59. Software Ltd Printing and Help S E p 26 Burford Crescent Wilmslow Copy Display bitmap Cheshire SK9 6BN Save File to Disk United Kingdom Print Current Spectrum Dispayd g Convert File Screen Preview for Display A Help About Casa XPS Open File to Clipboard L New File Experiment Frame The Button bars File Access buttons File Processing buttons Ko Variable Calibration controls Bi A mj veja Block comment info controls Display Properties buttons Display Modifier buttons Display Options buttons Display Scaling buttons Decrease intensity scale maximum Launch Annotate window Increase Energy scale range t Launch Library window Increase intensity scale maximum Reset Intensity scale to original maximum Launch Processing window Step zoomed energy scale left 7 olem i se nal Waal kad Era kad at File Processing buttons Display Scaling buttons Step zoomed energy scale right Launch Quantification window Reset to original scale L h Tile Displ i L ue i neti window Zoom out step back round history aunen Fage cayout Window Zoom in requires selection box rescales intensity to suit t adds 20 left and right Display Options buttons Display Properties buttons B S a o PERE g EEL t tri state buttton Insert many blocks into current scale Toggle Offset for multiple traces t Insert one block into current Display scale Toggle 2D 3D a
60. Tougaard can be only ap plied after accounting for such variations in the intensity The Intensity Calibration property page allows the inten sity to be adjusted using a function of a constant power of the energy The same page on the Processing dialog window allows the inclusion of a relative transmission function to be specified The ISO 14976 file format does not specify how the trans mission function for an instrument can be incorporated with the data however CasaXPS does allow the use of a second corresponding variable per block to define the transmission behaviour for the instrument See Appendix 1 ISO 14976 format files annotated on page 143 This is particularly important when a single relative sensitivity factor R S F is used to quantify results from a range of magnifications and pass energies Without first correcting for the differences due to operating modes of the instrument a single R S F would not yield consistent quantification results for the same sample If a valid corresponding variable index is specified the act of calibrating the intensity modifies the data by dividing the raw spectrum by the indicated transmission function Processing Processing History Each VAMAS file data block can be processed using any of the available options The Processing History property page provides the means of viewing the operations that have been applied to the data Further the processing may be re versed enti
61. aa ERS kya Hew DIARI DEL Se Be Figure 36 Al 2p Doublet from a monochromatic X ray source In order to fit spectra such as that shown in Figure 36 the pure Doniach Sunjic shape has been modified The lower KE asymmetry derives from the functional form for the Do niach Sunjic profile while the higher KE side of each com ponent is a Voigt type function A numerical convolution is applied to the combined profiles to produce the line shapes in Figure 36 Given an approximation to the Doniach Sunjic profile of the form shown in Figure 36 then the possibility of characteriz ing the line shape by the associated Voigt type profile be comes feasible The Voigt type portion of the line shape provides the position width and intensity while an asymme Line Shapes Available in CasaXPS Line Shapes and Backgrounds try index 6 measures the departure of the line shape from a symmetric form Such a regime fits well with the philosophy adopted by Shirley where the background shape was used to reduce an asymmetric step to a form that could be character ized by a Voigt function The asymmetry index is given by 7 fwhmiefi fwhm a right The asymmetry index is reported on the Components Prop erty Page for each line shape H a n GL p Hybrid Doniach Sunjic Gaussian Lorentzian product line shape H a n SGL p Hybrid Doniach Sunjic Gaussian Lorentz ian sum line shape The H form of the DS GL hybrid line shape is identica
62. abstract factors as a function of the ex perimental variable involved Here the term principal abstract factor means those abstract factors that contain sig nificant structure and are necessary for providing an ade quate description for the entire set of spectra Plotting the co ordinates for the spectra with respect to the subspace spanned by the principal components shows how each of these designated abstract factors contributes to the descrip tion of the spectra and depending on the nature of the ab stract factor trends can be identified No a priori model is required by PCA although pre processing the spectra can be used to assess the consequences of including for example a particular background approximation Principal Component Analysis is not a substitute for the con ventional quantification techniques and requires care before the results can be used to assess trends There have been at tempts to place physical meaning on the abstract factors but it should always be remembered that the abstract factors are truly abstract and owe their nature and size to a mathemati cal procedure for generating an orthogonal basis set for an n dimensional vector space The first abstract factor is chosen to be in the direction of the maximum variation in the entire set of spectra in the linear least square sense Subsequent factors are generated with the same criterion as the first but subject to the constraint that each new vector is orthogonal to
63. aks are defined in terms of intensity position and FWHM taken from the un derlying Voigt function This is not unnatural in view of the fact that the area beneath all of these curves would be infi nite without the introduction of some arbitrary cutoff limits DS a n The basic Doniach Sunjic profile is defined in terms of the asymmetry parameter a and a convolution width 61 n The profile defined above is numerically convoluted with a Gaussian whose width is determined from n to produce the final line shape Although the Doniach Sunjic line shape allows very good fits to experimental data the infinite intensity defined by the curve makes this profile difficult to use under practical situ ations Any scheme that limits the area through cutoff func tions or ranges introduces inconsistencies into the peak parameters that make relationships such as stoichiometry invalid The degree of asymmetry used to describe the data Doniach Sunjic Profiles 1 a 01n 0 a 0 2n 0 i meen ii g o T T m T n 948 950 952 954 956 958 960 962 Figure 35 Doniach Sunjic profiles for asymmetry parameter equal to 0 1 and 0 2 Note how the asymptotic curves cross over illus trating how intensity is transferred away from the peak maximum Line Shapes Available in CasaXPS Line Shapes and Backgrounds moves intensity modelled by the line shape away from the primary peak position and towards the cutoff regions Fig
64. although the full set of ISO file formats can be read by CasaXPS excluding the annotated form only a subset can be displayed and manipulated ISO 14976 File Format The ISO 14976 Surface Chemical Analysis Standard Data Transfer Format was designed to enable uniformly encoded transmission of data originating from a variety of instru ments which in turn might employ a variety of techniques to any appropriate processing system local or remote for its interpretation within specific applications A desire to unify descriptions for SCA data and so reduce the number of pro grams necessary to manage the increasing number of propri etary formats led to a specification of how data files should be ordered formatted and encoded This work identified and steered by a Technical Working Party of the VAMAS project see ISO 14976 VAMAS Files on page 17 in the early 1980 s was taken up by the International Stand ards Organisation ISO Technical Committee TC 201 Sub committee 3 and an internationally agreed standard was produced 2 ISO 14976 1998 Surface Chemical Analysis Data Transfer Format ISO TC 201 ISO Marketing Services Case Postale 56 CH 1211 Geneva 20 Switzerland http www ISO org CHAPTER 2 One natural consequence of defining a standard data format independent of data origin is that the file must contain all the information required to process interpret the data An ISO 14976 file is t
65. ame name but with a vms extension added The spectra will be converted and other experimental information ex tracted so that the ISO 14976 file may be quantified to pro duce identical results to those from the Vision systems That is transmission correction is included as part of the ISO 14976 file Element library files from the Vision 1 x system can also be converted to CasaXPS format and Vision 2 x element librar ies can be converted on request Using Different File Formats SSI M Probe Files SSI M Probe files may exist in an ASCII format where mul tiple regions appear in the same file or alternatively the data is described by a control file with the spectra stored in sepa rate reg files in a sub directory where the sub directory has the same name as the base name of the control file The first step to creating anew VAMAS file is to enable the Convert toolbar button The Convert option is disabled un less the selected sub frame window representing a file con tains no VAMAS regions If no empty Experiment Frame is amongst the list of windows then the New toolbar button must be pressed to create an empty Frame There are two methods for loading M probe files e ASCII format files must have an mrs extension and may be opened directly through the Convert Dialog win dow e The binary format consists of an mrs file together with a directory of the same name as the base name of the mrs file To load the information hel
66. ame that flags the type of conversion required however the new file created by CasaXPS will replace the entered extension by vms e The third method converts depth profile type file struc tures Using the Covert to VAMAS file dialog window move to the directory that contains the sub directories containing the spectral regions Whilst at the level of the sub directories enter the name of the new file name and add a mle extension The mle flags that CasaXPS should read each sub directory within the current direc tory and convert all the dts files found within the sub directories The new file name will replace the mle 124 Using Different File Formats extension with vms Figure 91 shows the dialog window these directories is shown in Figure 92 el a a Ripp Fouche ree Comoe j Figure 91 Convert to VAMAS File Dialog Window Bir o m M RE OL ly peme Dette teme Deeps er er rer bere Tre Pare ee Puta Hebe BENERA arbre eee Tiere Piers rer rors where the visible directories are about to be converted using the mle flag The file name profile mle indicates a that each of the directories form part of a depth profile Figure 92 VAMAS File Block Structure from a VG Eclipse and should be interpreted as the names for the regions Depth Profile found inside these directories The result of converting VGX900 Ron Unwin Ron Unwin developed the VGX900 system and so all the various forms of Ron Unwin s
67. amely Target Testing Target Testing Once a Principal Component Analysis has been performed the mathematical bridge between abstract and real solutions is Target Testing Individual spectra can be evaluated to as sess whether the corresponding vector lies in the subspace spanned by the chosen primary abstract factors The essen tial feature of Target Testing is to form the projection of the target vector onto the subspace spanned by the primary fac tors then compute the predicted target vector using this pro jection Statistical tests applied to the predicted and test vectors determine whether these two vectors are one and the same These tests serve as a mean of accepting or rejecting possible fundamental components of the sample Ultimately the goal of target testing is to identify a set of spectra that span the same subspace as the primary abstract factors Complete models of real factors are tested in the tar get combination step In the combination step the data ma Principal Component Analysis trix is reproduced from the real factors spectra rather than from abstract factors and by comparing the results for differ ent sets of real factors the best TFA solution to a problem can be determined Testing a target vector x with respect to the chosen set of pri mary abstract factors involves forming the projection t onto the subspace spanned by the PCA primary abstract factors The predicted vector x calculated using the co
68. ameters to ENA gt eT il Say Ri A XA Figure 70 shows a Monte Carlo simulation for the peaks i nS ar IRS KAS E Ar Be gt shown in Figure 67 Each Monte Carlo simulation step takes da Fe e E the envelope for the two peaks plus the background then su perimposes a normal deviate of mean zero and standard de ose viation one scaled by the square root of the counts in each ies data channel The envelope constructed in this way simu Two Peaks of Same intensity and Width But Separated by 16V i 0 50 100 150 200 250 300 350 400 450 25 Figure 69 Peak areas end points averaged 2 0 Not all peaks are isolated and so under these situations it ee falls to the acquisition phase to improve the statistics for the 3 data channels that determine the background An acquisition EEE system that accurately determines the points used to define the background will yield intensities with significantly more os precision than those that treat the end point as equal to any other in the acquisition region i 0 0 25 Intensity Of Peak 1 15 K Harrison and L B Hazell Surf Interface Anal 18 368 1992 Figure 70 Linear background using single end points Intensities Determined by Peak Fitting 104 lates a sequence of experiments performed on the same sam ple so that the only variation in the data is the random noise inherent in the acquisition process For each Monte Carlo envelope the peak parameters are refitted The points u
69. and Briggs D The XPS of Polymers Database CD ROM Surface Spectra Ltd 2000 Jones F et al Fluoride uptake by glass ionomer cements a surface analysis approach submitted to Biomaterials Press W H et al Numerical Recipes in C Cambridge University Press 1988 Cumpson P J and Seah M P Random Uncertainties in AES and XPS Surface and Interface Anal 18 361 1992 Appendix 5 Quick Reference Card following two pages for printing 151 Keyboard amp Mouse Shortcuts Function Shortcut New Experiment Frame Ctrl N Browser Open ISO File Ctrl O Print graphic Ctrl P Copy graphic to clip Ctr C board Set Normalise point Shift left click To Contact U S Zoom out cycles Ctrl right click R Tn Phone OMEA MENU right click 44 0 1625 535346 Edit a data point Ctrl Shift left click E Mail sda lace i eee QUICK REFERENCE CARD Overlay spectra F2 World Wide Web Zoom out F3 http www casaxps com Export Tab ASCII Reset Zoom F4 Export MetaFile Copy Display MetaFile F1 F4 are single press Function keys for the other commands hold down the control Ctrl and or shift key _ and press the listed key or mouse button Main File Access Bar Drag and Drop you may load ISO 14976 files by dragging them from Windows Explorer and dropping onto an open Crates CasaXPS Programme Frame m Casa
70. and side shows data that has been modified by ybl or VAMAS BLOCK index 1 where the indices start at vb0 72 The Calculator Figure 46 a The spectra selected in the browser will be modified by the spectrum displayed in the active display tile Normalizing the data which are the result of a calculator op eration with respect to one another is sometimes necessary One such situation arises when a peak lies on a plasmon loss structure Chromium oxide supported on SiO catalyst sam ples produces a spectrum where the O 1s line includes a pro nounced loss structure associated with SiO which impedes measurements for Cr 2p owing to the complex nature of the background particularly when the amount of chromium is small If a standard spectrum is available representing the SiO catalyst material prior to modification by the Cr sub tracting the standard may result in data for which common Using the Calculator and Comparing Spectra Figure 46 b Result of the operation shown alongside background types are appropriate Figure 47 shows two spectra taken from a Cr 2p region The O 1s plasmon struc ture is very evident in both spectra and clearly interferes with a consistent choice for the background to the Cr 2p lines The first step in quantifying the Cr 2p lines is to charge correct one spectrum with respect to the other Once per formed the scale factor representing the difference in count rates may be estimat
71. approx imations and there can be no doubt about the uncertainty as sociated with the different background algorithms It is therefore entirely possible to fit a set of synthetic compo nents with good statistics yet without any chemical or phys ical meaning To provide a guide to peak fitting an option on the process ing dialog window labelled Test can be used to replace the true data using one of a set of known peak structures These structures derive from work presented by Seah and Brown The relative intensities and separations are those presented in Table 4 in the Seah and Brown publication a GL 50 line shape has been used to generate the components It is useful to exercise the fitting procedure using these structures with different options for the synthetic components especially when a background has been added The following table provides the characteristics of the artifi cial peak structure Note that the data replaced determines step size energy position and count rate 4 M P Seah and M T Brown J Elec Spec 95 1998 71 93 Quantification Table 2 Artificial Peak Structures Carbon C 1s peaks are used to form data envelopes with the offsets and relative sizes list below PMMA PVA PVC PIB Numberof 4 4 2 3 peaks peak 2 2 2 2 offset eV 2 2 2 6 1 1 0 4 factor 1 238 1 04167 1 0 2 0 peak 3 3 3 offset eV 3 3 3 7 0 6 factor 1 238 1 04167 1 0 peak 4 4 offset eV 4 4 2 factor
72. are first selected is set to the colour just defined Each colour subsequently specified and the Add to Custom Colors button pressed will cause the next custom colour cell to change The cells are updated in a top to bottom left to right order However the colours assigned to the graphs are assigned on a left to right top to bottom order Some colour selections are for a single colour In these cas es Clicking on any colour cell so that the focus box sur rounds the intended cell followed by selecting the OK button on the dialog window will activate the Apply but ton on the property sheet for the tile parameters The colour changes only take effect when the Tile Display window is applied either by the OK button or the Apply button Figure 18 Colour definition window Data Display and Browser Fonts Typefaces fonts are managed in a similar way to colours Buttons are provided for fonts associated with axes labels the title and header text The procedure for adjusting a font is to select one of the buttons labelled Fonts The title for example is located on the Display property page To ad just the font press the button next to the text entry field that offers the current value for the title a font dialog window appears Then select the font parameters from the dialog window and press the OK button Once again like the col our dialog window this action will act
73. ata sets is the internal data structures used to maintain the experimental information The file structure is a link list of file records that hold a hierarchical description of the data Fortunately there is a utility called dump_dataset available from Kratos which converts the binary format to an ASCII version and it is the ASCII version that CasaXPS converts to the ISO 14976 standard To create an ASCII version of a Kratos Vision data set fol low this procedure at the command line level on the Unix workstation move to the data directory and type Sdump_dataset _ filename_lall gt new_filename kal where represents space and the command line prompt The extension kal is recognized by CasaXPS and tells the conversion routine to parse the data using the Kratos Vision option The first step to creating anew VAMAS file is to enable the Convert toolbar button The Convert option is disabled un less the selected sub frame window representing a file con tains no VAMAS regions If no empty Experiment Frame is among the list of windows then the New toolbar button must be pressed to create an empty Frame 128 SSI M Probe Files ER S a leg eais ierat Files od hepa Eita Fes bal Canoe Figure 96 Convert to VAMAS file Dialog Window The file display filter is set to show only kal files Selecting a kal file via the Convert to VAMAS file dialog window causes an ISO 14976 file to be generated with the s
74. ating the degree of uncertainty Given a distribution for each of the parameters the best way to describe the nature of the uncertainties is to offer an error matrix as well as a tabulated form for the distributions The error matrix provides numerical values from which the de gree of correlation can be assessed while scatter plots taken from some subset of these distributions allows visual inspec tion for the same information Ideally a single number for each parameter would provide the simplest means of stating the uncertainties but as the old adage goes To every com plex problem there is a simple solution that is wrong and so it is for correlated parameters The unfortunate fact is that if the peaks weren t correlated then synthetic models would be unnecessary Monte Carlo Data Sets Repeating an experiment many times is not always practical and for most XPS spectra peak models are developed based upon a single acquisition sequence Estimates for uncertain ties in the peak parameters must be made using assumptions about the noise typical of XPS spectra The essence of Mon te Carlo methods for estimating uncertainties is to take a data set remove the noise from the data then repeatedly add noise to the synthesized data to generate a set of simulated experimental results and apply the set of operations to that Monte Carlo Methods Uncertainties in Intensity Calculations artificial data The results represent a distribu
75. aukey CaraPl Ba da im ee ee eee ee 68 g0 ji im we oo iii Dinding ihany V3 Dining Bany is Figure 51 Calculator page setup up to subtract the standard data away from the Cr modified spectrum The standard is the spectrum labeled B and must be the spectrum in the active display tile The modified data A must be selected via the right hand side The Calculator 78 Using the Calculator and Comparing Spectra Limenzed io Naal Fanley Casa PT ma mi so Ha ma Hi i ii ii SA ma So sa dia di si dii Deming Mery aV Dmimg Burey Hi Figure 52 Cr 2p region after the equivalent data from the standard has been subtracted The Calculator 79 Principal Component Analysis Introduction XPS is a technique that provides chemical information about a sample that sets it apart from other analytical tools How ever the key information sought by the analyst is locked into a data envelope and as a consequence the need for powerful algorithms is paramount when reducing the data to chemi cally meaningful quantities Two approaches have been em ployed on XPS data e Curve synthesis and fitting see Quantification on page 42 e Techniques from multivariate statistical analysis of which Principal Component Analysis PCA is the most common form Curve synthesis is probably the most widely used method for data analysis practised by XPS researchers Unfortunate ly statistically good curve fits are not always physically m
76. be charge corrected but only those channels for which the charge correction lies within the original range of channels will the results be meaningful Figure 45 illustrates how the data is adjusted to account for charge shift corrections prior to performing one of the calculator operations The Calculator is designed to allow a single spectrum to act upon a set of spectra Any spectra selected in the right hand browser frame will be altered by the operation defined by both the Calculator property page and the current data in the active display tile Using the Calculator and Comparing Spectra Valid region of data nid Charge Calculator Shift operation AopB Figure 45 Calculator Operations and Spectrum Acquisition Channels after charge correction Figure 46 shows a set of spectra in various states of oxida tion The spectrum in the left side display tile Figure 46 a represents a metallic surface and may be subtracted from the set of spectra selected in the right hand side of the same ex periment frame to produce a set of difference spectra show ing the trend in the Al 2p oxide peak Figure 46 b illustrates the result of the subtraction operation shown on the Calcula tor Property Page where each of the spectra selected in the right hand side browser frame is displayed showing the re sidual Al 2p oxide peak following the Calculator operation Note the processing history indicates that the active tile on the left h
77. bility of well characterized RSF values coupled with a specific background and line shapes Some of the line shapes introduced in CasaXPS have been constructed to allow Doniach Sunjic asymmetric behaviour to be associated with an underlying Gaussian Lorentzian shape The Voigt approximation is used to characterize the area position and FWHM while the asymmetric form ap proximates the rise in the signal much in the same way that the Shirley background is used to reduce the data to symmet ric shape The advantage of retaining a separate asymmetry parameter in the synthetic model is apparent when a Tou gaard background is used to remove the extrinsic contribu tion to a metal spectrum To facilitate both trend analysis and basic quantification a wide range of line shapes is required 8 Castle J E et al J Electr Spectr Related Phenom 106 65 2000 Line Shapes and Backgrounds A List of Line Shapes The line shapes offered in CasaXPS are based around the following fundamental functional forms The Voigt functional form has been the basis for most quantitative analysis of XPS spectra and is simply a convo lution of gaussian and lorentzian primitives Unfortunately this convolution proves to be mathematically intractible and an exactly accurate analytical form for a gaussian lorentzian convolution is not available so prac tical systems have adopted two approximations to the true Voigt function as described below
78. butions may not be symmetrical about the initial value and asymmetric confidence intervals are possible Note that the error matrix is calculated from the distributions centered on the mean parameter value for the distribution not the initial values One of the real advantages of using Monte Carlo error anal ysis is that it highlights when a quantification parameter is poorly determined by the combination of model and optimi zation procedure It also allows the influence of constraints within a model to be evaluated Adding information about chemical shifts relative peaks widths and or peak areas can 101 Monte Carlo Methods Uncertainties in Intensity Calculations Quantification nes 1 Figure 66 Scatter plot between Area and FWHM parameters for Peak C 1s 1 in Figure 65 alter the manner in which noise adjusts the parameters from their initial value Monte Carlo derived scatter plots can of ten help to understand how rigid models based upon chemi cal knowledge can reduce the range of outcomes for a given set of parameters Quantification Quantification of AES XPS spectra is routinely performed using combinations of intensities from integration regions and synthetic components The purpose of this section is to point out some of the less obvious consequences that result from the specification of the background associated with the recorded peaks Monte Carlo End Point Determination Consider an all too common situatio
