Home

BRUKER AXS PACK - X-ray Diffraction Texas A & M University

1. 10 Can you imagine More than one series can be integrated in one single run 59 if necessary confirm the suggested values and after that click on Execute The other two options Change default parameters and Copy local parameters to default parameters are proscribed for beginners Open a New Project Project name Crystal Type Protein Small Molecule Use the file locator to help identity frames for SAINT to process Click on a p4p or frome filename wath the base name of the series of frames you wish to process J o alo Abn Filtype p4p files Annuller Erames path Quipudirectory Ework Fig 8 1 SAINT Open a New Project dialogue box At first sight Basic SAINT menu for analyzing Fig 8 2 seems really benign However stay very very cool if you open any of More options Sort Global or Filter at top right because everything could be chaotic there As can be seen in some of the following pictures expected names of your files are listed together with names of some other files usually but not necessarily with Ylide as the standard crystal in a happy mixture It is really incredible that some dialogue boxes containing file names are not updated and saved until you finish or close the job have we already told you something about a ruffle between the left and right hand Fortunately since Sort Global a
2. 14 An abc file is a copy of a raw file containing among other data corrected intensities and their s u s it is prepared by XPREP during a face indexed absorption correction T7 previous lengthy path by pressing the up arrow repeatedly and then editing the rightmost part to specify the intended file It is much better to start SADABS from the SHELXTL menu The third very simple way used in our laboratory to run SADABS is copy the program into the actual working directory directory containing your data and run SADABS exe During the SADABS run carefully read the screen outputs and follow the question and answer format The SADABS questions are marked by the usual sign and default answers are given in brackets Only parts of some long questions are given in the following text Step 1 Refinement of parameters to model systematic errors This step involves data input and modeling of absorption and other systematic errors e Maximum number of reflections allowed 500000 The number of reflections is not limited but the default number is more than enough for ordinary structures e Enter listing filename sad abs When prompted give a name for the log file that easily identifies your crystal If you omit the extension abs is added automatically The program then shows a list of Laue groups I Select the correct Laue group from the list e gt Test Friedel opposites Y The usual answer to this question is Y If
3. An Extract from SAINT Output Files Then decide what to do Obviously there are only two options repeat the integration or continue In the second case double click the SHELXTL icon and read the chapter SHELXTL programs Perhaps it would be better to read the text first The operations Sort Global and Filter see More options field at the top right corner in Fig 8 2 described below can be run independently of each other and of the integration procedure The profile fitting should work better with box size parameter 6x6x6 instead of 4x4x4 George Sheldrick private communication However we have not discovered how to set the box size parameter 65 sort and i E Input files Sort c rames CUST YUDO raw Browse Tite Bort Mid Cust Crystal C rames CUST YLIDI raw Boe Output file PADejanikirk ab work kirkabrn rew Browse Listing file Browse Point group fmmm orthorhombic Browse Statistics Browse Generate statistics Browse Orientation matrix file C fremes CUST YLIDI p4p Browse eA Lattice centering P primitive z Browse Generate diagnostic plot files Sor Open listing file Help Close Fig 8 5 SAINT Sort and merge dialogue box Sort and merge menu This program sorts one or more input reflection files ordering reflections on the HKL indices according to the user specified point group The most common use of SORT is to merge reflection fi
4. Parallel graphite 8 1 36 3 10 264 520 248 730 512 5 0 M Check for yes x Fig 2 4 EDIT gt Config uration dialogue box ee If the crystal is not discarded after the previous operation the next step is acquisition of o s an orientation matrix Before that invoke EDIT gt Config and check the diffractometer configuration Fig 2 4 Input proper Source kilovolts and Source milliamps values Check with your system administrator that you have the correct Sample detector distance and the correct Direct beam X and Direct beam Y center Check also if the proper FloodFld flood field correction and Spatial corrections are used For our detector at the bottom right in the SMART window both fields should contain the same code h423117 extensions are not visible Be sure to have a Dark field correction of appropriate time the same as you use for the exposures in Matrix See description of _ dk files in the beginning of this Guide If necessary load one from the c frames ccd_1k FILE gt Load Dark which has the most recent date After that invoke ACQUIRE gt Matrix or the equivalent CRYSTAL gt Unit cell These commands automatically collect three short series of scans at different starting angles clear Reflection Array invoke Threshold and store some collected spots into the Array perform Autoindexing procedures perform least squares on the reduced primitive c
5. 03044 02428 09017 05996 20 000 COCO OOCCOCfO 0 0 01106 01322 10567 10392 05717 08367 07984 05061 14043 14277 X3 Freq 00603 06229 06979 09673 04401 0 100 20 000 FOM H 1 H 2 H 3 Reflections fit tween solutions Retrying with next shortest vector max index truncated FOM H 1 H 3 H 2 Reflections fit Axis limi Length Angles be 1 17 20 Indexing CS 8 0 19 1 tween solutions 2 0 100 20 000 0 100 20 000 GONNWORNND 3 2 19 I 0 100 20 000 0 100 20 000 2 1111100111101001110111111110111101101101111011 Fewer than 3 solutions found Re trying Axis limits max index failed on difference vectors with first 100 full vectors 0 100 20 000 0 100 20 000 0 100 20 000 20 5 5 Length FOM H 1 H 2 H 3 Reflections fit Angles between solutions Solution number axis length figure of merit lower value is better direction of axes solution code 1 for each vector group that fits and O for vector groups that do not fit So be happy if you see a lot of 1s 1 10 11 23 24 9 275 0 17 1 0 O 11111111101111111101111111111111111111111000010000 1111111111110101110011111 6 487 0 09 0 0 1 12111111111111111111111111111111111111111011110111 1111111111111111101001111 106 4 9 705 0 17 1 0 1 211111111011111111011111111111
6. In the list of most disagreeable reflections F is much bigger than F for most cases such reflections often reveal a systematic trend e g in nearly all cases h 3n It must be emphasized that some warning signs listed have quite different origins and the reader cannot expect to see all or most of them together On the other hand many twins reveal more than one warning sign If you came here the crystal you are dealing with is very probably twinned After all maybe you should be interested in reading the above cited material Otherwise call for help from an experienced crystallographer or send your data to a twin skilled person Just in case that this one is not the first twinned sample you ever met we should mention here a specific complication which appears when attempting a twin structure refinement If you were using GEMINI for resolving non merohedral twins or other intergrowths you were offered by the program a very convenient mode of preparing the hkl file in the HKL 5 format This format is accepted by SHELXL as a means of including the whole pattern of overlapping reflections The problem is that you cannot apply an absorption correction in this case unless you do a face indexed correction Both XPREP and SADABS use raw files to produce hkl files GEMINI too This means you cannot prepare an absorption corrected raw file for GEMINI and thus you cannot get an absorption corrected hkl file out of it A special version
7. data collection performed by ACQUIRE gt MultiRun gt Hemisphere or gt Quadrant This directory could be on a networked off line computer while the working directory is always on the computer that is directly connected to the instrument Of course both directories can be located on the same computer After some unpleasant experiences with network breakdowns our current practice is measure everything on the directly connected computer transfer data afterwards to where you want to work on them The working and data directories also include smart ini or smartdef ini and specific calibration files used for a given project As usual file names consist of two parts jobname and extension that is name ext By default they follow the well known and very old DOS 8 3 convention but other setups are also possible in EDIT gt Config dialogue box Fig 2 3 see SMART manual p 6 3 The jobname is freely chosen by the operator to describe the crystal or project The combination of crystal name and crystal number forms Project name which must be unique It is stored in the ASCII project database file administrator prj located in directory c frames Projects or d frames Projects Note During the installation of SAINT this file may be named other than adminstrator prj consult system administrator if necessary During data collection the Bruker system will add some numbers to the file name field For instance the names of output
8. prepared by program SMART during data collection bg_snap 000 bg_snap 001 etc Files written by SAINT containing a snapshot of the background calculated during the integration cif CIF Crystallographic Information File prepared by the SHELXTL suite after the final stage of structure refinement see also pcf After editing this file is suitable for submitting a paper to Acta Crystallographica and many other journals or for deposition in CCDC ICSD and similar databases edx edy edz Files containing statistical results from program SAINT if Generate Diagnostic Plot Files is checked These files contain intensity deviation scatter plots showing intensity deviations vs position and can be viewed using PLOTSO SMART SADIE or FRAMBO EXX CYY CZZ EXZ eYZ exi Files containing statistical results from program SAINT if Generate Diagnostic Plot Files is checked These files contain positional deviation scatter plots showing positional errors vs position and can be viewed using PLOTSO SMART SADIE or FRAMBO fcf File containing structure factors in CIF format ini Configuration file where initial parameters as well as instrument and sample data are stored e g smart ini smartdef ini saint ini saintCL ini ins Command input file for SHELX programs hkl If you don t know about this file you probably are not a crystallographer Leave our session immediately Ist File produced by SHELX progr
9. u bigger than ca 15 mm the analytical corrections do not always give a d p s that are the best in the above sense Anyway if a part of the isotropic displacement parameters is small e g B less than 1 A U less than 0 013 A and if some of the a d p s later become non positive definite you have a problem with absorption correction The presence of some very elongated thermal ellipsoids could also be a reliable indicator of unsuitable absorption correction However thermal ellipsoids are sometimes elongated for good chemical reasons the list of examples and possible explanations is long and will not be given here In addition elongation could be caused by static and or dynamic disorder twinning etc Face indexed correction Technically the procedure for performing face indexed absorption correction using XPREP option A then F is very simple especially if description of crystal shape is already present in the p4p file Accept the default value for precision only However to use face indexed corrections one should measure the crystal shape very carefully Every serious 86 crystallographer can tell you that this pays off Your Rin might sometimes be worse after the application of this correction compared to the other types but your data will always be more realistic also refined a d p s and R factors should be better and the residuals in the difference map smaller when you do a face indexed absorption
10. 15 20 229 30 539 40 45 Inf H 81 0 0 0 0 0 0 0 0 0 K 80 1 0 0 0 0 0 0 0 0 L 81 0 0 0 0 0 0 0 0 0 Omega 31 16 15 6 3 4 3 al 1 ak O00 O 25 O50 Ost 1400 255 2550 eto 2500 2025 Int X 38 25 14 4 0 0 0 0 0 0 Y 28 25 9 11 5 3 0 0 0 0 H K L Omega X Y AVG 0 001 0 001 0 002 0 000 0 000 0 000 RMS 0 006 0 019 0 012 0 165 0 384 0 602 Mean differences and their s u s If AVG are bigger than RMS you may suspect misindexing or misalignment of the system A list of the 10 worst reflections follows Worst Agreements worst RLV error H K L dRLV GOMEGA 10 6 01 4 95 1 98 0 0046 0 203 28 3 01 2 99 1 04 0 0044 0 439 18 3 01 6 97 0 02 0 0042 0 331 65 1 00 1 05 5 00 0 0039 0 004 20 3 01 4 98 0 02 0 0038 0 364 17 4 99 3 04 4 01 0 0038 0 042 80 1 00 2 05 4 99 0 0037 0 016 136 2 01 12 02 1 03 0 0035 0 233 135 2 301 11 01 2 03 0 0035 0 241 61 1 00 2 96 6 00 0 0030 0 054 Detector corrections X Cen Y Cen Dist Pitch Roll Yaw additive 0 284 0 563 0 051 0 048 0 486 0 031 Detector corrections Pay attention if the first three values are greater than 1 0 for X and Y Cen and greater than 0 2 for Dist perhaps the crystal was not as good as you thought or maybe the diffractometer requires adjustment Contact administrator Eulerian angles 126 620 92 855 17 429 Omega Chi zeros 0 9928 0 3876 Crystal XYZ 0 0000 0 0000 0 0000 Pay attention to the Omega Chi zeros values that should b
11. 4 001 38 00 50 00 180 00 220 00 2 250 200 10 00 5 001 38 00 50 00 270 00 220 00 2 250 200 10 00 and represents a suitable starting point for strategy definition at 4 0 cm detector distance A template file rEdit txt is also prepared in order to avoid extensive typing each time you try to define a new strategy From some reason that only the Bruker guys can understand remember the saying The left hand doesn t know what the right hand is doing MultiRuns in SMART and in ASTRO have different formats and they are not interchangeable although according to the manual they should be MultiRuns in ASTRO has one column more called deltaD and no one knows what it means Thus the scan file for ASTRO corresponding to the file r40_25b txt looks as follows 0 001 38 00 38 00 00 27 00 3 250 720 10 00 1 0 1 001 38 00 38 00 180 00 27 00 3 250 720 10 00 1 0 2 001 38 00 50 00 00 220 00 2 250 200 10 00 1 0 3 001 38 00 50 00 90 00 220 00 2 250 200 10 00 1 0 4 001 38 00 50 00 180 00 220 00 2 250 200 10 00 1 0 5 001 38 00 50 00 270 00 220 00 2 250 200 10 00 1 0 Therefore for every MultiRun combination it was necessary to prepare two text files one for SMART and one for ASTRO ASTRO files are coded in the same manner as before the only difference is an additional capital letter A for ASTRO at the end Thus a file r40_25b txt for SMART becomes r40_25bA txt for ASTRO It should be easy to prepare a similar set of files for
12. 7 Strategy planning tools eee ce ceeee cee eeeeeeeeeeeeneeeeceeeeeseaeeeeeaeeeeesneeereeees 48 AS MRO EA E EA E ed Cea eens ea at E anette 48 COSMO ass seeded AE A da Alpe aa albeit opti haan unde aula ties 53 8 Data reduction strategy cccccccceceeeeceeeeeeeeeceeeeecaeeeeceeeeessaeseseaesaeeaesaeeereaaees 56 9 Crystal Shape iss sashes cee ssececcs easter erent hasbeen A A EE EEE a 68 10 SHELXTL programs idece sa ttial tata id ana eek OA 72 11 Absorption and other corrections by SADABS 0 0 eccceceeeeeeeeeeeeeeee 76 12 Notes on absorption correction 2 2 cece cece eee teee cee ceeeeaeeseeeeeeseeeaeeneeeaes 83 13 Notes on twins and tWiNning eee eee eeeceeeeeeneeeeceeeeeeeeeeeeaeseeeaeeaeeeeneeeneees 89 PDD OIG ID tec aos E E Gi HE ae ee gid 91 Some useful ShOrtuts ce ecccecscceceeeeeeeeeeeaeeececeeeseseeeaeeeceeeeeessaecnaeeeeeeeesensensanes 92 Crystal identity Card es sis2e ce cseteie cleadtegelacveuncuanidhincwnss iaaa aiie kai dea FaK aana aea 93 Key for understanding SAINT output files 2 0 0 cece eee eete eter teense ee eeeeaeeeeeenaeeeeee 96 Examples of MUultiRUN SCANS cccccceccesecceceeeeeseeeesaceeceeeeesesecceaaeeeceeeeeseesnsiaeeseees 98 Four circle diffractoMmeter c cccccccececeeeesecceeeeeeeeeseceenaeeeeeeeeesesencecaeeeeeeeenenes 98 Three circle platform diffractometer ccccceccessccececeeeeeseseenneeeeeeeeeeneennnes 100 K benhavns hus
13. A Circular cursor will appear if you invoke ANALYZE gt Cursor gt Circle or use the equivalent F7 shortcut Pressing ESC or clicking the right mouse button toggles between position change mode and radius change mode and allows adjustment of the circle radius see above Press the left mouse button or Enter key when finished The right side output is similar to that for the box cursor In addition radius of circle in 2 theta and in ngstr m as well as changes required to the phi axis position and to the goniometer head arc in order to move the diffracted layer circle to the beam center are shown If you are using a goniometer head with arcs this will help you to orient a desired and marked zero layer to coincide with the primary beam A Vector cursor appears if you invoke ANALYZE gt Cursor gt Vector or its F8 shortcut The primary use of this cursor is to measure distances on the displayed frame The cursor is of the rubber band type As before pressing Esc or clicking the right mouse button toggles between position change mode and size change mode In the latter case the vector origin stays fixed while its end point moves as the mouse is dragged The output to the right contains counts of pixels at vector origin and endpoint X and Y coordinates of the origin endpoint and midpoint length of vector in pixels degrees 2 theta and Angstr m If you wish to see a one dimensional slice or profile through pixels of the current frame us
14. FOD poe rasini tainai Optical goniometer control NER enea R Redetermine cell CUE e AEEA ERRE TTET Save CAF Wise ee ers Set screen contrast Ctl oak NA Index cell Cursor shortcuts PO nee ART de tee ead Mowe chee os Crosshair cursor FO iis a EAA ART Box cursor Feri A E Circle cursor FO SENEE A A EO E E Vector cursor a A A T A AN Set goniometer at zero angles Several additional SMART specific shortcuts are also defined For a full list of shortcuts see nowhere in the manual Many other shortcuts exist Practically all of them are identical to well known Windows shortcuts e g CtrltBreak interrupts an operation in progress TAB moves to the next field in a dialogue box Ctrl C copies highlighted text to the clipboard and so on See pages 1 8 and 1 9 in the SMART manual for a full list of Windows type shortcuts 93 CRYSTAL IDENTITY CARD COMPOUND mineral systematic IUPAC name Project name Date started Working directory CD code Data directory CRYSTAL DATA final values 7 Chemical formula M Crystal system Space group No h a A a t b A Pes bc A yes To Pia Asymm unit content Unit cell content h D g om D g em F000 u mm CRYSTAL DESCRIPTION Source Color and Shape Size points Size mm Mean size mm DATA COLLECTION Step one identification and initial unit cell determination SMART Name s of rotation picture s if any Matrix operation as 3 x of
15. SAINT adapt the orientation matrix during the integration The matrix should be determined as well as possible in advance see below and kept fixed together with the box size to insure the best possible intensities This means that the present stage LS in SMART is crucial for your lattice parameters and you should invest enough work in collecting good reflections for it and dismissing obviously bad ones In case you don t want to mess with suspect cases but use only decent crystals then your orientation matrix from SMART would probably be a good start for SAINT which is then supposed to update it during integration and make a final least squares parameter refinement As far as we have been able to discover there are three main approaches to this case We cite them all and let you decide according to your temperament etc 1 For the hasty type Use just your orientation matrix from the initial matrix collection but first clean out bad reflections those with large RLV errors and use for all runs in SAINT After the first integration you will most probably see in the listing that the matrix was not so perfect and then you will have to run SAINT again this time starting with the orientation matrix namem p4p from SAINT s post integration global refinement And maybe repeat this once more 2 For the believer You would choose to collect the best reflections from all your runs Crystal gt Threshold for all of them with a relatively hig
16. Software Reference Manual Version 5 0 Bruker AXS Madison USA 1998 SHELXTL Reference Manual Version 5 1 Bruker AXS Madison USA 1997 References for chapter 12 1 E N Maslen in International Tables for Crystallography Vol C 1995 Section 6 3 pp 520 529 2 K N Trueblood H B Burgi H Burzlaff J D Dunitz C M Gramaccioli H H Schulz U Shmueli and S C Abrahams Acta Cryst Sect A 1996 52 770 781 J de Meulenaar and H Tompa Acta Cryst Sect A 1965 19 1014 1018 P Coppens in Crystallographic Computing ed F R Ahmed S R Hall and C P Huber Copenhagen Munksgaard 1970 pp 255 270 R H Blessing Acta Cryst Sect A 1995 51 33 38 A L Spek Acta Cryst Sect A 1990 46 C34 b PLATON A Multipurpose Crystallographic Tool Utrecht University Utrecht The Netherlands A L Spek 1998 7 R H Blessing Cryst Rev 1987 1 3 58 b R H Blessing and D A Langs J Appl Cryst 1987 20 427 428 8 SADABS Area Detector Absorption Correction Siemens Industrial Automation Inc Madison WI 1996 9 H D Flack Acta Cryst Sect A 1974 30 569 573 b Flack H D J Appl Cryst 1975 8 520 521 c Flack H D Acta Cryst Sect A 1977 33 890 898 10 A C T North D C Phillips and F S Mathews Acta Cryst Sect A 1968 24 351 359 11 N Walker and D Stuart Acta Cryst Sect A 1983 39 158 166 12 S Parkin B Moezzi and H Hope J Appl
17. The principal face Miller indices are easily determined and even irregular plate shapes could satisfactorily be approximated by a selection of planes perpendicular to the plate or nearly so Of course the problem of how to measure the crystal thickness is very important here whereas the thickness is refined as a variable in the Lamina correction SADABS The program is specific because apart from absorption correction accomplished as described above it should be able to solve some other absorption problems shading by the mounting glass fiber absorption by a capillary and the solution in it etc which the other routines are not designed to cope with Because details on how to use SADABS are described in the previous chapter only a few comments will be given here Many maybe the majority of Bruker users employ SADABS but none of them can tell you what it actually does with the exception of the author of course The users are just satisfied that they obtain an improvement in their R values Our experience with SADABS is not so good since the program usually has given worse results than XPREP with an empirical psi scan correction applied However our samples often contain very heavy elements Bi Pb TI in a significant sometimes more than 50 amount and could not be considered typical It is well known George Sheldrick private communication that SADABS is not so good with strongly absorbing samples though version 2 07
18. The refined parameters principally affected by systematic errors due to absorption are the anisotropic thermal parameters more appropriately called anisotropic displacement parameters a d p s 2 Because absorption primarily affects the low angle data the net result is that the U tensor refines to a smaller value than the true value In severe cases it may even result in negative eigenvalues i e a non positive definite U tensor There are three basic methods for applying absorption correction to reflection data In decreasing order of theoretical rigor they are 1 Exact numerical corrections i e analytical Gaussian quadrature spherical cylindrical and needle 2 Semi empirical corrections i e psi scans CAMEL JOCKEY and multi scan 3 Refined corrections i e DIFABS XABS2 SHELXA 1 Numerical methods These methods provide mathematically exact absorption corrections providing of course that the input data are correct It is generally agreed that the best absorption corrections are provided by the analytical 3 or Gaussian quadrature 4 methods These two methods require that the crystal faces are well defined and can be accurately indexed and measured These conditions are not often met It can be time consuming to index a specimen with numerous faces and difficult to measure accurately the distances between faces For these reasons numerical methods are used less often than those which are easier to implement exper
19. Tip Even if you don t want to use SADABS for scaling your data George Sheldrick suggests you run the program in order to get and analyze diagnostic plots they may reveal more than you are expecting We must agree with this 83 12 NOTES ON ABSORPTION CORRECTIONS A short but very informative overview of different methods and programs used for absorption correction has been prepared by Louis J Farrugia it is available as help files absorb htm and abs100 htm distributed with the WingGX suite mentioned above L J Farrugia J Appl Cryst 32 837 838 1999 The first part of this chapter is mainly based on this summary which with some minor changes is reproduced here with the author s permission For most crystals containing elements heavier than the first row transition metals it is likely that absorption is the single most important contributor to the systematic errors A correction for absorption is usually required in such circumstances and is mandatory for an Acta Crystallographica submission Note If the ratio of predicted transmission factors Tmax Tmin is bigger than 1 10 the checkcif procedure generates the alert An absorption correction should be applied Predicted transmission factors are a function of crystal dimensions linear absorption coefficient u maximum theta value during data collection Omax and type of radiation There are several options for absorption correction 1
20. and other data required to perform data collection In that case the working directory contains smart ini three sets of matrix data as well as the orientation matrix unit cell parameters and other data needed in a proper matrix p4p file Tip 12 In collecting data for the preliminary cell determination and orientation matrix instead of MATRIX you can start with ACQUIRE gt EditQuad rant and edit the field to get the following lines O 001 28 00 5 00 70 00 110 00 2 300 n t 1 001 28 00 28 00 104 50 4 00 2 300 n t 2 001 28 00 28 00 20 50 26 50 2 300 n 3 QoL 28 00 28 00 9 50 16 50 2 300 n t where n is the number of frames in each series 40 is usually enough and t is the exposure per frame estimated as described above Then use ACQUIRE gt Quadrant with jobname MATRIX followed by Crystal gt Redtn Cell in an attempt to determine the cell automatically A general characteristic of this approach is that these four series of short scans are directed approximately toward tetrahedral faces with mutual angles of ca 109 and cover at least partially the whole reciprocal space not only a hemisphere So you could expect easier determination of the orientation matrix faster indexing better cell parameters much less parameter correlation and even refinement of crystal translations Some other combinations of angles could be suitable too Of course we do not expect a beginner to try this with her first or se
21. are corrected by SADABS Very likely this is a consequence of lower redundancy and not so good corrections at high resolution but could explain why SADABS is so popular We suspect you prefer to obtain as low R values as possible and we believe you do not like conflicts with referees most of them have not read this Guide If this is true you will very likely neglect rules no 1 and 2 and apply absorption correction respecting the other recommendations and following your own way This subject is very provocative Send us your comments and describe your practice We are very interested in comparison of the results after different absorption corrections 18 We reserve the right to change them in the next revision 89 13 NOTES ON TWINS AND TWINNING According to Giacovazzo et al 1995 twins are regular aggregates consisting of individual crystals of the same species joined together in some definite mutual orientation There are numerous ways to classify twins but if we consider their lattice geometry and diffraction patterns twins are divided into two main classes Twin lattice quasi symmetry TLQS twins with two or more differently oriented reciprocal lattices yielding double or multiple diffraction patterns and Twin lattice symmetry TLS twins with a single orientation of reciprocal lattice and therefore with single diffraction pattern In the old days some 30 or 40 years ago with the predominance of film me
22. box Read Help in order to make the proper choice but the mswinpr2 driver uses the Windows printer drivers and should work with any printer with raster capabilities e o e Usually you should repeat R the whole calculation at least once with a new resolution limit the best g value so far and possibly with particular scans left out The g value may need yet another cycle of adjustment but it is not easy to achieve any further improvement Repeat R write hkl file W or quit Q W Finally when the plots look as good as possible write W the corrected data to a file name hkl or sad hkl if you only confirm the default name This file is in HKLF 4 format i e contains F and o F You may also supply the value of uxr to apply an overall isotropic absorption correction ris the effective crystal radius weighting the smallest dimension more heavily Enter name of output hkl file sad hkl Mu r of equivalent sphere for additional spherical absorption correction Enter lt CR gt if none However we never have obtained any improvement when using this additional correction and the previous SADABS versions Actually the results were nearly the same with and without the correction Version 2 07 has not been tested yet 82 The last question means that data will be corrected for A 2 contamination The default value is suitable for a sealed tube Mo Ka SMART diffractometer e Lambda 2 factor 0 0015
23. but submit the correct value if your system is calibrated see sadabs htm file for suggestions how to do it the value should depend on the tube voltage After the SADABS run check the log file which contains most of the screen outputs you have already seen and some other useful data but only the ratio of minimum to maximum apparent transmission Tmin ANd Tmax respectively So how to get these very popular and often quoted numbers In SHELXL input the crystal dimensions using the SIZE command and in combination with the ACTA command you will find the Tmin aNd Tmax Values in CIF If their ratio is equal to the ratio listed in the SADABS log file then everything is OK If not stay cool nobody cares except perhaps referees of some journals especially of Acta Crystallographica To combine SADABS with a face indexed correction you should run the face indexed correction first The procedure is to integrate with SAINT use XPREP to apply the face indexed correction to the raw file and then feed the corrected file to SADABS In XPREP be sure to say yes when asked whether to copy the complete data file with corrected intensities and sigmas it is the file you want to use with SADABS By default it is named with the extension abc Then use SADABS with uxt 0 to do the other corrections and merging Important note When data are corrected with SADABS then the experimental absorption correction type should be defined as multi scan
24. chapter If you have decided to use the same matrix for all runs delete default names from the Matrix p4p Filename fields Use only the first field and enter by typing or browsing the name of a p4p file containing the best cell parameters so far Listed cell parameters are extracted from the default p4p file in the first line If you changed the name of this file the listed parameters will not change automatically unless you click the corresponding Cell button in order to update cell parameter values The first entry will be used automatically for all runs Integrate Sort Global Validate Qpen listing se Helo Gose Fig 8 3 Integrate dialogue box If you repeat the integration see the explanations above and below enter the namem p4p file meaning that the file created during the first SAINT run will be used as input for the second run If you collected control frames Run 9 see the end of the ASTRO chapter delete those entries so the program does not try to integrate them In an ordinary case no other buttons on this menu are important except Integrate from More options so we do not describe them Hence press Integrate now Integrate menu Find the Reflection size field in the upper left corner Fig 8 3 The boxes should be already filled with spot sizes determined automatically during the Threshold procedure If you are in doubt whether to use