79. ch while derived from char acteristic transitions within the electonic structure of atoms of a particular element can also reflect the exact chemical environment of those atoms that is provide chemical com pound specificity Three property pages within the module offer two views of the data held in the element library file and a means of loading and merging data into the current session The module is launched from the main toolbar as shown be low SHE Hoo 7S Figure 22 Library button in Main Toolbar In order to function correctly CasaXPS should have an ele ment library present in the system although some basic fea tures will function without one but for many commands and operations involving qualitative and quantuitative anal CHAPTER 6 ysis a library file must be present The file may be changed at any time using the controls described below so that sev eral library files may be held on disk but only one library of unlimited size may be active at any time CasaXPS is normally shipped with a relatively simple ge neric library Instrument or analysis specific libraries must be generated or imported as required by the user the module recognises both CasaXPS and Kratos proprietary formats Library File Structure The element library file is an ASCII plain text file As such it is open to simple modification or change by the user so long as changes are made within the casaXPS format out lined below
80. ch statements are themselves suspect and so re quire support from other data reduction techniques For example curve fitting using three sets of synthetic peaks all linked with the appropriate stoichiometric relationships would lend support to the hypothesis Curve fitting such structures is not an exact science and such fits themselves should be supported by statistics gathered from the fitting parameters 87 Principal Component Analysis Abstract Factors for cls_test1 vms Abstract Factors for cls_test2 vms nee ea 300 298 296 294 292 290 288 286 284 282 280 300 298 296 294 292 290 288 286 284 282 280 Binding Energy eV Binding Energy eV Figure 55 Abstract Factors and Eigenvalues Figure 56 Eigenvalues for cls_test2 vms Abstract Factors The second example illustrates the effects of experimental error on a PCA calculation Target Factor Analysis 88 Target Factor Analysis Principal Component Analysis Table 7 PCA Applied to Artificial Data with Simulated Noise Factor Eigenvalue RMS RE RSD IE IND 1000 Chi sq Calc Baas d C Is 1 10883620000 267 626 617 9728 195 4202 7629 294 50489 63 1800 C 1s 2 687172800 18 93302 47 15373 21 35612 746 152 3572 687 1592 C 1s 3 2714766 6 90718 26 01386 14 24838 530 8951 483 7729 1386 C 1s 4 946794 7 0 1649431 2 106561 1 332306 58 51557 0 5688374 1182 C 1s 5 5095 574 0 0
81. ction a 7 re ojassa es e ee Page Layout toolbar button Figure 11 shows the settings used to achieve the display layout for the spectra Figure 13 Tiling a printed page is achieved using the Page Layout dialog window Now that we ve completed the tour we ll consider all the CasaXPS modules in outline in the summaries below and in much greater detail in the chapters that follow Data Display and Browser Windows Modern surface analysis instruments produce data at a high rate that makes efficient data reduction an important consid eration when assessing the sample throughput for a labora tory The time spent processing the data can become a Introduction Element Library serious problem when an automated acquisition sequence has produced hundreds of spectra and spectrum regions all of which fall into sets with related features The data reduc tion often consists of the same processing quantification op erations tediously repeated for each of the spectra within a data set The results of these calculations often have to be graphed or tabulated to show the variation of concentrations as a function of some experimental variable e g etch time or sample tilt CasaXPS has been designed based on the ISO 14976 stand ard format The structure imposed by this data format per mits acquisitions to be saved with the ancillary information necessary for understanding the context of the data and s
82. ctra or spectral regions The Calculator is a means of managing and extracting information used to identify spectral features and then to take that information on to quantify regions and components Calculator operations may be performed in a one to one or one to many fashion Creating a Comparison File CasaXPS enables calculation operations on spectra provid ed that the data blocks are located in a single ISO 14976 file It is clearly not always convenient for data acquired at different times to be stored in the same file but spectra from one or more files may be easily copied into a single file in order to make such comparisonsor calculations The procedure for moving the spectra is as follows 1 Create an empty experiment frame window use the New button or shortcut CHAPTER 9 2 Load the separate files for which a comparison or calcu lator operation is required each in its own experiment frame 3 Select the ISO blocks of interest via the loaded experi ment frame browser windows To add to the selection the Control key must be pressed when new blocks within a file or from different files are selected 4 Move the cursor over the right hand side of the empty experiment frame window and click the right hand mouse button A dialog window will appear and offers the list of ISO blocks currently selected Provided these data are compatible in the ISO sense the selected spec tra will be copied into the
83. ctually present in the data Hence the dimensionality of the data should be three not ten 4 4 2 An additional twist to this example is that two of the under lying envelopes are similar in shape to each other though not identical see Figure 53 a and b The trend throughout the first data set may be seen in Figure 54 PMMA PVA PVC Artificial C 1s Spectra te 300 298 296 294 292 290 288 286 284 282 280 Binding Energy eV Figure 54 cls_testl vms C 1s Spectra Principal Component Analysis No noise is present in the data therefore eigenvalues be longing to the primary set of three abstract factors should be non zero while the remaining seven eigenvalues should be zero The results of applying PCA to these data sets Table 6 il lustrate the uncertainty associated in estimating the dimen sionality of the data matrix from the statistics The fourth largest eigenvalue in each case is small but non zero Also the statistics for IE and IND indicate a minimum at eigenval ues other than the expected result Chi square is not a valid Statistic since no noise is present in the data however it does show that three abstract factors are sufficient to reproduce the data to within reason 86 Target Factor Analysis Table 6 PCA report for file clstest vms Principal Component Analysis Factor Eigenvalue RMS RE RSD IE IND 1000 Chi sq Calc o d C Is
84. d by a set of eigenvalues and eigenvectors The eigenvalues provide a measure for the significance of the abstract factors with re spect to the original data Various statistics can be computed from these values that aid in identifying the dimensionality of the subspace spanned by the spectra The procedure for calculating the abstract factors has its roots in linear least square theory In fact the preferred meth od is to form a Singular Value Decomposition SVD for the data matrix D USV Where D is the data matrix formed from c spectra each con taining r channels U is the same dimension as D while S and V are c by c matrices S is a diagonal matrix the diago nal elements are the square root of the eigenvalues of the correlation matrix Z DD The abstract factors are computed from US The rows of V are the corresponding eigenvectors of Z the co ordinates of the eigenvectors represent the loading for the abstract fac tors and specify how linear combinations of these factors can be used to reproduce the original data Including all of 81 the abstract factors with the appropriate loading enables the data to be reproduced to an accuracy only limited by the pre cision characteristic of the Eigenanalysis procedure The essential feature of the SVD procedure is to compute the abstract factors so that the factor corresponding to the largest eigenvalue accounts for a maximum of the variation in the data Subsequent abstract factors a
85. d in this binary format the base name of the mrs file must be specified but with an ssi extension replacing the mrs extension Figure 97 129 Convert to YAMAS file 24 x Lookin C MPobe JAA Alfoil Alfoil2 Demopet1 E3 Alfoil mrs Alfoil2 mrs Demopetl mrs File name Files of type Jal Files 7 Cancel Figure 97 The binary data held in the file directory pair named Al foil will be converted to ISO format PHI MultiPak ASCII files PHI MultiPak writes an ASCII format Data in this format can be converted to ISO format through CasaXPS The first step to creating a new VAMAS file is to enable the Convert toolbar button The Convert option is disabled un less the selected sub frame window representing a file con tains no VAMAS regions If no empty Experiment Frame is PHI MultiPak ASCII files Using Different File Formats amongst the list of windows then the New toolbar button must be pressed to create an empty Frame The ASCII files may contain single regions so called mul tiplexed regions or depth profile data To convert a set of files in any of these formats into a single VAMAS file the asc files must be collected into a sub directory that contains only valid asc data files and then a new filename must be entered with an extension of qua Figure 98 A VAMAS file containing all the regions from the files in the directory with an asc extension will be created Lookin E Surve
86. d model offered by the three parameter cross sec tion provides more guidance in the regions most influential in peak fitting but will struggle to reproduce the asymptotic behavior for the cross section as well as the adjustments needed to describe the near peak region of the background Nevertheless a practical approach to the use of Tougaard backgrounds in peak fitting is to choose a background that works in conjunction with the synthetic peaks The custom form for the cross section allows the background to develop at the same time as the peak parameters and choice of line shapes Figures 36 through 38 all use a custom three param eter universal cross section to describe the background un der an Al 2p metal doublet typically researchers would choose Shirley background for such spectra 12 Jo M Surface Science 320 191 1994 67 Background Subtraction Adjusting the universal cross section The rational function for the three parameter cross section is defined in terms of three constants B C and D This approx imation is the quotient of a linear polynomial and a quadratic in terms of T2 but it is the denominator that determines the position and strength of the resonance in the cross section The parameters C and D both determine the position width and height of the resonance but to a first order small D the position is influenced by the square root of C and the sharp ness of the cross section is determined from the size
87. d when A Quick Tour of CasaXPS changed are made to the set of integration regions so there is no problem about out of date quantification results being re 14 ported over a spectrum If the Annotation History property page is the active page on the Annotation dialog window Figure 10 then each annotation item will be displayed with a small position box attached The annotation may be repositioned using these boxes This is achieved by pointing the cursor at a box and then with the left mouse button held down the cursor is dragged to a new position The annota tion will move once the mouse button is released Page Tile Format x c E E FS is GPa ros SS POS Te m Tiles Arranged in Tiles per Row Column i C Columns oO ow A e o og Ca g o o o oo s wog e 97o g r Number of Rows Columns 4 Cancel trov Figure 12 Page Layout dialog window The spectrum shown in Figure 11 is now ready to be printed Any spectra visible in the left hand side of the experiment frame in the Display window will be printed on a single page If the Display window contains a scrolled list of spec tra then each tile within the scrolled list will be printed as a separate page If more than one spectrum is required on the same printed page then the Page Layout dialog window of fers the means of arranging up to sixteen tiles on one page Figure 12 and Figure 13 Data Display and Browser Windows Introdu
88. data for such differences problems may arise due to changes in the system from aging or tweaks to lens functions by the op erator Both require recalibration of the instrument without which peak areas from a wide scan acquired using one pass energy cannot be used with intensities measured via other modes If a spectrometer is characterized by a set of relative sensi tivity factors RSF s applicable only to the survey mode of an instrument but the resolution of the survey mode is insuf ficient to provide the desired chemical state information then one way to provided a more detailed quantification is to reference the intensities from the to high resolution spectra to this survey mode Peak areas determined from the high resolution data provide the relative proportions for the chemical state intensities and therefore the corresponding concentration from the survey data can be subdivided using these proportions thus allowing detailed quantifications without the need for transmission correction Such a proce dure effectively takes out the transmission function from the quantification step and lessens the need for the time consum ing calibration of each operating mode Indeed for many in struments transmission function correction is not available and so this procedure represents the only way to combine data from different operating modes which would otherwise require the maintaining of multiple sets of RSF values Tagging region
89. ddition to the name field in each integration region or syn thetic component there is a Tag entry in the form of a string which is used to link intensities from synthetic peaks to a specific integration region If the information is available CasaXPS will enter into the Tag field the element and transition associated with the ac quisition and only those items with tags are included in the quantification table To remove an item from the quantifica tion step delete the string from the Tag field and press return The system will enter the key word NoTag when return is pressed to indicate that the item is not to be used in the quan tification Note that the NoTag entry is necessary to exclude from the elemental concentration table the integration region defined for the high resolution scan seen in Figure 80 The background to the peak fit for the C 1s spectrum requires a region to be defined but the intensity from this region must not be included in the results from the survey spectrum In this example the tag field in the region used to define the background to the peak fit must be set to NoTag while each of the synthetic components are tagged with the same name as the corresponding region in the survey quantification namely C Is Not all forms of quantification tables use tags For example the Component annotation option used to display the com ponent table in the upper spectrum does not discriminate be Quantification using
90. de is that most manufacturers systems perform the quantifica tion by this means The results from CasaXPS in the default mode for transmission correction therefore will agree with these other systems although it should be stated that correct ing spectra point by point prior to calculating the intensity is probably the better method An example of the way in which transmission can be encod ed within the ISO 14976 file format and thus used by CasaXPS is given in Appendix 1 see page 143 24 Data Display and Browser CasaXPS uses a multiple document interface MDI as de fined by Microsoft That is it conforms to the same style as programs such as Microsoft Word or Excel and therefore presents each file Document with a document view The main frame of the program manages a set of document file frames These file frames may be minimised or maximised or arranged using the Window menu allowing the frames to be tiled with respect to one another or to create additional views of the same file Understanding the Data The ISO 14976 files used by CasaXPS are viewed through a splitter window called the Experiment Frame The right hand section of the Frame the Block window represents the logical structure of the experiment showing the relationship of the spectra to each other and to the experimental variable The splitter terminolgy is a reminder that the relative sizes of the two components of the frame may be adjusted by
91. e Peak Parameters 47 Propagating Quantification 49 Data Editor t 0ciaa hence Geni eeeteee aed 49 Derivatives and Peak Envelopes 50 Quantification by Example 51 Report Files and Excel 54 Line Shapes and Backgrounds 55 A List of Line Shapes 0 57 Gaussian Lorentzian Product Form 57 Gaussian Lorentzian Sum Form 57 Alternative Asymmetric Line Shapes 58 Line Shapes Based upon Backgrounds 59 Line Shapes Available in CasaXPS 59 Further adjustments to the basic shapes 60 Asymmetric Line Shapes 61 Modifications to the Doniach Sunjic function 62 A New Line Shape 0235 22208 te Sean x os 64 Background Subtraction 65 Adjusting the universal cross section 68 Simulating Spectra using Tougaard methods 69 Using the Calculator and Comparing Spectra 71 Creating a Comparison File 71 Th Calculator cerre ei eain baean eae dhs 12 Principal Component Analysis 80 IniroduchOn ssie a ena i ee ia 80 Theory of Principal Component Analysis 80 Residual Standard Deviation Real Error 82 Chi sq are eosi opat ea EER oe eee 83 Target Factor Analysis 00 83 Target Testing cs issc sacre eie a ia paneas 84 Principal Component Analysis by Example 84 Principal Component Analysis and Real
92. e Simplex method fails to converge when the number of parameters exceeds a problem dependent value even for smooth functions The conclusion is don t try to fit too many peaks using the Sim Quantification plex method unless the parameters are already close to the optimal values Propagating Quantification Once a representative spectrum has been modelled using re gions and components the next question is how to transfer this model to similar spectra within the same ISO 14976 file and other files CasaXPS provides a propagation mechanism to accomplish this task The spectrum for which the component model has been con structed must be displayed in the active tile view Any ISO blocks that require the equivalent model must be selected via the Browser View for each of the files involved Once these conditions are met the mouse cursor is placed over the ac tive tile and the right mouse button pressed A dialog win dow appears for propagating the quantification objects Check boxes are offered on the Propagate dialog window These allow a choice of what actions are propagated through the selected ISO 14976 blocks To transfer the quantifica tion model the Regions and the Components check box should the ticked Then press the OK button A progress dialog window appears that allows the propagation process to be terminated see Processing on page 34 Data Editor Curve fitting assumes a statistical model for the noi
93. e of an experiment The challenge is to devise methods for identifying changes in the spectra and therefore provide a way of following these changes as a function of the experimental variable The usual tools available for reducing XPS AES spectra are simple quantification regions and or synthetic spectra deter mined by optimisation procedures with various forms of pa rameter constraints Implicit in both of these data reduction methods is the introduction of a background approximation and in the latter case rigid models for the synthetic line shapes The consequence of the assumptions involved in these data reduction techniques is hard to assess with respect to the results for individual spectra Techniques for estimat ing the errors due to noise in the data are available for both peak fitted parameters and quantification region values but there is still the question regarding what influence these as sumptions have on the trend itself An alternative means of following a trend through a data set is to perform a Principal Component Analysis PCA This technique is a linear least square procedure that transforms a 114 set of spectra when viewed as vectors into a set of orthog onal basis vectors abstract factors that span the same sub space as the original data set Trends within the original spectra can be assess by examining the so called abstract factors and plotting the co ordinates of the spectra with re spect to the principal
94. e preferred and hence at least for metals Doniach Sunjic profiles ought to offer better fits A practical solution offered by Ulrik Gelius QSA nil uses a Voigt like function as the underlying shape and modifies the lower kinetic energy electrons using the equations given above CasaXPS offers these line shapes in the form A a b n GL p Gaussian Lorentzian product formula modified by an asymmetric form define above The param eters a and b allow the asymptotic form of the asymmetric tail to change whilst also altering the shape of the asymme try prior to attaining the asymptotic shape Figure 34 shows one such class of profiles where the parameter a has been held fixed at 0 35 while b varies between 0 2 and 1 The third parameter n defines the width of a Gaussian used to convo 9 G Wertheim J Electron Spectrosc 6 239 1975 Line Shapes Available in CasaXPS Line Shapes and Backgrounds lute the basic shape of the profile A 0 35 b 0 GL 0 b 1 0 b 0 8 b 0 6 b 0 4 b 0 2 p R T T 1 948 950 952 954 956 958 960 Figure 34 Asymmetric form due to Ulrik Gelius The line shapes defined by A a b n GL p and GL p K b0 b1 are different from the Doniach Sunjic pro file in that the Doniach Sunjic profiles asymptotic limit is zero while both of the ad hoc forms allow steps to be mod elled within the line shape itself Another characteristic of these ad hoc line shapes is that the pe
95. eaningful and in many cases great care must be exercised when choosing the model to describe the data Any assist CHAPTER 10 ance in understanding the model is therefore of great value and it is with this end that Principal Component Analysis is offered as a supplementary tool Factor analysis is a field that is as broad as it is deep It is a mathematically challenging tool that requires knowledge of matrix algebra coupled with a feel for a statistical approach to data interpretation A true understanding for the subject can only be obtained by studying the literature and through practical experience Therefore the material presented here is only an introduction rather than a complete set of works Theory of Principal Component Analysis Factor analysis is a multivariate technique for reducing ma trices of data to their lowest dimensionality by use of orthog onal factor space The challenge is to identify the number of significant factors principal components and use this infor mation to model the data using techniques such as Target Transformations or curve fitting In XPS the data matrix is composed of spectra where each 80 acquisition channel is viewed as a co ordinate in an r dimen sional space r is equal to the number of acquisition channels per spectrum The problem addressed by PCA is that of de termining the number of distinct spectroscopic features present in a particular set of c spectra The following example
96. ear in the scrolled view Although the modern PC hardware is 32 bit some 16 bit code still remains in MS Win dows software in particular the class used by the routine that sets the scroll bar attributes does not like the logical scrolled Data Display and Browser area to exceed the range of a signed 16 bit number 32767 Sor ry If more tiles are required to view a set of spectra then the page layout mechanism can be used to reduce the overall size of the scrolled list Tile Display Many of the attributes for displaying the spectra can be ad justed The menu item labelled Tile Display on the Op tions menu displays a dialog window for modifying the appearance of the spectra Ranges for the axes labels and fonts used for the labels can be adjusted on the first two property pages found on the Tile Display dialog window The user may also choose between displaying the spectra using binding energy or ki netic energy as well as counts per second or recorded counts An option on the Y Axis property page allows overlaid spectra to be offset with respect to one another A check box enables the offset mechanism and a numeric field permits a percentage to be entered which changes the separation of the traces Other check boxes offered on the Y Axis property page alter the appearance of the display The axis label and scale may be toggled on or off also a spectrum may be plot ted with the value of the ex
97. ection is displayed in a single tile The principal component analysis is performed when the Apply button is pressed Each spectrum displayed in the active tile is replaced by the computed abstract factors The order of the VAMAS blocks containing the spectra is used as the order for the abstract factors The factor correspond ing to the largest eigenvalue is entered first Subsequent blocks receive the abstract factors in descending order de fined by the size of the corresponding eigenvalues A report showing the statistics for understanding the dimensionality of the factor space appears in a dialog window Principal Component Analysis A button labelled PCA Report allows the current PCA re port to be re displayed Care should be exercised since the values are subject to any additional processing including PCA that may subsequently be applied to any of the spectra included in the original analysis The PCA property page includes a button to reset the processing operations for every spectrum displayed in the active tile This allows a PCA calculation to be undone in one stroke It will also undo any processing previously per formed on the data PCA is aimed at the raw data the chi square Statistic is referenced to the raw data and has an un defined meaning when the data have been processed prior to performing factor analysis Target Factor Analysis in the form of target testing is also available on the PCA property page Follo
98. ectrum then the merit of attempting to model the result with theoretically correct synthetic line shape is less clear 65 Background Subtraction Tougaard and co workers L10 have devoted much time to understanding the backgrounds that are present in XPS spec tra The transport of electrons through a material after exci tation by X rays can be described by equations U1 that involve a one sided convolution of the recorded data with an energy loss probability distribution This loss function of TAF Linea ie LPs Parre ve Clik ye bate a Hale Ciak ah a i Alime A at Taga Cempoemris Dain dic Fem Seer omaa sebelia arija Ce a E H J F a er nek Figure 40 Three parameter universal cross section for Silicon fers a prescription by which electrons leaving the sample with an initial energy may have their characteristic energy altered by the interactions with the surface Each material 10 Tougaard S Surf Interface Anal 25 137 1997 11 Tougaard S Surf Interface Anal 11 453 1988 Line Shapes and Backgrounds has a characteristic loss function of varying complexity and Tougaard has written many papers describing methods for establishing both generally applicable approximations so called universal cross sections and specific forms for indi vidual materials The principal methods used by Tougaard for calculating the loss functions are theoretical dielectric respon