25. example of course without comments that are added in italics Some lines were omitted AUTOINDEXING SUMMARY A B Cc Alpha Beta Gamma Three vectors 6 487 9 275 14 067 89 949 87 959 73 631 Linear LS 6 518 9 371 13 770 89 988 88 536 74 231 Determined unit cell parameters before and after linear LS refinement Axis limits max index Axis a b and c limits followed by the maximum HKL index difference used to compute test solution Length Difference Vectors 0 100 20 000 0 100 20 000 FOM H 1 H 2 H 3 Reflections fit Angles between solutions Just one unsuccessful attempt 0 100 20 000 5 30 Group vector length reciprocal space components of difference vectors and number of difference vectors in groups Length 1 17 34149 2 13 77727 3 13 57940 truncated 6 10 21316 7 9 00571 8 8 55909 9 8 36116 10 7 51440 11 7 49792 12 7 48468 13 7 36564 14 6 81270 5 6 74176 truncated 51 4 27546 52 4 26849 X1 01333 03155 01536 09525 10372 04534 03602 11815 08936 08949 07051 02992 02958 19292 322215 Pay attention to difference vectors repeating many times OONO EO OOO 0 0 Retrying with next shortest vector Axis limits max index Program was not able to find set of three solutions it retries with next shortest vectors Length Angles be 0 100 X2 05557 01323 01654 01918 03738 01752 04485 02195 05291 05726 10349
26. file SADABS htm prepared by George Sheldrick is available too The accessory program GSView must be available to run in the middle of a SADABS run GSView freeware is able to view and print even on a non Postscript printer postscript and encapsulated postscript files extensions ps and eps prepared by SADABS or other programs An exe file for installing GSview is provided in the win directory of Sheldrick s site but the program can be downloaded from many different places try Yahoo or Google search SADABS can correct for diverse kinds of experimental errors like variation of the irradiated volume of the crystal incident beam inhomogeneity absorption by the crystal support crystal decay etc SADABS also improves the s u s of the intensities SADABS works using data present in raw or abc files and except for an interactive screen input no other data files or variables are required It is important to start with properly prepared raw files from SAINT for the individual runs with none of the fudges that SAINT can apply In particular no correction for decay no instrument error and no correlation filter should have been used but you already knew that if you read chapter Data reduction strategy carefully The program functions by comparing reflections that should have the same intensity either equivalents or the same reflection recorded at different instrument angles hence high redundancy much improves SADABS
27. frame usually the first in a series Now track a reflection from frame to frame remember Ctrl _ or Ctrl _ until you find the frame on which it appears the strongest Then execute CRYSTAL gt Pick which provides an interactive means for adding reflections to the Reflection Array This command is very similar to the previously described ANALYZE gt Graph gt Rocking and ANALYZE gt Cursor gt Box ones refresh your memory except that the reflection data are now written into Array As already stated you should be very careful when defining the correct box size Width Height Fig 5 6 For this you can use a dialog box screen editing or both options successively x Reflection 0 Quadrant 0 Frame halfwidth 5 Origin 259 Y Origin 249 Width 36 Height 36 OK Cancel Fig 5 6 Pick dialogue box Check that the value of zero is specified in the Reflection field This means that any new reflection will be placed in the first available empty line of the Reflection Array In case you specify a number above 0 the reflection is written in the corresponding position in the list overwriting what previously was there It is enough to have the number 3 in the Frame halfwidth field but if you are in the middle of a series put in a bigger number 5 for example here Reflections will be better centered Disregard the flashes which appear on one frame and occupy only a few pixels These are most
28. frames could be name1 ext name2 ext etc where underscored numbers are automatically set by the system The first number behind the jobname is the crystal number as set in the Project description It is of course possible to use the same crystal more than once or on the other hand more crystals of the same kind can be used for a data collection As shown above one additional number often appears in the name field This number just before the dot is Run and represents sets of data collected during MATRIX MULTIRUN efc procedures The numbers in the extension are Frame Thus name21 001 name21 002 name22 458 etc show jobname crystal 2 data sets 1 and 2 frames 001 002 and 458 respectively If the frame number is higher than 999 then letters a b etc will appear in that field If the letter m appears just before the dot for example namem ext it means that such file contains data merged from several different data sets 2 If the letter t appears just before the dot for example namet ext it means that such file contains merged decay corrected results from the SAINT program If the letter u appears just before the dot for example nameu ext it means that such file contains unsorted results Short description of BRUKER files sorted by extension Extensions of the most important files are given in bold letters 001 002 etc Files containing different measurement pictures or so called frames
29. i In non problematic cases when data collection strategy is carefully planned with as much redundancy as possible and a well defined orientation matrix none of any reasonably chosen see above integration parameters has a notable effect on the final agreement indices and atomic parameters Perhaps there is a slightly higher influence by I sigma threshold for least squares fit below vs simple sum above and Blend 9 profiles It must be emphasized again and again that there is no universal recipe for every case and don t blame us if something has gone wrong If SAINT finished its job successfully you should have a lot of different files in the working directory They are all mentioned in the chapter Directory folder name and extension of Bruker files The most important file is the reflection namem raw file containing output reflections merged from all runs and sorted according to the point group specified in the SAINT input The other important file is namem p4p file with orientation matrix unit cell and detector parameters refined in Post integration global refinement Listing log files describing integration and statistics results are name _Is and namem _ls More or less these files are copies of the output sent to the screen during the SAINT run Read with care the namem _Is file as well as all present name _Is files and compare your results with the results and recommendations given in the separately distributed
30. it could be suitable for a platform three circle diffractometer Even if you are disappointed four circle users do not give up immediately Obviously your strategy about refinement strategy was wrong Let us cite George Sheldrick and his story about COSMO personally find COSMO extremely useful but do not run it in automatic mode put in the runs think will give a good coverage of reciprocal space this could be a standard set of runs and then let COSMO taking the Laue symmetry and orientation matrix into account work out the completeness and redundancy as a function of resolution If am not happy with the result usually test extra runs myself or ask COSMO to suggest them In this way can take other factors into account for example COSMO does not know that SAINT rejects some reflections with high Lorentz factors which makes phi scans less suitable at high 2 theta I have suggested this to the author and can also take into account peculiarities in the way in which the crystal is mounted Note The way to put in your own scans additional scans not contained in the basis set is to invoke Edit gt Custom Run List Those scans are coded black in the Current Strategy window View gt Detailed Strategy and they will not be altered i e they always remain active by COSMO It is also possible to import Edit gt Custom Run List gt Import a run list from your library prepared for SAINT If you already have th
31. latter method is not something we would recommend and let it hereby be revealed that the authors of these lines are so old as to be educated in that ancient era before the advent of personal computers when students were still taught to plan their experiments The selection of single crystals by well documented trial and error methods will not be described here Drink enough to stop your hands from trembling but not so much as to make it worse But note that some countries prohibit the operation of X ray machines while under the influence of alcohol or some drugs In our system the single crystal should not be larger than about 0 6 0 7 mm The largest available collimator pinhole is 0 8 mm in diameter As a rule smaller crystals are preferable especially if they contain heavy elements and absorption is significant Recommended dimension for sulfides and oxides of heavier elements is ca 0 1 mm for common silicates and oxides containing lighter elements 0 2 0 3 mm for organics ca 0 5 mm In any event an optimal crystal dimension t can be calculated by the well known relation t 2 u where u is the linear absorption coefficient For organic and coordination compounds u usually lies between 0 1 and 2 mm for compounds containing heavier elements u is about 10 mm while for compounds with a high content of TI Pb or Bi it can reach 80 mm However the formula gives too large size for most organic and many organometallic compounds with no v
32. of width dmin Thinner slices allow insight into finer structure details For most structures such high resolution is unnecessary 0 8 A would be enough unless of course you are doing charge density studies A choice of desired resolution also depends on the scattering power of the investigated crystal If you take a look at frames collected during MATRIX operation you will see that in some cases the diffraction spots are concentrated close to the shadow of the beam stop while in some other cases spots are visible in the entire frame Use the crosshair cursor to determine 26 angles and resolution of the most distant reflections If the crystal does not diffract at high angles it is useless to require high resolution Keep in mind also that a 6 of 25 for Mo radiation or 67 for Cu radiation is the minimum requirement for publishing small molecule structures in Acta Cryst but use higher values if the crystal diffracts sufficiently As a first step you can try to cover the asymmetric unit unique volume by passing it trough the cap of the detector projection Use the F3 shortcut to select a view down the Z 51 axis With STRATEGY gt Drive or STRATEGY gt Scan Custom swing the detector projection to a position where one side just covers the X ray beam axis Then drive the other goniometer angles to put the asymmetric unit in an orientation where rotation about Z will cover as large a part of its volume as possible Sometimes it
33. one can expect spot shape distortion On the other hand a longer 3 distance improves the signal to background ratio and a shorter distance allows collecting more reflections in the same time Accordingly under normal circumstances distances between 4 0 and E 6 0 cm could be recommended Normal circumstances means that the biggest unit cell parameter is not above approximately 25 A for 4 0 cm and above 40 A for 6 0 cm f In case of a big unit cell the detector has to be moved away in order to avoid reflection j overlap Several slightly different equations have been proposed for that purpose In the simplest f approach the crystal to detector distance for a primitive lattice can be calculated from the relation distance in mm 1 5 x a in A gj where a is the ongest unit cell axis For high quality crystals with sharp reflections the distance could be even shorter For low quality crystals with broad reflections the spot size for integration must be very carefully determined or the detector distance must be longer Phi or omega scan In few words the choice of phi or omega scan is of no great 3 significance Although phi scans result in slightly lower R values in particular wR2 there is no big i difference in the quality of data collected by one or the other scan Actually a combination of phi and omega scans is very common in the case of four circle diffractometer while phi scan is almost j Never used on three
34. series Filename extension of frame read Number of reflections found in the frame Average deviation between observed and predicted observed minus predicted reflection positions If any of these numbers is systematically higher than 0 2 pixels you need to re refine the unit cell parameters and orientation matrix Root mean square RMS differences between observed and predicted as above reflection positions If values are consistently above 0 3 data are weak or there is a problem with the orientation matrix Average normalized intensity Average I SIGMA value of the reflections For strong data Sig is usually about 30 40 Weak data may have Sig lt 10 If 1 Sig is gt gt R sym consider data collection with a longer time Percent of reflections that are lower than 2 sigma For strong data this percentage is usually less than 10 for weak data it could be from 33 to 50 but in cases where the integration of a centered lattice is done with the P option would be accordingly many reflections here also for strong data Average coefficient representing the degree of correlation between 3D reflection profiles and the model 3D profiles computed from strong spots Usually strong data show an average correlation about 0 7 0 8 while coefficients for weak data are smaller Values consistently less than approximately 0 25 indicate that data are very weak or that something is seriously wrong e g wrong orientation matrix but remember the abo
35. signal A value near 1 indicates that the anomalous signal is very small Larger values and especially values above 2 indicate a very strong signal See SAINT manual p 7 7 for the corresponding equation Average signed observed minus predicted position in pixels for the observed reflections Absolute values consistently above 0 2 pixels over many frames show probable significant errors in the orientation matrix re refine the matrix Root mean square value of the errors in X Y and Z positions Values consistently above 0 3 pixels indicate problems with orientation matrix or spatial calibration Coverage or cumulative statistics tables Angstrms Obs Theory Compl Redund Rsym Pairs Pairs Rshel Sigma lt 28 Resolution in to which cumulative statistics are computed Cumulative number of unique reflections Cumulative number of theoretically observable reflections Cumulative percentage completeness of the data Cumulative redundancy factor R sigma on I Cumulative number of unique Bijvoet differences Cumulative percentage of theoretically possible Bijvoet differences Rsym computed to the thin shell ending at this resolution Mean I sigma Percentage of reflections less than 2 sigma l 97 Frames statistics tables File Ref ErrX ErrY ErrZ RmsxX RmsY RmsZ Inorm Sig lt 2S lt Cor gt Ful Frame number relative to the start
36. the left button RLATT will show the distance between two points in A Fig 4 1 a In this version of RLATT 3 0 there are two very good tools that help measurements Use them to increase the accuracy of your determination If you press the Gray key you will subdivide the current measurement with the new distance shown It is of course possible to repeat this operation many times In the opposite direction the Gray key reduces the number of subdivisions If you press the Page Up key you will add extra divisions on each side of the existing one Fig 4 1 b They are of the same length as the previous divisions The Page Down key reduces the number of extra divisions For fine tuning you can use the cursor keys which move the measurement tool or Insert and Delete keys which rotate the lattice around Z With RLATT it is also possible to see if the lattice is rectangular or not and to estimate the angles The program is also useful for testing lattice centering Try the following approach Turn the lattice until you notice a family of lattice planes with the largest spacing and orient it for an edge on view Measure the interplanar distance the length of the corresponding crystal lattice vector call it a Now turn the lattice parallel to the planes until another family of planes with a relatively large spacing appears Try now to adjust the picture so that the new family of planes is oriented horizontal or vertical and you still see bot
37. the redundancy is high and if you know what you are doing enter N but this answer halves the data to parameter ratio for determination of program parameters If you answer N it is further necessary to define the point group so that Friedel opposites can be correctly defined Enter filename if no more The input files are nameO raw name1 raw name2 raw and perhaps more the raw is added automatically if you omit it The merged namem raw file can also be used As noted above find the path if necessary then replace name abs with name or namem Successive data files appear automatically The program now reads and briefly checks your data and prints some characteristic parameters Pay attention to the redundancy and mean I sigma tables The determination of a model for the systematic errors starts now and some thresholds should be specified for excluding some reflections from the parameter refinement These reflections may still be corrected and included in the final output hkl file I Accept defaults given in brackets for the first pass e gt Enter mean I sigma threshold 3 but threshold could be lower 2 5 or even 2 if reflections are exceptionally weak e gt Highest resolution for parameter refinement 0 1 This small default value 0 1 excludes none of the data in the model fitting and is a good place to start e Factor g for initial weighting scheme 1 0 04 Note that a more precise va
38. the screen in the so called Status area you will see the current goniometer angles current drive speeds crystal Laue group name of the current p4p file detector to crystal distance X ray wavelength desired resolution and a visibility state indicator for objects in the diagram With the help of listed shortcuts which act as toggle on off buttons or with the same effect by commands listed in the EDIT menu you can choose which objects will be visible At the same time the last item at the right bottom side of the screen Fig 7 1 indicates object Masks i e status of its visibility Laboratory axes are direct beam or X axis BEAM the other horizontal or Y axis Y and the vertical Z axis that coincides with omega and 2 theta and is oriented upwards not shown They form a right handed coordinate system Goniometer axes are omega chi and phi The Ewald sphere is a sphere of radius 1 in units 1 A centered at the 1 0 0 point 49 OBJECT DEFAULT COLOR SHORTCUT Laboratory axes BEAM Y blue L Goniometer axes Omega Chi Phi pink G Detector projection on Ewald sphere red D Ewald sphere blue E Reciprocal axes H K L yellow R Toroids green T Reduced HKL data cyan H Unique volume white U Symmetry equivalents purple M It is possible to change default colors using the EDIT gt Colors command Available ASTRO menu commands and corresponding shortcuts are listed in the ASTRO Manual pp A 1 to
39. the three circle diffractometer Believe us it is worth while to put effort into this job If you understood the preceding text and if you finally believe you have a good strategy the result of your pains is revealed by the STRATEGY gt Coverage After selection of the command you will see a dialogue box with some strange questions Just press Enter and wait for a few seconds while calculations finish Then a new window with the following data appears Point group of the crystal Possible number of unique observations for submitted resolution Estimated coverage of unique observations and Estimated mean minimum and maximum redundancy In a satisfactory strategy the coverage should be 100 or very close to that Estimated mean redundancy should be between at least 1 5 for triclinic crystals and 3 7 for higher symmetry crystals Remember higher redundancy in combination with reasonable intensities increases quality of your data and precision of the final structure parameters However do not put too much faith in values calculated by ASTRO In limited testing we have found the program to be pretty good when it calculates the possible number of unique 8 User is free to choose run numbers Run and the first frame number Frame Thus the first Run could be 1 and it is not obligatory to use sequential numbers Some users prefer to number the first frame of a run as 000 53 observations and coverage However mean red
40. threshold Ang 1 0 test for example values of 0 9 and 0 8 Finally make the corrections Process data using the combination of parameters that yielded the lowest Rnt value Even in the case of a very large uxr value the psi scan method could give good results in many aspects However the resultant a d p s are usually less acceptable 87 Lamina corrections This empirical correction is suitable for plate like lamina crystals Start psi scan corrections A then P in the usual way and when the question Lamina L Ellipsoid E or Quit Q E appears suddenly change your mind and enter L instead of E The program then asks for indices of the principal face fortunately suggesting them at the same time and a minimum glancing angle where the default value 3 should be accepted Although according to the manual for strongly absorbing crystals which are good approximations to thin plates this method can produce spectacular improvements we have never obtained convincingly better results with it Also in one case of a low absorbing sample unfortunately the only one tested the ellipsoid correction also gave lower R factors than lamina correction May be we did not find the proper combination of I sigma threshold 20 and Resolution threshold Ang 1 0 It always turned out that we could still use the face indexed absorption correction in the case of plate shaped crystals
41. with the scale of the cell dimensions If very accurate cell dimensions are important leave the crystal to detector distance fixed during the least squares refinement Naturally in this case the crystal to detector distance must be very carefully calibrated with a test crystal For a complete list of constraint codes see SMART manual pp 4 32 to 4 33 An empty field means that all parameters are free to refine W Finally remember to check or define Frame halfwidth in the bottom right area of the dialogue box The default value 0 15 corresponds to the 0 3 frame width as defined during data acquisition Matrix MultiRun etc After execution of the LS command a scrolling output window is displayed The content of the LS output is given below together with some comments added in italics Orientation Matrix 0 08826239 0 01603561 0 10353260 0 12271879 0 01568373 0 03945471 0 04739063 0 07047869 0 01477626 37 Lattice parameters amp Standard deviations 6 5798 13 5205 9 3254 90 000 106 390 90 000 795 89 0 0026 0 0050 0 0034 0 017 0 014 0 015 0 81 Very important tabular histograms follow The first one shows differences between calculated and observed H K L and scan angle values in degrees The second one shows differences for detector X Y positions in pixels In both histograms the numbers should be concentrated on the left side The best case is with all numbers in the first column Histograms 00 05 10
42. without a grain or a lot of critical thinking After all please note that we pay considerable attention to environmental problems We recycle X rays whenever and wherever possible 1 So what if you are Who has never been confused by a computer program Table of Contents 1 Directories folders names and extensions of Bruker files 1 Short description of BRUKER files sorted by extension a se 2 2 Crystal mounting and orientation matrix 0 ce ee ee eiee ee cette teeter eneeeeeneeenes 4 General description and four circle diffractometer cccecceseeseeteeeeeeeeetetenteees 4 Three circle platform diffractometer cceccccecccceceeeeeeeeeenneeeeeeeeeeeseeneaeeeeeeeeeees 13 Notes on low temperature data collection cccccceceesecceceeeeeeeeeeseeaeeeeeeeeneetennanes 18 3 How to analyze crystal data oo cece cece eee ceeeeeeeeeeceeeeeeeeeeeeaeeaeeeesesereeees 19 4 Reciprocal Lattice Display Program 0 cccececceceeeceeeeeeeeeteeeeneeeeeneeereneees 23 5 If matrix operation fails sts 35 Savciccasivaerntasasi oorniayniaas aerate eneamncten 27 6 Data collection or data acquisition strategy 0 ccc eeeeeeeeeeeeeeees 43 General description and four circle diffractometer ccccccccccccccccceeeeeeeeeeeeeeeeeeeeess 43 Three circle platform diffractometer ccececcecccceceeeeeeeeeeneeeeeeeeeeeeseneenaeeeeeeeeeeeas 47
43. yes Disable dark flood cor Y N F Check for yes OK Cancel Fig 2 2 Acquire gt Rotation dialogue box Look at the pointers on the goniometer scales They can easily get 1 away from zero and Home Axis will not notice If the pointers do not show 0 use Manual mode and position the goniometer circles close to zero Then repeat GONIOM gt Home Axis Take a rotation photograph rotation around phi by ACQUIRE gt Rotation Set the time if necessary Fig 2 2 but the shortest time is 72 seconds and this value is a good choice if the crystal is not very small There is a conflict in the manual here see pages 2 13 and it is not quite clear what is the shortest time allowed by the goniometer rotation speed Leave default values for the other parameters Beware rotation photographs are not stored automatically but must be saved using FILE gt Save if you wish to keep them for documentation a common extension for rotation photographs is rot Some characteristic rotation photographs are shown in Fig 2 3 The intensities of spots can give you an idea about the measurement time required for a frame the number of spots gives you an idea about how many frames to collect for the initial lattice determination If you think you do not see anything in your rotation photograph try another one with an exposure that is at least twice as long c d Fig 2 3 Examples of rotation photographs a a poly and micro crystal
44. 0 34 ORTHORHOMBIC C gt gt gt 15 3456 6 518 18 092 13 770 89 485 91 464 94 516 1 00 0 00 0 00 1 00 2 00 0 00 0 00 0 00 1 00 The best solution according to the program since the highest symmetry is one of the criteria However the angles are too far from 90 for an orthorhombic crystal truncated 16 16 09 9 854 12 786 13 770 89 245 90 956 68 913 1 00 1 00 0 00 1 00 1 00 0 00 0 00 0 00 1 00 truncated ORTHORHOMBIC P 20 13232 6 518 9 371 13 770 89 988 91 464 105 769 1 00 0 00 0 00 0 00 1 00 0 00 0 00 0 00 1 00 MONOCLINIC C 21 3 39 6 518 18 092 13 770 90 515 91 464 85 484 1 00 0 00 0 00 100 425 00 0 00 0 00 0 00 1 00 truncated 53 19 81 16 798 17 065 6 518 75 326 102 443 71 174 1 00 1 00 1 00 1 00 1 00 1 00 1 00 0 00 0 00 MONOCLINIC I 54 10 32 16 798 6 518 16 654 97 573 114 519 77 557 1 00 1 00 1 00 1 00 0 00 0 00 0 00 1 00 1 00 truncated 65 19 81 16 657 17 065 6 518 75 326 100 020 115 121 0 00 1 00 1 00 1 00 1 00 1 00 1 00 0 00 0 00 MONOCLINIC P 66 T25 6 518 13 770 9 371 89 988 105 769 91 464 1 00 0 00 0 00 0 00 0 00 1 00 0 00 1 00 0 00 The proper solution for this case characterized by the next to lowest GOF value 67 13 27 6 518 9 371 13 770 89 988 91 464 1 1 00 0 00 0 00 0 00 1 00 0 00 0 00 0 0 truncated 69 19 98 6 518 9 854 13 770 89 044 91 464 113 767 1 00 0 00 0 00 100 Yks 00 0 00 0 00 0 00 1 00 5
45. 00 38 00 00 27 00 3 250 720 10 00 al 001 38 00 38 00 180 00 27 300 3 250 720 10 00 2 001 38 00 50 00 00 220 00 2 250 200 10 00 3 001 38 00 50 00 90 00 220 00 2 250 200 10 00 4 001 38 00 50 00 180 00 220 00 2 250 200 10 00 5 001 38 00 50 00 270 00 220 00 2 250 200 10 00 r40_20 0 001 38 00 38 00 00 34 00 3 200 900 10 00 1 001 38 00 38 00 180 00 34 00 3 200 900 10 00 2 001 38 00 50 00 00 125 00 2 200 250 10 00 3 001 38 00 50 00 90 00 150 00 2 s200 250 10 00 4 001 38 00 50 00 180 00 10 00 2 200 250 10 00 5 001 38 00 50 00 270 00 50 00 2 200 250 10 00 r60_20 0 001 38 00 38 00 00 27 00 3 200 900 30 00 1 001 38 00 38 00 180 00 27 00 3 200 900 30 00 2 001 38 00 50 00 00 220 00 2 200 200 30 00 3 001 38 00 50 00 90 00 220 00 2 200 200 30 00 4 001 38 00 50 00 180 00 220 00 2 200 200 30 00 5 001 38 00 50 00 270 00 220 00 2 200 200 30 00 99 r80_25 0 001 20 00 20 00 00 17 00 3 250 720 60 00 1 001 20 00 20 00 180 00 17 00 3 250 720 60 00 2 001 40 00 20 00 00 2000 3 250 720 60 00 3 001 40 00 20 00 180 00 20 00 3 250 720 60 00 rEdit 60 mmm 0 001 38 00 38 00 00 27 00 3 250 720 10 00 1 001 38 00 38 00 180 00 27 00 3 250 720 10 00 2 001 38 00 50 00 00 220 00 2 250 200 10 00 3 001 38 00 50 00 90 00 220 00 2 250 200 10 00 4 001 38 00 50 00 180 00 220 00 2 250 200 10 00 5 001 38 00 50 00 270 00 220 00 2 250 200 10 00 rControl see the
46. 01 30 00 30 00 0 00 54 70 2 0 500 50 30 00 11 5 hours 0 001 30 00 30 00 0 00 54 70 2 0 400 455 30 00 1 001 30 00 30 00 90 00 54 70 2 0 400 326 30 00 2 001 30 00 30 00 180 00 54 70 2 0 400 173 30 00 3 001 30 00 30 00 270 00 54 70 2 0 400 113 30 00 4 001 30 00 30 00 0 00 54 70 2 0 400 50 30 00 16 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 606 30 00 ii 001 30 00 30 00 90 00 54 70 2 0 300 435 30 00 2 001 30 00 30 00 180 00 54 70 2 0 300 230 30 00 3 001 30 00 30 00 270 00 54 70 2 0 300 150 30 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 30 00 The settings in the following section will cover a sphere They can be used for crystals of any symmetry They should be used for triclinic crystals 4 6 hours 0 001 30 00 30 00 0 00 54 70 2 0 500 364 5 00 1 00 30 00 30 00 90 00 54 70 2 0 500 261 5 00 2 001 30 00 30 00 180 00 54 70 2 0 500 364 5 00 3 001 30 00 30 00 270 00 54 70 2 0 500 261 5 00 4 001 30 00 30 00 0 00 54 70 2 0 500 50 5 00 10 5 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 606 10 00 001 30 00 30 00 90 00 54 70 2 0 300 435 10 00 2 001 30 00 30 00 180 00 54 70 2 0 300 606 10 00 3 001 30 00 30 00 270 00 54 70 2 0 300 435 10 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 10 00 2 The table was obtained from Dr C Campana at Bruker We are grateful for that 101 11 5 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 60
47. 11111111111100010000 1111111111110101110101111 39 29 66 25 truncated 19 148 0 19 0 2 O 11111111101111111101110111111111001101110001110110 1110101110001011001101011 125 5 59 9 102 7 47 3 125 5 114 9 denotes angles around 60 or 120 16 856 0 20 Ou x2 1 11111111101111111101110111101011001101110000110110 1110101110001011011011010 56 6 100 6 66 2 119 7 56 6 59 6 160 6 15 699 0 20 1 2 O 11111111111111111111111111111011001111110000011001 1111101110000001000001010 96 7 63 6 79 3 71 8 96 7 87 7 28 8 145 5 19 473 0 19 1 2 1 11111111111111111111111111111111011111111100010001 1110101110010101000101011 100 9 46 2 73 6 60 7 100 9 86 8 29 1 136 6 17 4 14 067 0 20 1 2 1 11111111111111111101111111101011101101110100011001 1111101110000011010001010 90 1 88 0 88 7 88 9 90 1 89 3 41 7 146 6 24 4 41 7 denotes angles close to 90 x truncated 19 053 0 18 1 0 3 11111111101111111101111111111111011111111101011100 1111111111000001001011010 78 5 27 8 38 6 72 2 78 5 58 1 76 7 85 4 66 6 52 0 87 9 62 1 46 1 82 0 27 8 65 3 12 8 57 6 110 0 38 6 100 4 27 8 18 794 0 23 1 2 3 11111111111101111101110111001011001101110100110110 0111101100000011001101000 78 8 131 6 105 0 118 7 78 8 92 3 77 7 119 8 68 0 85 4 43 6 63 6 80 3 44 7 131 6 138 6 119 9 125 4 18 5 105 0 29 9 131 6 125 7 Predicted reduced primitive axes are Soln X Y Z Length 2 1 89857 5 96602 1 69785 6 48696 1 7 52911 5 38987 0 53336 9 27484 11 3 27490 3 28218 1