99. ed Input see Library on page 141 allows the session element library to be modified by reading different element library files A browser button offers a file selection dialog and the chosen file may be load ed or merged with the current session element library If the Load button is selected the element library in the current session is replaced by the contents of the specified file Al ternatively the Merge button causes the contents of the file to be added to the existing information for the session A user can maintain a set of element library files and load merge only those appropriate for a type of sample or operat ing mode of the instrument In the absence of transmission function data this mechanism allows the user to maintain Loading an Element Library Element Library different R S F values for different acquisition conditions F ie pi TES WH Da aby ar op lad to cps a an 12 Lodo tw so w an Bhibg Deg V Figure 23 Element Markers The Element Table window provides a scrolled list of the contents of the Library File clicking on a name toggle transfers a labelled position marker to the display for the se lected item Position markers may be removed from the dis play with the Clear All Elements button Find Peaks adds coloured markers to the display for all major peaks in the displayed block according to a defined algorithm the Clear Markers button removes t
100. ed from the relative size of the oxygen peaks and is used as a factor that is applied to the standard spectrum before subtracting the two Cr 2p data intervals First the data regions for the unknown and the standard must 73 Using the Calculator and Comparing Spectra The Calculator Figure 47 Cr 2p regions acquired from a catalyst before and after an uptake of Cr be copied into a new experiment frame In Figure 48 six re gions from each experiment have been copied into an empty file initially called CasaXP1 This file may be subsequently saved with an appropriately changed name The steps used to calibrate the energy scale are as follows refer to Figure 49 and Figure 50 1 Overlay the two O 1s peaks 2 Select the Calibrate option on the processing property page window Use the mouse to drag out a box from the top of one peak to the top of the second The limits of the box define the Measured and the True energies entered on the Calibrate property page pressing Apply will calibrate the two spectra with respect to one another 3 Once satisfied with the O 1s calibration the values for the Measured and True energies may be applied to the corresponding Cr 2p region Note that only one of the two overlaid O 1s spectra will have an entry in their processing history following the calibration step The Cr 2p spectrum from the same file as the O 1s with the calibration entry is the Cr 2p spectrum for which these cal
101. ed using the Savitzky Golay polynomial in which a quadratic is used to approxi mate 5 data channels From a user s point of view integra tion is performed simply by pressing the Apply button on the Integration property page with the single option of be ing able to reset the zero point of the integraal if required Energy Calibration A spectral line may appear at an energy position that is not the expected value for a transition The reason for this may be due to spectrometer calibration or sample charging how ever for presentation purposes the energy scale can be ad justed using parameters on the Calibration property sheet Two values can be entered the recorded position of the line and the required energy for the peak On pressing the Ap ply button a shift is computed and the display is updated Note that if preferred the spectrum can be shifted by an amount by setting the measure value to zero and entering the required shift in energy into the field for the true energy 35 Intensity Calibration Intensity Calibration A spectrum recorded on a particular instrument represents only the electrons as a function of kinetic energy that the in strument was capable of sampling The transmission charac teristics for an instrument are important if the spectra are used to quantify the chemical composition of a sample This extends even as far as calculating the background for the da ta where techniques presented by
102. eighted aver age of the background limits chosen to tie in with the spec trum at the end points of the region The weighting is determined from the area between the background and the data Since the weighting is determined using the quantity being computed a sequence of iterations are required to ar rive at the desired result Tougaard has extensively studied the subject of back grounds to XPS spectra however the background generated by CasaXPS is simply calculated using the Universal loss function For this algorithm to work it is necessary to re move all instrumental contributions from the spectral shape before calculating the background This is seldom possible for practical situations To allow the procedure to provide a background under less than ideal conditions an adaptive procedure has been adopted that attempts to fit the back ground to the given spectrum The result is a background that looks plausible in situations when the data has not re 43 Regions ceived the necessary pre processing and equal to the back ground proposed by Tougaard for practical applications when the appropriate adjustments have been made to the da ta To delete a region first select the region in the scrolled list of regions Then press the Delete button Each region defined for a spectrum has a number of statistics gathered from the data and displayed in the scrolled list on the property page in question The raw intensity CPS eV bet
103. en be used to edit the drawing in remarkable detail The button to save these graphics files is indicated below red circle meri pasan Se Figure 19 EMF amp Annotation buttons in Main Toolbar Although the tile display options permit text such as the ti tle to be adjusted and enhanced metafiles allow sophisticat ed visual editing XPS spectra can sometimes best be annotated with information derived directly from the data The most obvious being tables related to the quantification of the chemical state CasaXPS therefore provides some op tions that allow the inclusion of such information as part of the annotation CHAPTER 4 The options for adding annotation are found on the main menu Options or via the main toolbar Figure 19 magen ta square Quantification Tables After quantification regions have been defined for a spec trum a tabulated form of the information gathered from data may be added to the graphical display The procedure to in clude region information is similar to that used for synthetic components Both types of tables are added using the Re gion or Component property pages found on the Anno tation dialog window The font used for the table may be selected before pressing the Apply button on the page of interest A table of the current values derived from the source in question will appear in the display area and it is then repositioned via the Annotation History property
104. error distributions which is cer tainly not true then one standard deviation in each of the pa rameters is given by the square root of the diagonal elements in the matrix Figure 64 Simulated C 1s data where noise has been introduced to the system Table 10 Error Matrix for the C 1s spectrum in Figure 63 1 Area 1 Pos n 1 FWHM 2 Area 2 Pos n 2 FWHM 1 Area 39 230 0 122 0 196 31 512 0 128 0 186 1 Pos n 0 122 0 001 0 001 0 120 0 000 0 001 1 FWHM 0 196 0 001 0 002 0 185 0 001 0 001 2 Area 31 512 0 120 0 185 38 330 0 125 0 202 2 Pos n 0 128 0 000 0 001 0 125 0 001 0 001 2 FWHM 0 186 0 001 0 001 0 202 0 001 0 002 Monte Carlo A Simple Example 100 It may be seen that the areas of the two peaks are anti corre lated where the quantity 31 512 y 39 230 x 38 330 char acterizes the degree of interaction between the two parameters The minus sign indicates the parameters are anti correlated A scatter plot constructed from the two dis tributions for the peak areas is shown in Figure 65 where the anti correlation is obvious from the bias in the scatter along a direction 135 to the positive x axis An alternative scatter plot Figure 66 for the area and FWHM parameters taken from peak C 1s 1 Figure 63 shows that these two pa rameters exhibit positive correlation The uncertainty for correlated parameters can be estimated from plots such as those shown in
105. eter 0 09 and no nu merical convolution applied wide Gaussian characterized by the number 380 and ac counts for the Lorentzian bias in the Voigt approximation Figure 39 shows how a pure Voigt type function SGL 70 is merged with the asymptotic form for a Doniach Sunjic Line Shapes and Backgrounds profile asymmetry parameter equal to 0 09 for a range of the linear mapping parameter The characteristic shape of the Doniach Sunjic profile can be retained to some degree but is supplemented by a class of shapes that go beyond what is possible with a DS profile The principal difference be tween the DS profile and the F function lies in the static position of the maximum for the class of profiles The mov ing maximum is a feature of the DS profile and can be achieved to some extent by use of the numerical convolution option Increasing the width of the Gaussian caused the line shape to lean in towards the asymmetric side of the peak Figure 38 A further variation on a theme is also possible where the F function can be replaced by 1 a 2 E x a m F E m Voigt x 1 m Voigt x The E function is merely allows the shape to the right to be a part of the asymmetric shape to the left Background Subtraction The line shapes described above are very dependent on the availability of background subtraction algorithms that com plement their use If a background is incorrectly removed from a sp
106. eters tabulated on the spectrum in Figure 63 The next step in the simulation is to introduce noise onto the data envelope that is consistent with noise found in experi mental XPS spectra i e variations about the channel inten sity of magnitude related to the square root of the counts Figure 64 shows the data envelope from Figure 63 after noise has been added and the peak parameters refitted This procedure yields the first set of simulation results If re peated many times the output is six distributions one for each of the peak parameters involved and these can be used to assess the influence of experimental noise on these quan tities Note that this differs in some respects from adopting a purely experimentally determined parameter distribution The initial stating point for the peak parameters will not be identical for an experimental data set since the experimental data may be subject to sample charging and any errors in the 99 Monte Carlo Methods Uncertainties in Intensity Calculations measurement procedure that can not simply be described by the random nature of the counting system will be omitted A Monte Carlo simulated data set only tests the stability of a model with respect random noise and therefore may neglect other factors present in an experimental data set Table 10 is the error matrix that results from a Monte Carlo simulation for the spectrum and synthetic model in Figure 63 If viewed as uncorrelated
107. fica tion Parameter dialog Components Property Page appears as a column of parameters in the scrolled list shown in Figure 76 These columns are headed A B C and so on To con strain the area of the component in column B to be half of the area of the component in column A the area constraint in column B should be set to A 0 5 Similarly to offset a component in column C by 0 2 from the component in column B enter B 0 2 in the position constraint field in column C Pre FEL Lh m E H LI TO a de Wiin LAT GUTS Sd A 22H L1G LT Shd Sa Tirsting Bree AV Figure 77 C 1s envelope from clean PAA acquired on a Scienta 300 RUSTI Daresbury Laboratory UK 109 The peak shapes from a Scienta ESCA 300 may differ from a VG ESCALAB 220i the source of the real data in Figure 74 or any other manufacturer s instrument but the essential structure should be suitable as a basis for the new model Copying the Beamson and Briggs pure PAA model into the data in Figure 76 leaves a residual that requires an adjust ment for the two non PAA peaks together with the introduc tion of a third peak The new peak in Figure 78 is constrained to be the same width and position as the saturat ed PAA C Is peak located at 285 eV BE The area of this new peak is allowed to adjust at will and accounts for carbon with the same characteristics as the PAA peak at that posi tion The consequence of introducing the new peak is tha
108. files can be converted in the same way The first step to creating a new VAMAS file is to enable the Convert toolbar button The Convert option is disabled un less the selected sub frame window representing a file con tains no VAMAS regions If no empty Experiment Frame is VGX900 Ron Unwin 125 VGX900 Ron Unwin amongst the list of windows then the New toolbar button must be pressed to create an empty Frame Files generated by the VGX900 system are ACSII files and each file may contain one or more spectral regions The method for loading these files into CasaXPS is to collect a set of the files in a sub directory and then enter a new file name in the Convert to VAMAS file dialog window but add an extension of unw to the specified name Figure 93 All files in the directory will be read and appear in a single VAMAS file Since the files have no characteristic file type it is essential that only files generated from the VGX900 system are present in the sub directory and no others Some formats of the VGX900 system do not include infor mation necessary for a proper description of the data in the VAMAS sense but the system does allow the writing of VAMAS files based on a user specified template This al lows data to written in a form directly readable by CasaXPS but the user must ensure that the correct fields have been used to describe the spectra Alternatively the conversion option in CasaXPS can specify some of the mi
109. format of the display Hovering over a button placing the mouse screen pointer on an icon without clicking produces a descriptive label for that button a Tool Tip and a slightly longer description in the status bar Screen Preview for Display Help About Casa XPS Print Current Spectrum Dispayd Function Buttons F5 F10 mirror the Processing buttons Page Layout Element Library Options Bar 134 Command Summary Decrease intensity scale maximum Increase intensity scale maximum Reset Intensity scale to original maximum Increase Energy scale range t Step zoomed energy scale left 7 al Oban kad Ed a Display Scaling buttons Step zoomed energy scale right Reset to original scale Zoom out step back round history Zoom in requires selection box rescales intensity to suit tT adds 20 left and right Display Properties buttons fe 9 T ao EH Tedd t tri state buttton Toggle Offset for multiple traces t Toggle 2D 3D and Factor Space Display t Toggle Counts and CPS Intensity scales Toggle Binding and Kinetic Energy scales Options Bar Toggle normalised display Toggle subtracted display Display Modifier buttons C ed EA BG ete S L Toggle background Toggle residuals display Toggle components display Toggle shaded region display NB Modifiers require prior function definition Comment Information amp Variable control ae Edit Block
110. function 55 annotation history 32 moving 32 repositioning 15 annotation dialog 11 artificial peaks test data 47 asymmetric line shapes alternative 58 Auger lines 39 B Background Subtraction 65 Backgrounds 43 Beamson and Briggs 107 block window 7 Briggs and Seah 61 Browser 15 16 browser 7 c CasaXPS 5 installing 6 starting 6 terminology 6 7 Colours 28 138 D data Zooming 26 Data Display 15 Data Editor 49 Define Custom Colors 28 Derivatives 50 Differential charging 55 display 3 D plot 28 Geometry 27 window 7 Display Parameters 29 Doniach Sunjic 56 58 59 61 150 DS800 Binary Files 123 E Eclipse Files 124 Element Library 16 loading 40 Element Library dialog 9 Enhanced Metafiles 31 Excel 54 experiment file 8 Experiment Frame 9 Index experimental variable 22 F F line shape 64 F W H M 40 File Formats 123 File formats DS800 Binary Files 123 fonts 29 G Gaussian Lorentzian Product 57 Sum 57 Gelius Ulrich 58 61 Geometry 137 Graph Annotation 17 H H line shape 63 l IBM 5 Identifying Peaks 41 IE 86 IND 86 Installing CasaXPS 6 integration regions 13 intensity calibration 23 ISO 14976 5 Binding vs Kinetic Energy 22 blocks 16 20 Data Blocks 21 Experiment header 21 line separator 20 parameter exclusion list 20 Partially Encoded 19 Selecting 26 spectrometer geometry 20 experiment 20 Experimental Variable 22 File Format 19 File Structure 20 Quantifica
111. g Dialog windows Command Summary Keyboard amp Mouse Shortcuts board Function Shortcut New Experiment Frame Ctrl N Browser Open ISO File Ctrl O Print graphic Ctrl P Copy graphic to clip Ctr C Set Normalise point Shift left click Zoom out cycles Ctrl right click Context menu right click Edit a data point Ctrl Shift left click Display spectrum F1 Overlay spectra F2 Zoom out F3 Reset Zoom F4 F1 F4are single press Function keys for the other commands hold down the control Ctrl and or shift key and press the listed key or mouse button Drag and Drop you may load ISO 14976 files by dragging them from Windows Explorer and dropping onto an open CasaXPS Programme Frame 142 Appendix 1 ISO 14976 format files annotated 1 Simple single block XPS region spectrum Item Description Line ISO 14976 Format Item in PET Cls vms file format identifier 1 VAMAS Surface Chemical Analysis Standard Data Transfer Format 1988 May 4 institution identifier 2 Acolyte instrument model i d 3 Kratos AXIS HS operator i d 4 AC experiment i d 5 PET Cis Test Spectrum comment lines 6 1 comment lines 7 Acolyte copyright 1999 experiment mode 8 NORM scan mode 9 REGULAR regions 10 experiment variables 11 parameter exclusion entries 12 manual items 13 future e
112. ginal sate Apply selection resets the displayed block then reapplies only the selected items from the existing history Use con trol and shift keys with the mouse to select non contigu ous or continuous items respectively Propagate flag sets the propagate flag to apply the current selection from the history to a series of selected blocks 140 Annotation see Graph Annotation on page 31 Controls placement and style of text for annotation Text Peak Labels Regions and Component summaries and Quantification tables in the display tile and printed output An Annotation History panel functions in a similar fashion to the Processing History enabling removal of unwanted items or selective application Each tab has provision for changing of typeface font size and colour Position for an notation items may be referenced to the display tile frame or the data itself for convenient placement when multiple blocks are displayed and labels may be orientated vertically or horizontally Positioning targets small squares with cen tral dots may be set with the mouse anywhere within the display tile and disappear when the annotation window is dismissed Processing Dialog windows Command Summary Library see Element Library on page 38 ema Tdte Pancdic Takis luc Fae In order to function correctly CasaXPS should have an ele ment library table of defined line positions widths shape
113. gion can be set to zero The name for the region should be chosen to be different from any that will be used to identify the components Let us say the only chemical state information required is that related to the Al 2p transitions For all other transi tions only the total intensity is needed to construct the pro file So for O 1s and Mg 2p repeat the following steps described for C 1s 1 Select all the spectra for C 1s by clicking the column header 2 Enter the spectra into the scrolled list by pressing the button that displays one spectrum for each of the tiles Figure 30 3 Scroll to the spectrum that best illustrates the character of the data and click in the tile with the left hand button of the mouse This makes that tile the active tile 4 Bring up the dialog window for quantifying the data Figure 24 5 Add a region via the Region property page and adjust Quantification by Example Quantification the parameters until the background and the integration limits look right whatever that may mean 6 Ensure that only the set of carbon transitions is still selected in the Browser view of the ISO file Then right click with the mouse cursor over the active display tile The Propagate Actions dialog window will appear 7 Select the check box for Regions and press OK The result is that each spectrum in the set of carbon spectra now has a region defined to be the same as the spectrum displayed
114. hanges are required The dialog window enables the name and or the formula to be modified or deleted Each entry in the Custom Report corresponds to the exper imental variable in the active data file The columns are en tered two per named formula one for the raw intensity calculated from the expression and one for the percentage concentration Thus a table as follows Table 1 would re sult in a report with five columns One for the experimental variable and two pairs headed Oxygen CPS eV Oxygen and Aluminium CPS eV Aluminium 9 Table 1 Oxygen O Is region Aluminium Al 2p Al 2p Oxide The variable names derive from an Oxygen region O Is re gion and three synthetic components two Al 2p from a doublet of aluminium metal peaks both named Al 2p and one Al 2p Oxide The Al 2p name was repeated so that the intensities calculated for both peaks for the metal doublet were implicitly summed before the explicit sum defined via a formula computes the total contribution from the alumini um 46 Regions Optimisation of the Peak Parameters Fitting synthetic components to a data envelope is probably the most useful tool in XPS data reduction It is also one of the more difficult procedures to perform owing to the com plex nature of the underlying peak structure and the question of how to account for the background structures in the data The line shapes used to characterise a peak are only
115. he peak areas within the PAA model are constrained to one an other but the synthetic component at 286 54eV shows a de gree of correlation with the intensity of the pure PAA area 110 At first glance the behavior of the GIC 2 peak at 286 54eV is unexpected since one might think that two peaks next to one another should produce anti correlated area distribu tions The constraints have altered the concept of next to since the PAA sub model spreads across the entire envelope and it becomes difficult to judge by eye what the influence of noise might be on the final result This type of insight can only help to understand what constraints do to a fitting pro cedure as well as provide a rule of thumb estimate for error bars multi dimensional error distributions can seldom be described by a single number Scatter Plot for Normalised Peak Areas 1 30 p C1sGIC1 C1s Saturated C1sGIC2 1 20 rypjen 1 10 oe re a tasi 4h iaa sts a aar fa t 1 00 0 90 0 80 HR 0 70 0 94 0 96 0 98 1 00 1 02 1 04 1 06 Noramlised Area for PAA Staturated C 1s peak Figure 79 Monte Carlo simulation results for normalized peak areas The three peaks associated with GIC are plotted against the saturated C 1s peak area from the pure PAA model The exact meaning for the model in Figure 78 is left to the Quantification using Tagged Regions Analytical Applications experimentalist
116. he layout and interpretation of these parameters is straight forward Data Blocks The parameters which apply to a block the business end of the format fall into the following groups in the order in which they appear identity of the block in the experiment identity of the sample date and time optional comments technique analysis analyser ISO 14976 File Format signal recording parameters sputtering sample orientation additional parameters and future upgrade parameters followed by the corresponding data values Each block thus provides the acquisition parameters togeth er with a set of corresponding variables The correspond ing variables may contain more than just recorded counts Additional corresponding variables can include values such as transmission function correction factors for adjusting the recorded counts with respect to a reference spectrum Use of a second corresponding variable is the method pre ferred by at least one commercial manufacturer to record transmission function data in a VAMAS format The transmission characteristics for their instruments accompa ny the data point by point within each block Although blocks contain a comprehensive set of parameters it is permitted to replace many by a value that indicates no information is present 1E37 10 is the defined value Use of this feature can reduce the ability of a target system to process the data but it can he
117. hen when a cre ate button is pressed the values for the synthetic component are read from the transition so specified If the element li brary page is not active then the ISO species transition label is used to pick out the parameters for the component from the element library Line shapes that are available at present are either Gaussian Lorentzian product or sum functions augmented by an asymmetric tail shape for more detail see Line Shapes and Backgrounds on page 55 These are specified as follows GL n where 0 lt n lt 100 for a product function or SGL n for a sum function A value of n 0 provides a pure Gaus 44 Regions sian while n 100 results in a pure Lorentzian A tail is in troduced by appending T x where x gt 0 and is typically between 0 5 and 10 0 For example an asymmetric line shape for Al 2p might be GL 30 T 2 0 see Figure 25 for an example of such a component Clicking on the corresponding item in the scrolled list and entering new values can specify peak positions widths and areas Alternatively when the Components property page is on top of the dialog window the mouse can be used to pick up a component and move it to a new position area or width If the cursor is located near the top of a peak then the position and area are adjusted If the cursor is located near the side of the peak then the width will change Note that the height also changes when the width is adjusted using the
118. herefore much more than just a conven ient way of uniformly encoding output from surface analysis instrumentation it is also a blueprint for anyone designing a data acquisition system The strength of the ISO file format lies in its ability to store all the information needed to process the data at a later time This is also a weakness since its completeness makes the de tailed description formidable to the uninitiated Neverthe less in the fullness of time anyone designing a file format for XPS data will surely move towards a structure that es sentially holds the same information as ISO 14976 albeit perhaps in a different order or layout Partially Encoded Format Versions ISO 14976 is based on the earlier VAMAS format but with three major changes File Structure e the spectrometer geometry description is now referenced to a right handed rather than left handed co ordinate sys tem e the number of entries in the parameter inclusion or exclusion list has been set to zero in order to simplify the format e and the line separator sequence has been changed from the single 7 bit ASCII character carriage return to the more commonly used two character sequence of car riage return followed by line feed The ISO format thus may be a sub set of the VAMAS defi nition Although both the VAMAS and ISO formats are very gen eral CasaXPS is only concerned with files generated from X ray Photoelectron Spectr
119. hese Take care to note the dis tinction between element markers from the library and peak markers from the Find Peaks routine 40 Identifying Peaks Identifying Peaks Peak identification is performed using the scrolled list of transitions and the Periodic Table property page A spec trum may be viewed with markers added at energies where peaks might be expected Figure 23 note that the position line sizes heights are scaled to reflect the RSF values stored in the library To add and remove these markers se lect an element through either the periodic table interface or from the element table The markers are toggled on and off by repeatedly selecting the same element An alternative method for examining a spectrum is to have the scrolled element table visible and then run the cursor over the tile display window with the left mouse button held Element Library down The energy order list will scroll with the energy posi tion of the cursor Release the mouse button when over a peak of interest and then use the table to toggle the peak markers until a match is established A more detailed example of how to approach peak identifi cation is given in A Quick Tour of CasaXPS on page 7 Peak and element markers are maintained within a session on a spectrum by spectrum basis Altering the markers on one spectrum does not change those previously chosen for a different spectrum block even within the same experiment Si
120. hole followed by a number of blocks of data often called data blocks followed by an experiment terminator Each block consists of a set of parameters that only apply to that block followed by a series of ordinate val ues which may represent a curve e g a depth profile a spectrum or a map The experiment header contains the context for the acquisition sequences blocks that follow while the information held in the blocks is specific to a par ticular determination e g a spectrum region recorded in the course of an experiment Macintosh computer users may see some resemblance to the resource fork data fork file structures of MacOS here An experiment described in this way is suitable for both complex and simple data sets The header and block struc ture may seem perhaps over elaborate for a single spectrum experiment yet even the most complex profiling analyses are well supported by the same organisation in the data the only difference lies in the number of blocks within each file 20 Experiment header Experiment header The parameters which apply to the experiment as a whole occupy the first section of a file written in the ISO format They fall into the following groups in the order in which they appear identity of the experiment in its analytical environment optional comments experiment mode number of blocks and how they are arranged pointers to manually entered parameters future upgrade parameters T
121. ibration values are appropriate Figure 51 and Figure 52 show how to define the calculation for removing the underlying trend from the Cr 2p peaks The quantification report printed over the wide spectrum Figure 48 for the unknown sample shows the results com piled using a Shirley background applied to the difference spectrum for the Cr 2p region Note that the quantification report is compiled using the results from the high resolution spectra even though it is displayed over the survey spectrum Without subtracting the reference spectrum the validity of the Shirley procedure would be in question for the Cr 2p da ta 74 Using the Calculator and Comparing Spectra Figure 48 Quantification based upon the Cr 2p data after subtracting the corresponding spectrum from a standard sample The Calculator 75 Using the Calculator and Comparing Spectra 1s nt w S H 1a in Figure 49 Calibrate the spectra with respect to one another The calibration point is determined from the overlaid O 1s peaks The Calculator 76 The Calculator Using the Calculator and Comparing Spectra E j Sie TEA EEMI SAE im Ea f anen SHeG 6l E 10 ra ma oi ao mn 0 Figure 50 Regions after the calibration has been applied to both the O 1s and the Cr 2p regions using the Apply to Selection button on the Calibration Property Page Figure 46 77 Using the Calculator and Comparing Spectra Limenzed to Meal F
122. ide shape from the av erage description represented by the first abstract factor while the metal line shape is enhanced by virtue of the neg ative structure beneath that part of the spectrum The adjust ment to the average shape reverses as the oxidation reaction progresses The PCA has managed to describe the trend us ing essentially a single loading and a careful choice of the basis vectors An interpretation for this trend could be a mi gration of the metal form of aluminium to aluminium bond ed with oxygen and or the correlated attenuation of the metal signal due to the build up of an oxide over layer A strong conclusion from the PCA is that whatever processes are in volved these processes occur in a linearly related fashion This statement does not exclude a hydride moving to an ox ide state with the same speed as the metal making the same adjustment It does however show that metal and oxide line shapes are sufficient to describe the oxidation process measured by the XPS spectra The next step is to construct a synthetic model that produces the same trend as identified by the PCA Synthetic Model for an Oxidation Sequence Photoelectric peaks for metals require synthetic models ca pable of describing asymmetry in the observed line shapes Various approximations have been introduced to account for the asymmetry seen in the aluminium lines ranging from ad hoc functions to theoretically based shapes As always re gardless of
123. imization routine nor any other errors introduced into the measurement proc CHAPTER 11 ess and these true values will lie inside the region of the N dimensional parameter space defined by the set of outcomes to this sequence of experiments Obviously if the synthetic model does not describe a set of parameters in tune with na ture the results may be in more doubt than the measured dis tribution might suggest However given that all is well then the task is to offer a means of understanding the uncertain ties in the peak parameters within the context of these pa rameter distributions Peak identification in XPS spectra represents a challenge since synthetic models more often than not involve overlap ping line shapes Figure 63 the consequence of which is correlated optimization parameters That is to say if a single data envelope results from two overlapping peaks and if one of these underlying peaks is reduced in intensity then in or der to describe the same data envelope the other must in crease in intensity Adjustments to peak parameters of this nature are inherent to any optimization procedure and the choice between the possible combinations of these peak in tensities is made based upon a chi square or root mean 97 Monte Carlo Data Sets square metric The problem is therefore to identify the point at which these goodness of fit metrics fail to produce results that can be believed and provide some means of illustr
124. in the factor analysis In addition each spectrum must have the same number of acquisition channels as the others in the set of spectra to be analysed The first step in the calculation replaces the values in each spectrum by the result of interpolating the data within the defined quantifica 94 Target Factor Analysis tion region for the spectrum This is designed to allow ener gy shifts to be removed from the data used in the factor analysis The quantification region also provides the type of back ground to the spectrum Performing the analysis on back ground subtracted data attempts to remove artifacts in the spectrum that derive from other peaks within the vicinity of the energy region Background contributions can be signifi cant in PCA Additional primary abstract factors are often introduced as a consequence of changes in the background rather than the underlying peaks within the region of inter est The presence of such abstract factors can be viewed as information extracted from the data although in many cir cumstances they can lead to incorrect synthetic models if background contributions are misunderstood A factor analysis is performed on the set of spectra displayed in the active tile Although PCA is offered as a processing option it is the only processing option that acts on a collec tion of spectra Any other option from the processing win dow would only act upon the first VAMAS block in a selection when that sel
125. ing attributed to custom backgrounds should be viewed in the same way a synthetic model for a peak enve lope is viewed That is to say only chemistry and knowledge of the sample allows a synthetic model to be constructed which has any meaning and this should also be the case for custom Tougaard backgrounds The use of these back grounds is entirely at the discretion of a researcher and bad backgrounds can be constructed in much the same way that synthetic models are open to abuse However a carefully constructed background can reveal features that may be overlooked or not accessible when a linear or Shirley form is applied Figure 42 shows an O 1s structure from PAA po 68 Line Shapes and Backgrounds ly acrylic acid after a custom Tougaard background has been subtracted The structures to the higher binding energy side of the peak may be real or an artifact of the background algorithm but a linear background applied to the same data would fail to ask any questions at all regarding these fea tures Paes a ah J Een BY Figure 42 High resolution spectrum from the same PAA sam ple shown in Figure 41 The custom Tougaard background has been subtracted from the data Simulating Spectra using Tougaard methods Given a set of synthetic peaks and a three parameter univer sal cross section it is possible to construct a spectrum with Background Subtraction an associated background The synthetic PAA spectrum in mha
126. introduced by the background calculation that occurs prior to performing the peak fitting procedure A comparison of the distributions suggests that the intensity determined form the peak fitting procedure will be more accurate than the integration region approach although there is a tendency to over estimate the total area when the sum of two fitted peak intensities is used and the background calculation is sensitive to noise Again the use of more acquisition channels to define the integration region end points improves the precision in the results for in tensities calculated from the sum of the fitted peak areas Figure 72 Monte Carlo Simulation Total Intensity Determined from the Sum of Two Peaks Twenty One Points Used to Define Integration Regions 1 30 1 20 1 10 0 90 0 80 0 70 T T T 0 100 200 300 400 Figure 72 Linear background with twenty one end points In addition when the uncertainty in the background is re duced asymmetry in the distribution of area ratios is im 105 Summary proved dramatically compare Figure 71 and Figure 72 Introducing additional information into the background cal culation does not improve the uncertainty in the intensities for the individual peaks determined by a peak fitting proce dure but it does remove some of the correlations in the peak intensities due to the background These correlations are vis ible in Figure 70 at the extremes of
127. ir Pik Frio Iki xa a Figure 43 Synthetic PAA spectrum constructed from a set of Gaussian Lorentzian line shapes and a custom Tougaard cross sec tion Figure 43 has been constructed from a set of Gaussian Lorentzian line shapes approximately positioned at the ex perimental values for the valance band two C 1s structures two O AES peaks and one O Is photoelectric peak The widths and intensities are consistent with results for such a sample measured using a Scienta ESCA 300 at RUSTI Daresbury Laboratory UK and the cross section has been chosen similar to the one used in Figure 42 A sequence of iterations using these peaks and the three parameter univer 69 Background Subtraction sal cross section shown in Figure 44 converges to the syn thetic spectrum in Figure 43 This technique for generating theoretical spectra offers a way to understand the consequences of choosing a particular cross section and if nothing else provides a basis from which real spectra can be viewed Surface morphology and nano structure causes most real spectra to exhibit background be havior that is not characteristic of a material but the environ ment in which the element is found Attempting to construct a spectrum based purely on a resonant cross section model clearly shows the limitations of such approximations when applied to real samples Although synthetic modeling has little direct use for real samples it does offer an analyst the opport
128. itioning the cursor towards an end point the mouse button is pressed when the cursor is towards the middle the result is both end points adjusted simultaneously A box out lining the region is displayed and moving the mouse causes the box to shift both end points for the region Backgrounds to spectra can be selected from Tougaard Shirley Linear or None To specify a new background type it is sufficient to type the first character for the background name For example to change from a Tougaard to a Shirley replace the name Tougaard by S then press return The regions will be recomputed with the new background type The initial and final values for the background are tied to the Quantification end points of the region Noise in the data often makes the value at the end point unrepresentative of the desired back ground limits To reduce this problem a number of channels may be used to define the background limits The Average Width parameter defines the number of data channels to be averaged when calculating the tie in points for the back ground The backgrounds are calculated using algorithms based on those presented in a number of published articles The Lin ear case is simply a straight line between the end points of the region None is just a constant set to the lowest data channel in the region The more complicated backgrounds are those due to Shirley and Tougaard The procedure due to Shirley is essentially a w
129. ivate the Apply but ton on the property sheet for the tile parameters The font will only be changed if the OK or the Apply button is pressed Display Parameters and Scrolled Tiles Tile display parameters can be applied to all the tiles in a scrolled list A property page headed Global allows a tog gle to be set that causes the settings currently active to be transferred to all tiles when the Apply or OK button is selected Two additional toggle buttons enable the transfer of ranges for the X axis and Y axis It is not always desirable to in clude these display parameters when the other characteris tics are globally applied but on occasion it can be useful For example setting the Y axis range to that of a specific size for a set of similar spectra can provide an interesting 29 Data Display and Browser visual effect when scrolling is used to view the data The ble is highlighted in this way variation in intensity as a function of an experimental varia Display Parameters and Scrolled Tiles 30 Graph Annotation An important feature of CasaXPS is the ability to write En hanced Metafiles This permits data to be visualised through the data system then exported in to other programs in order to provide the final form used in a document or report En hanced Metafiles supply the drawing information used to display the data in CasaXPS to a word processor such as Mi crosoft Word which can th
130. k without the use of the Shift key or the Ctrl key clears the current selection and replaces it by the indicated block Zooming the Data Once a spectrum has been entered into the left hand side scrolled display window the data can be examined under mouse control In order to zoom into a range of the data the mouse is used to mark a rectangular region and then the third button is pressed Figure 15 The result of this action is to cause the graph axes to be computed from the defined area and the data is redisplayed Zooming may be performed a number of times and the system maintains a list of zoom states These zoom states can be revisited using the fourth toolbar button To reset the list of zoom states press the fifth toolbar button The axes are returned to the state when the data was first displayed in the scrolled list Tiles of Spectra A printed page corresponds to the visible area associated with the left hand scrolling window Each click within the scroll bar moves the display by one page whilst dragging the scroll button allows the pages to be moved at will The arrow buttons on the scrollbar permit fine adjustments to the posi tion of the spectra in view Pages can be divided into sub units referred to as tiles The format for the page is defined using a dialog window avail able from the main frame menu headed Options If Page Layout is selected the dialog box presents property pages that allow the user
131. l event Noise spikes are super imposed on the true data and these should be removed before techniques such as curve synthesis are applied In this case the spikes appear in the background and are therefore obvious to the eye however similar non physical structure that occurs on the side of a peak is less obvious and could be missed 90 Target Factor Analysis The first step in performing the Principal Component Anal ysis is to define a quantification region for each of the spec tra to be used in the analysis These regions specify the acquisition channels that will be used in the data matrix Also any shifts in the data due to charging can be removed from the calculation by using an offset in the energy region for those spectra affected x 10 e am 82 80 78 76 74 72 70 Binding Energy eV Figure 59 Al 2p Spectra Next select the set of spectra in the Browser View and dis play the data in the active tile The processing property page labelled PCA offers a button labelled PCA Apply On pressing this button those spectra displayed in the active tile are transformed into abstract factors Figure 59 displays the spectra before the PCA transformation while Figure 60 shows the abstract factors generated from the eigenanalysis Note that the abstract factors are truly abstract The first fac tor Figure 60 looks like an Al 2p metal doublet however this is because the Al 2p metal envelope dominates the data
132. l to the DS a n GL p form except the position FWHM and area parameters are all determined from the Voigt type shape Compare the area values reported in Figure 37 to those from Figure 36 The area taken from the Voigt type portion is sig nificantly less than the intensity taken from the Doniach Sunjic curve where a cutoff has been used Another differ ence between the two forms is that the area reported by the H is much less sensitive to adjustments in the asymmetry parameter numerical convolution does move shape infor 63 mation between the two halves of the line shape i FEHH Lh m E Taisns ESL s a ITS ES SL Be Figure 37 Hybrid DS GL line shape characterized by the Voigt type portion A New Line Shape The hybrid form of a DS GL line shape is blended using a numerical convolution The maximum for the two shapes di verge for larger asymmetry values therefore characterizing the peak parameters by the Voigt type portion is more diffi cult as the asymmetry becomes more pronounced A new class of functions can be introduced to overcome this prob lem namely a set of line shapes that are a blend of the as ymptotic behaviour from the DS line shape with a Lorentzian The cosine term in the Doniach Sunjic profile rapidly ap Line Shapes Available in CasaXPS Line Shapes and Backgrounds proximates a step function as the asymmetry parameter in creases while for small values of this parameter the cosine
133. label includes one of the sub strings B E BE Binding or binding then the ener gy type is assumed to be binding energy Otherwise the en ergy type defaults to kinetic energy Experimental Variable The numerical value for the experimental variable is record ed in each block Associated information for the value en tered in the block is specified in the experiment header section of the file where a label and the units for the variable are found A proper use of the experimental variable is essential when multiple blocks are recorded in a file with the intention of quantifying trends in the data Reports generated by Transmission Correction and Quantification ISO 14976 File Format CasaXPS rely on these values to organise the blocks so that profiles are meaningful Transmission Correction and Quantification Modern XPS instruments are relatively complex devices characterised by numerous operating modes and methods An ability to control and configure analyser input lens oper ation enables use of a range of energy resolutions spatial resolutions and magnifications but for each operating mode of an instrument transmission characteristics will dif fer and also perhaps vary in a complex fashion within any particular scanning mode causing energy dependent differ ences to occur between spectra taken from the same sample but using different operating conditions These differences in data translate in
134. lex algorithm once close and suffi ciently constrained will march straight to the exact fit The Marquardt method works best when it has a clear view of the target such as with PVC Figure 27 Quantification The option for switching between the two optimisation methods is on the Components property page x10 E 296 294 292 290 288 286 284 282 280 Binding Energy eV Figure 27 PVC artificial structure The optimisation procedure labelled Marquardt is actually not a pure Levenberg Marquardt method Linear and non linear parameters are separated Each step of the optimisa tion procedure includes the solution of sub problems intro duced by the presence of constraints The Marquardt method is used to establish the next set of non linear parameters for the current set of optimal linear ones Although no optimisa tion procedure should be viewed as universally applicable these procedures provide a robust method for determining the peak parameters The Marquardt method uses information about both the function and its derivative There are situations where the in formed steps produced by this method are slower that the lets guess here approach of the Simplex method In fact once the Marquardt method stops making significant im provements in the chi square value there is no harm in 48 Regions switching to the Simplex method for one last try This can sometimes push the parameters away from a local minimum tha
135. log window where a scrolled list of region en A Quick Tour of CasaXPS Introduction ajan a aeri aca at l d d Element Table Periodo Table input Fie Figure 8 Peak Labels Property Page on the Annota tion Dialog Window tries is offered Each column in the scrolled list represents the set of parameters that define a quantification region The annotation for the peaks is used by the Create from La bels button indicated at the bottom of the Quantification Parameters dialog window in Figure 9 Regions created in this way have the name and RSF set from the entries in the element library The background type for each of the newly Introduction created regions is determined from the last background type entered on the Regions property page Juatidvabon Pam amnhers Regions Companerts Date Ector Repot Spec rossiv22 1 2 3 4 5 6 7 8 J Figure 9 Creating Quantification Regions ing a field within the table using the left hand mouse button and the cursor position If a field can be altered then select Step 6 Adjust region start and end points ing that field with the mouse will both highlight the column The Regions Property Page is used to manually adjust the of parameters for that region and cause the selected field to values for the integration regions The table of parameter become a text edit field Any changes are accepted only af fields offered in the
136. lp to limit the complexity of writing a valid ISO file for an acquisition system Comment lines may be included in each block These are useful for summarising technique specific information asso ciated with a particular instrument For example names cor responding to lens modes magnification could be entered as a comment line Operational parameters could accompa ny such text where the state of the charge neutraliser or any other relevant descriptive information may be added CasaXPS provides an option on the Tile Display dialog window for the display of block comments as a header for 21 the graphical display Suitably formatted comment lines may be used to annotate the display with e g acquisition pa rameters for a spectrum Binding vs Kinetic Energy An important piece of information that is not explicitly stat ed in the ISO format is the nature of the energy scan CasaXPS needs to know whether the energy is in binding or kinetic energy The ability of the display to switch between binding and kinetic energy is dependent on the analysis source characteristic energy containing the correct value and the correct assignment for the energy type A REGULAR scan is defined by abscissa start and ab scissa increment energy values To differentiate between the two types of energy the associated abscissa label field is used by CasaXPS as an indicator of which type has been used to describe the data If the
137. m the Al 2p set the other members of the set can be given simi lar models automatically Ensure that the Al 2p set is se lected in the Browser view then in the tile view right click on the active Al 2p spectrum Choose both the Region and Component check boxes then press the OK button Each spectrum within the set is fitted with the same region and components as the model which was prepared earlier When components are propagated the parameters are auto matically fitted to the target data If there is a trend in the da ta for example the oxide components dominate at the surface but smoothly diminish with depth then the compo nent parameters for the previous spectrum in the set of Al 2p narrow scans are a better starting point for the opti misation process The Propagate Action dialog window includes a check box labelled From previous block If the check box is ticked the parameters for successive blocks are taken from the preceding block Otherwise the parameters derive from the block displayed in the active tile Again it is advisable to view the results of these automatic fits by scrolling through the spectra in the tile view Once satisfied the ISO 14976 file is now ready to generate a quantification report The first step is to ensure that all the spectra to be included in the report should be selected within the Browser view A full report detailing parameters such as position F W H M R S F raw area and
138. may have col extensions associated with these data directo ries but later versions of Eclipse generate such directories without this naming convention Similarly depth profiles or angle resolved experiments appear as sets of directories one per acquisition region where each sub directory contains the dts files for one of the acquisition regions involved in the experiment Again older versions of Eclipse use a nam ing convention where each sub directory was given a mle extension The first step to creating anew VAMAS file is to enable the Convert toolbar button The Convert option is disabled un Using Different File Formats less the selected sub frame window representing a file con tains no VAMAS regions If no empty Experiment Frame is amongst the list of windows then the New toolbar button must be pressed to create an empty Frame There are three ways to convert VG Eclipse data e If an individual file is required then selecting the dts file via the Covert to VAMAS file dialog window will result in the creation of a VAMAS file that contains the spec trum in that dts and no others e If on the other hand all the dts files are require in the same VAMAS file the Covert to VAMAS file dialog window should be used to find the directory containing the spectral regions The name of the new file entered in the dialog window must be specified in the text field but with a col extension It is the extension given to the file n