48. 2 1 2 wR2 Bo Fe F a F3 also denoted wR ee l a w r FY n ph where n is the number of reflections and p is the total number of refined parameters 104 Literature All BRUKER reference manuals have been used more or less successfully during the preparation of this text Readers should be warned that some manuals are out of date In many cases it is better to use help files that are more current than the manuals ASTRO for Windows NT Software Reference Manual Version 5 0 07 Bruker AXS Bruker Analytical X ray Systems Madison USA 1998 COSMATIC NT v1 40 User Manual Bruker AXS Madison USA 2002 Available as cosmatic pdf file after program installation COSMO NT v1 40 User Manual Bruker AXS Madison USA 2002 Available as cosmo pdf file after program installation GEMINI Twining Solution Program Suite Version 1 0 User s Manual Bruker AXS Madison USA 2000 Bruker AXS Madison USA 2002 Available as cosmo pdf file after program installation RLATT Reciprocal Lattice Display Program Version 3 0 Bruker AXS Madison USA 2002 Available as rlatt_help pdf file after program installation SAINT Software Reference Manual Version 4 Bruker AXS Madison USA 1994 1996 Sheldrick G SADABS Version 2 03 Bruker Simens area detector absorption and other corrections University of Gottingen Germany 2002 Available as sadabs htm file SMART for Windows NT and IRIX
49. 3 28097 14 06705 X Y Z components of each vector and vector length Reduced primitive cell is 6 487 9 275 14 067 89 949 87 959 73 631 Least squares cell parameters 6 518 9 371 13 770 89 988 88 536 74 231 809 06 32 Full vectors Vector number number of the closest vector indices and reciprocal space vector components H K and L values should be close to integers e NEAR H K L X1 X2 X3 1 33 2297 0 99 0 09 0 15390 0 41132 0 12468 2 22 3 98 L97 1 75 0 16351 0 59921 0 04324 3 33 1 99 0 98 2 06 0 10050 0 31619 0 06073 4 26 wd 2 22 105 0 20108 0 72561 0 34113 5 7 3 93 2 65 1 45 0 08570 0 61437 0 05200 6 30 1 98 0 06 6 03 0 07347 0 16333 0 51612 7 1 3 95 2 00 0 14 0 13477 0 57320 0 15351 8 0 1 00 0 00 1 02 0 07086 0 11087 0 12017 9 4 5 26 2 82 92 0 13067 0 73831 0 34843 10 4 5 92 1 98 4 82 0 23040 0 74725 0 59538 truncated 74 57 0 01 4 06 3 99 0 35796 0 21303 0 33867 75 34 0 06 Po 3 18 0 32667 0 06574 0 18256 Press Enter or click OK to exit the window A message box with question RE INDEX REFLECTION ARRAY ARE YOU SURE will be displayed Confirm Yes button to write HKL indices into the Reflection Array and to set the H flag of each indexed reflection Then the program displays the next scrolling window with output very similar to the previous one That does not require additional explanation If these attempts result in reasonable unit cell parameters just follow t
50. 6 12 00 1 001 30 00 30 00 90 00 54 70 2 0 300 435 12 00 2 001 30 00 30 00 180 00 54 70 2 0 300 606 12 00 3 001 30 00 30 00 270 00 54 70 2 0 300 435 12 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 12 00 13 5 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 606 15 00 ii 001 30 00 30 00 90 00 54 70 2 0 300 435 15 00 2 001 30 00 30 00 180 00 54 70 2 0 300 606 15 00 3 001 30 00 30 00 270 00 54 70 2 0 300 435 15 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 15 00 16 4 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 606 20 00 1 001 30 00 30 00 90 00 54 70 2 0 300 435 20 00 2 001 30 00 30 00 180 00 54 70 2 0 300 606 20 00 3 001 30 00 30 00 270 00 54 70 2 0 300 435 20 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 20 00 20 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 606 25 00 1 001 30 00 30 00 90 00 54 70 2 0 300 435 25 00 2 001 30 00 30 00 180 00 54 70 2 0 300 606 25 00 3 001 30 00 30 00 270 00 54 70 2 0 300 435 25 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 25 00 23 hours 0 001 30 00 30 00 0 00 54 70 2 0 300 606 30 00 1 001 30 00 30 00 90 00 54 70 2 0 300 435 30 00 2 001 30 00 30 00 180 00 54 70 2 0 300 606 30 00 3 001 30 00 30 00 270 00 54 70 2 0 300 435 30 00 4 001 30 00 30 00 0 00 54 70 2 0 300 50 30 00 102 K BENHAVNS HUSKESEDDEL COPENHAGEN S MEMO A quick guide through steps in a non problematic single crystal diffraction measurement 1 Inform the l
51. 769 1 00 TRICLINIC P 70 0 00 6 518 9 371 13 770 89 988 88 536 74 231 1 00 0 00 0 00 0 00 1 00 0 00 0 00 0 00 1 00 This solution with the best GOF is identical to the input and solution no 66 71 3 30 6 518 9 854 13 770 89 044 88 536 66 233 1 00 0 00 0 00 1 00 1 00 0 00 0 00 0 00 1 00 truncated 96 19 58 6 518 12 786 16 798 102 818 102 443 135 146 1 00 0 00 0 00 1 00 1 00 0 00 1 00 1 00 1 00 Sometimes the Bravais output may contain 500 solutions upper limit and it is hard to find the best one suggested by the program Do not worry pressing Enter or clicking OK exits the Bravais output and a new window containing the number of the best solution appears Just note the number press Esc or click Cancel and you will be returned to the output window allowing you to analyze the results in more detail After you make the selection press Enter again type in YOUR OWN choice and continue The reflections in the Array will be re indexed accordingly Note Input zero if you for some reason want to discard The other option is to input the number of the solution identical to the input this is the one with Sig 0 00 see the above example 35 We are now ready to refine the unit cell parameters by invoking CRYSTAL gt Redtn Cell gt LS Ctrl L shortcut This operation refines orientation matrix unit cell parameters and many instrumental corrections such as detector position goniometer zero
52. A 3 During the scan the lattice is rotated about the goniometer center and the volume of the reciprocal lattice that passes through the detector projection on the Ewald sphere will be recorded The detector projection is like a cap on the Ewald sphere Reciprocal axes are A B and C crystallographic axes directions coincide with H K L ASTRO Area detector D V5 007 c 1997 Bruker AXS Eor E Fie Edt Viewpoint Stategy Symmetry HKLs Hep a b Fig 7 1 ASTRO screen without a and with toroids b shown Toroids green can be described as possible scans The aim of the data collection strategy is to fill an asymmetric unit or unique volume of the reciprocal lattice with them The unique volume represents an asymmetric unit of data chosen in accordance with the Laue group specified in the Symmetry menu Symmetry equivs is active if the Symmetry gt Map_symmetry option is enabled With this option portions of the scan toroids outside the unique volume are symmetry mapped into it The viewing direction can be easily chosen using VIEWPOINT menu commands or corresponding shortcuts see the ASTRO manual for a detailed explanation The view approximately down the omega Z axis is a good viewpoint to be used when performing scans interactively Use the F3 shortcut to specify this view The views down the crystal axes F9 F10 and F11 shortcuts are especially useful for examining completeness of coverage after the scans have be
53. A BEGINNER S GUIDE TO THE BRUKER AXS PACK and other noble time wasters A never finished beta test version 4 0 Prepared by Dejan Poleti with help from Ton i Bali uni H kon Hope and Ljiljana Karanovi and other possible and impossible rights belong exclusively to the authors and GEOLOGISK INSTITUT K benhavns Universitet This guide is mainly devoted to the instrument control and data acquisition program SMART V5 054 However some useful comments about other Bruker programs RECIPROCAL LATTICE DISPLAY PROGRAM V3 0 ASTRO V5 007 COSMO NT V1 42 SAINT V6 28A SHELXTL V5 1 XPREP V6 13 and SADABS V2 07 are also included Copenhagen Belgrade 2003 A SHORT HISTORICAL SURVEY The very first shorter version of this text was written in the summer of 2000 during our D P and Lj K three month visit to the Geological Institute University of Copenhagen Denmark We suppose some readers wonder about our motivation to prepare the Guide No secrets or urban legends no dream appearances or touch of destiny It was indeed necessary to orient ourselves and to organize our minds when we for the first time faced a Bruker Nicolet Syntex Simens Bruker Nonius AXS four circle diffractometer equipped with a 1000 K CCD detector and a lot of its technical documentation unjustly referred to as Manuals For almost one year the first version of our Guide was in use only in Copenhagen and Paris Ecole Central
54. AINT run If multiple runs have been integrated together there is also a merged file with the name ending in m see above results of intensity integration done by SAINT gt Integrate for example a file name1 001 will become name1 _rb etc 2 CRYSTAL MOUNTING and ORIENTATION MATRIX General description and four circle diffractometer The work starts with mounting what we think is the best crystal in the batch we obtained lucky girls and guys who work with or are themselves synthetic chemists and can choose among a mass of good crystals or what is the only available rubbish we have in the case of mineralogists or not so lucky synthetic chemists Then comes the moment of truth a preliminary check on the diffractometer the crystal will turn out to be something we just throw away and choose another for the above mentioned lucky girls and guys or will offer us a glimpse into the problems that await us while trying to solve its structure In the old days some crystallographers would use film methods for this part of the crystal checking before turning to measurements on a diffractometer So why not spend an hour or two carefully examining the preliminary data before you start filling your computer with megabytes of data collection On the other hand with the current instrument it is possible to start data collection right away and then by inspecting some tens of the first frames to decide whether to proceed and how or not The
55. Cryst 1995 28 53 56 13 SHELXL Suite of Programs for Crystal Structure Analysis Release 97 2 G M Sheldrick Institut f r Anorganische Chemie der Universit t Tammanstrasse 4 D 3400 G ttingen Germany 1998 A O O oOo 105 Finally some discussions and recommendations are based on results published in the articles G rbitz C H 1999 What is the best crystal size for collection of X ray data Refinement of the structure of glycyl t serine based on data from a very large crystal Acta Cryst B55 1090 1098 Ruhl S amp Bolte M 2000 Strategies for data collection on a CCD diffractometer Z Kristallogr 215 499 509 Nota Bene For those preferring INTERNET documents and a quick start a very brief set of instructions for the Bruker SMART APEX system is available from the web site http www msg ku edu xraylab Acknowledgement Partial financial support of the Danish Natural Science Research Council as a part of the project Complex Sulfides No 9901772 KU 56347 is gratefully acknowledged 106 Availability of the Guide The Guide was originally written using Word 6 0 then it was translated into and printed from Word 97 Word 8 Even with Word 2000 we have had no more trouble than expected but more than we need For Copenhagen users only Copies of the document are located in all laboratory computers directory for all places c dejan guide under names BRUKAp1 doc and BRUKAp2 do
56. Detector M OCD detector Fractonsl faceplate transmittance at normal incidence 983000 Number of pixels per cm 81 920000 Distance from detector face to phosphor or wire g d cm 800000 Fiducial plate hole spacing cm fo 254000 Read noise electrons 13 000000 Electrons per analog to digital unit f2 500000 Electrons produced per rey photon 28 000000 Base offset per read out analog to digital units 16 Full scale value analog to digital units 838848 Volidate Help Qose Fig 8 8 SAINT Instrument Configuration dialogue box 68 9 CRYSTAL SHAPE Although the main use of the diffractometer is to measure diffraction of course the attached parts microscope or TV camera make it possible also to measure crystal shape This should not be neglected because an accurately measured shape is the best basis for a successful absorption correction In addition the morphological characteristics the crystal habit are among the most important physical properties so why not determine them together with the structure Even if you have an irregular fragment with only some or no planar faces you could try to approximate the shape by a combination of crystal faces There are at least two known to us programs for the crystal shape optimization from the intensity relation of equivalents These are X SHAPE from STOE and EUHEDRAL The first program is commercial The latter one is written by M Lutz and A M M Schreurs from the Univers
57. Finally set constraint Crystal translation Other constraints can also be chosen if you have determined the appropriate parameters and included them in the input p4p file from measurements on a standard crystal Filter menu This operation modifies or deletes certain reflections according to the list of operations and values listed in Operation type Reflection Record Item Minimum and Maximum fields Fig 8 7 It is possible to apply up to eight filter operations simultaneously Types of operation and reflection record items are described in the help file Since actual programs for crystal structure determination have the capability to filter reflections it is unnecessary to use the FILTER program from SAINT Anyway the recent point of view on crystal structure determination recommends use of all available recorded reflections Instrument menu This menu is used to change detector and monochromator parameters Normally during the initialization of SAINT detector parameters are extracted from the header of a frame file while monochromator information is obtained from the smallmol ini Do not even open this dialogue box otherwise you can commit the sin of unintentionally or intentionally changing some parameters It will be enough to take a look at Fig 8 8 3 Monochromattor 24heta degrees 12 170000 Roll degrees 0 000000 Absorption Beam path linear absorption coefficient 1 cm 0 001250
58. INT and SADABS so called TWINABS have been released recently though we have not seen the programs yet but visit the web site http 128 104 70 72 SHARE1 WWW Meeting_2003 Presentations htm with user name bruker and password elements 35 92 and 36 case sensitive The new versions will integrate and scale data from twinned and incommensurate structures Il PRELIMINARY BUT VERY IMPORTANT NOTES It is assumed that you are familiar with the Windows NT operating system It is also important to know something about the good old DOS Bruker pack is a counter intuitive mixture of both operating systems prepared in a very fashionable patchwork style Since the above mentioned programs are not connected in any useful way you should be prepared for various troubles especially with project management and file manipulation as well as with limited sizes of DOS and other windows Have fun Many chapters of this document can be used as a checklist Some documents given in the Appendix can be printed separately and used as a reminder Crystal Identity Card is of special importance here If you fill in this form regularly with a lot of good luck all necessary data will be kept together when the crystal structure determination is finished Do not be confused although you probably will be Some operations can be done in an identical or very similar manner starting from different menu commands Obviously Bruker s Staff prefers to have a l
59. T determines its angular range in a 9x9x9 integration volume SAINT maps the submitted values to cover 7 9 or 78 of a 9x9x9 profile Since a Gaussian spot shape is assumed the optimal width is the full width at 10 of maximum nput values should be large enough to accommodate every profile with some room for positional errors Therefore it is 58 better to make the estimate slightly larger rather than too small On the other hand if the box is too large the program complains about too little background being updated The exceptions from this rule are cases with very close or partially overlapped reflections but then consider the question Why was the detector so close during data collection It is true that the new versions of SAINT extract reflection sizes from a p4p file but can we trust computer programs Furthermore if you have collected reflections manually by handpicking not by the threshold operation the sizes remain unknown to the program Reflection sizes are crucial parameters for integration and incorrect values are the most common reason for the need to repeat an integration The Reflection size values estimated automatically during the SMART thresholding may be suitable but you can adjust them if they do not fit well As a rule the program estimates X and Y to be equal while Z is much lower In our experience with respect to the estimated values the X and Y sizes should be slightly increased while the di
60. T diffractometer is equipped with a video camera the black tube from the right that replaces the telescope The video camera provides a clear magnified picture on the computer screen of the area near the crystal This picture makes it easy to adjust the crystal position It also allows for reasonably accurate measurements of crystal shape and dimensions Using this machine is in principle and also in practice not much different from what has been described above for the four circle diffractometer Initial project start up is the same crystal selection and mounting are the same data collection is pretty much the same But the video camera requires some new commands and the crystal centering process is a little bit different As with the four circle diffractometer the first thing you do is to open the SMART application and start a new project as already described above Do this then keep the SMART window open e 14 File Edt Grab View Tools Window Hep ejaj fe tmt ale New image Zoomin Start Point Circular capture cursor grid Vector stretch cursor Fig 2 7 Top of Video window with explanation When you have finished setting up SMART you will start the video camera On your desktop there is a binocular shaped icon named Video Double click this icon The camera window opens At the top of the window is a menu bar Below that there normally is a toolbar If the toolbar does not appear open the View menu and select
61. Tables note that UNIX is case sensitive XPREP performs automatic crystal system and space group determination absorption corrections scaling and merging of different data sets index transformation reflection statistics etc XPREP reads the raw data file namem raw and the parameter file namem p4p both written by SAINT XPREP also prepares the command file name ins and reflection data file name hkl which are needed to run the other SHELXTL programs In addition all operations are logged to the name prp file while some experimental data are written in CIF format into the file name pcf At the end of the structure refinement the last file should be combined with the name cif file Li If you are repeating XPREP for some reason and this happens quite frequently do not forget to change the name of the previous prp file before you start It would be overwritten and in this way much important information about what you have been doing is lost just because you wanted to change a small thing in your data When you start SHELXTL open a New project for the last time At the very beginning you have to choose between the normal and huge version The normal XPREP works with up to 450 000 reflections which is usually enough even for very large structures If you have problems starting XPREP from the SAINT or SHELXTL window very likely some system variables are not set properly copy the program to the current working directory and ru
62. Toolbar Fig 2 7 shows the top of the video window with the toolbar open The functions of the most commonly used buttons are also indicated in the figure To start the video running click the button with an icon that looks like a dog eared page to open a new image then click the button with the black and green arrow shaped icon to start capture Options xl n Pixels Cancel Microns fico 0 ae Reticle Color Get Pete from Micron Scale Select Color from Table Drign Afp Gf B fico x aoo Pirols Reticle TR E y 2360 Pixels oo Major 1000 microns Select Origin Minor 20 0 z Camera Zero Offset Magnification Ange 0000 degees Lens 100 x Fig 2 8 The Video Tools gt Options dialogue box You can select the shape of the grid that appears on the screen Most people like the circular grid see Fig 2 10 below but you are of course free to select any grid you want The grid is often referred to as a reticle You can also set the orientation of the grid To do this select TOOLS gt Options menu In the box that appears Fig 2 8 you can type in the tilt in degrees The fixed chi angle is 54 7 so if you set the reticle tilt to 35 3 the grid x axis will be parallel to the phi rotation axis when phi is perpendicular to the optical axis of the video camera This tilted grid orientation makes centering more intuitive for most users When done click OK At this stage it is a good idea to verify that the center of the re
63. ab staff that you are going to measure they prepare the instrument and assist you Drink a beer not more than one or a cup of green tea Mount the crystal Center the crystal SMART Optical Project defining SMART New project Configuration Check for time and step SMART Rotation Trial MultiRun MATRIX collection SMART Unit cell or Matrix A chat with Helene while you wait for results 9 Unit cell determination SMART Index Bravais LS RLATT 10 Data collection planning ASTRO and or COSMO 11 Data collection SMART MultiRun 12 See you tomorrow 13 Unit cell for data integration SMART Threshold Pick Index Bravais LS 14 Crystal shape determination SMART Faces 15 Integration SAINT 16 A chat with Helene while you wai 17 Data preparation XPREP 18 From now on you are left alone with your destiny SHELXTL ONO aA RON t 28 23 Published without permission of the domestic Tourist Council because we do not include the Little Mermaid and other you must see in Copenhagen lf a Czech beer is not available use a local brand 25 Recommended by Emil Makovicky and a billion Chinese friends 6 If Helene is not available use the local lab technician 27 No beer or tea at this time 8 See footnote in step 8 103 Some useful definitions Rn 2 F F mean 3 F Rema Z o F 2 F Ri x F F F also denoted R4 2
64. advisable to check this field In the field Fractional lower limit to generate active mask try different parameter values usually in the range 0 1 to 0 3 to get a better active pixel mask According to the SAINT manual this has no influence with a CCD detector but according to George Sheldrick it is very important if the beam stop or low temperature nozzle throws a shadow on the detector Unfortunately we have had no opportunity to test it SADABS Change Instrument error from 0 005000 to zero Unless your computer is short on memory and especially if the number of rejected reflections given under the title Exceeding frame queue size is big compared to the Good 64 reflections written name _Is files increase Active frame queue half width to about 13 This will result in fewer reflection with large Lorentz factors being rejected If everything is properly set it is time to press the Integrate Sort Global button As you could see in the previous Figures this button is available in all three previously described dialogue boxes Whichever button you press all three operations will be performed sequentially using the current parameter setup If Enable global least squares refinement and Sort by Laue class in the Integrate dialogue box Fig 8 3 are activated and they should be there is no need to press Sort and Global in the Basic SAINT menu Fig 8 2
65. al gt New Project Crystal gt Edit Project etc Fig 2 1 XPREP will extract the chemical composition from the p4p file In step C the program calculates some very important data Mw Z Dx non hydrogen atomic volume F 000 and u All of them should have reasonable values for your sample consult an experienced person if you are in doubt Tip In many cases the non hydrogen atomic volume should be about 16 A but it can be as low as 12 when only very light atoms are present Conversely for compounds containing numerous heavy atoms the volume can even reach 25 A 74 In cases of high tretragonal or above symmetry it is a good idea to test for merohedral twinning option M somewhere in the XPREP run Before the last step and after the cell content description it is highly advisable to apply an absorption correction but for this you have to take your own initiative and choose from the program menu Different absorption corrections are described in chapters 4 7 and 4 8 pp 4 21 to 4 23 of the SHELXTL manual In XPREP they are hidden under option A and should be run between C and F There are two options FACE indexed absorption correction input A then F PSl scan absorption correction input A then P The second option also includes correction for laminar plate like crystals Pay attention to the intensity agreement index Rin before and after absorption correctio
66. ame every time a new frame is displayed Note A box displayed on the screen means that the reflection center RC is in the scan range of the current frame a circle means that RC is in the preceding frame a cross means that RC is in the following frame It is sometimes helpful to compare two or more up to four frames on the screen Go back to the FILE gt Display dialogue box Fig 3 1 and pay attention to the Quadrant window Default value 0 Full means that only one frame will be displayed at the time using full screen With Quadrant window opened it is possible to display half sized frames in one of the quadrants low left low right etc e If you are not sure about the current orientation matrix use o e e o e e 20 FILE gt Read p4p in order to check or replace the orientation matrix A command that is very similar to the FILE gt Display is ANALYZE gt Display gt HKLs at first sight not a very useful command since this time an overlay will be drawn only over the currently displayed frame You can use this command to adjust Spread i e width in degrees of the predicted frame but the default value 0 75 is a good choice However on occasion if the spots are very weak or very intense it is difficult to analyze displayed frames In this case it could be helpful to use the previous command for increasing decreasing the thickness Lineweight of the overlay
67. ams and containing results of crystal structure solution or refinement 4p VERY IMPORTANT FILES Parameter files containing crystal data unit cell parameters orientation matrix reflection arrays etc The file is often updated with additional data during MATRIX operations and after data collection pcf A part of CIF prepared by the XPREP program This file contains crystal system space group some experimental data etc in CIF format see also cif prj A file actually administrator prj containing Project names and located in c frames Projects or d frames Projects directories prp Login history and results of program XPREP Some files are in binary format and you will probably never find a use for them Taw rot sim _am _dk _br _ff fl 3 File containing reflections sorted as symmetry equivalents after integration by SAINT Contains practically all information about the reflections The format of the raw file is explained in SAINT help See files INTEGRATEHELP PDF or INTEGRATEHELP HLP Optional but common extension for a rotation picture obtained by ACQUIRE gt Rotation or CRYSTAL gt Evaluate commands in SMART Script file Active mask or active pixel mask created from the initial background frame program SAINT Dark current frame An old dark current frame file can be adequate for a MATRIX procedure However it is highly advisable to collect a new dark current correction file b
68. and refinement Just remember that the name hkl file contains F not F and corresponding sigma values XS is the program for structure solution by direct methods or Patterson synthesis XL is the least squares structure refinement program XP is the program for interactive molecular graphics XCIF is the program for preparing tables for publication from CIF XPS is the program for structure solution by fragment search XPRO is the interface for protein applications XWAT is the program that performs automatic water divining for macromolecules Tip Try to solve your structure by direct methods if the crystal consists of light atoms only organic compounds or if the crystal contains many heavy atoms that could have different atomic numbers Try to use the Patterson heavy atom method if you have one or two atoms that have significantly higher atomic numbers than other atoms Direct methods are usually more successful for noncentrosymmetric the Patterson method for centrosymmetric structures For beginners and crystallographers with not so much experience with the SHELXTL programs we recommend Allen Hunter s Youngstown State University X ray Structure Analysis Lab Manual A Beginner s Introduction The manual is available in the form of a pdf file from the web site http Awww as ysu edu adhunter Y SUSC index html Some worked examples can also be found at the same site Instead of the SHELXTL suite some of us readily u
69. and with a reasonable say I sigma gt 30 gt 50 or even gt 100 limit to accumulate a Reflection Array with several hundred up to 999 strong centered reflections with a C flag Then continue with CRYSTAL gt Index gt Bravais and gt LS commands as described above in chapter If matrix operation fails option no 1 Tip Because of the limited number and sometimes low intensity of reflections collected in the MATRIX procedure it is easy to overlook the correct crystal symmetry If after data collection it is not possible to reproduce the Bravais lattice and unit cell parameters obtained in MATRIX they were most likely wrong In such a case the true crystal symmetry is usually lower than that determined before and you should not be surprised Again attempts to find unit cell parameters sometimes fail even if you submit known unit cell parameters Sounds familiar If you get mad handpicking could help you cool down Just remind yourself of the procedure described in the chapter If matrix operation fails option no 4 and go on However the exact procedure depends very much on what tactic you would apply In some instances the unit cell parameters are defined prior to integration and will not be updated non merohedral twins or intergrowths of several phases are typical examples It is because one would expect many partially overlapping reflections during integration and then it is not recommended to let
70. at similar method called CAMEL JOCKEY 9 uses a trigonometric series expansion of the diffractometer angles but is little used since so many more experimental measurements are required The method most commonly used with scintillation counters is the azimuthal scan or psi scan method of North et al 10 This is the simplest method involving the measurement of the intensities of a usually small number of reflections with chi values close to 90 at different psi phi values An averaged absorption surface is thus computed and used to calculate the transmission factors It works remarkably well but is unsuitable for crystals with large uxr values Note In checkcif procedure if uxtmia is bigger than 3 0 tnia medial crystal size you will receive the message Alert A Crystal and compound unsuitable for non numerical corrections 3 Refined corrections These methods are the least scientifically valid methods in that they require a refined model hence the structure must be solved and isotropic a d p s should be known and provide a correction for the data based on the assumption that any consistent differences between the calculated and observed structure factors are due to absorption This type of absorption correction the so called refined corrections DIFABS 11 XABS2 12 and SHELXA 13 have fallen out of favor in the recent past The various programs differ in the exact mathematical functions used to model these differen
71. ately According to our experience this correction is not so simple as could be naively concluded reading the previous text and a very short report given in the manual By changing the uxr product and some of the other parameters as described below you can attain a significant improvement in your results Follow the procedure Chose option A then P and press Enter three times Type in 0 001 as the Tolerance for direction cosines 0 00000 Confirm default values for I sigma threshold 20 and Resolution threshold Ang 1 0 After the message showing number of reflections selected for the refinement confirm E as the default answer to the question Lamina L Ellipsoid E or Quit Q E and input actual uxr product Then the program calculates Rin values before and after corrections Accept the default answer to the question Print intensities Y or N N and chose R as the answer to the question Process data P Repeat R or Exit E P Circle through the E gt N R E loop changing the uxr product within reasonable limits say between 0 1 and 3 0 Generally speaking the uxr product should be increased for low absorbing and decreased for highly absorbing crystals Note the value that yields the lowest Rint After that choose Exit E and repeat the whole procedure several times with different I sigma threshold 20 test at least values of 10 and 30 and Resolution
72. best or not 2 Transmission factors Tmin ANd Tmax If the values are close to each other and or close to 1 the correction is very likely insignificant Too small values are not promising see below Psi scan corrections what to do The values of Tmin and Tmax and or their ratio should be close to the values predicted from crystal dimensions Those theoretical values will be calculated by SHELXL if you add the SIZE and ACTA commands to the ins file or in the checkcif procedure However we do not see any good reasons why Acta Crystallographica editors insist on such theoretical values since they do not account for specific experimental details and the exact crystal shape 3 Number of inconsistent equivalents It is easy to check this number since it is given in every SHELXL Ist file As a rule a smaller number means better correction although once again there is no direct relationship between the number of inconsistent equivalents and final R factors As usual SADABS is the best in this sense very often giving zero number of inconsistent equivalents but not always the best corrections as measured by the final results 4 Weighting scheme In ordinary circumstances the first parameter in the weighting scheme WGHT is small less than 1 and the second one tends to zero If the second parameter is 5 10 or even bigger carefully re check your data One possible reason for such a high value could be inappropriate absorption correc
73. bottom corner 63 xd Model profiles Beam monitor synchrotron systems ZK Vsigma lower limit for reflections used to update model profiles F Enable beam monitor normalization F Normalize all runs to the first Pari Fraction of model profile maximum used to generate simple sum limits Multiwire detectors Ht sigma threshold for least squares fit below vs simple sum above M Enable reference correction AA Resolution lower limit A above which simple sum is always used F Blend 9 profiles Output listing and diagnostic files l Generate diagnostic plot files Background and Active Mask I Keep temporary files Base 2 log of background update scaling factor X bebe b b XN p T Append listing file Ve 1 F Use pre existing active pixel mask Eponge Background file output frequen 100 Fractional lower limitto generate active mask fraction of average intensity 0 000000 Ehala Eala ahad geat plae Active mask _am file Bima Corrections to intensity esd s Instrument error fraction of intensity 0 005000 Spatial ct ST pana comecson Factor multiplying intensity esd s fi oo0000 F Apply spatial calibration correction from separate indexed fiducial _ix file Frames stored to monitor reflection overlap Spatial _ix file Browse Active frame queue half width Override selected input frame header information Crystal F Overide frame header Scan axis Frame width Starting exposure time hours 0 000000 2 theta Omega Phi Chi Batch Crystal In
74. butions most of the numerous other interventions and suggestions were smaller and will not be explicitly mentioned here Still they were important and we are grateful for having received them Dr Carpenter was kind enough to provide us with his instructions for using SADABS to send several MultiRun scan examples and to give many very useful suggestions Dr Ibers informed us about practices in his laboratory and donated a few tips on how to perform data collection integration and absorption correction in order to get better results Finally on a suggestion from Dr Hope the paper size of the document has been set to a combination of Letter US length and A4 European width Now all readers should be able to print the Guide without any tedious reformatting but this was only briefly tested for US standard 8 5x11 inch paper Nevertheless the paging has been optimized for HP1000 LaserJet 600 dpi If you do not know what this means see Post Scriptum at the very end of the Guide For printing on A4 paper aesthetic reasons suggest you set File gt Page Setup top margin at 3 25 cm and paper height at 28 69 cm Of course some other options exist including for example an entire reformatting while you wait for data collection to finish In some way Dr Hope Dr Carpenter and Dr Ibers are liable for the appearance of version 3 0 which was distributed in about 130 copies in June of 2002 Meantime several new programs or new version
75. c Explanation in our language bruka means shame scandal If you really insist the Guide is available from Dejan Poleti Dept of General and Inorganic Chemistry Faculty of Technology and Metallurgy Karnegijeva 4 I 11000 Belgrade Serbia and Montenegro E mail dejan elab tmf bg ac yu and Ton i Bali uni Geological Institute University of Copenhagen ster Voldgade 10 DK 1350 Copenhagen K Denmark E mail tonci geo geol ku dk Post Scriptum Since Word is one of Microsofts products its behavior is unpredictable and depends strongly on your printer and its configuration Always check the document in Print Preview mode before printing the Guide In addition it happens not so seldom that some figures float appearing at a completely wrong place or that many of the figures disappear leaving only a nice empty rectangle sometimes with a red sign similar to the letter x Advice complain to Microsoft We hope you will see these suggestions before you see a messily printed Guide
76. ces but all suffer from the same philosophical problem in that the data are being modified to fit the model Use them to discover discrepancies between calculated and observed structure factors If corrections are significant then you can suspect the quality of previous absorption corrections Remember that refined corrections are based on a very simple hypothesis Check everything experimental data structural model and try another correction program All methods and programs mentioned are available in the WinGX suite Some of them are diffractometer specific Enraf Nonius CAD 4 but some require only name ins as well as name hkl and or name fcf file and Bruker users can employ them with no problems In some cases it is necessary to run SMART SAINT Import Utility DATA gt SMART SAINT to convert and concatenate SAINT raw files first default file name is concat_hkl sortav Then use this file as input in the DATA gt KappaCCD gt Process Data menu where you can perform empirical SORTAV or Gaussian quadrature corrections for absorption We have no experience with data correction programs available in WinGX but if you are desperate Anyway let us know if the results are extraordinary A comparison of different methods using Na WO 2H 0O with u Mo Ka 19 mm as a model crystal is also given in the absorb htm file As you already know from the previous chapters Bruker AXS Pack offers several options for absorption corrections Th
77. check re check your data 73 After that you will see the main menu which depending on the stage should be more or less similar to Fig 10 1 If you follow the program s own suggestions for the sequence of data preparation steps you will be guided through the usual necessary operations checking the reduced cell and a possible higher lattice symmetry determination of the space group extinctions calculation of the reflections statistics a suggested unit cell content with determination of calculated density and linear absorption coefficient as well as preparation of the ins and hkl files So ina typical XPREP run you will follow the order H Search for HIGHER metric symmetry S Determine or input SPACE GROUP D Read modify or merge DATASET C Define unit cell CONTENTS F Set up shelxtl FILES Q QUIT program Be ready to see many submenus and sub submenus and to make proper choices or input some requested data Fortunately the default options indicated by XPREP are the best choices in most cases XPREP Ver 6 11 W95 98 NT 2000 ME Copyright Bruker AXS 2002 Current dataset Wave length Original cell Esds PgUp PgOn scrolls text only graphics window may be resized Fig 10 1 XPREP window showing the main menu Keep on hand the chemical composition of your crystal you need it for option C However if the chemical formula was entered at the time you defined the current project in SMART Cryst
78. circle diffractometers f Redundancy On a conventional point detector the intensities of reflections and their angular j adjustments are very precise This is not the case with an area detector and individually collected f data could be of lower quality Nevertheless area detectors have many advantages data collection is very fast j a complete diffraction picture is recorded and a wrong initial assignment of unit cell parameters could be easily corrected impossible with a conventional detector j it is common practice to collect symmetry equivalent reflections up to the whole diffraction f sphere i By collecting many symmetry equivalent or identical at different 2 theta angles reflections E which means great redundancy the quality of the final data set containing averaged reflections will approach that from a point detector Therefore it is really worth spending time measuring equivalent reflections R hl amp Bolte 2000 showed that wR significantly decreased up to a redundancy of about 6 and after that slowly increased At the same time R and R sigma f j decreased in the whole interval and did not reach a minimum even for a redundancy as high as 25 j Of course the extent of possible redundancy depends on crystal symmetry and unit cell volume it correlates with crystal to detector distance too In the case of triclinic and monoclinic F crystals data collection strategy must be very carefully pla
79. compares your entries with some pre defined reasonable limits and warns you if a value is outside the expected range Press Validate from time to time especially if you have changed many default values Check if Laue class corresponds to your crystal we suggest you leave Lattice centering as P primitive except if you are quite certain you have a centered lattice It is always better to check the integrated reflections for absences and assure yourself about those that are missing Have you really looked through all frames and checked that no missing reflections are present Tile Laue dass Lattice centering Resolution limit for output Cell parameters C 2 theta degrees A 8 6212 Alpha 30 000 C sin thets lambda 1 angstroms B 26 0533 Beta 119 246 d spacing angstroms C 8 8055 Gamma 30 000 Integration fles Maximum wait for frame file seconds f0 00000 Storting Frame Filename of Frames Matrix p4p Filename Output Filenome IDA Dejon kirkob kirkeb0 001 Browse 650 D Dejen kikeb kirkkebl 001 Browse 100 Dejan kirkab Browse 100 De kak Browse 100 Deji kot 4 00 Browse 100 Browse 0 More integration files Increment last run integrate Sort Global Yalidate Qpen listing fle Help Close Fig 8 2 Basic SAINT menu for analyzing dialogue box The value in the Resolution limit for output field means that reflections with
80. cond or third or even fourth crystal If the attempt to find unit cell parameters fails then you can a Choose to spend your time with something easier than crystallography or b Find the chapter If Matrix operation fails somewhere below or c Ignore your results and perform data collection anyway 13 Three circle platform diffractometer If you are using a platform diffractometer with fixed chi and a video camera this section is for you Be aware however that only operations specific for this type of diffractometer are described here Therefore your idea to skip the first part of this chapter was totally wrong Most SMART diffractometers are built with the phi mechanism set at a fixed chi mounted on a platform diffractometer The diffractometer is shown in Fig 2 6 Fig 2 6 Three circle platform diffractometer The phi rotation mechanism is housed in the block that sits on top of the platform off center The phi rotation axis is tilted 54 74 from the horizontal In case you wonder the 54 74 angle is the same as the angle between a body diagonal and a face diagonal in a cube when the two diagonals begin at the same corner This design allows the exploration of most of reciprocal space and affords convenient access for add on attachments such as a cooling apparatus In short you get the functionality you normally need but with a simpler machine than a four circle diffractometer The typical SMAR
81. correction Nevertheless if the crystal is very small or needle or plate and opaque it is very hard to index and measure its faces properly with the Bruker optical system on the P4 goniometer Keep in mind that the shortest distances are the most critical Fortunately in this case absorption effects are not very pronounced and the other techniques may give satisfactory results Psi scan corrections Ellipsoid corrections This method of empirical absorption correction in XPREP should give accurate results if your crystal has an ellipsoidal like shape It could also do well if you have a crystal with irregular shape that has well developed faces provided that it is not a thin plate It is not necessary to collect real psi scan data since the program itself selects suitable sets of reflections for refining the absorption parameters to define a pseudo ellipsoidal surface which is further used to calculate corrections For ellipsoid corrections you have to input the uxr product where u is the linear absorption coefficient and r is one half the mean crystal dimension Therefore note the u value calculated by XPREP calculate the mean crystal dimension before the run and make sure you keep cm and mm straight for both the measurements and the units of u According to the SHELXTL manual during this calculation the smallest crystal dimension should be heavily over weighted and at the same time the uxr product does not need to be determined very accur
82. could be useful to select another possible asymmetric unit This can be done with SYMMETRY gt Increment Ctrl Shift l shortcut Try rotating about Z with phi at chi 0 until the asymmetric unit is inside the Ewald sphere with its leading edge just behind the detector projection By making rotations of phi and omega the usual rotation angles it will be possible to find a good starting position Keep in mind that phi is left handed while omega is right handed In this position a left handed rotation about Z would move the asymmetric unit through the sphere Next perform a scan Select Omega Scan from the STRATEGY menu Drive the scan angle omega or phi with the STRATEGY gt Omega AlttNum2 shortcut or STRATEGY gt Phi Num3 shortcut Move the scan axis until the entire asymmetric unit has just passed through the detector projection Then select STRATEGY gt End Scan and the scan will be stored in the run list You can check and edit the run list using STRATEGY gt Editruns The possibilities of ASTRO are enormous If for example you add a new run into the run list and press Enter a new group of toroids automatically appears on the screen Therefore it is not really necessary to run the simulated goniometer With a little experience you can guess a good scan add it into the run list and see the results If you are going to collect a full sphere of data or as much as possible the best way is to have a full phi scan and suppleme
83. d is designed to adjust this slide so that the crystal is well centered after repeated 180 rotations Positions A B and C D are rotated to each other by 180 around chi the first two hold the goniometer head pointing up the latter two pointing down For crystal centering we recommend the following procedure Start with the A button depressed and activate the AXES PRINT button to drive the goniometer to the starting position Adjust the crystal both in the horizontal and the vertical directions note the appropriate drivers on the goniometer head Then depress B followed by AXES PRINT After this the crystal should be clearly visible it is now in focus so you can adjust it precisely in the horizontal direction The extreme left and right points of the crystal should be equidistant from the center of the microscope cross keep the graduated scale horizontal it can be rotated by gently turning the microscope eyepiece do not push the microscope sideways By repeating AXIS PRINT check the center point If it exactly coincides with the vertical line of the crosshair the crystals outermost points should be equidistant from the center after the 180 rotation If the crystal is now shifted to one side adjust it in the opposite direction for half of the displacement and note the position of the center of rotation Repeat AXIS PRINT with B depressed until the crystal stays on the axis Now depress A and do the final centering in this position b
84. deration in ASTRO or COSMO So after briefly playing with suitable ASTRO COSMO commands it should be possible to find an adequate strategy for data collection In our laboratory in Copenhagen such a library is located in directory c frames and will be described later Even if you decide to use some old scans always check completeness and redundancy of data by ASTRO or COSMO ASTRO ASTRO is an interactive and useful graphics program for determination of an efficient data collection strategy during MultiRun scan ASTRO is excellent for educational purposes too The program uses the well known Ewald sphere concept in order to visualize and calculate the volume of data in the scan The area detector is presented as a projection on the surface of the Ewald sphere It is possible to simulate goniometer movements as well as to display and record the segment of the reflecting sphere that passes through the detector projection It is also possible to calculate the completeness and redundancy of data corresponding to the currently displayed or input scans and of course in accordance with the chosen Laue group Indeed many possibilities exist However only elementary operations will be described here When ASTRO is started you will see a graphical screen containing much visual and textual information in different colors Sometimes it is not easy to orient oneself in all those data and the table above could be useful In addition on the right side of
85. e GONIOM gt Manual then click OK Click in the Video window Select the vector cursor see Fig 2 7 Because the diffractometer is in manual mode you can adjust phi and perhaps omega with the manual box for best viewing You measure a desired length by clicking at the starting point then move the cursor to the end point Below the lower right area of the picture is a window that shows some geometric parameters among them is the length you want Write it down and continue measuring until all desired dimensions are done At this stage you should quit Video On some systems there seems to be interference between Video and the SMART display Rotation photo You are now ready to take a phi rotation photo Most of the time you can skip this step without loss of information The crystal quality information can just as well be obtained from MATRIX frames however by taking a rotation photo you can save some time in the case of bad crystal The Rotation procedure is similar to that described above Rotation is reached by Acquire gt Rotation A menu similar to that shown in Fig 2 2 will appear Set the exposure time to 60 180 s and set the other parameters as shown in the figure except that making chi 54 74 has higher aesthetic value it is probably preset anyway Click OK See above for other aspects of the rotation photo and compare Figs 2 3 and 2 11 Note that the axis of the picture is tilted 35 3 relative to the hor
86. e ANALYZE gt Graph This command is very much like the preceding ANALYZE gt Cursor gt Vector Vector start and end points are fixed in the same manner When you have specified the cursor position the program plots intensities along the cursor path in an X Y graph Fig 3 2 a of intensity versus vector line The right side output contains details very similar to those displayed after execution of ANALYZE gt Cursor gt Vector e 22 The command ANALYZE gt Graph gt Rocking draws a graph of the so called rocking curve or reflection profile The graph shows the integrated intensity of a specified rectangular area as a function of scan angle Execution of this command requires that you have a contiguous series of frames available for input The rectangular area is chosen by a Box like cursor see the ANALYZE gt Cursor gt Box command described above You have to be careful to choose the correct box size in the appropriate dialogue box or to correctly resize the box on the screen The other important parameter is Frame halfwidth which defines the number of frames on either side of the current frame to be included in the curve It should be at least three but preferably more If it exceeds the number of frames available for processing an error message will appear Just start again and define a lower number for Frame halfwidth but a curve obtained with Frame halfwidth parameter less than three is not
87. e Readers mainly beginners like we were in ancient times three years ago described the document as very helpful and much better than the original documentation No doubt about it they can be judged guilty for putting that foolish idea into our heads The challenge was big and we finally decided to offer the Guide to all users The revised text containing many new recommendations and description of the meanwhile installed version 6 02a of SAINT was prepared during the summer of 2001 in September about 70 copies were distributed worldwide as version 2 0 Once again the resonance was excellent but the number of suggestions and remarks was lower than we expected Very likely the term beginners in the title repelled some experienced users and kept them from reading the text However three members of our crystallographic and Bruker users community made great and un selfish efforts to improve the content of the Guide They are Dr Hakon Hope Department of Chemistry University of California Davis USA Dr Gene Carpenter Department of Chemistry Brown University Providence USA and Dr James Ibers Department of Chemistry Northwestern University Evanston USA Dr Hope edited the text in order to bring it closer to standard English He also made the Guide more universal by writing completely new sections with a description of the Bruker three circle platform diffractometer with the video camera In comparison to these two contri
88. e and you are done If it is not find the midpoint between the current tip position and the current origin Then move in a vertical direction on the screen in your mind until you reach the x axis This point should be the correct origin From the Tools menu select Options Click the Select Origin button then OK Position the cursor at the determined origin and click the left mouse button Be careful not to click prematurely e e 16 Repeat the whole procedure to double check that the origin has been properly set When done return omega and phi to the default start positions by pressing D then Axis Print on the manual box Remove the calibration pin from the goniometer head and store it in its proper place Now is the time to mount your crystal Select the best crystal you can find and attach it to the mounting fiber Position the mounting pin in the goniometer head Make sure the video window is active An example is shown in Fig 2 10 Fig 2 10 Screen image of a crystal The crystal is not yet quite centered Center the crystal on the phi axis and adjust the height The procedure is similar to that described above in steps 1 and 2 Because you know the true origin of the reticle it is not necessary to rotate omega When you are done the center of gravity of the crystal should remain at the origin at any phi value At this time you should measure the crystal dimensions Click in the SMART window and select manual mod
89. e black lines on white background contrast and brightness adjustments are also needed or play a game with the colors in EDIT gt Colors or CRYSTAL gt Faces menu The colors are defined using the RGB red green blue convention for example 000000000 means black 255255255 is white 255000000 is red 255000255 is violet etc If you decide to use another instrument like an optical goniometer for crystal face indexing and measurement then you can add face description directly to the p4p file which is also in ASCII using any text editor The format is FACE 0 00 1 00 1 00 0 055 FACE 0 00 1 00 1 00 0 055 etc i e the format is identical to that in the Crystal Faces window Fig 9 2 Type in these lines just before the reflection list In this way you can save some time and avoid typing errors when using XPREP for face indexed absorption correction 72 10 SHELXTL PROGRAMS The SHELXTL suite consists of several major modules of which XPREP program for reciprocal space exploration should be used first The last version of XPREP should be available from the official BRUKER site as a part of the SHELXTL suite However to obtain an up to date demo but fully functional version the latest one is 6 13 you can go to the web site http shelx uni ac gwdg de axs In that case you need to specify user axs and give the password 7 characters which is the last hexagonal space group no 194 in the International
90. e close to zero Once again these parameters indicate quality of the crystal data or misalignment of the detector Crystal translations were not refined in this example 38 Nonlinear refinement StartingRes FinalRes GOF Cycles Code 2 21751E 01 1 74753E 01 15 67 2 0 This line contains starting and ending residuals goodness of fit number of refinement cycles and a completion code GOF value higher than 10 indicate that the refinement is not so good If the code is not 0 rerun refinement with negative constrains The list of all reflections in the Array follows Flags H K L Swing Omega Phi Chi X Y Inorm I Sig 1 ACHS 3 0 1 340 000 330 252 0 000 330 000 236 74 208 85 6954 5 106 00 01 00 0 003 0 18 0 44 2 ACS 4 2 2 340 000 329 051 0 000 330 000 287 50 140 04 3994 7 80 02 38 01 1 214 3 71 10 90 truncated 5 ACS 4 1 3 340 000 337 153 0 000 330 000 287 44 140 04 2597 6 65 0253314633 3 877 5239 EAL 6 ACHS 2 6 0 340 000 334 444 0 000 330 000 221 89 402 64 3227 5 72 01 01 02 0 313 0 54 0 77 truncated If you at last have obtained good unit cell parameters answer to the Frequently Asked Question the meaning of good will be presented in the next edition of this Guide take a not too long break skip all other options and jump to the chapter Data collection or data acquisition strategy Chances for success almost better than none If you came he
91. e commands are Left button drag rotate the lattice Middle button drag backward or forward zoom in or zoom out Ctrl Alt keys left button emulates the middle mouse button Right button drag pan the lattice Ctrl key left button emulates the right mouse button The keyboard commands are The arrow cursor keys rotate the lattice around the X 1 and Y lt axes Rotation around Z is done with the Insert and Delete keys Note that X is horizontal on the screen Y is vertical on the screen and Z perpendicular to the screen Gray and signs means zoom in and zoom out respectively Press the Shift key to speed up rotation or the key to slow down rotation Available shortcut keys are F1 rotate 90 about X horizontal on the screen F2 rotate 90 about Y vertical on the screen F3 rotate 90 about Z coming out of the screen Shift F5 F8 save the current orientation 24 F5 F8 restore a saved orientation RLATT is extremely useful and you should run the program as often as necessary to check the regularity of the reciprocal lattice and the values of the unit cell parameters If everything is OK you should see regular rows parallel to the reciprocal axes The phrase as often as necessary means each time you determine or re determine the unit cell parameters First of all in case you see curved rows or even duplicated spots do not blame your crystal or y
92. e corresponding data will be extracted from the frames themselves Fi Reflection disposition y Rot I D e found 250 9 133 5 363 7 3583 6 1 4 o p on first or last frame 1 182 4 323 5 368 1 10448 114 1 9 c in X exclusion range 1 251 2 133 2 361 4 2308 53 1 4 Cc fa outside radius limit 0 176 0 211 2 371 5 8951 105 2 0 C outside res limits 0 155 4 203 6 371 2 7388 96 2 2 B lt I sigma threshold 209 105 9 419 3 368 1 3512 66 2 0 C spots too small 0 183 5 322 9 373 6 2587 56 1 9 C bifurcated 3 83 9 413 0 36 3 2619 57 2 Cc satellites 1 7 173 7 374 7 10244 113 33 S spots too large 0 182 8 325 0 363 6 4819 tf oh E with 1 L lt 1732 0 105 9 418 365 8 3008 61 2 0 E Total useable 8 84 0 411 8 364 9 2578 S 2 2 E centered for IS 8 5 8 393 5 362 8 1 5 3 0 20 Total unus 214 471 0 47 7 365 8 1065 36 08 10C XY spot size 1 471 1 47 6 372 3 618 8 0 8 10C Stored XY spot size 1 0 471 0 47 7 370 1 237 17 0 8 10C Z FWHM spot size 1 53 14 374 5 50 8 0 8 10C Stored Z FVHM spot size 1 53 495 5 15 68 8 42 7 08 10C Z FWHM gt frame vidth NARROW frames x 495 5 14 2 367 2 17 S 0 8 10C v Fig 5 5 Threshold output Print means print on printer Write write data to disk Threshold output written directly to the screen is not complicated Fig 5 5 Number of observed centered accepted etc reflections is listed first The Stored XY spot size and and Stored Z FWHM spot size b
93. e library of standard scans if you do not then prepare it and if you accept the strategy recommended by George Sheldrick then COSMO has many advantages over ASTRO For example some of them are good graphical presentation showing the contribution of every series of scans ability to sort runs Sort Runs for Completeness in cases where crystals are sensitive to X rays etc Besides when you are satisfied with the result press File gt Export Strategy the file produced is in standard format and will be read in SMART Edit MultiRun window Fig 6 1 with no problems Do you remember that ASTRO and SMART MultiRun files are not interchangeable In a limited test we have found that COSMO is as good as ASTRO when it calculates a possible number of unique reflections and completeness However the mean redundancy although not perfect is always estimated better Another reason for using COSMO instead of ASTRO 56 8 DATA REDUCTION STRATEGY Ugh X ray data have now been collected and you can finally proceed to the next very important stage usually called data reduction It is time to click on the SAINT icon Wrong In most cases in order to obtain the best quality data you have to re refine the unit cell parameters using SMART and all intensities available after the data collection step Unfortunately this time it will not be done automatically as it was in the MATRIX procedure So run the CRYSTAL gt Threshold comm
94. e more than one trial Also give a short title that will appear on each page of the diagnostic file 16 g value for use in su k o I g lt gt where k is a scaling factor o I originally comes from SAINT but in the final stage is substituted for l where is the corrected intensity 79 Short lt 21 chars title for Postscript plots Test Three pages of plots are then constructed automatically but not yet displayed and you are given a choice whether to add pages of scatter plots Spatial display of I lt I gt su greater than 3 0 0 for none for which we find no use However according George Sheldrick quote there are several fairly obvious things that can show up there a an image of the beamstop because the active pixel mask is wrong b a shadow of the low temperature nozzle same reason c rings of constant resolution caused by ice on the crystal d hot spots caused by bad pixels if they always show up in the same place they should be entered into the bad pixel table and e red in the middle and blue at the edge or vice versa bad floodfield e g using a MoKa floodfield with an AgKa X ray tube Overall scale and R int variations for Test Normalized Scale Factor Smoothed R int Run Frame gt View the diagnostic plots before continuing with SADABS by clicking the icon for the program GSView Then open and study the diagnostic file b