139. mouse These two parameters are modified simultaneously so as to preserve relationships between peak intensities Constraints may be specified for position width and area These may be either an interval for example 0 5 1 3 or a constraint that is measured with respect to another compo nent In the latter case the parameter for one peak can be fixed with respect to another by entering the column label into the constraint box For example to fix the position of a peak in column B with respect to a peak in column A type A for the position constraint in column B The sys tem responds by setting the constraint to be A x where x is equal to the initial offset between the two peaks The value for x can be modified by hand To provide an offset of 2 eV column B constraint item should read A 2 Area may be constrained similarly but instead of plus use times e g A 0 666 Read column B component is 0 666 times the area of column A Quantification Quantification Calculation Percentage atomic concentration X4 calculations are per formed using the formula y EARTE 4 E E R T E Where R is the relative sensitivity factor for the measured intensity J and T E is the transmission function for the in strumental operating mode used to measure the intensity at kinetic energy E The term involving raising the kinetic en ergy E to a power allows traditional adjustments for ana lyzer behavior for th
140. mply select the tile used to display the data and the period ic table will update to reflect the state of the markers chosen for that spectrum 41 Quantification Data can be quantified at a number of levels The most basic form of quantification is using regions applied to a wide scan spectrum the results are added to the display using an annotation option Alternatively both regions and synthetic components can be prepared over a set of narrow scan spec tra and the results specified through a set of expressions in volving the names used to label the objects A report generated in this way can be viewed within CasaXPS or ex ported as a TAB separated ASCII file for use with a spread sheet program such as Excel Figure 24 shows the main toolbar button for the dialog window used to perform these two extremes plus a number of reporting mechanisms in be tween Disks OG s HS Figure 24 Main Toolbar The three property pages used to quantify spectra are la belled Regions Components and Report Spec CHAPTER 7 Regions The first step in quantifying a spectrum is defining one or more energy regions These specify the data channels that are to be included in the intensity calculation Each region requires the following information e Name The name given to a region is signifi cant as a quantification object A report is specified in terms of these names and intensities are summed for regions with identical names
141. n Windows 98 7 adjusting 68 xX v X Axis 136 VAMAS x ray line shape 55 definition 18 VAMAS Files 17 Y VGX900 files 125 Y Axis 136 Vision 1 x 2 x ASCII files 128 Voigt 57 59 61 Z Zoom Reset 14 zoom states 26 LN 2001 Casa Software Ltd 26 Burford Crescent Wilmslow Cheshire SK96BN United Kingdom CasaxPS Tel 44 0 1625 535346 E Mail info casaxps com World Wide Web http www casaxps com aE
142. n regions for a survey spectrum The Element Library dialog window is invoked or brought to the front of any CasaXPS dialog windows by pressing the toolbar button indicated in Figure 6 The first property page of the element library dialog window is a scrolled list of line positions where the list is ordered by the energy of the en tries If the mouse is left clicked on the spectrum displayed in the left hand side of the sub frame shown in Figure 6 then the scrolled list of element library entries will scroll to the energy indicated by the mouse click Therefore by pointing at the peaks in the spectrum via the mouse the names for the various features can be brought into view and if the name field displayed in the scrolled list is selected az CasaXPS Licensed to Neal F auley earlpox sci vma Jt Ete Yoon Window plone ep Introduction again using the mouse a set of markers will appear on the spectrum positioned at the energies at which spectral lines should appear for the chosen element Figure 7 shows the spectrum after the O 1s and C Is lines have been activated via the element library scrolled list piss ale olala ajej hlebdajajolala Krem omele see ee odd 1 2 3 s 5 6 8 9 1 Figure 7 Peak identification via the element library A Quick Tour of CasaXPS 10 If the name field of an element library entry is selected a second time the element markers are removed from the dis play
143. n the expected value and the computed value is the basis for determining the number of principal components Both Xn and its expected value decrease as n increases Y_ initially is larger than 7 expected but as n increases a crossover occurs The true dimensionality of the data matrix is chosen to be the value of n for which Ke 1S closest to its expected value Note that smoothing the data will alter the characteristics of the noise Performing such pre processing therefore invali dates the YX statistic Target Factor Analysis Principal Component Analysis provides a set of basis vec tors that describe the original set of spectra Although useful as a means of characterising the data these abstract factors 83 Target Factor Analysis are in general not physically meaningful Target Factor Analysis is concerned with identifying vectors that can also describe the data but with the additional property that they are recognisable as spectra rather than simply abstract vec tors in an r dimensional space There are numerous methods for transforming the PCA ab stract factors to provide vectors that are more open to chem ical interpretation These involve constructing abstract rotation transformations that map the abstract factors into one of the infinite number of alternative basis sets for the factor space Fortunately there is a technique which when coupled with curve synthesis lends itself to the analysis of XPS data n
144. n where a pair of unre solved peaks is the subject of interest and the total intensity is the value that is calculated from the data The data in this example Figure 67 are purely synthetic two Gaussian peaks of equal intensity and equal FWHM of 2eV are offset by leV ona linear background This provides an envelope open to analysis by Monte Carlo simulation that will provide visual insight into the errors associated with noise in the data Integration Region Limits The purpose of this analysis is to illustrate the consequences of choosing inappropriate end points for the integration re gion Note that the peak fitting procedure is not at issue since only the data are used to calculate the intensities for the combined peaks 102 Quantification ee ee ee ee m m nae m lt r a oH ma Ha Ji ii E ma m Misting ery Figure 67 Two unresolved peaks Figure 66 shows a scatter plot where each point is a normal ized area plotted against simulation index These areas are calculated from the original data envelope after the introduc tion of pseudorandom data and the computed area is refer enced to the initial area before noise was superimposed on the spectrum Monte Carlo Methods Uncertainties in Intensity Calculations Each calculation includes determining a new linear back Peak Area determined from Integration Region One End Point Defines Region Limits 1 20 oo 145 bd re r3
145. nal column is the percentage concentration for each set of quan tified items with the same experimental variable When the report is based upon regions and or components each item is included in the report on a separate line even if a region has the same name as another region or component This allows the statistics associated with each quantification item to be included in the list Items with the same name are treated in a different way when formulae are used to gener ate the report If a report is based on regions and or components and not all the regions are to be included in the calculation for the percentage concentration then an R S F of zero should be entered for those items that are to be excluded from the re sults The second type of report is generated using the table of named formulae see for example Adjusting a Custom Quantification Report on page 120 Each formula takes the form of an arithmetic expression Variables in these ex pressions may include names given to the regions and com ponents such as the data blocks which are currently selected Quantification in the Browser view of the active data file The list entitled Quantification item names shows the set of names current ly available for generating a report Clicking on the column header buttons labelled Name or Formula creates a new name formula pair The entry is ad justed via a dialog window Right click over the name item for which c
146. nalysis Principal Component Analysis secondary factors are those factors that can be associated with the noise and in principle can be omitted from subse quent calculations It is not possible to completely disasso ciate the true data from the error within the measured data however the statistics guide the analyst in choosing the most appropriate number of abstract factors that describe the data and therefore the best guess dimensionality for the data matrix In the case of XPS spectra the experimental error is known to be the square root of the number of counts in an acquisi tion channel Under these circumstances where the experi mental error is known a number of statistics have been proposed for determining the size of the true factor space Residual Standard Deviation Real Error An alternative name for the residual standard deviation RSD used by Malinowski is real error RE The RSD is defined to be where E is the jth largest eigenvalue n is the number of ab stract factors used to reproduce the data c spectra each with r channels are used to construct the data matrix RSD must be compared against the estimated experimental error If the value computed for RSD is approximately equal to the estimated error then the first n abstract factors span the factor space The dimensionality of the original 82 Target Factor Analysis data matrix is therefore n Two further statistics may be derived
147. nd C 1s In addition to the integration regions the O 1s spectrum is fitted using two synthetic components and both of these components are named using the same name as the integration region for the O Is spectrum A Standard report generated from these inte 120 ak a jompi 3 j Haa Fea PIEM Li Ami l 33205 3905 Gua Sod tS k 3305 20003 Gua srr HOO iid Been Aj Heat Pai FEHH mma bad pf ds Seas 2 ee le Ck T E H E H t E NNA u el Dadig Bap pti Figure 88 Three High resolution Spectral Regions gration regions is as follows Table 11 Standard report generated from the quantification regions shown in Figure 88 Name Energy FWHM R S F Area Pin O Is 532 35 2 38589 0 78 13299 4 54 52 Al 2p 72 7 0 740902 0 193 2192 6 36 33 CIs 286 3 1 39428 0 278 795 3 9 148 Adjusting a Custom Quantification Report Analytical Applications A custom report using the same names as those shown in Table 11 is as follows Table 12 Custom report using the same names as those used in Table 11 Ols Al2p Cis CPSe cpsev cpsev 078 Alap E ls 34075 5 11360 8 2860 9 70 5538 23 5227 5 92353 Note that the O 1s percentage concentration is larger than the equivalent value reported by the Standard report The Hm Pra FAHM Loh m h Fak Il 332025 LAMI GLUE Bee FT F g Peake SAL IEE GLM erma oE Of Of 1 af
148. nd Factor Space Display t Toggle Counts and CPS Intensity scales Toggle Binding and Kinetic Energy scales Display all selected blocks in a tile L Display one block per tile The Toolbar upper buttons in general provide ac cess to menus or dialog boxes The Options bar lower buttons in general execute actions immediately or reversibly change the for mat of the display Hovering over a button placing the mouse screen pointer on an icon without clicking produces a descriptive label for that button a Tool Tip anda slightly longer description in the status bar Function Buttons F5 F10 mirror the Processing buttons Page Layout Element Library Toggle normalised display Toggle subtracted display Display Modifier buttons C 3 Bal BG ete E L Toggle background Toggle residuals display Toggle components display Toggle shaded region display NB Modifiers require prior function definition Comment Information amp Variable control ae Edit Block Comment For Display sof mf salt Set Experiment variable linear a Edit Species Transition ind Edit Source Analyser parameters Edit Block Information ae 104 Index Symbols GL 50 99 Dayta System Files 127 GL p 60 GL p K b0 b1 60 GL p T k 60 SGL p 60 gt SGL p K b0 b1 60 SGL p T k 60 Numerics 1E37 21 A AET apparent error in the test vector 84 Analyser response
149. ned for germanium U Poly Tougaard short form U Poly Three parameter cross section determined for polymers U 4 Tougaard short form U 4 Custom three parameter cross section where the parameters are entered on the Re gions Property Page Figure 10 in the Cross Section text field The four numerical values entered in a comma separat ed list correspond to B C D and TO in the energy loss func tion above C 4 Tougaard short form C 4 Alternative custom three parameter cross section where the parameters are entered on the Regions Property Page Figure 10 in the Cross Section text field The four numerical values have the same meaning as those found in the U 4 Tougaard form however TO is used to shift the resonance structure rather than simple acting as Line Shapes and Backgrounds a cutoff limit It is therefore possible to retain the shape of the distribution but move the position of the maximum with out changing the C and D parameters The custom Tougaard cross sections offer a chance to change the background in response to the spectrum under in vestigation Such an approach is not unprecedented 12 Direct simultaneous determination of XPS background and inelastic differential cross sections using Tougaard s al gorithm by Dr M Jo describes a method that extracts loss functions from the XPS data using optimization procedures applied to a cross section defined in terms of spline polyno mials The rigi
150. nformation should not be underestimated Proper intensity calibration opens up the opportunity for consistent quantification not just across op erating modes of a single instrument but also for instruments from the same manufacturer and possibly instruments from different manufacturers Analysts reliant on data taken from a range of instrumentation will be aware of this problem Although there is no prescription for transmission functions within the ISO 14976 file format standard the mechanisms are in place that enable recording of more than one corre sponding variable per ISO block and therefore the energy dependence of the transmission can easily be saved along with the energy dependence of the signal ASCII files have come of age with the increasing size of hard disks and zip technology There is no longer a serious concern about the size of XPS AES spectroscopic files and the benefits of re cording open and explicit data outweighs any drawbacks that may accompany ASCII format spectra 23 Quantification Historically transmission correction has been performed via adjustments to the RSF values used to quantify the peak in tensities While at first sight this would seem a reasonable approach the nature of transmission functions means that errors are introduced by this procedure The RSF attempts to allow for the set of factors that cause peak intensities to vary for reasons other than sample composition The influences of the t
151. ng oxidised in a vacuum chamber as a result of combination with low level residual water and oxygen 1 x 10 mbar These are particularly good data for a PCA since the spectra have been recorded over the Al 2s as well as the Al 2p photoelectric lines and both structures should vary to gether as a unit over the course of the experiment A PCA for the data set shown in Figure 82 using correlation about the origin produces the abstract factors and associated eigenvalues displayed in Figure 83 The first abstract factor is clearly a vector in the direction of the maximum variation in the overall data set There are shapes within this first ab stract factor that belong to what can be identified as alumin ium bonded with oxygen The second abstract factor exhibits structure associated with each of the aluminium peaks and what is more the structure is of similar form The third abstract factor contains almost no significant structure and all the PCA statistics point to the need for two Trend Analysis for Metal Oxidation Analytical Applications Figure 83 First three abstract factors generated from the spec tra shown in Figure 82 principal components when reproducing the original spec tra The trends within the data set can therefore be assessed by plotting the co ordinates for the spectra when projected onto the two dimensional subspace spanned by the first two abstract factors At this point it would be easy to conclude that any syn
152. ng propagated to other blocks the history item can be flagged so that it is not propagated along with other appropriate actions To exclude a processing operation from a propagate action select the item in the history list then press the button la belled Propagate Flag The entry in the list for the selected processing option will change to indicate that the flag for propagating the action is set to false don t perform this Repeat the procedure toggle to remove the exclu sion Processing Please note that although the Calculator PCA Prin cipal Component Analysis and Test Data tabs appear rightly in the Processing module they are of sufficient im portance and power to merit their own topic sections see Using the Calculator and Comparing Spectra on page 71 Principal Component Analysis on page 80 and Optimi sation of the Peak Parameters on page 47 The same is true of the Monte Carlo procedures the entry to which appears under the Components tab of the Quan tification module see Monte Carlo Methods Uncertain ties in Intensity Calculations on page 97 37 Element Library The Element Library dialog window provides the means of managing the information used to identify spectrum fea tures and then create regions and components with prede fined values Strictly it is not an element library at all but a list of spectrocopic lines whi
153. ng variables within a data block Some manufac turer s data acquisition systems routinely provide output in this important form Principal Component Analysis and Target Factor Analysis are also available as a processing options as is a Monte Car lo method for uncertainty analysis Quantification Spectra can be divided into quantification regions which form the basis for estimating the background shape associat ed with the recorded data A sample may be analysed based on these user defined energy ranges or fine structure within a data envelope investigated using synthetic components Intensities for the elemental composition can be combined from regions components or via an arithmetic expression that involves any combination of the region or component intensities Tables generated from the quantification process are saved to disk in a TAB separated ASCII file Spreadsheet pro grams such as Excel will read these files and provide many options for printing the data in either tabulated form or as a graphical profile ISO 14976 VAMAS Files The ISO Surface Chemical Analysis Standard Data Transfer Format ISO 14976 is a prescription for presenting the ac quisition parameters and data for a range of instruments and techniques CasaXPS is only concerned with X ray photo electron spectroscopy XPS and as such whenever the term ISO spectrum or VAMAS file is used in this manual it is intended to mean that subset of
154. nnotation History property page Entries on the Annotation History property page are in the format used to save the annotation in the ISO 14976 file A block of data within the file includes a number of comment lines It is within these comment lines that processing anno tation and quantification information is stored The idea of offering the full description used to save the data to file is that the parameters can be understood if desired It is then possible to save the data and include CasaXPS information from the start Peak Labels Spectra can be labelled using the names from the element li brary The element markers currently displayed are made available for annotating the spectral features on the Peak Label property page of the Annotation dialog window The element marker text is entered in the scrolled list found on the page of interest To annotate the spectrum the ele ment names must be selected within the scrolled list and the desired attributes chosen The annotation text entries are added to the annotation history list when the Apply button is pressed Again the exact location for the annotation will need to be assigned through the normal positioning proce 32 Text Annotation dure Note that Peak Labels derived from the entries in the Element Library are different from Peak Markers generated by the Find Peak button in the Element Table tab of the Element Library module see page 1
155. nor Tick Marks is a simple on off control Display Axis Scale similarly turns off the y axis scales and labels Normalise Display sets y values equal at indicated point on the x axis To move normalisation point use shift left click in the display tile uses the foreground trace scale in a multi trace display Control mirrors the normalise button Display Exp Variable Left Right labels each trace at its right or left extremity with the corresponding value of the experimental variable e g etch time Non exclusive Display Background Subtracted includes or excludes background from display Greyed when not available Geometry Tile Display Geometry selects either 2D 3D or Factor Space The parameters immediately below apply to 3D Back Plane and Front Plane and Factor Space Scatter dis Processing Dialog windows Command Summary plays only 2D has no further options Reverse Display mirrors the order of plotting traces in a 2D or 3D display top to bottom or front to back but not the traces themselves spectra are always plotted with in creasing KE left to right Back Plane Parameters provide scaling and offset and thus perspective for the furthest back trace of a se ries of traces visualised as slices of a cube subject to plot ting within the tile area Values are percentages Offsets may be positive or negative and relate to the leftmost or
156. ns links for experimentalists and so it is appropriate that means are provided for transport and display of ISO data over the web Standard methods of text file compression and transfer work well with the format because it is text based Display of spectra in a web environment on a browser page for example is a non trivial matter which is of interest in many contexts and there exists at least one simple Java applet which enables display of XPS spectra live on a web page The applet is also designed to provide a basic practical mechanism for platform independent that is universal transfer and display of XPS spectra across the world wide web See for example http www acolyte co uk JISO MIME types Multimedia Internet Mail Extensions are the designators which enable applications such as web browsers and their helpers to decide how to handle interpret an otherwise unknown file type The vms file extension is used for the ISO 14976 spectra and is in process of submission through the IETF as a Chemical MIME type The present status is experimental i e the type sub type chemical x vamas isol14976 should be set Since the format is solely text based however http systems and browsers communicating with them will in general react correctly if configured to serve or read unknown MIME types as text plain In most cases this will be the default action 148 Appendix 3
157. ntation Errereaerre S Preceee a eeeree ta tt tt ror irtitt Enana m Dm O at ie Te ie he a ETS eee ie er Thit pe be ce Get i Karaan F plors When more than one item of the same type is present in a display e g spectrum traces then custom colours are cy cled through in order Global This tab provides a means of applying the changes made with the other tabs to all the tiles in the current Experiment Frame including or otherwise x and y axis ranges and the title the Tabs also include buttons to enable saving the de fault parameters for colours settings and fonts typefaces A directory called casaxps def should be created at the Processing Dialog windows Command Summary same logical level on the currently operational system disk as that at which the file casaxps exe is stored CasaXPS will then create and store files for all the default parameters for acquisition colours and so on here Quantification see Quantification on page 42 Gages kapran panid Fasa tpar Regions This tab provides access to the mostly self explanatory pa rameters for any new or existing region for the displayed block or blocks Note that all the parameters in the region list are editable A selected region as a whole click on the header letter has a blue background an individual parameter selected by a mouse left click has a plain white background and the e
158. nvelope where three GL peaks have been added then fitted using a Marquardt Levenberg optimization algo rithm No constraints have been applied and the result is a reasonable fit to the experimental data but the fitting param 19 Jones F et al Fluoride uptake by glass ionomer cements a surface analysis approach to be published 20 Press W H et al Numerical Recipes in C Cambridge Uni versity Press 1988 107 Analytical Applications Figure 74 C1s high resolution spectrum taken from a polymer Figure 75 An initial fit to the C 1s spectra shown in Figure 74 based sample using an ESCALAB 220i at University College London PAA The PAA stoichiometry is still doubtful but the essen tial positions for a pure PAA envelope as indicated by Beamson and Briggs have more or less appeared Further input is required to make sense of the new synthetic model eters are not readily open to chemical interpretation For ex ample the FWHM are much bigger than would typically be expected from C 1s profiles the given instrumental resolu tion The three peak model suggests further deconvolution is required before the sample can be fully understood To constrain a parameter so that it does not adjust during an op If it is assumed that the FWHM for a C Is photoelectron timization step set the constraint interval to have the same value peak is 1 1 eV only a guess then applying peaks with said as the fixed value
159. o processing software now has the opportunity to operate on the data with the same degree of automation that is available to the acquisition routines The Browser in CasaXPS presents the ISO blocks in a tabu lated matrix form in which each row of the table repre sents data acquired with the same experimental variable e g ion gun etch time The columns are spectral regions that have the same chemical species label element symbol or formula and transition detection energy or level nota tion Individual spectra or complete sets of them can be selected and displayed in a scrolled list and all processing and quan tification operations can be applied to a current selection The Browser thus forms the basis for automatic processing of the data It is also through Browser operation that quantification re ports are generated again reducing the time to complete the analysis of both simple and complex samples T a aTa Te i iii H E U U A a amp Banding Eey hT Pasir Figure 14 The CasaXPS Program Frame containing the Toolbars and a single Experiment Frame which shows both display and browser splitter windows Element Library An element library is maintained by the CasaXPS system It allows the user to manage the information that describes the X ray induced transitions such as binding energy peak widths or relative sensitivity factors These quantities once entered can be accessed through