95. e perhaps slightly more readable Just in case you are curious and relaxed the fundamentals are published in W Kabsch Evaluation of Single Crystal X ray Diffraction Data from a Position Sensitive Detector J Appl Cryst 21 916 924 1988 SAINT requires one or more sets of frames collected for example by MultiRun a spatial and a flood correction file _ fl and _ix a dark current frame _dk and at least one p4p file created by SMART Any operation can be interrupted with Ctrl C Unfortunately SAINT does not recognize projects defined in SMART therefore the first operation must be to open a New Project Fig 8 1 Input Project name why not use the same as defined in SMART and select Crystal type protein or small molecule Then find the directory containing your data and click on the corresponding p4p file A new subdirectory work is offered as the default choice for the output directory We support the idea of separating experimental data from integration results and probably from the results of the crystal structure determination In this version of SAINT the information about the file names and basic integration parameters should be appropriately extracted from frame headers p4p files SMART ini and default SAINT initialization files Saint ini and SaintCL ini That will be done when you run Initialize from the SAINT menu Therefore press Initialize every time you open a New Project
96. e pin with the tip attached Adjust the goniometer head so the image of the tip just touches the y axis y points more or less in an up direction of the reticle Then use the perpendicular slides of the goniometer head to center the tip so it appears stationary when phi rotates it is not yet necessarily on the reticle x axis You control phi rotation from the manual control box Pressing AXES PRINT alone causes phi to rotate 180 To rotate phi 90 depress the currently undepressed button C or D and press AXES PRINT Make sure the tip of the needle remains stationary during a 180 phi rotation when you are done with this step The needle tip is now positioned on the actual x axis It may or may not coincide with the reticle x axis If the actual and reticle x axes coincide the needle tip is at the origin skip to step 4 If needed position the reticle x axis correctly Go in Video to the Tools gt Options dialogue box click the Select Origin button then OK Carefully position the mouse pointer at the tip of the needle then click Be careful not to click prematurely The x axis should now be correctly set and the needle tip should be at the current origin Next you will determine the position of the y axis and with that the true origin To do this you rotate omega 180 by pressing A and then Axis Print After the omega rotation note the position of the tip on the screen If it is still at the origin then everything is fin
97. e run do e ACQUIRE gt Hemisphere Enter jobname and click OK Let the diffractometer collect the data set Further processing is described below in chapter Data reduction strategy 7 It is possible to perform a phi scan on the three circle diffractometer and there are cases where it may give better data e g for a strongly absorbing needle like crystal mounted along phi However generally there are many disadvantages of phi against omega scan 48 7 STRATEGY PLANNING TOOLS The simplest way to prepare a data collection strategy is to skip this annoying step Of course you must have some parameters in the above described MultiRun or Hemisphere Run dialogue box but it is always possible to leave the values that have been who knows when and why written in the smart ini file as defaults The results will most probably be questionable Also you can utilize blindly some old scan data from a previous data collection for a crystal with similar symmetry and unit cell parameters Not a bad concept but still not fully recommendable since orientation matrices would not be the same However it is a good idea to prepare a library of MultiRun scans that should be coded according to detector distance and frame widths That can make your life much easier especially if possible scans are related to the crystal data may be in your notebook Data from the library could be used later at least as a starting point for consi
98. eflections Rin Rasma No of observed reflections Criterion for observed 1 gt o CRYSTAL STRUCTURE SOLUTION AND REFINEMENT Crystal solution method a Direct b Patterson c Other Refinement on F F or I Treatment of occupancy factors Treatment of hydrogen atoms R Ri a No of refined parameters wR wR a No of reflections used in refinement S GOF No of constrains used in refinement Restrained GOF No of restrains used in refinement A O min Weighting scheme A O max w AP min A Ap max A Extinction factor Extinction correction method if applied Scale factor Source of atomic scattering factors Int Tab C 95 INSTRUMENT DATA Diffractometer type Bruker AXS 4 circle Radiation source fine focus sealed tube Wavelength A A MoKa 0 71073 Monochromator graphite flat Collimator mm Measurement device CCD 1000K area detector Number of pixels 512 x 512 Active detection area 6 25 x 6 25 cm Detector distance cm COMPUTING Computing data collection Bruker SMART Computing cell refinement Bruker SMART Computing of integrated intensities Bruker SAINT Computing data reduction Bruker SHELXTL Computing absorption correction a XPREP from SHELXTL b SADABS c other Computing structure solution SHELXS97 Computing structure refinement SHELXTL97 Computing molecular graphics Computing publication material Bruker SHELXTL ADDITIONAL DETAILS 96 Key for understanding SAINT outp
99. efore data collection see Data Collection Strategy below These files are regularly located in c frames ccd_1k folder The dark current frame used is also automatically saved in the working and data directories The usual convention in our laboratory is to give an eight digit name in the format timeyymmdd _dk For example 60000703 _dk means 60 seconds of background collection taken on 2000 07 03 3M990722 means 3 minutes dark frame collected on 1999 07 22 Brass plate image a calibration file forget it Flood field image file a calibration file forget it Flood table flood field correction or flood correction file Similar to the dark current frame file but containing pixel by pixel uniformity correction Does not need to be updated so you can almost forget it However verify with the system administrator or some other experienced person that you are using the correct one Initial background providing the starting point for preliminary background refinement in program SAINT File containing table used to transform pixels from raw to corrected values information only forget it File containing table used to perform reverse transformation with respect to the _if file information only forget it File containing indexed fiducial spot positions for spatial correction It does not need to be updated See the note for _fl files Log file present only if login option is on Intensity integration results log file of S
100. efore deciding what to do next Fig 11 1 An example of the first page of SADABS eps diagnostics file Runs scans are delimited by vertical lines The first plot in Fig 11 1 shows the variation of the normalized scale factor as a function of run and frame number it is sensitive to the value used for the restraint esd entered in Step 1 It should be smooth with small high frequency ripples If it is too smooth the esd should be increased slightly on the next run looser restraint on the variation between adjacent scale factors if it is too noisy the restraint esd should be reduced slightly The second plot of R int against frame number Fig 11 1 may be useful in showing that some part of the data is poorer than the rest but usually there is no change to be made because of this plot The third plot page 2 of diagnostic file shows R int and R sigma as functions of resolution The data should be cut off on the next run where the R sigma curve begins to rise sharply say somewhere around 0 75 A for the example shown in Fig 11 2 80 Intensity statistics for Test R int blue R sigma red Rint oc R sigma E P ant 40 20 15 1 2 1 0 0 9 0 8 Resolution d Fig 11 2 An example of the third plot from SADABS diagnostics file R bag o S a IE2 L for general hkl S p S N 40 20 15 1 2 1 0 0 9 0 8 Resolution d Fig 11 3 An example
101. ell determine the Bravais lattice perform a least squares refinement of the resulting unit cell and finally write the new cell parameters into a p4p file 11 Choosing proper values in the corresponding dialogue box Fig 2 5 should be possible by the checking procedure described above but be sure that some skilled crystallographer is on hand In short the values are Frames number of frames in a series preferably between 20 and 60 Frame Width preferably between 0 2 and 0 3 Seconds frame preferably from 10 to 60 s but it could be even 180 s for very small or poorly diffracting crystals Indexing HKL tolerance and LS RLV tolerance should be 0 1 and 0 01 respectively for expected excellent quality crystals or 0 2 no more than 0 25 and 0 02 respectively for expected problematic cases x Frames 60 Frame width 0 3 Seconds frame 45 Title Orientation Frame Job name matrix Max display counts 1 Lower axial limit 3 0 Upper axial limit 200 0 Indexing HKL tolerance 0 2 LS RLY tolerance 0 01 Fig 2 5 ACQUIRE gt Matrix dialogue box If you did not change Job name why should you the collected frames will be named from matrix0 001 matrix1 001 matrix2 001 to matrixO xxx matrix1 xxx matrix2 xxx where xxx is the number of frames in each series defined by the Frames argument If the MATRIX operation finishes successfully you will obtain initial cell parameters
102. en defined 50 The left and right arrow keys can be used for rotating the viewpoint around the screen Y axis vertical With the up and down arrow keys it is possible to adjust the zoom up zoom in down zoom out The best way to familiarize oneself with the ASTRO graphical window is to turn different items on and off using shortcuts from the previous table You can start with an empty window and then make any combination of items Finally select the red green stereo view with the EDIT gt Stereo Mode command shortcut S put on red green glasses and play with the cursor keys in order to obtain a good viewing angle Also try different views using commands from the VIEWPOINT menu The data required to run ASTRO are crystal orientation matrix which should be read from the corresponding p4p file with FILE gt ReadMatrix Orientation matrix can also be entered manually using EDIT gt Configuration command In addition you have to check and possibly adjust some configuration parameters X ray wavelength Crystal to detector distance desired Resolution this directly influences the coverage calculation Frame Width and Frame Time Finally it is important to know the Laue group class of the crystal which is determined during the Bravais lattice command in SMART If you are not quite sure about the Laue class use the lowest symmetry group for the corresponding crystal system Of course it is also very
103. end of chapter ASTRO 9 000 28 00 00 00 00 3 300 1 20 00 0 000 28 00 14 00 00 307005 23 300 1210 20 00 9 001 28 00 00 00 4005 3 300 1 20 00 1 000 28 00 28 00 00 60 00 3 300 610 20 00 9 002 28 00 00 00 00 3 300 1 20 00 2 000 28 00 2 00 00 60 00 2 300 166 20 00 3 000 28 00 2 00 90 00 60 00 2 300 166 20 00 9 003 28 00 00 00 00 3 300 20 00 21 Because the detector is too heavy many people do not like to change the detector position 2 theta angle during the data collection 100 Three circle platform diffractometer Columns are Run Frame 2 Theta Omega Phi Chi Axis Width Frames Time The settings in the following section will cover at least a hemisphere They can be used for crystals of higher symmetry 1 3 hours 0 001 30 00 30 00 0 00 54 70 2 03 00 61 30 00 1 001 30 00 30 00 90 00 54 70 2 03 00 44 30 00 2 001 30 00 30 00 180 00 54 70 2 03 00 23 30 00 3 001 30 00 30 00 270 00 54 70 2 03 00 15 30 00 4 4 hours 0 001 30 00 30 00 0 00 54 70 2 01 00 182 30 00 1 001 30 00 30 00 90 00 54 70 2 01 00 131 30 00 2 001 30 00 30 00 180 00 54 70 2 01 00 69 30 00 3 001 30 00 30 00 270 00 54 70 2 01 00 50 30 00 9 5 hours 0 001 30 00 30 00 0 00 54 70 2 0 500 364 30 00 i 001 30 00 30 00 90 00 54 70 2 0 500 261 30 00 2 001 30 00 30 00 180 00 54 70 2 0 500 138 30 00 3 001 30 00 30 00 270 00 54 70 2 0 500 90 30 00 4 0
104. ery heavy atoms and too small size for highly absorbing crystals From the results published recently by Ruhl amp Bolte 2000 follows that narrow 0 2 and 0 3 4 mm collimators should be avoided since with all other conditions kept constant they give low f quality data and the worst R indices This can be attributed to the inhomogeneity of the collimated beam The best results are obtained with a 0 5 mm collimator but a 0 8 mm collimator yields very i similar results It is interesting to emphasize that crystals much larger than the collimator pinhole can be successfully used for data collection for example if it is risky to cut a smaller piece G rbitz 1999 Ruhl amp Bolte 2000 However the greatest surprise was the conclusion that better results are j obtained for larger crystals R hl amp Bolte 2000 It seems that the effect of a higher number of f scattering centers more than compensates for the negative influence of absorption This is in accordance with the authors belief that highly diffracting crystals would give good data under nearly all experimental conditions so the most important thing is to obtain strong reflections j j The reader must be warned here that low diffracting and low absorbing u lt 0 2 mm organic crystals with oxygen as the heaviest element so called small organic molecules were used in the R hl amp Bolte 2000 and G rbitz 1999 studies Bu
105. est old diffractometers Commands that allow you to check create or edit a run list are AQIURE gt EditQuad AQIURE gt EditHemi and AQIURE gt EditMulti The utility programs ASTRO and COSMO described in the next chapter could be used to optimize the run list for different crystal to detector distances wavelengths hardware configurations and crystal symmetry and unit cell parameters See below and the next chapter for more detailed discussion of this subject Several examples and default values provided and described in the SMART manual pp 2 1 to 2 5 could be helpful Each of the commands mentioned has its own separate run list but they are very similar and always contain the following data Fig 6 1 Run Frame 2 Theta Omega Phi Chi Axis Width Frames Time Run is simply the run number that will be used for the series Frame is the starting frame number in the series 2 Theta and Chi are goniometer angles for the frame series scans Omega and Phi are goniometer angles at the start of the frame series scans In the next Axis field 2 for omega or 3 for phi scan should be specified depending on which of them should be changed during the run Width is scan width in degrees of each frame to be acquired Frames is number of frames collected in the series while Time is exposure accumulation time for each frame in seconds As seen in Fig 6 1 the first two series represe
106. ey are numerical face indexed Gaussian in XPREP and empirical psi scan in XPREP multi scan in SADABS corrections This should be enough since the refined corrections cannot be recommended in any event It is first of all necessary to learn how to recognize good or suitable absorption corrections Several parameters are generally utilized for this purpose They should always be combined otherwise you can easily come to the wrong conclusion The parameters are listed in order of appearance and approximately increasing importance It is assumed that your hkl file contains the complete set of reflections including all collected equivalents 1 Rint and or R values Most programs for crystal structure solution and refinement calculate these values Of course in a typical procedure you will already have compared the values in an XPREP or SADABS run If after correction the values are not significantly lower something is wrong with your data or with the applied corrections In a good correction you could expect that R R values will be halved and even more gain is possible in case of high 85 redundancy and high linear absorption coefficient However there is no simple or linear relationship between Riny R and final agreement indices meaning that for example a lower Rint Value does not always result in a lower R value SADABS almost always gives the lowest Rin R values no matter whether the other indices are the
107. fference between these values and Z should be smaller than the program suggests e it is advisable to increase Z significantly Finally you can double click the SAINT icon You will see an almost empty window with an encouraging message No project opened and the toolbar containing Project SAINT SADABS XPREP Utility with programs PLOTSO FRAMEREN SUMMARY and XYTOANGS and Edit menu All specified programs could be started from this window meaning that SAINT is only a traffic regulation utility Here only SAINT will be described in detail SAINT performs all necessary steps to reduce raw crystallographic data frames taken with SMART to Lp corrected intensities their standard uncertainties and HKL s The operation includes determination of background active pixel mask and spot shape profiles integration of spots and elimination of bad spots arrangement of reflections by point group equivalents preparation of reflection statistics final refinement of unit cell parameters and if necessary time decay correction Do not try to understand how the program really works although it would be good if someone did Maybe its authors are not quite sure either black box is the best and also a very popular description for SAINT The SAINT manual is a very good example of how to prepare a very bad text Even for experts reading the manual is a true nightmare The help files usually located in the SAINT32 subdirectory ar
108. frames x frame width x time No of reflections in Reflection Array and criterion J o Initial cell parameters a b c a pe i Bravais lattice Step two data collection and new unit cell parameters SMART Coverage and redundancy from ASTRO Resolution required A Estimated nos of possible and collected reflections coverage Redundancy as average minimum maximum Temperature K T Dark frame 94 Type of scan MultiRun Hemisphere Quadrant add conditions printed on a separate peace of paper Number of frames Frame width Measuring time s Collection method and scan No of reflections in new Reflection Array and Criterion o New cell parameters a A b A c A a bOs yo Step three integration data reduction and final cell parameters SAINT and XPREP Spot sizes determined by Threshold command as X Y Z Spot sizes used for integration as X Y Z 20 range for final lattice parameters No of reflections for lattice parameters and Criterion J o Decay correction No of standards used reflections and interval Decay Decay correction parameters Absorption correction a Empirical w scan b Integration Gaussian c Other Absorption correction parameters Tmin Tnax Rin before Rin after No of measured reflections Omin Omax Amin limax kimin kmax Lmin li No of independent unique r
109. g frames with the help of the following commands that belong mainly to the ANALYZE menu ee First of all choose the desired frame to be displayed on the screen with FILE gt Display ANALYZE gt Display or simply Ctrl D Sometimes as shown in Fig 3 1 it is necessary to enter the full path together with a frame name Actually if you choose any frame from a set of data the whole set becomes available for analysis For example if the current frame is name1 009 all frames from name1 001 to name1 xxx where xxx is the number of frames in set 1 can be analyzed Leave the other parameters in the dialog box as they are Images of the next or preceding frames can be obtained by pressing Ctrlt and Ctrlt lt respectively Tip When one frame in a series is displayed Altt and Alt lt runs rapidly like a movie through the series a really nice review of collected data x Frame filename c peteribc9p6 matrix0 048 Level Low counts 1 High counts f Region Quadrant 0 Full Magnification fi X Center 256 Y Center 256 l Autoincrement l Preserve Graphics I Use orientation in header M Add HKL Overlay Fig 3 1 FILE gt Display dialogue box If the lattice has already been determined you can check the small Add HKL Overlay box in the File gt Display menu dialogue box An overlay of predicted spot positions that uses the current see below orientation matrix from a p4p file will be drawn on top of the fr
110. get to define New Project later or some instrumental corrections could be out of date A project can also be defined using CRYSTAL gt Copy Project and CRYSTAL gt Config To Project commands Once defined the project can be activated using CRYSTAL gt Switch To Project Center crystal optically with GONIOM gt Optical Ctrl O shortcut or CRYSTAL gt Evaluate The latter command automatically performs optical alignment and takes a rotation picture For some slightly obscure reason which we experts wish to keep to ourselves but see Preliminary notes it is better to use the first option Do not forget to note crystal dimensions maximum intermediate and minimum during the optical alignment operation In any case a dialogue box appears with some pre defined diffractometer angles First of all check that they correspond to the goniometer used From time to time and for e 7 unknown reasons SMART forgets the pre defined base angles For our P4 goniometer they should be 2 theta 0 omega 0 phi 60 and chi 30 or 330 Then press ENTER and find the manual box on a sidewall of the diffractometer cage When using the GONIOM gt Optical command with default angles marked A B C and D on the manual box one of the slides on the goniometer head is always perpendicular to the viewing direction The 180 rotation around phi when activating the AXIS PRINT button while keeping the same angular button depresse
111. h I sigma level otherwise there could be too many the limit is 999 This is what you would then use for a really good orientation matrix and then apply it for the integration in all runs allowing for update Even in this case it could be worth while or necessary if the results are well outside of recommendations given below and in Extract to repeat the integration 57 But what is a really good orientation matrix As we already stated above the Reflection Array for a good orientation matrix should contain several hundred not fewer than 500 and not more than 999 the upper limit reflections from all parts of reciprocal space Never start integration unless the RLV tolerance is smaller than 0 002 or even 0 0015 Discard all reflections that have an omega error larger than 0 2 and positional X Y errors larger than 0 5 pixels Also pay attention to the GOF value it should be as low as possible preferably about 1 0 or even lower 3 For the incurable skeptic fan of Agatha Christie For each of the data collection runs you should collect separately the best reflections a few hundred if possible and prepare the orientation matrix in a separate p4p file Of course during the integration for every run you will supply this run s orientation matrix obtained in SMART Regardless of the actual crystal system you would apply no constraints triclinic and you would keep the same option for updating the orientation matrix in SAINT and
112. h families edge on not an easy task Measure the second interplanar spacing crystal lattice vector call it b If you have some applicable angle measuring device you can try to measure the angle between the two plane families vectors gamma on the screen Now turn parallel to the second family of planes with the help of the cursor keys one click turns exactly one degree or with F1 or F2 a 90 degrees rotation until you see another good family of planes edge on Counting the clicks with the cursor keys measures The interplanar spacing of the third family gives the c parameter angle measuring on the screen gives a Positioning the third family of planes vertical or horizontal and then turning by cursor keys until the first family of planes reappears edge on gives a measure of y a can now be measured by rotating the first family of planes until the second reappears again If this was a triclinic case let s hope you remembered to keep your axes oriented properly right handed and know how to use the elementary lattice mathematics to calculate the final values 25 Fig 4 1 Examples of the RLATT screen output a A triclinic reciprocal lattice with measured distance between two layers b The same lattice with added divisions showing that the measurement is not so good Pay attention to the super structure effects indicated by the presence of weak reflections between the main layers Therefore the true c axis is doubled wi
113. h updated and sigma l N appears during the face indexed correction By default the file obtained has the extension abc and is suitable for input into SADABS I At the end of an XPREP run option F do not accept the default answer when the question Do you wish to over write the intensity data file name hkl N appears Type in Y es otherwise a new hkl file will not be written After the run inspect the XPREP log file name prp This file contains all XPREP results The XPREP output file name ins can be directly submitted as input for crystal structure determination by direct methods Among the numerous XPREP features RECIPROCAL space displays option R Contoured PATTERSON Sections option P or Generate powder pattern option D from main menu then option G from submenu should be noted as being very helpful e g for checking the lattice symmetry during structure solution by the heavy atom method or if you had enough material to collect a powder diffraction pattern The ideal powder pattern will be generated if you submit an ins or res file as input for XPREP The other programs listed below will not be described in detail We assume that you are an experienced user of the SHELX programs that have been around in so many versions for over 30 years Of course you can use a name hkI file prepared by XPREP and any other 75 program for crystal structure solution
114. have loaded a file the brightness of screen spots corresponds to the reflection intensities it can be adjusted with the slide bar on the right side of the screen While reflections from raw or hkl file are always cyan reflection from a p4p file are coded according to the ACSH flags flags are described in the next chapter white reflections with all flags magenta missing C and H flag yellow missing A flag red missing C flag green missing H flag blue missing S flag This list is also available from the Help gt Quick Color Reference menu but read the RLATT manual for the order of priority A lot of colored spots usually implies an incorrect cell or a composite crystal Reflections shown on the screen could be toggled or filtered Toggling reflections on and off is simple just press the first letter of their color Filtering Filter menu is a little more complicated but offers more options reflections could be filtered by any combination of flags indices and resolution parameters If you do not like the blue default background color change it from the Display gt Background Color menu It is however more important that from the Display menu you can choose to show direct space reciprocal space or laboratory axes The projection can be rotated in real time by dragging the mouse with the left button depressed or with the cursor keys Just open the Display menu and check Rotate mode The mous
115. he sequence listed in the CRYSTAL gt Redtn Cell submenu of SMART window It means that you have to execute CRYSTAL gt Redtn Cell gt Bravais or Ctrl B This operation determines possible Bravais lattices that will be ranked according to the symmetry and goodness of fit GOF Normally this command is invoked after the Index and before the Least squares operation but all three commands store their results in the same p4p file Therefore unit cell parameters and their standard uncertainties are always up to date and available to each other and it is not always necessary to follow the listed order E g if after indexing the standard uncertainties of the lattice parameters are too high and their accuracy too low it sometimes causes an enormous list of possible Bravais lattices many of them totally stupid One can overcome this by first performing LS on the reduced cell parameters lowering the standard uncertainties e g by excluding reflections with too high RLV error and then trying Bravais The number of solutions will be drastically reduced Lattice type P X Max sigmas any 20 0 Max sigmas best 6 0 typ Input Unit Cell Current A axis 4 036 Current B axis 18 954 Current C axis 23 798 Current ALPHA 90 000 Current BETA 90 000 Current GAMMA 90 000 Input Standard Deviations ESD of A 0 001 ESD of B 0 004 ESD of C 0 005 OK Cancel Fig 5 2 Bravais Lattice dialogue box The program will suggest the best so
116. ical mode It might be better to wait with the precise determination of crystal dimensions until you believe the crystal is satisfactory i e after the MATRIX operation 4 SMART is not as smart as one could expect 5 The left hand buttons on the manual control box are usually labelled with both letters and angles e e e e 8 After the crystal has been centered lock the goniometer head slides and recheck the crystal centering Then press Esc to quit the operation Close the doors adjust until all lights are green and press the Reset button Again set the diffractometer angles to zero GONIOM gt Zero or F10 shortcut Confirm Yes If necessary raise voltage and current to the desired values by GONIOM gt Generator Adjust the voltage first Both parameters have to be increased in steps of 5 until the normal values for our system 40 kV and 35 or 37 mA have been reached Mo tube Perform GONIOM gt Home Axis which checks the calibrated position controls of the goniometer circles so that the reported positions are exactly consistent This command should be repeated four times to home all four axes on the P4 goniometer however since default values are listed sequentially it is enough to press Enter to advance to the next axis Exposure time i Zero 2 theta omega Y N Check for yes Starting phi 0 00 Chi 0 00 Max display counts 1 Suppress correlation Y N I Check for
117. icate a strong correlation between frame width and exposure time For example two measurements one with 30 s time and 0 3 scan width and the other with 10 s time and 0 1 width yielded nearly identical R values It means that very similar data quality could be expected for a constant scan a speed s i Our diffractometer refuses to collect data with a time longer than 50 seconds too much work probably Thus if the choice is for example 60 s the diffractometer halves the time Of course the trick is to double the time and to put 120 instead of 60 s in Scan MultiRun list Fig 6 1 45 It can be concluded that a frame width of 0 1 is probably too small as well as that the j f default frame width of 0 3 is probably a little too big In normal everyday experiments it is advisable 3 to use 0 20 to 0 25 frame widths If the selected frame width is 0 2 and the crystal is of optimal i thickness it would be possible to collect the data using 30 20 s time in most cases Crystal to detector distance Because the data quality depends on the reflection intensities it i is reasonable to expect that shorter distances are preferable In a series of experiments with crystal to detector distances varied between 3 0 and 9 0 cm Ruhl amp Bolte 2000 obtained the best results j for 6 0 cm However the differences in R values for 3 0 4 5 and 6 0 cm were practically negligible f although at shorter distances