160. o provide a standard graphical user interface as well as much of the data management required by the system The software suite as a whole may be regarded as being structured from five interacting components e Data display and browser an experiment window e Element library a dialog window as are the others e Graph annotation e Data processing e Quantification This manual provides a detailed description of the system its operation command and function set and its use in prac tice in real life analytical applications Getting Started Getting Started Installing CasaXPS You must install the application from the CasaXPS CD onto your hard disk Drag and Drop installation is all that is re quired for Windows 98 Instructions are available if re quired in the Read Me file on the installation CD Installation should occupy no more than SMB on your hard drive You will need perhaps an equivalent space for storage of data files Make sure that you have your serial number to hand when you install the application or else the package will revert to Demonstration Mode and will not allow you to save the results of your work The serial number is avail able from your registration documents which may be e mailed to you or included with the CD You can validate a registration at any time after installation turning a demo version into a fully licensed product by using the About button
161. of each compound varies throughout each set of ten VAMAS blocks The data is located in the files clstestl vms clstest2 vms and clstest3 vms The underlying trends intro duced into each file are as follows peaks corresponding to PMMA and PVC obey quadratic adjustments in intensity over the set of ten spectra PMMA decreases while PVC in creases The difference between the three files is the pro portion of PVA in each data envelope The first file clstestl vms has a constant level of PVA Figure 54 the second file clstest2 vms varies linearly first increasing then decreasing the third file clstest3 vms includes a lin ear increase in the level of PVA The objective is to show how the statistics used in PCA be have for a known problem Data matrices constructed from the three sets of spectra should have a dimensionality of three Principal Component Analysis PMMA a CPS nN D a a a DRT a T T 5 el ees 296 294 292 290 288 286 284 Binding Energy eV PVA 95 10 b i b 15 m A gt 10 5 0 294 292 290 288 286 284 Binding Energy eV PVC x 10 14 12 c 10 mn 8 A gt 6 4 2 0 294 292 290 288 286 Binding Energy eV Figure 53 a c Artificial C 1s Data 85 Target Factor Analysis Note that although each compound is constructed from a number of C 1s peaks PMMA 4 PVA 4 and PVC 2 the stoichiometry of these compounds masks the true number of synthetic components a
162. of D xn ci ma He o Dar Pasting Berg 1 Figure 41 Wide scan spectrum from a PAA sample with a custom three parameter cross section applied The distance of a complex pole from the real line in the do main of a complex function is instrumental in determining the width of a resonance when the function is plotted with Line Shapes and Backgrounds respect to a real variable The D parameter principally moves the poles of the Tougaard cross section away from the real line while C changes the projection of the pole onto the real line Small D corresponds to near singular behavior in the cross section and therefore results in shape structures located to some degree by C while the magnitude of the function can be moderated by adjusting the B parameter It should be noted however that B adjusts the behavior of the cross section in the region close to zero energy loss This is precisely the region of interest for modeling a peak and so the values given to B and D can be influential in obtaining a reasonable synthetic model The three parameter universal cross sections for specific materials may be used as a basis for determining a back ground A wide scan spectrum allows the essential form for a background to be determined in the case of materials for which cross sections are not available The resulting back ground is easily transferred to any high resolution spectra from the same sample Figure 41 and 42 The mean
163. ons and components then to define the way these quantities should be combined in the form of a report The multiple document interface MDI used by CasaXPS requires anew document before a file can be selected When CasaXPS is first run the initial state is an empty document and the Open menu option on the File menu is in an ac tive state If all the documents have been used or none are on offer then the Open option as well as the corresponding toolbar button will be inactive To activate the Open op tion press the New menu button an empty ISO document frame will be created Osea ame Ss Figure 29 Main Toolbar A file dialog window enables browsing of the file system and choice of an ISO file Take care within the file dialog It is the standard MFC dialog with drag and drop functionality It is therefore possible to move files around simply by drag ging the icon for the file over the icon for a folder Once read the ISO file structure is displayed in the right hand side of the Document view Each transition appears as a column and each row is populated by spectra acquired with Quantification by Example Quantification the same etch time Table 3 Structure of the Browser view for a depth profile Paan Cis ois Al2p Mg2p 0 Cis O1s2 Al2p3 Mg 2p 4 100 Cis Ois7 Al2p 8 Mg 2p 9 200 C 1s 11 O 1s 12 Al 2p 13 Mg 2p 14 300 ae J Pa 2000 C 1s 101 018 102 Al2p 103
164. or less than the data range for the selected tile Type in decimal values values relate to the scale se lected above so take care with max and min values for BE Set Scale enables a linear remapping of the x axis for use in e g depth profiles Most usefully employed when the output report is saved to a fresh file with vms extension then re opened in the system and the scale changed Display Minor Tick Marks is a simple on off control Display Axis Scale similarly turns off the x axis scales and labels Rescale Axis changes the scale to KE and also changes the label to whatever is typed in the adjacent box Y Axis Dependent variable provides a radio button choice be tween counts per second and counts counts per bin i e counts accumulated in each energy interval acquired for y axis display a means of labelling the axis as desired the 136 type in boxes and of setting the typeface Font for this label Min and max enable setting the span of the display which may be greater or less than the data range for the selected tile Type in decimal values which relate to the scale select ed above Offset spectra provides an automatically calculated shift in the y axis for each overlaid spectrum The value sets the size of the y axis scale in relation to the available area not to the offset The offset is then calculated to divide the remaining space equally Display Mi
165. oscopy XPS data That is to say while CasaXPS will read files in the format defined by ISO 14976 display and processing within the system are strictly only available for a subset of the Format This should pose no significant difficulty for virtually all practical situa tions Partially encoded versions of the Format essentially tem plates for the most common and most useful instances were developed as annexes to the VAMAS definition and other early reports e g Appendix 2 of the National Physical Laboratory Report DMA A 164 July 1988 This is main tained in the full ISO standard Annex C The first of the partially encoded formats C2 is particularly appropriate to CasaXPS It describes the fields for defining an experiment involving a number of regular scanned spectra or spectral regions for one technique as a function of one experimental variable the analysis not being at a specifically addressed ISO 14976 File Format point on the sample This partially encoded format is suita ble for experiments such as XPS depth profiles either using angle resolved methods or ion bombardment etch sequenc es or surface chemical analysis over time with or without temperature variations see Appendix 1 ISO 14976 format files annotated on page 143 File Structure What is defined by ISO 14976 is an experiment This con sists of a series of parameters which apply to the measure ment procedure as a w
166. ose who wish to quantify spectra according to a manufacturer s specification The value for the exponent a is entered on the Regions Property Page of the Quantification Parameters dialog window and is used whenever the accompanying check box is ticked If the check box is ticked then both the exponent and the transmis sion function located in the VAMAS block will be automat ically used during quantification however a transmission function must be present for either of these corrections to be included in the calculation Note that intensity calibration for a spectrum can be performed using the Intensity Calib processing option where an exponential adjustment to the data can be made regardless of the presence or absence of at ransmission function Quantification Report There are several types of quantification report The proper ty page labelled Report Spec allows peak intensities to be 45 Regions compared to one another either via information derived from regions components or both regions and components In ad dition a set of named formulae allows intensities to be mixed and matched in any way that an arithmetic expression can define A report generated from regions and or components is pre sented in the form of a scrolled list For each experimental variable from the ISO file the set of quantification items is listed Information including peak position F W H M R S F and raw area appear in separate columns The fi
167. p 3 4727872000 82 5236 159 3106 34 76442 398 2765 26032 99 2600 Al 2p 8 65086870 3 784555 23 78739 7 340948 65 89305 136 4193 2451 Al 2p 13 393178 7 5 875731 20 74929 7 842494 64 04102 133 7949 2304 Al 2p 18 305001 5 3 48073 17 85783 7 793798 61 79182 78 0949 2159 Al 2p 23 156694 8 3 249367 16 25038 7 929371 63 47803 59 6866 2016 Al 2p 28 98359 06 2 757161 15 2192 8 135007 67 64091 51 1035 1875 Al 2p 33 86168 29 2 48836 14 18397 8 189118 72 36718 41 82519 1736 Al 2p 38 65267 54 2 22333 13 35424 8 242415 79 01916 37 98182 1599 Al 2p 43 53613 14 2 247765 12 61316 8 257255 87 59142 42 46731 1464 Al 2p 48 43569 08 0 1744253 11 97161 8 261198 98 93895 0 1869893 1331 Al 2p 53 32387 23 1 710532 11 52946 8 344409 115 2946 29 01139 1200 Al 2p 58 28174 98 2 021671 11 12658 8 410907 137 3652 33 42906 1071 Al 2p 63 24742 1 261896 10 75487 8 461885 168 0448 15 17705 944 Al 2p 68 23980 27 0 40768 10 29759 8 407944 210 1548 1 649807 819 Al 2p 73 20710 42 1 113217 9 867347 8 33943 274 093 10 69842 696 Al 2p 78 18345 12 1 155456 9 424948 8 226769 376 9979 15 75993 575 Al 2p 83 16109 71 0 7358818 8 960655 8 062218 560 0409 5 605129 456 Al 2p 88 13003 76 0 7303461 8 600543 7 962556 955 6159 5 882052 339 Al 2p 93 12307 15 0 7049177 7 99876 7 608339 1999 69 6 193881 224 Al 2p 98 9285 948 0 000443667 7 554815 7 372745 7554 815 2 12747E 06 111 92 Target Factor Analysis Principal Component Analysis Table 8 Report generated by a PCA for Al
168. pectra highlight trends within the data set and the relative impor tance of the abstract factors can be examined A plot in which the axes are defined by unimportant factors generally appear random while factors that are significant when de scribing the data typically produce plots containing recog nisable structure 96 Monte Carlo Methods Uncertainties in Intensity Calculations Optimization routines are completely deterministic in that the same set of parameters used in the same functional forms applied to the same set of data with the same initial and final conditions will result in the same set of parameters on termination Vary any of the above conditions and the re sult from the optimization routine will change in some re spect One method for assessing the uncertainty in the parameters for a peak model is to vary these optimization conditions by repeating an experiment on what are hoped to be identical samples Then for each set of data apply the same optimization routine to the same synthetic model and so determine a distribution for the set of parameters used to quantify a sample Such a procedure will vary almost every aspect of the measurement process and so result in a distri bution for the parameters that truly represent the nature of the experiment The basis for such an approach as described above lies in the assumption that there exists a set of parameters only known to nature that does not depend on any opt
169. percentage concentrations for all the quantification units is obtained by pressing the Quantification by Example Quantification Combined button on the Report Spec page Regions and components can be reported in a similar format by pressing the buttons on the same property page but baring the appro priate names A custom report for the same data is usually in a format more suited to showing trends To generate such a report it is nec essary to complete a table of names and formulae The first column for a custom report is a list of the experi mental variable values In this example the column will con tain etch time in seconds the units are defined by the VAMAS file format Subsequent columns are defined by the formulae and are labelled by the corresponding names Pressing the Reset button initialises the list of names for mulae Table 4 The entries are taken from the regions de fined for the spectra currently selected within the Browser view Typically this provides the right number of quantifica tion names but additional ones may be created or existing ones removed via the edit dialog window Right click the mouse button over a name field to display the edit dialog window Table 4 Name Formula List Name Formula C ls C ls O Is O Is Re Al 2p Re Al 2p Mg 2p Mg 2p 53 Report Files and Excel In the current example three of the entries are appropriate for the custom report
170. perimental variable displayed at either side of the graph area and the spectrum background may be subtracted from or included with the display see deatail of command options on page 136 An alternative way of displaying overlaid spectra is via the Geometry property page Here the user can select between 27 Colours 2 D and 3 D views of the spectra The latter option plots the spectra as a function of the experimental variable The first spectrum in plotted in the front plane of a cube while the last spectrum in the selection is plotted at the back plane All spectra in between appear at a plane that is determined from the value of the experimental variable for the VAMAS block The 3 D plot is adjusted using parameters that shift and scale the two planes associated with the front and back of the cube Figure 3 shows a set of spectra plotted against etch time The front plane is 75 of full size in both X and Y di rections while this reduced axes area has been shifted left by the maximum amount by specifying 100 for the shift Note that the front plane can only be shifted around in X while the back plane may be shifted in both X and Y A third option for data display allows a set of spectra to be viewed in terms of abstract factors This involves perform ing a Principal Component Analysis PCA on the spectra Please see the section on Principal Component Analysis for a description of PCA and the use of this display option
171. quantification of high resolution spectra If on the other hand it would be more convenient to separate spectra by an experimental variable then the new file name entered in the dialog window should be followed by a col string e g newfile bri col omit the quotation marks The 127 experimental variable is the index number determined form the order in which the files are read Convert to VAHAG fiis HE en face Gl olay S ited OSB Erce OESP Et By Eide OT BP Mj Eride WASP hicap M Eda H EL KJ Ea F Evade 5 F_ BE ceeds OR EP_ a Eride 0175P_ Erais Dn SF Mj Eida E F E Er 125 P_ Evade MSP era E F C Fis of i fa Files Canoe Figure 95 Convert to VAMAS file dialog window for Dayta A bri file extension has been specified The result is a file called newfile vms containing all the SP_ data files seen in the window Kratos Vision 1 x 2 x ASCII files The Kratos Vision data sets are probably the closest in logi cal design to the ISO 14976 file format A single binary file maintains the context for an experiment where an experi ment may include multiple acquisition regions characterized by so called state changes These state changes are equiv Kratos Vision 1 x 2 x ASCII files Using Different File Formats alent to the experimental variable fields used in the ISO 14976 standard and the Vision Objects are equivalent to the VAMAS blocks The principal problem with converting the Vision 1 x 2 x d
172. ra displayed in the ac tive tile The loading used to compute the predicted spectra are listed in the report The report may be written to file us ing a similar procedure to the TFA report described above Viewing the Data in Factor Space CasaXPS offers an option on the Geometry property page on the Tile Display dialog window labelled Factor Space If selected the VAMAS blocks displayed in a tile are used to define the axes for a subspace and the original data are plotted if possible as a set of co ordinates with re spect to these axes The plot represents a projection of the data space onto the subspace defined by a set of two or three abstract factors Principal Component Analysis 1 0 432072 D 574209 DE64039 2 DIO6 D512376 DDI E461 13 0 2223 DD 0737 DLOI41226 14 0 24226 DD 74342 DD330708 13 0 242002 D8I4353 D0672406 16 0 24923 DDPOGI 99 DDI 37312 17 0 244408 D 100229 D D199300 12 0 24393 DICI WI DDI 20846 19 0 247212 D 1D36I9 D 191 008 2D D24604 D10734 D 1 422877 21 D24047 0 121289 D 212746 32020 72 76 74 N D Binding Energy eV Figure 62 Projection onto Abstract factors 1 2 and 4 The abstract factors defining the axes are graphed together with a list of the co ordinate values for each of the spectra projected onto the subspace spanned by the chosen abstract factors Figure 62 A 3 dimensional plot provides a visual interpretation for the spectra Patterns formed by the s
173. ransmission function are usually assumed to be con stant across a peak but this assumption is often false in the case of the relative transmission functions The response of the instrument can vary rapidly across a peak width partic ularly when portions of spectra are compared at opposite ends of the resolution or energy scale Empirically determined RSF values necessarily compensate for instrumental response in the reference mode Alternative modes need to be adjusted relative to the reference mode by dividing spectra point by point by the transmission function rather than choosing a representative value for the transmis sion typically evaluated at the peak maximum The differ ence between intensity values determined by these two Transmission Correction and Quantification ISO 14976 File Format approaches can be outside the precision estimates for the quantitative values A consistent approach to transmission correction therefore goes hand in hand with proper proce dures for monitoring the response of an instrument Provided a transmission function is included in the data file CasaXPS offers both methods for adjusting spectra with re spect to the transmission function The default method is to use a single point in the energy scale to evaluate the trans mission function This single value is then used to scale the RSF value before applying this value to the calculated peak area The main reason for offering this as the default mo
174. re amongst the list of windows then the New toolbar button must be pressed to create an empty sub frame Figure 90 shows the Convert to VAMAS file dialog win CHAPTER 13 dow where the Files of type filter allows only mpa files to appear in the listing Simply select a DS800 file and press Open or double click the desired file CasaXPS will convert the binary file into the ISO 14976 file format where the new file will have the same name as the original mpa file but with a vms extension added Lomee Bo VABAS ihe HEA Femme porearea oooO Fievolime Di0 Fn OOO O N a E Figure 90 Convert to VAMAS file showing DS800 files 123 VG Eclipse Files VG Eclipse Files VG Eclipse files are binary files one file per spectral region organised in directories where files with common acquisi tion characteristics appear in the same directory or inside sub directories within that directory CasaXPS will convert these directory structures into a single ISO 14976 file where experimental information and relationships are maintained Quantification in CasaXPS can be performed in an identical way to the Eclipse data system transmission correction is accounted for by CasaXPS whenever the information is in cluded in the original files Individual spectral regions are stored in files with a dts ex tension Sets of regions are stored in individual files all lo cated in the same sub directory Older versions of Eclipse
175. re each data region is la in Figure 2 To selecta VAMAS file press the toolbar button belled by the VAMAS block identifier string and these block indicated in Figure 3 to invoke the Open VAMAS file dialog labels are arrange according to the element transition strings window recorded in the VAMAS file Figure 4 shows the state of the CasaXPS windows after a spectral data file has been loaded Once the data file has been selected via the Open VAMAS Open YAMAS file 20 x Look jn ja ScientaFiles2 x Gl cl ScientaD ataset2_files WebPage El earlyox sci vms af earlyos1 sci vms Filename fearyoxl scivms 0000 Files of type Vamas Files vms x eae Figure 2 Open VAMAS file Dialog Window Figure 4 CasaXPS frame windows after a file has been loaded Figure 3 Toolbar The Open file toolbar button is circled Note the Multiple Document Interface MDI architecture When opened the VAMAS file appears as sub frame win dow which is managed by the CasaXPS main frame and it A Quick Tour of CasaXPS in turn displays the file as a logical set of spectral blocks These logical blocks labelled by name allow spectral regions to be selected and displayed in the left hand side of the sub frame in the Display window The current selection is that set of labels highlighted blue default colour scheme and when the file is first opened the top row of spectra regions will be highlighted and displayed in the left hand s
176. re generated such that 1 as much variance as possible is accounted for by each new factor and 2 the newest axis is mutually orthogonal to the set of axes already located The procedure therefore com putes an orthogonal basis set for the subspace spanned by the original data matrix that is oriented with respect to the data in a linear least square sense In principle the number of non zero eigenvalues is equal to the number of linearly independent vectors in the original data matrix This is true for well posed problems but even the presence of errors due to numerical operations will result in small eigenvalues that theoretically should be zero Nu merical errors are an insignificant problem compared to the one presented by the inclusion of experimental error in the calculation Noise in the data changes the underlying vectors so that almost every data matrix of c spectra with r acquisi tion channels where c lt r will span a c dimensional sub space This is true even though the underlying vectors should only span fewer than c dimensions Various statistics are available for identifying the most like ly dimensionality of a data matrix These statistics are de signed to aid partitioning the abstract factors into primary and secondary factors The primary factors are those corre sponding to the largest n eigenvalues and represent the set of abstract factors that span the true subspace for the data The Theory of Principal Component A
177. rea File List provides a numbered list of the file names for the open windows Options Launches one of six dialog windows for the major com mand sets and processing activities Page Layout deals with the way in which the graphi cal display area of the active tile is arranged in rows and columns Tile Display deals with the way that parameters e g energy scale intensity are displayed within a tile and sets attributes such as colour Quantify provides access to the Regions and Compo nents specifications enables definition of Report format standard or custom with defined names and tags and also provides a numerical Data Editor for spike removal Elements controls the setting and display of the basic line position database for the system accessing Element Table Periodic Table and Load File dialog 132 Toolbar Processing provides access to the major processing routines of the system Smoothing Integration Differ entiation and Principlal Component Analysis PCA as well as Test Data including spectrum synthesis Energy Calibration Intensity Calibration a spec trum Calculator e g for addition and subtraction of whole spectra and the Processing History for the selected tile Annotation provides Peak Labels and general Text annotation for the display control for Regions Compo nents and Quantification Table storage and presenta tion and access to the Annotation History for
178. rely or selectively The Reset button removes all the processing operations and restores the data to that originally supplied in the ISO 14976 file Alternatively by selecting a set of lines within the scrolled list then pressing the Apply Selection button the indicated set of processing operations are performed on the original data Selection of items follows the normal Microsoft conventions left mouse button for a single item left click then shift left click for an inclusive list or control click for separated items Thus the previous processing history is replaced by only those actions selected prior to pressing the button Processing for a block of data can be globally applied to a set of ISO blocks Select these blocks within the browser view then right click on the corresponding display view to reveal a dialog window for propagating operations such as curve fitting and processing Choose the required actions then press the Apply button A second window appears showing the progress of the propagation and offers a Stop button If the Stop button is pressed the propagation will terminate following the completion of the current action Note that some actions for example curve fitting may take a significant period of time to complete Not all actions for processing the data may be applicable to 36 Processing History all the selected blocks To prevent for example the energy calibration for the current block from bei