118. imentally In cases of severe absorption however they are the only really effective methods The spherical and cylindrical corrections also provide numerically accurate corrections for crystals adopting the requisite morphology but are less useful since grown crystals are not usually spheres or cylinders Crystals may be ground to spheres but many materials do not survive such treatment 2 Semi empirical or in short empirical methods These methods provide an approximate correction based on an experimentally determined absorption surface The semi empirical methods rely on making extra intensity measurements The multi scan method of Blessing 5 is of most use when there is a large redundancy in the data set Equivalent intensities are analyzed in terms of a multi polar spherical harmonic expansion and the method is implemented in the programs PLATON 6 SORTAV 7 and SADABS 8 This program requires the data set to contain an extensive set of equivalents and is primarily of interest for data collected using area detectors which routinely have a large References for this chapter are listed at the end of the Guide see Literature 84 degree of over determination redundancy It models the absorption surface using a spherical harmonic series based on differences between equivalent reflections and will probably not be very effective for triclinic and monoclinic crystal systems since these have fewer equivalents A somewh
119. important to consider collision limits at the specified crystal to detector distance Although it is possible to use slightly broader goniometer collision limits for our diffractometer they are defined in degrees as Positive 2 theta limit 110 Negative 2 theta limit 110 Positive omega limit 45 Negative omega limit 45 Positive omega minus 2 theta limit 40 Negative omega minus 2 theta limit 20 Both phi and chi limits are set to 0 meaning free rotation around these axes You can see the limits using EDIT gt Limits or GONIOM gt Limits commands in SMART or using STRATEGY gt Limits command from ASTRO Do not change the limits defined in SMART and put the same values if necessary into ASTRO They should be kept in mind when you are planning a data collection strategy Now you are ready to continue with ASTRO hunting Select different resolution limits in EDIT gt Configuration and note how the unique volume increases Input shorter crystal to detector distance and you will se how the detector projection increases too Required resolution min is a very important parameter here It represents a lower limit in d values for observed reflections and is related to the upper limit of the theta angle in the Bragg equation A 2dminS N Omax The best resolution that usually can be attained with Mo radiation is about 0 6 A Omax 35 2Onax 70 For an explanation we can imagine that the unit cell is cut into many slices each
120. in the list will be restored from these files Very likely you will make a hard copy of the screen showing your crystal A negative of the standard display with a white background would be preferable For this you should leave the Faces menu don t worry the picture of your crystal will not disappear and go to EDIT gt Config gt Colors Choose PRINT LUT as the Color table filename and there Fig 9 4 you are Finally run FILE gt Print or FILE gt Save or both and do not forget to change Color table filename to the default when you are done 12 This is one of the smartest parts of SMART 71 SMART Bruker Molecular Analysis Research Tool V5 054 Coppi 1997 98 BrukecAXS Joj x Fie Edt Crystal Acque Analyze Goriom Detector Level User Help Co phen pyr je2ca0 631 37702 02 15 18 33 Created 07 02 02 Phi 157 50 width 7 Counts 101 ard Time s Distance 3 850 Size 12 Jelcal d ir n jecal Distance FloodFld Spatial ar S1z Gain Fig 9 4 SMART screen showing crystal habit and faces a negative of the standard display Tip Tip with screen contrast and brightness further adjusted using Ctrl T shortcut Crystal shape is not very legible with the default Color table in use In order to improve the picture you can leave the Crystal faces window and adjust contrast and brightness to get white lines on black background or use Color table filename PRINT LUT to always se
121. iometer see page 4 4 in the SMART manual these settings usually appear as default a OK Cancel Print Write Read Fig 9 2 Crystal Faces window After acceptance of the settings the Crystal Faces window is opened Fig 9 2 In the beginning the list is empty Unless by some miracle or previous measurements on an optical goniometer you know the shape of your crystal skip it by pressing OK which brings you to the final window Fig 9 3 in which the shape will be presented graphically Of course in the beginning this window is empty too and do not expect to see your crystal before you define a closed polyhedron isi SMART Bruker Molecula Analysis Research Tool V5 054 Coppi 1997 98 BrukewAXS AAE Project Jelcal DataDev d DataDir n jecal WorkDev g WorkDir n jecal 9 008 2 10 562 2 11 558 2 4 6 4 8 5 Distance 3 FloodFld 423117 Spatial h423117 10020630 o Siz Gain 52 4 el top face indices 1 75 1 00 1 21 Omg Phi Chi Psi 0 00 320 00 330 00 113 14 Fig 9 3 SMART screen showing crystal habit and faces original colors Here you can use various shortcuts described at the top of the window to analyze the shape For example you can rotate the picture around the phi axis using the left and right arrow keys and rotate around chi not omega as written on the screen which is valid for a three circle diffractometer using the Page up and Page down keys It
122. ion html http www msg ukans edu xraylab notes twin html http www cryst chem uu nl lutz twin gen_twin html http www xtlox ac uk twin html http www chem umn edu services xraylab twin_workshop pdf In addition to many useful definitions this material contains numerous illustrations of twinned structures and some worked examples Perhaps you will need not right now to read them From a practical point of view twins are classified in several groups Merohedral twins including racemic twins belong to the TLS class 159 of the 230 space groups are candidates for merohedral twinning Pseudo merohedral twins where reflections from two or more twin components almost exactly overlap Common examples are triclinic crystals with two angles close to 90 monoclinic crystals with B 90 with a b or with a b and B 120 orthorhombic crystals with a b etc Non merohedral twins belong to the TLQS class They are usually recognized already after the MATRIX procedure by the simultaneous presence of sharp and split spots This group 19 Giacovazzo H L Monaco D Viterbo F Scordari G Gilli G Zanotti amp M Catti Fundamentals of Crystallography International Union of Crystallography Oxford University Press Oxford 1995 p 83 90 also includes reticular pseudo merohedral chameleon twins observed in the case of rhombohedral crystals The best way to deal with twinned crystals is
123. is no doubt better On the other hand refinement of the structure of glycyl t serine u 0 134 mm crystal size 2 2x2 0x0 8 mm G rbitz 1999 showed how effective SADABS could be in reducing Rint values even for very large crystals e g crystal bigger that the collimator pinhole especially when there is great redundancy Nevertheless ten sets of data collected under different experimental conditions with crystals of different sizes and treated not treated with SADABS revealed no significant variations in molecular geometry and atomic displacement parameters These results agree with one of our studies on a large crystal size 1 15x0 14x0 13 mm with approximately ten times bigger u value When running SADABS you have to choose the resolution limit twice first for the parameter refinement and second after inspection of the diagnostic file Fig 11 2 for data truncation At first sight it seems reasonable to input the same value used for data truncation of course in both places However in most cases the choice does not have a significant influence on the results of the structure refinement Therefore confirm the default values 0 1 in the SADABS run and truncate reflections if necessary with the SHEL command during structure refinement by XL or SHELXL 88 Epilogue The aim of this chapter was to clarify the role of absorption and the corresponding corrections in single crystal structure analysis If we have not been very s
124. is also possible to rotate the picture by moving the mouse with the left button depressed 70 All the time the program automatically finds Miller indices of the plane that can be seen horizontal in your microscope and is oriented upwards Thus to measure the distance of the plane from the center of the crystal you should turn the scale in the microscope vertical While the Faces window is activated you can drive the goniometer angles manually from the manual box Normally you need just phi and chi for a P4 goniometer Choose the circle with a button in the lower left of the box and drive it by the right side buttons for forward and backward slow and fast rotation When a plane is in the proper position horizontal in the microscope and pointing upwards press the AXIS PRINT button this displays the Miller indices of the plane together with the corresponding angles in the lower part of the screen The plane can be entered in the list and displayed graphically on the screen if you now press Enter You are asked about the plane indices first here you can enter the correct value for the plane distance from the center of crystal The distance should be in mm so do not forget to convert the reading from the microscope scale The scale in our microscope has been calibrated by a precision micrometer and the conversion table for the microscope reading can be found at the operator place Tip If the screen picture of your crystal is too small multi
125. ity of Utrecht and can be seen obtained by visiting the web site http www crystal chem uu nl distr euhedral These programs achieve optimization by comparing the intensities of equivalent reflections but why not try to do it by an optical inspection You could start from a combination of real of hypothetical faces the more the better drive the goniometer to the appropriate position and measure the distance of the surface from the crystal center Or drive the crystal around and define a lot of faces until the constructed figure is close to what you see Read in the following how to do it i Before you start the measurement of morphology you MUST have the lattice orientation defined in p4p file otherwise the program could not give you Miller indices of faces Fastslow speed ratio 4 Base Omega 0 000 Base Phi 30 000 Base Chi 30 000 Fixed telescope VIN F Check for yes Telescope angle 0 000 Eyepiece rotation 0 000 Une color 255150000 Fig 9 1 Options for Crystal gt Faces dialogue box e gt The measurement is accomplished through the CRYSTAL gt Faces submenu This opens a dialogue box Fig 9 1 with settings Note that they are quite different for platform three circle and P4 four circle goniometers In the second case use the following Base Omega 0 0 Base Phi 30 0 Base Chi 30 0 Fixed Telescope unchecked Telescope Angle 0 0 69 Eyepiece Rotation 0 0 In the case of a platform gon
126. izontal direction on the screen 17 Fig 2 11 Examples of rotation photographs obtained on three circle platform diffractometer a low quality crystal picture supplied by Dr Marilyn Olmstead b good quality crystal note the weak odd layers Unit cell determination Initial determination of the unit cell is essentially the same as explained above so it will not be discussed here The commands are e gt ACQUIRE gt Matrix gt OK gt Options A typical value for the number of frames entered under Options is 20 If the crystal is very small or it diffracts weakly a larger number 36 40 can be used Be sure to save the results as described for the four circle diffractometer 18 Notes on low temperature data collection In the picture of the platform diffractometer Fig 2 6 you can see the low temperature nozzle above the crystal The figure below is a close up photo of the region near the crystal Fig 2 12 Arrangement of mounting pin and cooling nozzle In the background is the face of the detector Top pointing down is the tip of the nozzle for the cold stream The outside of the tip is held at room temperature by electric warming No shield gas is used The goniometer head holds a copper mounting pin with an empty glass fiber The pin is made from copper to allow enough heat to be transferred to the tip of the pin to prevent frost In the lower middle of the picture one can see a small amount of fog around the c
127. keseddel Copenhagen s MEMO cceecceeeeetteeeeeenteeeeeeeeeeees 102 Some useful definitions ce ccesccccceceeeeeeeeeeeaeeececeeeeeseeceeaeeeeeeeessneseseseeeeeeeeneess 103 LtGrature y csssvevetece EE ea aie ed a elas dered A E eee eta eve eee me siearedied 104 Acknowledgement seriinin eiiaoe a kaa KEEKO a EN TEAR iE eia 105 Availability of the Guide niire rsi niaii Ronee eh A aii eee i aeaa 106 POSE SCM CUI ss eee ETET Rae ac CAT eA CAE a CAN ee UNE en geiees 106 1 1 DIRECTORIES folders NAMES AND EXTENSIONS OF BRUKER FILES The files used by a project normally reside in one of three directories working data or calibration directory The calibration directory contains the detector parameter file and different calibration files This directory is shared by all projects For everyday work it is important to remember that all measured dark current files _dk see below are located in this directory from where they can be reused The other files in the calibration directory are used automatically and rarely require updating or need no updating at all The location of files in the other two directories could be slightly confusing The working directory contains files written during crystal orientation and unit cell determination that is when ACQUIRE gt Matrix ACQUIRE gt Rotation CRYSTAL gt Unit Cell etc commands are run The data directory is the destination for frames collected during scan runs e during
128. l nothing could stop you anyway os o 39 This option means that you will try to use already collected data frames but with a new set of reflections that should be added to an empty Reflection Array In a sense the effect should be similar to editing of the Reflection Array if some spots below the selected limit are deleted manually You can run the command Reditn is an abbreviation for Redetermine CRYSTAL gt Redin Cell Ctrl R shortcut This command sequentially performs all operations needed to go from one or more series of collected frames to the orientation matrix and unit cell The steps included are clear the Reflection Array threshold the frames to build a list of spots autoindex perform least squares on the reduced primitive unit cell determine Bravais lattice perform least squares on the resulting unit cell and write the necessary data to a p4p file However in keeping with our step by step philosophy we recommend you start with CRYSTAL gt Clear in order to empty the Reflection Array Confirm the size of frames 512 that will be used in subsequent operations Then think fora moment and click Yes when the question Erase Reflection Array Are you sure appears After invoking the CRYSTAL gt Clear command be sure that all potential errors have been eliminated which means that all appropriate corrections have been updated and applied dark X Y Dist In Configuration The
129. l value If you press Constraint mask on the right side a new window opens giving you an opportunity to define constraints for one or several variables Usually only crystal translations are constrained as indicated in Fig 5 3 by code 512 Tip Sometimes you can speed up refinement and avoid wandering into false minima by constraining the system in the early stages Similarly if the least squares procedure seems to go in the wrong direction try to refine only part of the parameters at a time for example detector corrections only or cell dimensions only If you are measuring crystal after crystal with the same detector distance then it is a good idea to determine the 6 detector corrections by measurement of a standard crystal which should have accurately known lattice parameters and could be easily centered You would then refine only detector corrections and use them subsequently for other crystals as known parameters with code 768 You would have to repeat the standard measurements regularly to assure that your corrections do not change by virtue of some black magic We have ourselves rarely attempted standard measurement this way only for two cases of complex large crystals Usually the lattice parameters come out satisfactory and detector corrections refine to reasonable values even with only limited MATRIX data Tip The correction for the crystal to detector distance as calculated by SMART and SAINT is strongly correlated
130. lections i e reflections already present in the Reflection Array perhaps with some above mentioned editing check the number of reflections in Array Remember a maximum of 999 reflections can be stored in Array 25 is an absolute minimum for indexing and in accordance with BRUKER recommendations a number in the range 50 100 is about optimal at this stage All reflections should be centered e have a C flag Delete all non centered reflections and all reflections with I sigma less than 30 or even 50 if you have too many suspect reflections in the Array An alternative approach is to identify hot spots using RLATT as described above and to remove them from the Array After that you can go to e CRYSTAL gt Redin Cell gt Index or Ctrl X which is also called Autoindexing This command uses reflections stored by Threshold or by manual picking in the Reflection Array to determine indices a rough crystal orientation matrix and a reduced primitive cell The autoindexing procedure is based on an algorithm operating on difference vectors which are classified in groups derived from the input reflections If the procedure with difference vectors fails the program tries indexing on full vectors If you really believe that you know the true unit cell parameters maybe they were determined previously e g with the program RLATT or you think you know them by some providence or those from MATRIX were good and you are in trouble af
131. les generated from different runs of INTEGRATE e g on different series of scans of the same crystal or on different crystals of the same kind Another use is to change point group symmetry if the point group specified during integration was not correct The last operation will also be done by XPREP see later so that as a beginner or even as an experienced user you will need SORT only rarely and in special cases Choice of parameters Fig 8 5 is simple intuitive and does not require special explanations Global least squares orientation matrix refinement menu This program is useful if you need to re refine unit cell parameters e g if Code parameter at the end of namem _Is file is not zero and you do not wish because it is not necessary to repeat the integration Here a binary _ma file from the last integration should be used as the input file Fig 8 6 Also enter the name of the p4p file used during the INTEGRATE run in Used matrix p4p file box whereas the choices of Output matrix p4p file and Listing file names are free Check Frame width degrees and Frame size pixel per side values during INTEGRATION these values are properly extracted from the frame files but in a separate run they can be wrong Global Refinement Fig 8 6 SAINT Global least squares orientation matrix refinement dialogue box Fig 8 7 SAINT Filter dialogue box 66 67
132. line sample b a polycrystalline sample containing several big and several small crystals c a good single crystal sample d the same as c but with photo converted to a film like view Note that the last picture is not produced in SMART but by screen capture and further handling After making a preliminary choice also make a preliminary scan of 10 frames Continue as follows Open the ACQUIRE gt Edit MultiRun menu and edit to get only one line 1 001 00003 0 3 10 t t should be your estimated exposure time per frame in seconds After that start ACQUIRE gt MultiRun give the name Trial Test or something like that this would be the name of the frames and press Enter When the operation is finished check the collected frames for appropriate exposure time Maybe you are not quite satisfied with the intensity of the spots and wish to increase the exposure time or perhaps the spots are so intense and nice that you can use a shorter time But do not go below 10 seconds per frame If the spots tend to appear on only one or at most two sequential frames you have to decrease the step size called Frame width in SMART Increasing the step size to more than 0 3 is not advisable even if some spots continue through many frames After choosing the step size make the following calculation N 160 number of spots x selected step size This gives you an estimate of the number of frames N needed for a go
133. lines A value of 1 5 or 2 really helps At the same time you can try to adjust screen contrast and brightness using EDIT gt Contrast or Ctrl T shortcut Dragging the mouse left or right with the left mouse button depressed changes the contrast Drag to the left to increase the contrast and to the right to reduce it This is analogous to a contrast control achieved by changing aperture or sensitivity of a photographic film Dragging the mouse up and down changes brightness analogous to different exposure times of a photo Notice changes in the color scale on the right side of the SMART screen Note It is possible to use the arrow keys or a combination of arrow keys and the Ctrl button for the same purposes Release the left mouse button to exit and save changes press Esc or the right mouse button to exit and reset color mapping to the default level Very similar to the FILE gt Display command is ANALYZE gt Load but there you can optionally make and display a linear combination of two frames Header of the current frame can be viewed using ANALYZE gt Frame Info Any square area on the display can be zoomed using ANALYZE gt Zoom when a crosshair cursor appears Once the center of the area to be zoomed has been selected press Enter or release the left mouse button in order to see a pop up menu from which you can choose the magnification factor 1 2 4 8 and 16x or Exit Press the right mouse button to escape from the
134. lue of g is determined later in step 2 However if the new value is very different from 0 04 this stage should be repeated e Restraint esd for 0 005 15 w 1 o7 N g lt i gt where lt l gt is corrected mean intensity o o o o o o o s 78 This value is almost always in the range 0 002 to 0 005 but could be higher in the case of highly absorbing samples the best guide is appearance of the Postscript plot see the explanation below Suitable spherical harmonics orders 6 3 in the case of moderate absorption but should be 4 1 for weakly and 8 5 for strongly absorbing crystals Special treatment of plate like crystals N It seems that this option does not work at least in versions 2 03 and 2 05 In a limited test using version 2 03 answer Y es always resulted in a small increase in the final agreement indices Number of refinement cycles 15 Sometimes it helps to increase this number to 20 The program then prints the number of reflections employed effective data to parameter ratio the initial value of R int refines parameters against the selected data and reports the resulting value of R int Normally you accept option A this result and proceed to the next step Repeat parameter refinement R or accept A A Step 2 Reject outliers and establish error model The error analysis continues and correct standard uncertainties s u for the co
135. lution meaning the highest symmetry lattice whose goodness of fit is better than an input threshold But you are free to choose any of the 33 listed solutions Finally Reflection Array will be automatically re indexed according to the selected unit cell A Bravais lattice window is shown in Fig 5 2 Unit cell parameters and their standard uncertainties Current A axis Current ALPHA ESD of A are automatically filled in with the values supplied by the Index procedure If you insist but it is not necessary you can set Max sigmas any and Max sigmas best fields The first parameter is the GOF limit for the listed solution so that all solutions with GOF above this value will be rejected It is hard to imagine that a solution with GOF gt 10 could be correct The second parameter is the GOF limit for the best solution and solutions with GOF lower than this value are potential candidates The best solution is flagged with a gt gt gt sign in the Bravais lattice output Leave Lattice type as P of course after Indexing which determines a reduced cell However the program also runs on centered lattices if e g you changed your mind after accepting some centered solution An example of a Bravais lattice output is listed below with added comments in italics As usual some lines were omitted Sigmas for acceptance of any soln best soln 20 0 6 0 Input lattice type P Input cell and standard deviati
136. ly achieve a completeness of 100 and the desired redundancy Answer the question is there any difference between astronauts and cosmonauts 54 COSMO is the first Bruker program or maybe the second if we also include GEMINI with a worthy Users Manual containing even a short but informative Tutorial The manual is available in pdf format therefore print and read it first Here we will describe only a few characteristic points and give some recommendations on how to use COSMO COSMO 1 42 Copyright 1999 2002 Bruker AXS Inc Completeness 9 Redundancy Distance fem 3 85 d Ar 0 80 5 00 Wavelength A 0 71073 26 52 75 Mosaicity 0 50 sin 6 A 1 A 0 63 4 00 Bijvoet Pairs merged Laue Class 1 soul 3 00 Lattice Type P anh 2 00 afAy 9 01 a n 114 34 b A 10 56 BI 92 94 c A 11 55 Yi 114 52 1 00 172 137 120 109 10 095 090 086 083 Total Reflections 6356 Unique 3510 afAl Predicted Reflections 6356 Collected 0 Completeness O incomplete axial data Redundancy Runs 7 Frames 2800 100 00 5 00 Disk Space MB 727 75 00 4 00 0 00 3 00 Current Target Priority Completeness 97 99 100 25 00 2 00 Redundancy 2 60 4 00 20 Time h 13 18 24 00 5 Strategy Custom mall 0 00 2 1 00 T T z TE Tr lt 1 072 219 365 512 659 805 952 1099 1245 Refine Strategy 4 Time h Dataset completeness and redundanc
137. me for the second integration Always check Crystal translation constraints in both fields If you start with a carefully prepared orientation matrix it is a good choice to leave Crystal system as triclinic unconstrained for Periodic orientation matrix updating allows for possible small inaccuracies in your instrument and crystal mounting but immediately choose the right crystal system for the Post integration global refinement Check Crystal translation constraints in both fields Even in this case we recommend repeating the integration The only exception is for triclinic crystals since no significant improvement can be expected in that case Anyway integration must be repeated if the results do not agree with recommendations given in examples of output files or if the results are well outside the expected values or if the reflection size was wrong or if Finally there are some girls guys who determine the detector corrections with a standard crystal before they collect actual data We bet they would choose to check all the upper six fields as constrained Detector center X Detector center Y Detector pitch etc together with Goniometer zeros and eventually Crystal translations the lowest fields SADABS Uncheck Enable correlation filter Leave the other parameters as they are default values are a good choice at least for the first integration and press Advanced integrate button somewhere in the left
138. meters For Max pixel separation we still have not found what it can help but Raw count threshold could be changed to an estimated background level helps sometimes in case the reflections are rather weak The value 1 lets the program guess on its own You could try to navigate through some of the frames with ANALYSE gt Cursor and mark the typical counts for the background and the diffraction spots Then use a value that is high enough to exclude all of the background and low enough to register the spots Minimum and Maximum excluded X are used to exclude spots falling close to the vertical plane of the primary beam in standard MATRIX runs it is close to the left edge of the frame check with the shadow of the beam stop These spots belong to reciprocal lattice vectors that pass very slowly through the Ewald sphere and are less accurately centered Radius cutoff and Minimum Angstroms could be used to exclude regions which are without reflections anyway Maximum Angstroms cuts reflections with low diffraction angles less accurate for lattice parameters Wonder if you can get out something by manipulating these values but a smaller number in the Minimum Angstrom field helps to get more reflections in the Array It seems useful to decrease the temperature factor its default value is highly overestimated but the effect of decreasing the value is practically negligible Leave default number 9999 for the last six parameters It means that th
139. ms least squares orientation matrix refinements Sometimes you can enter a bigger value here in order to take larger fluctuations into account However what you choose under Periodic orientation matrix updating and Post integration global all data refinement depends on your tactic as explained at the beginning of this chapter Note that when low temperature data collection is used there is no need to redetermine the orientation matrix or to apply decay correction If you fix the orientation during the integration then uncheck Enable periodic updating and do not worry about the rest under it Supposedly for the same reason you keep a fixed orientation you would not expect to supply a better centered set of reflections from integration than you already have used and you would uncheck Enable global least squares refinement and not worry about the rest here If you are the hasty type and have only an approximate orientation matrix then do the following For the first integration find and choose Triclinic or Unconstrained in both Crystal system fields For the second run leave Triclinic in the Periodic orientation matrix updating middle field but choose the proper Laue class or check Constrain refinement by Laue class in the Post integration global all data refinement field right side Do you still remember that the file namem p4p should be submitted as Matrix p4p Filena
140. n The value of Rin Should be significantly lower after correction Do not forget to calculate the new reflection statistics after the correction Press or confirm in sequence D Read modify or merge DATASET S Display intensity STATISTICS A Merge ALL equivalents then compare the columns A int and R sigma before and after correction The corresponding tables are also logged into a prp file Here we emphasize that a face indexed absorption correction is preferred from a physical point of view Read more about this and detailed instructions in the Notes on Absorption Correction below It is important to note that the quality of correction depends also on the accuracy of the absorption coefficient applied It is set up in the definition of the unit cell content in XPREP and this should be checked carefully both the chemical composition and the Z number Sometimes the structure solution might show that your assumption about the chemical composition was not completely right and then it is advisable to repeat the absorption correction with the new absorption coefficient calculated for the correct chemical formula For area detector absorption and other corrections it is also possible to use a separate program SADABS described later in chapter 11 If you wish to combine a face indexed correction in XPREP with SADABS which is a common sequence remember to answer Y when the question Copy whole of input file wit
141. n it directly double click on the program name or from Windows Run command XPREP is an easy to use essentially interactive and menu driven program At each step the program presents all available information selection of possible answers and the recommended choice Just read the screen output carefully However we strongly recommend going trough the SHELXTL manual pp 4 1 to 4 25 and examples listed there i It is very important to note that if any ambiguity exists about the space group you should assign different filenames ins and hkl for the data prepared in each space group Then the options have to be considered and tested through XS or any other structure solving program during future work before making a final choice Usually a CFOM of less than 5 is a decisive indication that the proposed space group is correct and a value grater than 10 is rather unlikely for the correct space group Chances of success could be increased if you accept one of the probable space groups then perform absorption corrections and repeat the space group determination If the Laue class is changed remember to perform absorption corrections using the true Laue class and initial data set When you run XPREP the program asks for name raw beware extension hkl is assumed and format of input output files then checks preliminary the lattice centering 13 It means that one can run XPREP using any hkI file as input useful if you want to
142. n run CRYSTAL gt Threshold which performs automatic picking of reflections from a series of frames and places them in the Reflection Array In the First input frame name type in the first frame name Fig 5 4 from one of the existing series followed by the full number of frames frames to process in the series Repeat the operation for all available series First input frame name d danticosb cosb10 001 frames to process 620 Max pixel separation 1 Raw count threshold 1 Sigma threshold 50 Minimum excluded X 0 0 Maximum excluded X 30 0 Radius cutoff 520 Minimum Angstroms 0 5 Maximum Angstroms 200 Temperature factor 10 2 Theta override 9999 Omega override 9999 Phi override 9999 Chi override 9999 Scan axis override 9999 Frame width override 9999 Fig 5 4 Threshold dialogue box When performed automatically during the MATRIX procedure Threshold sets too low a value only 10 for the I sigma threshold Very often this is the main cause of your our troubles since some spots belonging to the impurity phase s are present or even so 40 called hot spots nonexistent spots have been picked You should play with the argument in the I Sigma threshold field which is of special importance here Try to change this value in order to pick up over 30 reflections from a MATRIX series about 100 from all series If you still do not obtain a satisfactory result try adjusting some other para
143. nd Filter are only rarely used as independent options if the file names in the first Basic SAINT menu for window are OK you usually can forget the other incorrectly specified files Usually means that you do not have any exotic ideas about how to name the various files that are produced in the various stages of the integration but happily accept the default composed of the first part of your frame names plus typical additions If you do have ideas God save you in your attempts to try them in Sort and Filter menus but do not ask us about possible results First of all you have to decide whether you will use SADABS to make a multi scan absorption correction because it requires quite different responses in several cases These changes are labeled SADABS On the other hand you can always perform the integration using SADABS settings Namely our investigation of four different compounds and five different crystals showed that the quality of your data will not be seriously reduced and the results will be approximately the same no matter how you proceed i e you can later change your mind and use XPREP instead of SADABS for corrections or completely skip absorption correction 60 Basic SAINT menu for analyzing small molecule area detector frames Let us first explain the function of the Validate button located at the bottom of the dialogue box Fig 8 2 After you press the Validate button SAINT
144. ng is suppressed and the non linear refinement will start from the lattice parameters as written in the dialogue boxes It is advisable to use 1 Triclinic only the first time and then to continue with the 1 Triclinic option or negative options for other systems Unfortunately the program always returns to the default positive value It is sometimes necessary to try the positive option again if the suggested detector corrections get suspiciously high Try setting them back to zero and starting a positive system constraint especially if you are changing from triclinic to other systems Note If you select 0 the program performs only a linear refinement not recommended Max RLV error defines the maximum allowed difference between the observed and calculated Reciprocal Lattice Vectors If the difference is bigger than Max RLV error the reflection is removed from the least squares calculation and its H flag is cleared If some reflections are outside the tolerance a box is displayed containing the message lt N gt REFLECTION S REMOVED FROM LS OUTSIDE HKL TOLY Some times the number of removed reflections is too large click Cancel and try with a looser tolerance In the first stage of refinement Max RLV error could be bigger say up to 0 02 especially if least squares removes too many reflections Then it should be gradually decreased to 0 005 and preferably to a lower number 0 002 or 0 0015 as the fina
145. ngth tol is equal to 1 or bigger the lower limit is defined as 0 01 Try with values of 20 or 25 A for axes and 1 for Length tol or with the best guessed values obtained for example during previous analysis of frames ANALYZE gt Cursor gt Vector do you remember As already stated HKL tolerance should not be bigger than 0 25 or 0 3 At this stage do not worry about Expected angle values Alpha Beta Angle tol and Assume known cell fields But the Angle tol should never exceed 0 2 otherwise the results could be really bizarre A default value for Minimum vectors is 2 and this is a good starting value Default value for Length fraction is 0 3 and for Fraction which must be fit it is 0 8 You can change these values but not too much Length fraction should not exceed 0 4 Fraction which must be fit not less than 0 7 or maybe 0 6 In short these three parameters define the minimum number of difference vectors in a group manner of grouping and the number of difference groups that a trial axis vector must fit to be accepted see SMART manual pp 4 21 to 4 22 for more details Finally press Enter or click OK and wait for a few seconds If the message No acceptable set of 3 solutions Fewer than 3 solutions found does not appear take a look at your results which will be shown on the screen in a scrolling window The results will be similar to the following
146. nned in order to reach full coverage and 3 Maximum redundancy In conclusion it is advisable to achieve as much redundancy as crystal symmetry and time j available permit Therefore in practice you will always choose runs that cover at least a hemisphere f of reciprocal space f According to a suggestion of Dr Ibers private communication instead of making a single measurement of one reflection in time it is better to make N measurements each in time t N Of j course the t N value must be some reasonable number of seconds This is very probably true for j large and well diffracting crystals but they require a face indexed absorption correction i 46 I Before you run any type of scan ensure that it agrees with collision limits defined in GONIOM gt Limits for your particular system If you are trying a data collection strategy that has never been used before on the instrument a check of scan angle limits with the help of the manual box is COMPULSORY I Do not forget to acquire a new dark frame immediately before the start of data collection In order to obtain the proper background correction a dark frame MUST BE MEASURED using the same time as the time for frame scans Give the dark frame a proper name following the convention described in chapter Directory folder names and extensions of Bruker files e If this has not already been done run GONIOM gt Home Axis which checks the calibrated positi
147. nt it with some omega scans that result in full coverage the calculation of coverage is described below So you would start by entering the full phi scan in Editruns By adjusting chi for this scan so that the calculated coverage is largest typical values around 30 one can find a good starting run and continue with several omega scans that give the maximum coverage and desired redundancy As already stated above some old scan data could be a good starting point for the strategy determination They can be read and written from the Edit MultiRun dialog box in Smart Fig 6 1 and the corresponding Editruns box in ASTRO Our library of scans do you remember in c frames directory is organized as text ASCII files extension txt The names are coded as rdd_fwl txt that means run detector distance in mm _ just a separator frame width in hundredths of a degree letter necessary to differentiate scans with the same combination of distance and width For example r40_25b txt represents a MultiRun scan for detector distance 40 mm frame width 0 25 option b 52 Reminder Columns are Run Frame 2 Theta Omega Phi Chi Axis Width Frames Time The above mentioned r40_25b txt file contains the following lines 0 001 38 00 38 00 00 27 00 3 250 720 10 00 1 001 38 00 38 00 180 00 27 00 3 250 720 10 00 2 001 38 00 50 00 00 220 00 2 250 200 10 00 3 001 38 00 50 00 90 00 220 00 2 250 200 10 00
148. nt phi scans Each series comprises 186 i e together they make more than a 360 scan around the phi axis This is a common situation and nearly all MultiRuns involve such combinations of scans It is of course possible to collect full 360 data in a single series The only disadvantage is that after number 999 frames will get a combination of letters and numbers in the extension a little more difficult to follow 44 Rung Frane t 2 Theta 0 001 28 00 1 001 28 00 001 140 30 00 140 30 00 140 30 00 140 30 00 001 001 001 Cie WN Fig 6 1 Edit MultiRun dialogue box As you see it is possible to Print data on printer to Write them as an ASCII file on disk and to Read previously saved text files Frame Width and Time can be identical to those employed in ACQUIRE gt Matrix command if that finished successfully or they can be adjusted after inspection of frames already present in order to collect data that will be as good as possible Reminder Recommended Frame Width is from 0 1 to 0 3 whereas exposure time usually varies between 10 and 60 seconds If the crystal is very small the only way to obtain an acceptable data set is to use a time that is longer than 60 seconds Keep in mind however that some especially organic crystals are not always stable under prolonged exposure to X rays The best way to avoid crystal decomposition is to collect data at low tem
149. o change the tension of the locking screws In any event always use both hands when you carry the goniometer head box In other laboratories the goniometer head is always left on the diffractometer Locking tension is set for smooth movement and stability and is rarely adjusted Only the mounting pin is moved You can cancel any diffractometer operation by pressing the Ctrl Break or Esc keys In the first case SMART will interrupt the run instantly The second case gives a smooth break meaning that the current operation will be finished An interrupted MultiRun scan can be continued using ACQUIRE gt Resume Check diffractometer angles If necessary set them to zero by running GONIOM gt Zero or F10 shortcut then confirm YES Mount the goniometer head on the goniometer Find a mark on the bottom part of the goniometer head In order to mount the head properly the mark should be away from you There is a secret slot on the base of goniometer head which should match the small nipple on the screw base of the goniometer but do not try to see the slot with your crystal mounted In order to have appropriate menu items you should switch to Level 2 in the LEVEL menu During work look at the bottom part of the SMART window Sometimes it is possible to find useful information there 3 This set of instructions is appropriate for a four circle diffractometer and room temperature work only but our diffractometer is not equipped wi
150. o convert and the one from the other run that has the proper orientation and note the necessary changes Remember that the components of the reciprocal cell edges are written COLUMN WISE and find out which one has to go in which place and with which sign the numbers are almost but only almost the same in the two matrices Of course they were determined from different sets of data and of course some critical signs have changed Now enter the correctly permuted matrix and hope that you have done everything right Run another LS of course 1 We don t like to confuse you we are believers except when we are not meaning that in most cases we recommend and follow procedure 2 Always analyze your data with an overlay of predicted spots displayed on the screen and check the regularity of the lattice with RLATT As a rule the Reflection Array contains only medium to strong reflections which may not be enough to guarantee unit cell parameters and crystal symmetry In addition to a good orientation matrix you have to know typical reflection spot sizes We bet that you forgot to measure them Go back to SMART and use CURSOR gt Vector to measure the full width of several strong spots preferably not right on the rotation axis and note the max d deg value as XY spot size Then use GRAPH gt Rocking and in a similar manner determine the spot width Spot rocking curve from the displayed rocking curve For each reflection SAIN
151. od MATRIX operation see later However values lower than 20 or higher than 60 frames are not recommended Tip 10 Also check the crystal quality with the ANALYZE gt Graph and ANALYZE gt Graph gt Rocking commands described in the next chapter They can reveal splitting of spots due to a fragmented or composite sample Your crystal is poor and yielded only a powder diffraction pattern as in Fig 2 3 a Do not give up immediately Use a vector cursor described in the next chapter to measure the diameter of at least three of the most prominent circles read d values in ngstr m to the right of the image and divide them by two If any powder diffraction database is on hand use these d values to search through the base Who knows maybe the structure has already been described or unit cell parameters have been determined before or you have another example of isostructural compounds and isomorphous replacement or Configuration filename User name Calibration data directory Sample detector distance Low temperature device Y N X ray target material X ray source wavelength Source kilovolts Source milliamps Filter monochromator Characters in base frame name Characters in Run Base of Run Characters in Frame Base of Frame Direct beam X Direct beam Y Size of created frames Auto command mode timeout Reference detector Y N smart ini Tonci Balic Zunic C frames ced_1k 6 000 IM Check for yes
152. of SADABS called TWINABS should overcome this problem but the program is still in the beta test stage 2 intergrowths are similar to non merohedral twins but they consist of two or more different phases In each case the biggest problem is the presence of fully or partially overlapped reflections APPENDIX 91 92 SOME USEFUL SHORTCUTS Only the most important shortcuts are listed By alphabetic list of commands Bravais lattice Ctrl B Cursor shortcuts Crosshair CUrSOL 0 0002c00e F5 BOX cursor occian F6 Circle CUISOLM cccccccceeeeeeesaeeeeeees F7 Vector CUISOF o cceceeeeeeeceesee ees F8 Display frame n Ctrl D Next frame Ctrl gt Previous frame l Ctrl Edit reflection array eee Cth E Frame information o ae Ctrl F Set goniometer at zero angles F10 LEASt SQUAMES minnena Ctrl L Manual goniometer control Ctrl M Optical goniometer control Ctrl O Re determine Cell cacce Ctrl R DAVE act tues a a a hh lee adel rate Ctrl S Set screen contrast cccccceeeeeeee Ctrl T lides cel Ren ann on Ctrl X By alphabetic list of shortcuts EB oaas atk ete a e Bravais lattice GUN aa D E E T Display frame Ctrl eee Next frame Ctrl eee Previous frame CULES clin ists aA Edit reflection array CUR E neta Frame information Cli st errs ax sc stesraceds A Least squares CE M inntrinn Manual goniometer control CM
153. of the fourth plot from SADABS diagnostics file The fourth plot Fig 11 3 shows the mean E 1 versus resolution For large organic compounds the curve should be close to the 0 968 line for a centrosymmetric or close to the 0 736 line for a non centrosymmetric structure For coordination compounds containing heavy atoms and for compounds with heavy atoms in special positions the plot is less reliable The fifth and sixth plots page 3 of eps file show chi squared as a function of resolution and of intensity Fig 11 4 If the parameters have been correctly adjusted both should be nearly flat around one the g parameter may occasionally need to be increased if the plots drift away from unity at high intensities It is not unusual to see deviations from unity at low resolution 81 Chi squared distributions for Test Chi squared Mean of NSum I lt L gt N 1 Sum su L N equivalents 5 Chi squared uw Ps N 40 20 1 5 1 2 1 0 0 9 0 8 Resolution d Chi squared w N OL 02 03 04 05 06 OF 08 09 Intensity quantiles Fig 11 4 An example of the third page in SADABS diagnostics file It is advisable to print out the plots for easier comparison with later trials This is done directly from GSView Set proper paper size Media gt Letter or A4 and then print using the File gt Print command It is necessary to mark a suitable printer driver Device in the Printer setup dialogue
154. old stream The temperature at the crystal position is near 90 K The X ray beam tunnel collimator and the beam catcher are also visible Note the way the copper pin is attached to the goniometer head The goniometer head is a standard xyz Supper head that has been modified to allow side entry of the mounting pin The side entry is necessary when the cooling apparatus is in place The pin is held securely in place with the little screw containing a small spring loaded steel ball For mounting the pin just snaps in place it is easily removed without loosening any screws With this set up both crystal and crystal mount remain completely free of ice for as long as is needed also in very humid weather Crystal mounting for low temperature work tends to be very fast and simple There is no waiting for an adhesive to set Data collected with the crystal near liq N temperature are generally much better than room temperature data Crystal decay and crystal movement during data collection are eliminated Intensities from a cooled crystal can be much higher than from one at room temperature displacement parameters are to a reasonable approximation proportional to the thermodynamic temperature in K 19 3 HOW TO ANALYZE CRYSTAL DATA No matter how successful you were during unit cell determination and data collection procedures it is a good idea to examine the crystal data in more detail This can be done by careful inspection of the correspondin
155. on control of the goniometer circles so that the reported position is exactly correct Job name cosb1 Title cos Max display counts 1 Suppress correlation Y N F Check for yes Suppress bias detn Y N IM Check for yes XENGEN output format Y N I Check for yes Sequence of starting run 1 Sequence of ending run 9999 Oscillate Y N I Check for yes OK Cancel Fig 6 2 MultiRun dialogue box Now you are ready to collect data that is to run one of the commands e ACQUIRE gt Quadrant e ACQUIRE gt Hemisphere or e ACQUIRE gt MultiRun Just take a look at the MultiRun dialogue box Fig 6 2 change if necessary Job name and press Enter or click OK A MultiRun scan interrupted for any reason can be continued using the ACQUIRE gt Resume command Beside the commands summarized above under the ACQUIRE menu you can collect orientation matrix ACQUIRE gt Matrix and rotation photograph ACQUIRE gt Rotation These two options are described above A single still frame at a fixed goniometer position ACQUIRE gt Still and an omega rotation picture with the crystal rotated around the a b or c axis ACQUIRE gt Axis command could also be acquired See SMART Manual pp 2 14 to 2 17 for more detail 47 Three circle platform diffractometer After a reasonable cell has been determined you can start data collection The three circle geometry results in some changes to the measurement st
156. ons 6 518 9 371 13 770 89 988 88 536 74 231 809 06 0 077 0 125 0 253 Number of solutions 96 Sig A B C Alpha Beta Gamma Number of solution Sig is GOF goodness of fit and unit cell parameters RHOMBOHEDRAL OBV HEXAGONAL R SETTING Lo 1962 6 518 9 854 43 248 105 821 77 401 113 767 1 00 0 00 0 00 1 00 1 00 0 00 2 00 1 00 3 00 3x3 matrix that would transform the HKL indices from the input lattice to the lattice corresponding to the solution the nine elements are in the order T11 T12 T13 T21 T22 T23 T31 T32 T33 TRIGONAL P NO C AXIS SYMMETRY HEXAGONAL P C AXIS SYMMETRY 2 16 43 6 518 9 371 13 770 89 988 91 464 105 769 1 00 0 00 0 00 0 00 1 00 0 00 0 00 0 00 1 00 3 18 45 6 518 9 854 13 770 89 044 91 464 113 767 1 00 0 00 0 00 1 00 1 00 0 00 0 00 0 00 1 00 TETRAGONAL I 4 18 71 16 798 16 654 6 518 82 427 102 443 114 519 1 00 1 00 1 00 0 00 1 00 1 00 1 00 0 00 0 00 ORTHORHOMBIC F 5u 401s 077 6 518 18 092 28 137 89 459 101 926 94 516 1 00 0 00 0 00 1 00 2 00 0 00 1 00 0 00 2 00 truncated 10 18 52 O 371 13 883 29 088 99 1130 108 781 65 081 0 00 1 00 0 00 2 00 1 00 0 00 0 00 1 00 2 00 ORTHORHOMBIC I 11 18 47 6 518 16 798 16 654 65 481 82 427 77 557 1 00 0 00 0 00 1 00 1 00 1 00 0 00 1 00 1 00 truncated 14 19 88 6 518 17 065 16 657 64 879 79 980 75 326 1 00 0 00 0 00 1 00 1 00 1 00 0 00 1 00 1 0
157. ons and hot spots and then try to re index reflections If this does not help go to option no 3 e e 43 6 DATA COLLECTION or DATA ACQUISITION STRATEGY General description and four circle diffractometer SMART allows data collection as a single run or as a series of runs named multiple runs The single run is usually preferred for specimens mounted in capillaries It is invoked by pressing ACQUIRE gt Single Run command and it will not be described in detail The multiple runs are Quadrant Hemisphere and MultiRun There is no important difference between the corresponding lists since Quadrant and Hemisphere can be considered as subsets of the MultiRun procedure While Quadrant is suitable for collecting a wide scan data set for an initial structural analysis and will be only rarely or more likely never used the Hemisphere means that at least half of the Ewald sphere will be collected The MultiRun might be set up for a complete Ewald sphere collection with a high redundancy We strongly suggest that you consider collecting a full sphere in any case It still would take just a fraction of the time required for a routine data collection on a diffractometer with a point detector At the same time the accuracy of your lattice parameters and intensities of the merged symmetry equivalent reflections would significantly improve over the quadrant hemisphere measurements and be at least comparable to those of the b
158. or other corrections if you like to see and publish low R factors and flat AF maps 5 If the crystal is low absorbing u lt 1 5 mm and larger than the X ray beam collimator pinhole do not waste your time with cutting and avoid the risk of loss or damage to the sample Just collect the data and correct them by SADABS 6 When the crystal is of irregular shape try both ellipsoid corrections in XPREP and SADABS You can expect better results with SADABS if there are more equivalents symmetry higher than monoclinic and u is lower Very important here is Fig 11 1 Chapter 11 the more flat and smooth curves especially the first one indicate good corrections 7 When performing ellipsoid corrections in XPREP do not forget to play a game as described earlier with the uxr product and the other parameters 8 Check reflection statistics and number of refined parameters in order to find the proper limit of resolution Although only indirectly related to this story it should be emphasized here that truncation of reflections has a significant effect on the R factors For low and middle absorbing crystals after changing the upper limit of resolution from 0 6 more or less the complete set of data to 0 7 and finally to 0 8 A you could expect a decrease of R values for 0 3 in each step The gain could reach even 0 5 for highly absorbing crystals It is interesting that the observed decrease of R values is almost always biggest when data
159. ot of menu items or perhaps their programmers were fascinated by automation However crystallography is NOT and will NEVER be brainless routine work Therefore we favor better control which means step by step operation whenever possible Be careful Think Do not trust your memory keep your laboratory notebook up to date with as much detail as possible This is of great importance for your present future and especially past work Remember the Chinese saying Faint ink is better than the best memory Note that some data such as file or directory names are laboratory and or user specific They are usually listed for our laboratory in Copenhagen or for Dr Hope s laboratory in Davis Some other details could also depend on practice and standards in your laboratory as well as on laws in your country Finally the proper strategy strongly depends on properties of the crystals under study This is sometimes but not always mentioned in the text This manual should not be a substitute for thorough discussions with experienced crystallographers and your system administrator Disclaimer Well it is not reasonable to expect that we can go farther than Bruker Therefore the authors shall not be liable for errors nor for damages nor for nor for In simple words the use of this Guide is at your own risk Examples given in figures and computer outputs may not represent the best solution for your problem and should not be used
160. oth are given in degrees are transferred through the p4p file as MOSAIC parameter to SAINT They are maximum values and can be a good estimate for the reflection size required by SAINT Before them typical values are given Reflection statistics follows Inorm means integrated intensity normalized to 1 minute Sig is I sigma ratio D represents resolution Code C hexadecimal system means that the reflection is accepted while reflections with the other codes are rejected If the entire action results in a reasonable number of reflections that is at least 25 of them and preferably between 50 and 100 total in the Reflection Array you should sequentially execute the commands listed above under option 1 e e e 41 Chances for success do not be dissatisfied This option means that you have decided to ignore that blind useless stupid _ etc program So you will try to pick a new set of more representative reflections manually by so called hand picking Once again the first step is to erase the currently used reflections stored in the Reflection Array Run CRYSTAL gt Clear Now that you are sure you know what to do confirm once again the size of frames 512 that will be used in subsequent operations and confirm Yes when the question Erase Reflection Array Are you sure appears Open a set of frames Fig 2 1 with FILE gt Display Ctrl D choose the desired
161. our whiskey most often it turns out to be an artifact of inaccurate detector correction values X Y for the detector center too far off or more often a wrong detector distance so check and try to correct this If you sometimes see a nice circular orbit of spots in close succession it is an almost unmistakable sign of hot spots being included in your list of reflections If you are the type that likes to read telephone books or such just go to your reflection list Edit gt ReflArray in SMART and look for repeating almost identical Xs and Ys Delete them maybe back up your list first and check again Bet your orbits have disappeared Another easier approach is to mark graphically the ghost spots in RLATT see later how to change flags and colors of the reflections save the results in a file edit it by throwing out the marked reflections and then use the edited file in further work It is not obligatory to have the true unit cell parameters in the p4p file you use It is still possible to verify the lattice regularity and measure distances between layers Although the measurements are not very accurate the obtained values are good estimates for the next round of unit cell determination How to measure distances between layers Turn on Measurement mode Display gt Measure distance first and zoom in the picture if necessary Then drag the mouse with the left button depressed between the chosen start and end points After you release
162. pends on how you are going to use the results If you are still in the phase of finding the real crystal system are the angles really 90 120 or just very close are parameters equal or just almost equal then it is best to start with a triclinic choice no constraints eventually get rid of the bad reflections large RLV and omega errors in a few refinement cycles and then check the results did angles converge to 90 120 or are they still just close are parameters equal or still just almost equal After you make up your mind about the real crystal system you would probably run the final refinement with the proper crystal system constraints and check the standard uncertainties of parameters once more just in case If you are preparing a p4p file for input to SAINT then your decision on constraint depends on what tactic you would apply during the integration but in any case we recommend that you obtain the best lattice parameters and orientation matrix possible See more in chapter 8 In any case note that all options appear with a positive or a negative number 1 Triclinic or 1 Triclinic Positive values force the program to start with a linear least squares refinement where only lattice parameters are refined From this result it 36 continues with a non linear least squares if the detector corrections or crystal translations are specified for refinement Negative numbers mean that the linear least squares fitti
163. perature The following recommendations are based on the already cited paper of R hl amp Bolte 2000 fj and our experience Exposure time It is not easy to estimate this very important parameter If you are in doubt f and if you are not in a hurry define a longer rather than a shorter time As a guide you can utilize the intensity of spots in a rotation photograph as well as in frames collected during Trial see i Chapter 2 or MATRIX operations If many say more than 20 of the reflections in Reflection Array have I sigma value about 100 or even higher feel free to select a shorter time If the crystal is of proper size see the beginning of Chapter 2 then for a constant frame width of 0 3 data quality increases R values decrease exponentially with time up to 60 seconds i Above that limit some pixels become saturated that results in loss of quality It is also known that if f one keeps all other conditions constant the time can be lowered from 60 to 20 seconds if the frame f width is halved from 0 3 to 0 15 Frame width Narrower frame widths mean that diffraction angles and reflection intensities will 3 be more precisely determined Indeed for a constant exposure time there is a linear relationship j between R values and frame width It can therefore be concluded that an optimal value for the scan f width is 0 1 However the previous discussion and data published by R hl amp Bolte 2000 ind
164. ply all crystal dimensions by a factor of 5 or 10 This really helps But do not forget to re scale to the true dimensions before you finish Otherwise the wrong dimensions will be written to the p4p and ini files With the other shortcuts you can perform the following operations CNTL F Ctrl F brings up a window where you can enter the Miller indices of a plane and then the goniometer will orient the plane face to the measuring i e upward position If for any reason the chosen plane is not clearly visible you can try driving to the face with all of the signs for the indices reversed The desired face will then be oriented downward and maybe this position will be better HOME opens the face list window where you can edit face data e g if you have parallel faces on your crystal you can just add the second after you have measured the first INS toggles between a full projection of the crystal showing also the back faces and a front projection where face indices are displayed too END drives the goniometer so that the resulting view in the microscope should be the same as the screen picture ESC exits the Faces menu The display on the screen allows you to see how realistic your crystal description is It rotates with the crystal and gives a view corresponding to what could be seen through the microscope On exit the face list is saved to the smart ini and p4p file Naturally if you later invoke CRYSTAL gt Faces aga