179. ric Voigt approximations The table headed Components Only shows the atomic concentrations where the intensities are calculated from the line shapes These concentrations do not agree with the elemental results calcu lated from the integration regions alone where the Al 2p 113 Tie p yan kan pe alz ae Ka wt a Ji Af ES Tj mm Figure 81 Tag mechanism applied to differing line shapes The Al 2p has been modeled using Doniach Sunjic profiles while the O 1s spectrum is fitted with Gaussian Lorentzian line shapes concentration is computed to be 85 38 while components only reports 88 24 for the same quantity The differences are due to the cutoff criterion used to limit the infinite area under a Doniach Sunjic profile Using the tag mechanism the peak fit can be used to subdivide the concentrations from the integration regions using the individual components from the peak models The concentrations for the synthetic components are now consistent with the results of the inte gration regions but are in the proportions determined by the Trend Analysis for Metal Oxidation Analytical Applications models Trend Analysis for Metal Oxidation Introduction Many XPS AES experiments are performed to monitor how the chemical composition of a sample varies with one or more variables These variables may be time tilt angle of a sample depth ion gun etch time or any other quantity that characterises the stat
180. rightmost point in the dispay subject to the constraint of plotting within the viewable area in the tile Front Plane Parameters are similar except that y 0 is al ways the front baseline no offset Colours for 3D back ground and fill are set in the Colours tab Scatter Parameters provide for rotation in degrees posi tive or negative about the three axes x y and z and shift percentage of diplay area in x and y constrained as above Display Display provides simple checkbox on off controls for items to be included in the display Traces are normally dis played as lines continuous curves unless Draw Points is checked when actual point by point values are potted with no joining lines useful e g for curve fitting overlay ing envelopes lines and points Typefaces for the Header Title and Axes labels are controlled here a fresh header may be added in the input box here and the line width in pixels of the display defined e g thicker lines for clearer 137 AV presentation or graphical capture and scaling Colours provides a means of setting colour for Spectra Spectrum Background Tile Background 3D Background Region Background 3D Fill Residual Trace and Synthetic Compo nents Clicking any of these items brings up a Standard Custom Colour swatch for selection and a further click on Add to Custom Colours extends this to a Colour Picker with HLS RGB and visual prese
181. s component is introduced to account for the oxide structure seen on the other spectra The position and width for the ox ide component is determined from the last spectrum in the data set Figure 86 and then fitted across all the spectra in the experiment Figure 86 Oxidised Aluminium data envelope Note that the ox ide line shape has been chosen to be the same as the metal even though the asymmetry of the metal should not exist in the oxide This has been done to so that the intensity measued by the oxide syn thetic line shape is in some sense comparible to the metal A trend for the oxide peak intensity can be identified but the true intensities involved are difficult to assess for the reasons given above Figure 87 shows a comparison between the re sults from modelling the spectra using two synthetic compo 119 nents and the principal component analysis The two PCA trends correspond to covariance about the origin with PCA BG and without PCA background subtraction Note how the trend PCA BG contains elements from both the PCA and the normalised peak fitted Normalised PF results Clearly the background approximation influences the nature of the trend computed using PCA and therefore similar influences must also be present when peak fitting is used to extract the trend So ee Normalised PF PCA PCABG Figure 87 Comparison of trends identified using Curve fitting and PCA The
182. s senisitivity factors excitation source and labels present in the system This table may list any single or composite transition appropriate for the analysis undertaken it is not limited to chemical elements and may be changed at any time so that several files may be held on disk but only one library of unlimited size may be active at any time Input File above enables loading changing or combining the line position data file which forms the basis for feature identification File selection of either CasaXPS or Kratos basic type may be direct typing ifn a file name or using a file browser and the selected file may be merged with or may replace the existing file if any The Element Table window provides a scrolled list of the contents of the Library File clicking on a name toggle transfers a labelled position marker to the display for the se lected item Position markers may be removed from the dis 141 play with the Clear All Elements button Find Peaks adds coloured markers to the display for all major peaks in the displayed block according to a defined algorithm see XREF xxx the Clear Markers button removes these iima Ta avs Takis Gur is The Periodic Table window provides a simpler means of selecting elemental transitions from the current element li bary if any for display labelling Each element symbol is a toggle click on click off button amare Takis Pac Thi gr Ti Processin
183. s 129 PHI MultiPak ASCII files 130 Command Summary eceeeeceees 131 Main inentbar 2 2 2 s9 4ss 6 ees Ree ewe 131 Pile yay anced By ne Le ee ea oe whoa RS 131 VLIW e g acted n es eee ca ny el tes heey Pes an 132 AV INO Wiel 252 Zante NS Eee afk te 132 Opuons 5g fe See E ETa 132 CU sedyo e pa T tab aha tat aoe i 133 Toolbars 3 024 reas hae Ds UE OSE ERS 133 Options Bat sr prespat te uated eite ETE oi 134 Processing Dialog windows 136 Pare Layout 24546 pha kad teami eteni 136 Tile Display Parameters 136 Quantification ey cea ae 138 PROCESSING oinen oid way eed Pees 140 Annotation o dibs dba eee awe habe ds 141 LADRary 2 45 veh es Be re Se eae 141 Appendix 1 ISO 14976 format files annotated 143 Appendix 2 ISO 14976 and the World Wide Web 148 Appendix 3 Definitions and Formulae 149 Glossary of terms 004 149 Formulae Peak shapes 149 Gaussian Lorentzian Product Form 149 Gaussian Lorentzian Sum Form 149 Contents Doniach Sunpier 2 5 5 2 ie tees Cowes eka bs 150 Appendix 4 References and other Resources 150 References from the Text Footnotes Appendix 5 Quick Reference Card REN 151 pE 151 Introduction Computer Aided Surface Analysis for X ray Photoelec tron Spectroscopy CasaXPS has been designed and writ ten for analysts and research scientists who
184. s in a survey scan Figure 80 shows a typical situation from an XPS measure ment where a survey spectrum provides the overall elemen tal quantification but a narrow C 1s scan offers a wealth of Quantification using Tagged Regions Analytical Applications arare savai saimi ma Figure 80 C 1s high resolution data used to proportion the quantification for C 1s in the survey spectrum information lost by the limited resolution used for the survey spectrum In the case of relatively pure samples such as the one in Figure 80 it might be possible to run a set of high resolution spectra for each of the peaks identified from the survey data but many samples include so many elements that full quantification via high resolution spectra would be both time consuming and costly Degradation of a sample from exposure to X rays and financial considerations can make the use of high resolution spectra unattractive as a rou tine analysis regime The data in Figure 80 are quantified via a table that shows the elemental concentrations plus a further breakdown of the 112 C 1s elemental composition into chemical state concentra tions The proportions determined from the peak model for the high resolution envelope are used to show how the ele mental concentration is subdivided into chemical state in tensities This is achieved by assigning a tag to each of the quantification items used in the analysis That is to say in a
185. s may be determined either from peak inten sities or from the inelastically scattered background Anoth er manufacturer uses a polynomial form for the logarithmic shape of the transmission as a function of retard ratio where the coefficients of the polynomial are determined using peak area measurements rather than background information A third manufacturer again uses peak area measurements to determine two analyser dependent constants a and b as fol lows Eqn 1 2 b TRIS ee aa TA le F R is the retard ratio NPL offers an empirical intensity calibration standardisa 3 Tanaka A J Surf Analysis 1 189 1995 Transmission Correction and Quantification ISO 14976 File Format tion procedure where the output takes the form of a rational function of two polynomials The background of inelastical ly scattered electrons measured from pure gold silver and copper provides a vehicle for comparing the response func tion of an instrument under test with that of a well character ised metrology spectrometer Given spectra acquired at the various operating modes of the instrument to be calibrat ed and standard spectra from the reference machine then relative transmission functions are readily constructed and consistent quantification becomes possible It is clear that there is no absolutely correct way of represent ing transmission characteristics for a given set of intensities although the importance of this i
186. s to a larger intensity than is calculated from the data region in question Truncating the function to the integration region destroys stoichiometric relationships between peaks or rela tive intensities for multiple peak models The underlying line shapes must extend far enough to reduce the influence of offset peaks within the same region but whatever choice is made for the extent of the Doniach Sunjic line shape the 22 Trend Analysis for Metal Oxidation Analytical Applications intensity will not be consistent with the direct integration over the data region If the data region is used to determine the limits for the theoretical line shape then the results for the intensity will vary with the acquisition region Alterna tively a more extensive but arbitrary limit will result in more repeatable results for intensity values however these intensities will not compare to quantification values derived from different synthetic models The Voigt type synthetic line shapes are more ad hoc in na ture than the Doniach Sunjic approximation but have the merit that they make no attempt to model the background re sulting from the primary peak The form for the background must derive from one of the standard approximations nor mally applied Linear Shirley Tougaard etc Modifying the Voigt function by an exponential tail revisits the problem associated with the meaning of the intensity when compared to the actual acquisition region i e
187. se models and empirical estimates computed from spectra acquired using reflected energy loss spectroscopy REELS Tougaard fol lows these calculations by fitting the results to a rational function which models the resonance structure in the calcu lated loss functions and allows the essential distribution to be described by a simple formula These loss functions are defined in terms of four parameters three of which describe the shape of the rational function and one to allow for vari ous band gaps in different materials Tougaard refers to the distribution below as a three parameter universal cross section and has established values for a number of materi als including aluminium silicon silicon dioxide and others 10 The form of this universal cross section is BT op 2 P T B C D 4 C T br 0 T lt To CasaXPS offers several ways of using the three parameter universal cross sections The background type defined on the Regions Property Page of the Quantification Parameters dialog window Figure 40 may be chosen from the follow ing list 66 Background Subtraction U Si Tougaard short form U Si Three parameter cross section determined for silicon U SiO2 Tougaard short form U SiO2 Three parameter cross section determined for silicon dioxide U AI Tougaard short form U Al Three parameter cross section determined for aluminium U Ge Tougaard short form U Ge Three parameter cross section determi
188. se re corded with the data This model can sometimes be invali dated by the presence of unrepresentative spikes that are due to the detector system and have nothing to do with the true electron yield It is best to remove such artefacts before at tempting to optimise the parameters for the synthetic com 49 Regions ponents A data editor is provided on the quantification dialog win dow The abscissa and ordinates for the data displayed in the active tile are listed on a property page labelled Data Edi tor The value for an ordinate can be changed By right clicking the mouse whilst the cursor is over the correspond ing abscissa a dialog window is brought up that allows the ordinate to be edited A check box allows the user to specify that an ordinate value should be permanently altered That is to say if the data is written back to disk then the adjusted value will be used in the file It should be noted that some actions in the process ing window can cause edits made to the data to be undone This occurs if the check box is not ticked and the processing history is used to change the state of the processing either by resetting or applying a selection of processing to the da ta The history mechanism always refers to the unprocessed data before taking the requested action Derivatives and Peak Envelopes A number of authors have proposed methods for identifying the underlying peaks responsible for a measured spectrum
189. sed in the scatter plot are determined from the intensities for the two peaks after automatically fitting the peak parameters Area FWHM and Position to the new data envelope these intensities are normalized with respect to the initial values for the peak areas It is apparent from Figure 70 that the un certainty in the individual peak intensities may be in error by as much as 100 although the total area is determined much more accurately Figure 71 The scatter in the computed in tensities in Figure 71 should be compared with that for the intensities shown in Figure 68 Monte Carlo Simulation Total Intensity Determined from the Sum of Two Peaks 9 2 A e gt s 2 o hd SN gon h oe oe 0 lt oo oe 7 oe of Cd 10 te ean Ne oe Ley Le PA sa Ms fes l A po O L SNS 20 ry ee be FE 8 OO AS HAV SS t a ete E SM wE e beg te entre Mt aes OY lt afer oe ae Fai hte es e Sens 4 oo 6 Oe 09 08 07 0 50 100 150 200 250 300 350 400 Figure 71 Total intensity linear background from one end point Intensities Determined by Peak Fitting Monte Carlo Methods Uncertainties in Intensity Calculations In both cases the background used for the Monte Carlo sim ulations shown in Figure 68 and Figure 71 are determined using one end point for the integration region The asymme try in the distribution derives from the errors
190. set and therefore a vector having a similar shape accounts Principal Component Analysis for most of the variation in the overall set of spectra A more even weighting between the underlying line shapes would produce abstract factors that are less physically meaningful in appearance 3 x10 CPS Sse8S8 SSS SLL SAAR Gnas a Lana A De 82 80 78 76 74 72 70 Binding Energy eV Figure 60 Al 2p Abstract Factors The only real use for the abstract factors is judging their sig nificance with respect to the original data Abstract vectors that derive from noise look like noise factors that contribute to the description of the data contain structure The dividing line between the primary and secondary abstract factors can sometimes be assessed based on the appearance of the ab stract factors Analysing the Al 2p spectra generates abstract factors and eigenvalues that represent the PCA fingerprint for the data Table 8 is a report of the Al2p data set generated by CasaXPS and formatted using a spreadsheet program Each row of the report is labelled by the ISO block name that con tains the abstract factor corresponding to the listed eigenval ue 91 Target Factor Analysis Table 8 Report generated by a PCA for Al 2p Profile Principal Component Analysis Factors Eigenvalue RMS RE RSD IE IND 1000 Chi sq Calc Chi sq Expected Al 2
191. sition of the maximum intensity for the line shape It is therefore difficult to relate optimization pa rameters determined from the Doniach Sunjic profile to sim ilar quantities determined from Voigt type line shapes F 1 a 2 cos 32 1 atan 5 Do wEB ee F x E Nevertheless the Doniach Sunjic profile offers an asymmet ric shape that is particularly appropriate for non monochro matic X ray induced transitions the profile is potentially present in both the photoemission process as well as the ex citation source Line Shapes Based upon Backgrounds The above profiles are assumed to be entirely due to intrinsic electron energy variations where a background subtraction algorithm is required before these line shapes can be used to model the spectra An alternative approach is to include the background shape as part of the model 8 An analytical Line Shapes Available in CasaXPS Line Shapes and Backgrounds form for the Shirley background can be determined for each of the line shapes and the sum of these backgrounds plus line shapes is used to approximate the variation in the spec trum A simple constant background is all that is required for this procedure although other forms for the background are still an option Castle et al 8 have developed a Shirley type adjustment to a Voigt line shape The Shirley approximation is calculated from the current gaussian lorentzian shape and a polynomial bo b
192. ssing information using flags The X ray an ode defaults to Aluminium and the energy scale defaults to Binding Energy Appending the following strings to the file Using Different File Formats name at conversion time can alter these two options anode Mg specifies Magnesium anode energy KE specifies scans with increasing KE sweep Coreei to VAMAS fle HE wan aan sl A ce ey ba cl 1 bax eo 121 fa cl 12a wig ajig m ga Fesas aceiro Mg Fiesole ia Fin i Corcel Figure 93 Convert to VAMAS file dialog window VGX 900 Note the unw extension used to indicate that the set of files con tain spectra acquired using Ron Unwin s data system The an ode Mg is only used to force the X ray source to be Magnesium instead of the default setting The latest release of Ron Unwin s system includes this information 126 Figure 94 The set of spectra contained in the sub directory shown in Figure 93 are offered in CasaXPS using the browser and spectrum display frames Dayta System Files Bristol IAC system Spectra acquired by the Bristol Dayta system are stored as sets of ACSH files with SP_ file extensions These files may be connected to a particular experiment through an associat ed seq file Sets of SP_ spectrum files are listed in the seq file together with the any experimental information e g etch times Dayta System Files Bristol IAC system Using Differen
193. t the PAA synthetic model can adjust without breaking the stoichiometric relationships for pure PAA while differences in the intensity of the saturated peak from the PAA structure are allowed for by this additional constrained component Figure 78 Synthetic models such as the one in Figure 78 can be tested using Monte Carlo simulation techniques to assess the sta bility of the peak parameters with respect to noise in the da ta Once a set of peaks and constraints has been developed the Monte Carlo procedure simulates repeated identical ex periments on the same sample and for each simulation a new fit is determined for the peak parameters The result of this procedure is a set of distributions for the individual pa rameters from which scatter plots may be constructed that 21 Cumpson P J and Seah M P Random Uncertainties in AES and XPS Surface and Interface Analysis 18 361 1992 Organic Polymers and Curve Fitting Analytical Applications Figure 78 Final form for the synthetic model The Glass Ionomer Cement GIC 2 C 1s envelope containing three peaks from PAA plus three additional peaks not seen in a clean PAA spectrum highlight the way noise influences the data model Figure 79 is a scatter plot for the normalized peak areas of the non PAA peaks against the saturated C 1s peak from the pure PAA model The peaks at 288 4eV and 285eV have areas that are anti correlated with the pure PAA model all t
194. t File Formats The first step to creating anew VAMAS file is to enable the Convert toolbar button The Convert option is disabled un less the selected sub frame window representing a file con tains no VAMAS regions If no empty Experiment Frame is amongst the list of windows then the New toolbar button must be pressed to create an empty Frame There are two methods for loading Bristol files e The simplest method is to convert files that are listed in a seq file All that is required is for the seq file and all the SP_ files listed within that file are located in the same sub directory then the seq file is selected through the Convert to VAMAS file dialog window If for any rea son the set of SP_ files does not match the set listed in the seq file then the seq will need to be edited to remove or add entries as necessary The VAMAS file that results will include an experimental variable for each spectrum loaded e The alternative method for loading the SP_ files is to collect a set of these files in a sub directory and then enter a new file name in the Convert to VAMAS file dialog window but add an extension of bri to the speci fied name Figure 95 All SP_ files in the directory will be read and appear in a single VAMAS file The default action is that no experimental variable it defined for each spectrum As a consequence spectra read will all appear in the same row in CasaXPS i e in a format suitable for
195. t has trapped the Marquardt method or on occasion serve to confirm the optimum has been found Marquardt method with constraints can help with the progress towards physically significant peak parameters However if the constraints prevent the synthetic peaks reaching an obvious optimum without an alternative availa ble then the algorithm tends to be slow That is if no natural optimum lies within the range of the constrained parameters and yet an optimum is visible outside the parameter range then the algorithm will labour The problem lies in the nu merous probes outside the range that will cause backtrack ing to the boundary values Understanding this fact is useful A well posed problem will tend to converge quickly while an inadequate model may result is sluggish behaviour The Simplex method on the other hand works based upon maintaining a set of function values at the vertices of an N dimensional simplex One of a set of prescribed transforma tions for the simplex is employed depending on what func tion value is founded at a probe point As it is prescribed there is very little that can go wrong with the Simplex algorithm However convergence to the opti mum set of N parameters is not guaranteed especially when a non smooth function is used i e when constraints are in troduced It does seem to work well for the situations typi cally found by CasaXPS although it has been pointed out that for some optimisation problems th
196. t is assumed throughout that the user is familiar with the normal IBM PC and Microsoft Windows operation and terminology The following section should be used as a hands on intro duction to CasaXPS to provide an overview of key features Subsequent sections enlarge briefly on the role of the system modules and the remaining chapters deal in depth with spe cific topics and applications Introduction alternative names in brackets C CasaXPst Program Frame Display Window HF a E 6 HH t amp amp amp a Block ee Bindisg Exsargy al q j seed Splitter bar isi o a E Carat PE Licenced tn 21 hilo Pii EFE HACLAH Comoe E ja Figure 1 A Windows 98 screen showing CasaXPS in operation quantify a survey spectrum in CasaXPS The intention is to A Quick Tour of CasaXPS explain how the interacting components of the system aid the analyst rapidly to produce a quantification table suitable This section describes one of the many possible ways to for inclusion in a customer s report A Quick Tour of CasaXPS 7 Introduction Step 1 Load the experiment file Step 2 Select a spectrum for processing CasaXPS is designed for ISO 14976 formatted XPS spectra A data file containing spectra stored in the ISO format may file dialog window the spectra held within the file will be be selected via the Open VAMAS file dialog window shown displayed ina logical array whe
197. ted intensity as well as FWHM for the indi vidual peaks These factors emphasize the importance of comparing like with like when quantification results are used in practice The choice of line shape is determined by the nature of the problem The best fit is not always as important as producing quantification results that can be compared to historical re sults On the other hand if the results represent a sequence of experiments for which a trend is more important than the ab solute values then a good model may take precedence over the need to supply numerical values consistent with the past Provided the models are self consistent the use of Doniach Sunjic line shapes is usually acceptable Analysis driven by precise quantification lies at the root of 6 Doniach S and Sunjic M J Phys 4C31 285 1970 7 Evans S Surf Interface Anal 17 85 1991 56 A List of Line Shapes many ad hoc procedures used in XPS Shirley backgrounds were introduced precisely 8 to remove as much asymmetry as possible from recorded data Figure 32 in a well pre scribed fashion so that near symmetric synthetic models can be used to characterize the intensity under a peak A Gaus sian Lorentzian line shape is finite and with the appropriate relative sensitivity factors RSF can be used to compare in tensities from fitted peaks to those calculated from integra tion regions The key factor that has popularized Shirley backgrounds is the availa
198. term is responsible for the movements in the DS maximum and for a 0 results in a pure Lorentzian form The formula for the F profile combines these characteristics by blend ing the asymptotic behaviour of the DS profile with one of the Voigt type functions using a linear mapping Him Pea ren 5h imi II Tl ee aT Os AHA TIT DOW TRS 4 EY see pete epee nee pee en eee neces eneee pace tee e peee aae E m E iT 1 FL Eiig Bang ii Figure 38 A new line shape that blends the asymptotic behav ior of a Doniach Sunjic profile with a Voigt type function F a m n GL p Asymptotic Doniach Sunjic Gaussian Lorentzian product linear mapping line shape F a m n SGL p Asymptotic Doniach Sunjic Gaussian Lorentzian sum linear mapping line shape 64 Background Subtraction The linear mapping between the two functional forms is de fined as a percentage through the m parameter Figure 38 shows the same Al 2p data envelope as is seen in Figure 37 but the line shapes are defined using the F functional form In Figure 38 the same asymmetry value has been used as the one in Figure 37 however the linear mapping param eter is set to 32 DS asymptotic form 68 Voigt type func tion The resulting profile is convoluted with a relatively DS Voigt Linear mapping f 948 950 952 954 956 958 960 962 Figure 39 F function plotted over a range of values for the linear mapping parameter a constant asymmetry param
199. the selected tile Help About launches a window giving the system version number license status and a dialog enabling change or update of license Help launches on screen html file based help Command Summary Toolbar Export Tab ASCII Export MetaFile Copy Display MetaFile Main File Access Bar g Copy Display bitmap Save File to Disk Convert File Open File to Clipboard New File Experiment Frame The Button bars File Access buttons 4 File Processing buttons Variable Calibration controls m f p Aly PE Display Options buttons Display Scaling buttons m e A Sila 2 oja T Display Properties buttons gt Block Block Ex hy bel eae Block comment info controls Display Modifier buttons 133 Command Summary Options Bar Launch Library window Launch Annotate window gt Display Options buttons Launch Processing window gleja see page 135 i i Insert many blocks into current scale File Processing buttons Insert one block into current display scale Launch Quantification window Launch Tile Display window L Launch Page Layout window Display all selected blocks in a tile L Display one block per tile The Toolbar upper buttons in general provide access to menus or dialog boxes The Options bar lower buttons in general execute actions immediately or reversibly Printing and Help E E P change the
200. thetic model for the data envelopes requires two fundamental line shapes The structure associated with the aluminium peaks is positive for the region where the oxide peak should occur and is negative in the region of the main metal line shapes The second abstract factor represents an adjustment to the 116 average shape manifested in the first abstract factor that is required to describe each spectrum in the data set to within experimental error The loading for each of the abstract fac tors demonstrates that the average form is almost constant across the data set but that the adjustment smoothly moves from a negative contribution for the second abstract factor to a positive loading as the oxidation proceeds An interpreta tion for this trend might be the consistent conversion of one form of Aluminium to another It AF s mA Figure 84 PCA loading for first two abstract factors The graph in Figure 84 shows a plot for the loading associ ated with each of the first two abstract factors The loading factor for the overall shape seen in the first abstract factor is virtually constant across the set of spectra All the variation within the data set is accounted for by the second abstract factor This is an interesting result in the context of the struc ture seen in the second abstract factor The initial negative Trend Analysis for Metal Oxidation Analytical Applications loading represents a removal of the ox
201. tion 24 Transmission Functions 23 K Kappa 59 L Line shapes 44 Asymmetric Blend 58 Doniach Sunjic 58 F profile 64 Gaussian Lorentzian 57 H form 63 line shapes including background 59 M Macintosh computer 20 Microsoft Foundation Class 5 Microsoft Windows 95 5 Multiple Document Interface 8 25 N Name Formula List 53 NPL 20 Index 0 offset spectra 27 P Page Layout dialog 15 peak labels 32 Peak Parameters optimisation 47 PHI MultiPak ASCII files 130 Phonon broadening 55 polymer database 109 Processing 17 Q Quantification 17 Calculation 45 Custom Report 46 Propagating 49 Report 45 Quantification Parameters dialog 11 13 quantification regions 31 R R S F 40 REELS 66 Regions 42 name 42 Regions Property Page 12 13 REGULAR scan 22 Relative Sensitivity Factors 22 REP real error in the predicted vector 84 RET real error in the target vector 84 RSF 112 RSF values 24 RUSTI 69 S Shirley 42 43 59 Shirley background 74 Simulating Spectra 69 splitter 25 bar 7 25 window 25 SSI M Probe Files 129 Starting CasaXPS 6 Synthetic Component 44 T Tables quantification 17 Tag field 113 test data 47 text annotation 33 Tile Display 27 Tiles 26 tiles preferred layouts 27 reducing scrolled list 27 rows amp columns 27 Tougaard 42 43 66 transition name 39 transmission correction 24 transmission encoding 146 trend analysis 114 typefaces 29 Index U WwW universal cross sectio
202. tion for each of the parameters from which an error matrix and tabulated pa rameter distributions can be extracted Error Matrix The result of a Monte Carlo simulation for a peak fit takes the form of m optimization parameters from each of the n simulated data sets An error matrix derived from these pa rameter distributions is defined to be an m x m matrix as fol lows 1 n ij n m sik T Xi X sjy Xi k 1 Where the x are parameter values calculated for each of the simulation steps and each distribution is centered with respect to the mean rather than the initial parameter value The standard error in each parameter is given by e and the P eij correlation between parameters i and j is given by 1 An alternative method for estimating uncertainties in the peak parameters is to quote the inverse of the Hessian matrix used in the Marquardt Levenberg optimization routine So the question is why bother with Monte Carlo simulation when this very information desired is offered by the optimi zation procedure itself The answer comes in two forms Firstly the error matrix derived from Monte Carlo is not limited to the problem of just fitting the peaks The simula tion procedure can also take account of the background as 98 well as the peaks and so determine a global error matrix for the calculation rather than one that focuses on the stability of the peak parameters alone Secondly the Hessian ma
203. to choose the number of tiles per page as 26 Tile Display well as how many rows and columns of such tiles should ap pear on the page These pages represent an initial format for the number of tiles per page however the user is at liberty to adjust these as seen fit There is a maximum of sixteen tiles per page The layout is specified first by selecting whether the tiles are to be dis played in rows or columns this permits the layout to be transposed at a click of a button and then by specifying the number of rows or columns and finally by indicating using the radio button array the number of tiles in each row or column see Page Layout on page 136 Sixteen prede fined layouts are provided corresponding to arranging spec tra in symmetrical rows up to the maximum 16 allowed All these predefined formats may be changed to provide a custom option library of preferred layouts as required Deke Heme He Figure 16 Main Toolbar Figure 16 shows the Main Toolbar The dialog boxes avail able from the main menu Options are also displayed using buttons on this Toolbar see e g Options Bar on page 134 Move the cursor over a toolbar button a hint de scribing the action associated with the button will pop up and a slightly longer description will be displayed in the main window status bar see page 132 Problem The first release of CasaXPS limits the number of pag es that can app
204. to quantification results that depend on the instrumental settings as well as the sample itself and so an intensity calibration procedure is required if consistent re sults are expected from the same instrument operating with in its range of appropriate settings The most common method for comparing peak intensities is to use tabulated Relative Sensitivity Factors RSF to adjust the calculated values for physical effects Without intensity calibration procedures based upon transmission characteris tics these RSF values are only appropriate for a specific op erating mode It is therefore essential to map out the transmission functions that accompany the lens modes and pass energies before consistent results can be obtained from a particular sample 22 Transmission Functions Transmission functions are dependent on instrumental set tings that can be adjusted by the user and other settings that change with age A useful calibration procedure should therefore be easy to perform and apply Moreover since transmission functions vary with time the transmission function for a particular spectrum should be included as part of the data file State of the art instruments typically record the measured intensities and extract some functional form for the trans mission characteristics both pieces of information are saved to disk The methods adopted vary for example a piece wise linear approximation is used by one manufacturer where the node
205. trix has a tendency to be near singular when the Marquardt algo rithm terminates Indeed the optimization routine follow a sequence of steps that switch between a direct solution of the Hessian matrix and a solution of a regularized form of the matrix The regularization step occurs when the direct Hes sian inversion fails to produce an improvement in the fitting parameters and a common reason for this to occur is that the Hessian is ill conditioned Under these circumstances the co variance matrix derived from the Hessian should not be trusted 1 a HJ Hur Fas FAHH Lh ered id T i Serle Fae fe Figure 63 Simulated PVC C 1s Data Envelope Monte Carlo A Simple Example Monte Carlo Methods Uncertainties in Intensity Calculations Monte Carlo A Simple Example The best way to introduce the ideas behind Monte Carlo methods and how the results can be interpreted is to look at a simple example Note that the Monte Carlo button may be found on the Components tab of the Quantification mod ule Consider the data envelope in Figure 63 The spectrum is a synthetic envelope created from two GL 50 line shapes without any background and where the peaks are separated by an energy gap consistent with C 1s lines in a PVC spec trum this is actually the PVC data available from the Test tab of the Processing module In the absence of noise and experimental error the optimization routine always returns the peak param
206. trum will be displayed showing the full set of data and any integration regions currently defined for the spec trum will be entered onto the Zoom List Now pressing the Zoom Out toolbar button also shown on Figure 10 will cycle CasaXPS Licensed to Neal Faitey eartpox eci vme ELS oe ee ee i Introduction the set of zoom states and therefore sequentially display the regions Step 7 Print the results Once the elemental composition has been defined using the integration regions the results may be added to the spectrum display This is achieved via the Annotation Dialog window Figure 11 where the Regions property page allows a table Pile E3 alaix xn i Tams Pos FWHM Axa Ob 532 263359 66310 1031 Cis 2S 2130195 26270 119 Al2p 73 1 70239 87700 74 43 s0 Ar 242 338805 21870 3 28 A Drag Box 5 Ready x TEI m ojja s ees e le Etch Time Scant AL VALENCE Ois ROSTIT 3 ROSIYT3 ROWSTYIS A EALS MTAA REENER i KUSTIS A PROSSTy 15 EISES GIAA ETATS ROBT TS DONDA wN Annotatnn Quaniiicaton Annotation Hino annot regons Deplay 0 annot text peak Data 95 01s annot text pesk Data 12 CIs annot tat peak Data 14 Al2p annot text peak Data 12 Figure 11 Annotating the spectra using the Quantification results of atomic concentrations to be positioned on the spectrum The entries in the quantification table are update
207. tside the acquisi tion region or energy shifts due to sample charging have al tered the characteristics of the data 1 0 432072 D 534209 DDFING2S 4 0 13439 D 197209 D0944976 6 D20977 DOBI619 DDI 33207 7 D 216023 D0097377 DD04167 ose 2 DIVEIA D 0046427 DIOH 3066 s aa gt D272019 DDI 21267 DD746693 RERE 1D 0 23083 0 297536 D93071 1I 0 22369 00331997 D0709123 N W 73 76 74 N D 1 D2J6I09 D06749 D D34009 Binding Energy eV 13 0 2383 DD 0747 D003I229 14 0 242269 DDI74342 DDIID33 D3 100734 D 104200 2 D24417 0 121233 DD727527 Figure 61 3 D factor space The PCA report in Table 3 includes the spectrum labelled Al 2p 48 in the data matrix The consequence of not remov ing the spikes is apparent in the 3 D factor space shown in Figure 61 where the abstract factor with third largest eigen value clearly contains spikes and the projection point number 10 derived from the Al 2p 48 spectrum is obviously a Statistical outlier PCA and CasaXPS Principal Component Analysis is offered on the process ing window The options on the property page labelled PCA allow spectra to be transformed into abstract factors according to a number of regimes These include covariance about the origin and correlation about the origin Each of these pre processing methods may be applied with and with out background subtraction Quantification regions must be defined for each spectrum included
208. uantification Mae he cathe 24 Data Display and Browser 260 25 Understanding the Data 25 Selecting the ISO 14976 Blocks 26 Zooming the Data 222s20s42 0y2e0 fet x 26 Tiles OF Sp ctra S sasG i e sAateaadaowes eee nes 26 Tile Display 222 4200 enten Beds ee ole eae 27 COloUTS ow tats ger ets eeiw wise E eis 28 FODS Geog k tee E aye Fae ene one 29 Display Parameters and Scrolled Tiles 29 Graph Annotation ccc cece ee ecees 31 Quantification Tables 0 31 Moving Annotation and the History Mechanism 32 Peal LAD e ls ois raa a US one Oe oe eek SL 32 Text Annotation 2 2 aeedivtle sn vet eet Rea 33 PROCCSSING sfeeeesseeicawediwae venta vawe es 34 SMOE sek hes oh ote Ae edd Ae ho die 34 Differentiatone x2 2 acest ten ee ethene 35 Ile Stallion 6 4 noae ou eee ees R E OERA 35 Energy Calibration 00 35 Intensity Calibration 000 36 Contents Processing History 2 0005 36 Element Library cccccccccccecees 38 Library File Structure 38 Loading an Element Library 40 Identifying Peaks 0040 41 Quantification esseoessessessoesoesee 42 REGIONS o s Ayan 82S E E RRS RR 42 Synthetic Components 44 Quantification Calculation 45 Quantification Report 0 45 Optimisation of th
209. ub frame If more than one spectrum appears on the first row of the Browser then the spectra will be displayed in the Display using a scrolled list of display tiles The left hand mouse button and the cursor are used to make a new Selection Point the mouse at a block label within the array of labels on the right hand side and left click Any blocks currently selected will be deselected and the block under the mouse will become the current selection Press the toolbar button indicated in Figure 5 to display the selection in the scrolled list on the left hand side of the sub frame Figure 5 Toolbar button to show spectra in the Display window left hand sub frame In this example the spectrum of interest is the last in the se quence of survey spectra recorded in the data file Figure 6 shows the state of the CasaXPS windows after a specific sur vey spectrum has been selected and displayed Also note in Figure 6 that the user has maximized the Experiment Frame within the main frame of CasaXPS hiding others below A Quick Tour of CasaXPS Introduction Ji oe pe ere Gee fe a aE oE Saia iedolainlolalad aif cieted cil lt i Figure 6 CasaXPS ready to identify spectral features Step 3 Identify peaks CasaXPS will load an element library called CasaXPS lib located in the same directory as CasaXPS exe The element library is used to identify the peaks seen in the data and is central to rapidly creating quantificatio
210. unity to investi gate algorithms used to probe real spectra For example nu merous techniques have been proposed from Multivariate Statistics for reducing XPS spectra to meaningful chemical information 3 I4 Procedures for generating simulated spectra allow data sets to be constructed with similar charac teristics to the experimental data but with well defined in formation These algorithms can be applied to the simulated spectra in order to establish what information is extracted for 13 Fiedor J N Proctor A Houalla M and Hercules D M Surf Interface Anal 20 1 1993 14 Do T McIntyre N S Harshman R A Lundy M E and Splinter S J Surf Interface Anal 27 618 1999 Line Shapes and Backgrounds a perfectly understood data set Everyday procedures such as curve fitting can be evaluated using simulated spectra and therefore the ability to produce spectra with known shapes can enhance understanding for students and researchers alike neta mhk Lin for brr Pi jprm cE Leal k on Ho Et a Binding Ere Figure 44 Initial synthetic components used to construct a PAA spectrum green The background curve brown represents the first iteration of the cross section with the model data 70 Using the Calculator and Comparing Spectra CasaXPS provides a spectrum Calculator which enables the normal arithmetic operations adddition subtraction multiplication an d division to take place between spe
211. w kratos com http www phi com http vacuumgenerators com http www vauum org http dmxwww epfl ch ecasia http www surrey ac uk MME QSA http www chem qmw ac uk surfaces http www iop org http www rsc org http www acs org http www aip org http www vacuum org iuvsta default asp http www mrs org 150 References from the Text Footnotes 17 18 Tanaka A J Surf Analysis 1 189 1995 M P Seah and M T Brown J Elec Spec 95 1998 71 93 Shirley D A Phys Rev 55 4709 1972 Doniach S and Sunjic M J Phys 4C31 285 1970 Evans S Surf Interface Anal 17 85 1991 Castle J E et al J Electr Spectr Related Phenom 106 65 2000 G Wertheim J Electron Spectrosc 6 239 1975 Tougaard S Surf Interface Anal 25 137 1997 Tougaard S Surf Interface Anal 11 453 1988 Jo M Surface Science 320 191 1994 Fiedor J N Proctor A Houalla M and Hercules D M Surf Interface Anal 20 1 1993 Do T McIntyre N S Harshman R A Lundy M E and Splinter S J Surf Interface Anal 27 618 1999 to be added K Harrison and L B Hazell Surf Interface Anal 18 368 1992 P J Cumpson and M P Seah Random Uncertainties in AES and XPS Peak Energies Areas and Quantification NPL Report DMM A 26 May 1991 Surf Interface Anal 18 345 1992 and 18 361 1992 19 20 21 2d 23 S Evans Surf Interface Anal 18 323 1992 Beamson G
212. ween the background and the data an estimate for the F W H M and the position of the maximum count rate re corded are all list below the parameters that define the re gion In addition a percentage concentration is shown at the bottom of each column The raw intensity is calculated di rectly from the data however the percentage concentration includes R S F adjustments The results of quantification based upon a wide scan and a set of regions are therefore available through the Regions property page Synthetic Components The property page labelled Components manages the cre ation and optimisation of synthetic components Synthetic components are specified by name line shape R S F position F W H M and area CPS eV The name is the means of identifying a component A quan tification report specified using expressions makes use of the name to define how the intensities are to be combined Further components with the same name are summed to gether before an expression is evaluated Quantification 76 74 72 70 Binding Energy eV 82 80 78 Figure 25 Region plus Synthetic Components Line shapes and R S F values are stored in the element li brary and these values are used when a new component is created A new component is created using the buttons above the list on the Components property page If the el ement library dialog is active and a transition is selected from the Element Table property page t
213. wing a PCA can didates for the physically meaningful components may be assessed individually or collectively Choose an abstract factor from the PCA and entering this factor into the active tile Then select the number of primary abstract factors for use in the target test procedure A text field is offered on the PCA property page for this purpose and is found in the sec tion headed Target FA Next select the target test spectra in the Browser view and press the button labelled TFA Ap ply A report detailing the statistics calculated from the TFA procedure will appear in a dialog window The TFA report may be written to file in an ASCII format with TAB separated columns When pressed any of the but tons above the columns on the report will display a file dia log window from which the output text file can be specified This method for saving a report to file is used by the PCA report above and the Linear Regression Report described 95 Target Factor Analysis below Once a set of target spectra has been identified these spectra can be used to reproduce the original set of spectra through a linear regression step Enter the set of target spectra into the active tile then select the original spectra in the Browser view Press the button labelled Linear Regression A re port shows the RMS differences between each of the origi nal spectra and the predicted spectra calculated from a linear combination of the set of target spect
214. wn in Figure 21 DSHaE Ome y a Figure 21 Main toolbar processing options CHAPTER 5 Smoothing Data smoothing may be performed using a range of Savitz ky Golay algorithms or by weighted averaging the data us ing a set of weights that are a normalised Gaussian distribution Smoothing algorithms should always come with a statistical health warning Smoothing can seriously affect your curve fitting The act of fitting synthetic peaks is not only the best form of smoothing but also makes assumptions about the distribution of errors within the data as well as the statistical independence of the measurements Smoothing the data be fore curve fitting invalidates these assumptions A Savitzky Golay filter is derived by approximating the data with a polynomial of degree less than or equal to the chosen number of 2n 1 channels The coefficients for the selected polynomial are determined in the linear least square sense and the datum at the centre of the 2n 1 channels is replaced by the value for the polynomial at that point To perform a smoothing operation using Savitzky Golay 34 Differentiation methods it is necessary to select the degree of the polynomi al and the width of the data channels over which the approx imation is to be made Gaussian smoothing merely needs the width of the data channels to be used when averaging the spectrum Differentiation Differentiation of spectra is also achieved using the Savitz
215. x E is used to scale the background in order to pro vide a fit to the observed spectra The procedure yields a Kappa parameter given by bo that characterizes the in trinsic step in the spectrum observed for a particular sam ple Line Shapes Available in CasaXPS Curve fitting in CasaXPS is performed via the Quantifica tion Parameters dialog window where the synthetic line shapes are defined from the scrolled list on the Components 59 property page Figure 33 Regon Conporents Cats iida Ppor Sees Figure 33 Property Page used to define synthetic line shapes The line shape used to describe a photoelectric transition is entered in the row labelled Line Shape and takes the form of a text string GL p Gaussian Lorentzian product formula where the mixing is determined by m p 100 GL 100 is a pure Lorentzian while GL O is pure Gaussian Line Shapes Available in CasaXPS Line Shapes and Backgrounds SGL p Gaussian Lorentzian sum formula where the mix ing is determined by m p 100 GL 100 is a pure Lorentz ian while GL O is pure Gaussian GL p T k Gaussian Lorentzian product formula modified by the exponential blend SGL p T k Gaussian Lorentzian sum formula modified by the exponential blend GL p K b0 b1 Gaussian Lorentzian product formula modified by a Shirley type background prescribed by Castle etal A linear polynomial determined from b0 and b1 adjusts the step in the
216. xist 138 ing value is left justified Changing an entry requires a re turn enter key at the end of the entry to validate it merely clicking on a new parameter restores the old value Note also that correct spelling is required for background type linear Shirley Tougaard though case is insignificant Create brings up a new region with as far as possible the parameters entered for the displayed peak Create From Labels provides a region whose parameters are taken from the library entry with the same label as the displayed peak see Step 5 Create quantification regions on page 11 Delete deletes the selected region and its parameters from the Quantification window and thus from calculations and display Calculate Error Bars completes the standard deviation calculation and related procedures Save Region As VAMAS File records the region parame ters in the block header see page 145 Intensity Calibration provides either automatic incorpora tion of the transmission function data held in the Kratos style VAMAS file see also page 145 or manual entry of a specific factor transmision exponent in the type in box If CasaXPS detects the presence of transmission function cor rection values in the data block an appropriate correspond ing variable then the automatic checkbox will be ticked and the quantification will proceed using the correction Processing Dialog windows Command Summary
217. xperiment Frame Ex periment Frames are created using the New option on the file menu Once an empty Frame is selected the Open op tion from the same menu offers a File Dialog window for choosing the ISO file The Browser view right hand pane of splitter window of fers the data as described above Selecting the data is achieved via the mouse in combination with the Shift and Ctrl keys Positioning the cursor over a name in the brows er view then clicking the left hand mouse button selects that block If a range of contiguous blocks is required as seen in the browser view then select the first block in the range followed by holding down the Shift key and selecting the last item for the range All blocks between the first block and the last will be selected The range selection works across columns as well as within rows or columns That is to say rectangular sets of block can be selected by this mechanism Additions to the current selection require the use of the Ctrl key If the Ctrl key is held down and a new name is selected the existing selection is retained in addition to the block just chosen If the block indicated with the mouse is already selected then it is removed from the current selec tion The Shift key and the Ctrl key when used together allow rectangular sets of blocks to be added or removed Data Display and Browser from the current selection Selecting a bloc
218. xperiment items 14 future block entries 15 blocks 16 D OO 0O 0O blockid 17 Clis1 sample i d 18 PET year 4 digits 19 1999 month 20 06 day 21 24 hour 22 14 minute 23 45 second 24 00 hours GMT 25 1 lines in block comment 26 0 technique 27 XPS source 28 Al source energy 29 1486 6 source strength 30 300 143 source width x mu source width y mu source polar angle source azimuth analyser mode analyser resolution characteristic magnification of analyser analyser work function target sample bias analysis width x mu analysis width y mu analyser axis polar angle analyser azimuth species transition state charge of analysed particle abscissa label abscissa units abscissa start abscissa increment corresponding variables corresponding variable label corresponding variable units d none signal mode signal collection time channel scans for this block signal time correction sample normal tilt sample normal azimuth sample rotation angle additional params ordinate values min y max y first data point 399 data points last data point terminator 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 465 466 500 500 54 180 FAT 20 1 4 5 0 200 200 0
219. y File name fsi Survey qua Files of type All Files Cancel Figure 98 MultiPak ASCII files are converted by entering a filena me with an extension of qua 130 Li f yes pie inden Hairs s are da bore Poa Ei Ger Fini eer 1am Pi Lea ri ee Jibi 4ieme Eg e a Command Summary The CasaXPS system has three levels of control the fa miliar Windows style Menu bar a system specific Tool bar and Options bar providing commands or command windows accessed by clicking button icons and keyboard shortcuts This chapter provides brief one line descrip tions indicating where a command is how it is accessed what the command does and the part of the system or data to which it applies Main menubar The main system window Menu bar contains five menus File View Window Options and Help File New opens a fresh blank Experiment Frame inside the CasaXPS main window Programme Frame Open brings up a file dialog window so that a cho sen existing file is opened and displayed in the current CHAPTER 14 top tile Convert greyed out disabled unless the current Expeiment Frame is empty brings up the Convert file browser see Using Different File Formats on page 123 Close dismisses closes the current Experiment Frame providing the file it represents is saved Save As enables

Casa XPS User's Manual - The Molecular Materials Research Center

Contents

Download Pdf Manuals

Related Search

Related Contents