165. probably just hot spots Picked reflections are automatically stored into Reflection Array after a left mouse button click Unfortunately you will not be able to check the reflection before its profile is displayed on the screen In case of a bad not properly centered split very broad too weak etc reflection use EDIT gt ReflArray or Ctrl E shortcut immediately and delete the reflection from the list o 42 Move to the next frame Ctrl and later to the next series of frames Ctrl D as described above and repeat CRYSTAL gt Redtn Cell gt Pick command until you have sufficient 100 should be enough data It is important to have reflections from different parts of reciprocal space Use all available series of frames Go to CRYSTAL gt Index and continue the work as described in option no 1 Chances for success not so bad but could be better If all previous attempts fail and you decide to use the same crystal single or not so single that s the question now then run ACQUIRE gt Matrix again This time you have to be very very cautious during the choice of parameters in the corresponding dialogue box see above Usually you have to increase the number of frames to be collected as well as the exposure time Very often this has to be accompanied by a decrease in Frame width Also you should consider detector to crystal distance usually set between 4 and 6 cm In case of high symmetry crys
166. rategy Beside the fixed chi angle the main difference is in scan axis which is almost always omega in the case of a three circle diffractometer SCAN Hemisphere Run List 8 lines x Run Frame 2 Theta Onega Phi Chi Axis Width Frames Tine 27 00 26 00 0 00 54 70 2 0 300 606 10 00 33 00 21 00 88 00 54 70 2 0 300 435 10 00 33 00 23 00 180 00 54 70 2 0 300 230 10 00 Cancel Print Write Read Fig 6 3 The Edit Hemi window Enter all the required run parameters here Values can be conveniently taken from the run table given in Appendix The main set up window Fig 6 3 is reached by e Acquire gt Edit Hemi This is where you enter the data that determine the data collection run The table in the appendix lists different parameters for different total measurement times and a crystal to detector distance of about 4 cm To use copy one of the run set ups into the Edit Hemi window Note The 3 frame width will usually give a structure solution but the R index tends to be higher than most people like Other frame widths typically have worked well including the one at 1 The estimated data collection times are for the frame times given Different frame times can be selected for example 10 s for a 0 3 frame width is often a good choice of course it depends on how well the crystal diffracts m Be sure the dark current time corresponds to the frame time entered After you have finished setting up th
167. re it means that your crystal most probably is a complicated case a twin or even an intergrowth of different phases Congratulations Simple cases are for less talented people Real crystallographers like it hot OK it seems easy now to play clever after the magical program GEMINI appeared but we are not such cowards we really did some intergrowths by optically cleaning reflection arrays in RLATT and refining two lattices separately then calculating their overlap ourselves and integrating However after GEMINI appeared we would never try that again GEMINI really works wonders We still have not found the limit on the number of lattices it can resolve E g just from a matrix data collection on a microcline fragment a triclinic mineral with about 90 reflections total we easily determined both the twin law and the oriented intergrowth with quartz So try it We are not going to give procedural advice here the manual for GEMINI is very informative in itself It even includes the full explanation of the indexing philosophy tutorial and procedures for both SMART and GEMINI We would wish the program were just slightly more flexible It is designed exclusively to handle non merohedral twins and defragmented phases but it can also resolve all kinds of multiphase intergrowths and as mentioned we use it for all kinds but then have to handle some of the files our way Chances for success if you did not throw away such a crysta
168. representative 328 4 328 8 329 2 329 6 Distance Pixels Omega Deg a b Fig 3 2 a One dimensional slice obtained with ANALYZE gt Graph b rocking curve obtained with ANALYZE gt Graph gt Rocking The spot in the center of the screen has been analyzed after an appropriate zoom Note that the spot is bad looking and that reflections should not be split The screen output shows X and Y coordinates of the box center width and height of the box sum of the pixel values in the box the largest and mean pixel value in the box the standard uncertainty 3D three dimensional integrated intensity of the spot integrated intensity standard uncertainty ratio X and Y centroid in pixels as well as Z centroid in degrees of the integrated intensity 23 4 RECIPROCAL LATTICE DISPLAY PROGRAM A very powerful program for data checking is Reciprocal Lattice Display Program RLATT It displays measured data in reciprocal space Various reflection files can be used for input p4p from SMART indexing not necessary raw from SAINT hkl from SHELXTL and some less important files for the complete list see RLATT manual In order to have the proper orientation matrix before a raw or hkl file you must read a parameter p4p file Use Reflections menu to choose the file to add reflections more than one reflection file can be displayed at the same time or to remove Clear reflections Once you
169. resolution higher better than the shown value will be omitted from the output file Why omit a lot of reflections Put a low value there 0 75 or 0 70 is usually good enough but even 0 5 could be an acceptable entry in some cases mark d spacing circle You can always truncate reflections later A good reason to type in a higher number there is if your crystal does not diffract at high 2 theta angles The other good reason is the reflection statistics Under normal circumstances Rn values decrease with resolution pass trough a minimum somewhere around 1 5 A and then steadily increase up to 0 75 or 0 70 A However below these values Rint usually increases abruptly Do not forget that the final number of unique reflections in the hkl file should be at least ten times greater than the number of parameters varied Starting Frame Filename and Output Filename in Integration files field are usually correct However notice that a new subdirectory work is the default choice for output directory The most important here is what is written in the Matrix p4p Filename fields There the starting matrix data for each series are defined and so its integration destiny Check the proposed names and enter if necessary by typing or browsing your choice It would depend 61 on what you did to collect the best matrix data in your opinion see the description of the three temperaments characters at the beginning of this
170. results all available data should be used avoid the FILTER option in SAINT It is also possible to perform an additional absorption correction for a spherical crystal The namem raw or namet raw file may be read instead of the individual scans but namet raw is not recommended since SADABS can scale better and rejecting reflections prematurely screws up the statistics There is no disadvantage in reading namem raw instead of the individual scans since the namem raw file traces run numbers and any series of scans can be excluded in the next SADABS run if there is a good reason for it e g if that series is poorer than the rest of data In order to function correctly SADABS requires the Laue class for the crystal If the choice is uncertain at this stage you will have to pin it down with the use of XPREP operating on a raw file The program proceeds by first making a preliminary pass through the data and then adjusting parameters to achieve improved plots of several quantities against resolution or intensity This is the reason why GSView must be available It is also useful to review the documentation as you proceed because otherwise it is hard to tell what is going on There is a possibility that the new SADABS cannot be run from the SAINT menu but has to be run from its own directory consult the system administrator In that case a complete path must be given for any files it uses This is easier than it sounds because you can find the
171. rrected intensities are derived using some of the results from Step 1 High resolution limit 0 10 This value is usually but not necessarily the same as in Step 1 Remember that only corrected reflections satisfying this condition will survive i e they are written in the sad hkI file lt I gt su ratio for rejection 4 0 Very likely reflections with I lt I gt su gt 4 0 suffer from serious systematic errors g value used for rejections 0 04000 If the number of unique reflections left after the I lt I gt su test looks reasonable accept the parameters used The program now calculates a new g value which gives the best error model and should normally be accepted A table of tentative results appears next once again choose accept because there will be an opportunity to try changes later But if the results indicate that a particular scan is much worse than the others do not forget to leave this scan out when you repeat the whole calculation Step 3 Output Postscript diagnostics and corrected data The diagnostic plots prepared in this step are essential for verifying that the parameters in the error model have been correctly adjusted Write Postscript diagnostic file Y or N Y Name the diagnostic file something like name_n eps perhaps with a full path when prompted If you omit the extension eps is added automatically The n after name is a serial number because you will likely mak
172. s and crystal translation Usually the refinement is performed by the program in two steps The first is a simple linear refinement and the second is a nonlinear refinement with positional corrections and possible constraints Note After the calculation all values are updated so you do not need to enter new values The default output file extension Output p4p file see Fig 5 3 is _pr but after refinement a p4p file is also written Since the p4p file contains more information of interest and this file should be supplied to the SAINT SHELXTL and the other programs feel free to define a name p4p file Of course the name could be that of an already existing matrix p4p file will be overwritten or something in connection with the Project name But remember to make a note of it Output P4P file d tonci cosb cosb10 p4p Constraint 5 Ortho v Max RLY error 0 01 Constraint mask 512 Unit Cell A axis fi 000 B axis 1 000 C axis Alpha 30 000 Beta 90 000 Gamma 9 0 1 Detector Corrections X beam center 0 000 Y beam center 0 000 Distance cor Detector pitch 0 000 Detector roll 0 000 Detector yaw 0 000 000 0 000 000 Eulerian angle 1 0 000 Eulerian angle 2 0 000 Eulerian angle 3 0 000 Crystal Xtrans 0 0000 Crystal Y trans 0 0000 Crystal Z trans 0 0000 Omega zero 0 0000 Chi zero 0 0000 Frame halfwidth 0 15 Fig 5 3 Least squares dialog menu The choice of the crystal system constraint Fig 5 3 de
173. s around watching and waiting for you to fall in its trap After you finish editing the Reflection Array it is normal to click OK to accept its new contents Everything will be OK However if you meanwhile get the idea very clever to press Write in order to save data as an ASCII name txt file for example on disk just for reasons of insurance possible future usage or just for documentation and if something in Array is accidentally marked blue background white letters only the marked part of the Array will be written to disk A really amusing extraterrestrial phenomenon It is indeed so easy to lose all of your several hours effort especially if data in the current matrix p4p file were later spontaneously as defined by Murphy s law rewritten reindexed etc The same is true for printing If the problem is wrong or missing indexing there are at least six ways to proceed 28 First of all you can try to use and index reindex perhaps with some modifications reflections already present in the Reflection Array Sometimes only a small push is required to help the program find the proper solution s Here the Reciprocal Lattice Display Program and procedures described in the previous chapter could be very helpful If there is a suspicion of a composite crystal twin or another intergrowth we strongly suggest you try indexing by GEMINI as a second option The third and fourth options imply that you will collect au
174. s have been released and George Sheldrick was so benevolent as to send a lot of comments and recommendations which have been included in version 4 0 Personally we have accumulated more experience and have performed several studies on data integration and absorption correction strategies These results are partially presented in this version as a collection of new instructions and recommendations Once again Dr Hope was so kind to edit the complete text making many improvements and some new suggestions Starting from version 4 0 examples of SAINT output files prepared in landscape format with good reason are distributed as a separate package called An Extract from SAINT output files We believe this change will make the primary text more compact and easily readable and the examples should be more useful As you see many crystallographers contributed to this text and we are very indebted to them But we don t run away from our responsibility all suggestions comments remarks and why not praise should be sent to the signed authors Share your knowledge with the rest of the crystallographic community You should keep in mind that a final version of the Guide has to be at least twice as heavy as the Bruker materials At the suggestion of several readers we have decided also to prepare the Guide as a pdf file and this will be available soon Belgrade Copenhagen Jun 2003 The Authors P S Very likely major updates of SMART SA
175. se OSCAIL a program system developed by P McArdle Crystallography Centre Chemistry Department NUI Galway Ireland J Appl Cryst 1995 28 65 The latest version is always available from the web site http www nuigalway ie cryst software htm An excellent integrated system WinGX of publicly available Windows programs for the solution refinement and analysis of single crystal X ray diffraction data collected and connected by L J Farrugia Department of Chemistry University of Glasgow UK J Appl Cryst 1999 32 837 838 is available at the INTERNET address http www chem gla ac uk lous software wingx 76 11 ABSORPTION AND OTHER CORRECTIONS BY SADABS This chapter contains suggestions for the use of SADABS V2 07 As a result of many improvements and fine tuning this version gives lower Rn values especially for highly absorbing crystals but not always better refinement results Instructions were mainly prepared by Dr Gene Carpenter but some additional explanations are taken more or less directly from the SADABS help file The latest version of SADABS should be available from the official BRUKER site However to obtain the fully functional demo version you can also go to the web site http shelx uni ac gwdg de axs In that case you need to specify user axs and give the password 7 characters which is the last hexagonal space group in the International Tables note that UNIX is case sensitive The help
176. t axes 27 5 IF MATRIX OPERATION FAILS That happens more often than anybody wants and it is always very frustrating The question is How to recognize the failure The first case is very simple if MATRIX was run in automatic mode any screen output after the procedure is missing Apparently SMART was totally confused about your crystal and its reflections and probably about you as well However sometimes the procedure offers really funny solutions with unit cell parameters of say 80 or even more ngstr m Finally the results could be quite different from what you expected reconsider expectations It could be worth noting that some experienced crystallographers pay little attention to the MATRIX procedure which serves them chiefly to verify that the crystal is suitable for data collection even if perhaps twinned However even they must pass through determination of unit cell parameters and orientation matrix after data collection and before starting the integration Therefore the procedures described here are universal and they wait for you here there or in both places As a beginner you have to analyze frames and reciprocal lattices as described above as well as reflections picked automatically or manually during the MATRIX operation They are stored in a Reflection Array which contains reflections used or to be used for unit cell determination Running any of the following commands e EDIT gt ReflArra
177. t the opposite working philosophy If you remember when running ASTRO the point was to define several series of scans and to check possible coverage completeness and redundancy of experimental data for a desired resolution limit In COSMO you first define target values of completeness and redundancy as well as data collection time each with its own priority After refinement of data collection strategy COSMO offers a scan list that satisfies or not your requirements How does the program work It has a large set of scans several thousand according to its authors which are geometrically valid for supplied crystal to detector distance and contain different goniometer detector angles Using a generic refinement algorithm whatever this means which mostly looks like a trial and error procedure COSMO generates an optimal scan list Finally the list should be exported and read into the MutiRun editor Fig 6 1 of SMART Actually planning data collection strategy in COSMO is very similar to the recommendations we gave for using ASTRO There is a library of MultiRun scans associated with crystal to detector distance Depending on circumstances symmetry unit cell parameters crystal orientation completeness redundancy and time available the list is further optimized by a human being in ASTRO or computer in COSMO As usual humans are faster but less precise A typical COSMO run lasts more than 45 min with ASTRO you will only rare
178. t we also confirmed their results using a very i large crystal of one borate mineral with u 1 3 mm In contrast in our laboratory sulfides of heavy f elements with u typically over 40 mm are often measured and absorption correction usually i decreases Rin by 2 3 With the transmission factor getting as low as 0 001 one usually does not use f crystals over 0 1 mm in diameter f j In a radical approach of Dr Ibers private communication one should use the largest crystal commensurate with the collimator pinhole Of course this could be recommended only if physically meaningful face indexed absorption correction can be applied j The overall length of the glass fiber should be about 1 4 cm After mounting in the brass pin the visible part of the fiber should be ca 0 7 cm long Mount the fiber properly in line with the brass pin axis using wax Fix the crystal by dipping the top of the fiber in glue do not take too much of it and picking up the crystal In order to minimize absorption effects it is advisable to fix the sample with its smallest surface attached to the top of the fiber but not directly along the main axis Mount the brass pin in the goniometer head I Do not forget to unlock the head Find three small screws ask an experienced person if you are in doubt But also see next paragraph It should be mentioned that in some laboratories beginners are not allowed to remove or mount goniometer heads or t
179. tals and short crystal lattice periods a shorter distance e g 3 cm allows more data to be collected and increases the chance for success If reflections are overlapped at least one cell parameter big try with a longer distance This time set in advance Indexing HKL Tolerance to 0 2 0 25 and LS RLV Tolerance to 0 02 Chances for success if the crystal is indeed of good quality maybe even slightly better than you expect Find a better crystal define a New Project or edit the existing one and start from the beginning Follow procedures described in chapters CRYSTAL MOUNTING and ORIENTATION MATRIX Good luck Chances for success depends on the quality of your crystals and weather forecast but if you really are a gambler why not try for the jackpot If you read this chapter carefully it was impossible to miss the observation that the sum of probabilities for success is very close to 100 perhaps a few percent above Believe it or not Finally it is of course possible to take some combination of the above described approaches Almost everything depends on your imagination It is also very helpful if you know what you are looking for But if data do not support your prediction do not reject the data for that reason Revise your theory or leave this science and turn to religion Tip In our experience when you use SMART and RLATT together you should first of all clean up the Reflection Array from bad reflecti
180. tegrate Sort Global Validate Open listing file Help Close g p g Fig 8 4 SAINT Advanced Integration dialogue box Advanced Integrate menu The most interesting field here is Model profiles in the top left corner Fig 8 4 The first box I sigma lower limit for reflections used to update model profiles is a threshold for reflections that are candidates for updating the model spot profiles and orientation matrix The default value 10 is suitable for big strongly diffracting crystals Enter 8 in this box if intensities are low You can play with the number in l sigma threshold for least squares fit below vs simple sum above That number defines which spots below the threshold will be refined by a least squares technique the others are simply summed In a typical run this variable should be set to about one half of the I sigma lower limit to update model profiles value In our experience changing this number from 8 to 5 or vice versa could change the final R value by about 0 2 Unfortunately this happens in a non systematic way Very likely the behavior of investigated samples depends on reflection profiles but we never have analyzed it in detail According to the SAINT manual if in some regions reflections are very weak you can try to use Blend 9 profiles However in everyday practice Blend 9 profiles enabled usually improves Rint values and for most of the crystal types it is
181. ter data collection go to the Autoindexing menu Ctrl X input the expected values for cell edges and angles enter Length tol 0 1 Angle tol 0 1 and check the small box Fig 5 1 near the Assume known cell How the latter option influences the behavior of the program it is very hard to guess but anyway only a few seconds will be enough to reach extreme happiness or total devastation 29 If the attempt with known cell parameters fails change some parameters shown in Fig 5 1 and run the program again Yes it is much easier said than done What parameters If you do not have any idea about the parameters and your guess did not bring any result then try playing blindly xs Expected Unit Cell A axis 10 B axis 19 C axis 24 Alpha 90 0 Beta 90 0 Gamma 90 0 Tolerances Length tol 0 8 Angle tol 0 2 HKL tol 0 3 Difference Vectors Grouping Minimum vectors 3 l Assume known cell Length fraction 0 3 V Yerbose output Fraction which must be fit 0 6 OK Cancel Fig 5 1 Autoindexing dialogue box First of all the default values 80 A for expected axis lengths Expected unit cell A axis B axis are very stupid well sorry protein folks In combination with Length tol erance with a default value of 0 75 they define a possible interval for axis lengths The limits are between EAL 1 FCEL and EAL 1 FCEL where EAL is expected axis length and FCEL is axis length deviation if Le
182. th a low temperature device Define a New project with o CRYSTAL gt New Project Define the Project Name and type in all known data Fig 2 1 At least Crystal name Crystal number Working and Data directories must be defined If you leave Data directory blank it will default to the Working directory Do not worry too much the remaining items can be filled in later with CRYSTAL gt Edit Project As stated previously the combination of Crystal name and Crystal number must be unique If it is not SMART will increment the crystal number We strongly recommend that you use a new directory for each crystal Otherwise old data can be easily overwritten Press Enter or click OK when you enter data and choose Small Molecule from the next menu Crystal Name 32 X chars NewProject Crystal Number up to 4 digits 1 Title Chemical Formula Crystal Morphology Crystal Color al Maximum Dimension Intermediate Dimension Minimum Dimension Collection Temperature Measured Density Density Method Working Directory d dejan trty Data Directory m dejan trt Backup Work Directory I Check for yes Fig 2 1 Crystal gt New Project and Crystal gt Edit Project dialogue box Note It is also possible to use the present configuration and to enter your temporary data by the above mentioned CRYSTAL gt Edit Project command However we do not recommend it because in that case it will be easy to for
183. th respect to the value shown Change of reflection flags or colors see above is also possible if the graphical editing mode is in effect In the Edit menu you can choose to change Edit gt Delete or Add a flag one or more flags When Selection box editing is activated with the left mouse button depressed one can draw a box enclosing one or more RL points Then press Enter to change flag s of spot s inside the box this should be followed by color changes too It is assumed here that you are in trouble after the automatic MATRIX operation since as mentioned above SMART picked up too many hot spots and was not able to index your reflections Try to identify these spots and delete their A and or S flags Since reflections present in the Reflection Array always have A and S flags deleting just these flags will later help you recognize and erase changed reflections Finally save your job using EDIT gt Save results to a p4p file There is another option for graphical editing if Lattice overlay editing is invoked from Edit menu see RLATT manual for description but so far we have found no use for it 26 Some limitations of the program have to be emphasized It is not possible to read out the angles or indices of individual lattice spots although the latter can be guessed by counting in 3D Also it would be very desirable to define shortcut keys that will display projections along either reciprocal or direc
184. them or the values that you hopefully tried to determine yourself reread the beginning of this chapter However if necessary adjust the sizes after the SAINT run and inspection of the output name _Is files Instructions on how to analyze reflection profiles are given in the extract of SAINT outputs provided with the Guide Box size optimization if you check the Enable box size optimization box sometimes converges and sometimes not Anyway the refined box size is always much smaller than most people like to see There is an opinion that box size optimization results in slightly better intensities however we have not been able to confirm this So do not refine the box size at least not in the final integration If it is not already there enter 8192 in Limit on number of reflections to refine Apply decay correction should be used for organic crystals measured at room temperature but is not necessary for most inorganics It would be good however if you included and analyzed the above mentioned control frames Note if Apply decay correction is checked the original reflection file is left as is and an additional file namet _Is containing time decay corrected intensities is prepared analyze this file and decide whether the time decay is significant or negligible 62 SADABS Uncheck Apply decay correction The number in the Periodic updating frequency field describes how often SMART perfor
185. thods it was relatively easy to recognize TLQS twins and index data but difficult due to lower data quality lack of computers and programs to treat them properly After the appearance of single crystal diffractometers equipped with point detectors that collect only individual reflections twins and twinning became slightly obscured Twinning was frequently overlooked making the structure unsolvable while many obviously twinned samples were simply rejected With the advent of CCD area detectors that collect all of the reflections in reciprocal space twinning has again become visible being now a steadily increasing but often solvable problem for crystallographers Although twins and twinning attract more and more attention the number of programs for checking indexing and solving twinned samples together with the number of published articles grows steadily solving the structure of a twinned sample is not an easy task Therefore we do not expect beginners to be able to do it It is however essential to have some basic knowledge about twins in order to detect such samples Some kinds of twins TLQS could be identified already in the matrix determination step while for others TLS it is often necessary to solve and partially refine the crystal structure This chapter is based on the paper of R Herbst Irmer and G M Sheldrick Acta Cryst B54 443 449 1998 and some INTERNET available material http shelx uni ac gwdg de rherbst definit
186. ticle on the video screen corresponds to the diffractometer center For this you can mount a pin with a sharp tip in the crystal position In our laboratory we use the tip of a fine sewing needle attached to a e o e 15 regular mounting pin but any pin with a well defined tip will do including most mounted crystals Detailed step by step instructions follow This procedure should be done quite frequently but not necessarily for every crystal Orient the phi box so that it is in a good position for crystal mounting For this you make the SMART window active and select Goniom gt Optical In the window shown in Fig 2 9 define starting values for two theta omega and phi The values we use are 30 150 0 respectively These angles will position the phi axis in a plane perpendicular to the camera optical axis Click OK On the manual control box push D and then the AXES PRINT button This will bring the angles to the values in the Optical window if it does not get the person responsible for the diffractometer to help you Options for Goniom gt Optical xi Fast slow speed ratio 14 Base 2 theta 30 Base Omega 150 Base Phi 0 000 Base Chi 54 740 ce Fig 2 9 The Optical dialogue box Determine the actual x axis Make the video window active and select Tools gt Options Check if the reticle tilt is 35 3 Fig 2 8 if not type in this value Make sure the video capture is active Mount th
187. tion Unusual weighting schemes are often seen when the data have been corrected by SADABS 5 Final agreement indices R wR2 etc and residual peaks in AF map These indices are always strongly correlated and any one of them can be used as a reliable indicator for good corrections If the sample contains very heavy elements it is not unusual to see residual peaks of up to ca 4 e even with a satisfactory absorption correction Do not panic without corrections such peaks could be as high as 10 e A 6 The ultimate and probably the most important sign of a good absorption correction are atomic displacement parameters a d p s On the other hand in ordinary and non problematic cases they attract little attention According to L J Farrugia see above The accepted wisdom is that absorption errors cause the a d p s to be somewhat smaller and more anisotropic than the true values in this context the true value is assumed to be that obtained from the analytical correction The main problem with a d p s is how to estimate the true values if it is not possible to perform a face indexed correction Namely a list containing standard a d p s which would be similar to the list of standard average bond distances and angles does not exist And if the true values are already known from the analytical correction why would you be interested in something else An additional problem is that for middle to high absorbing crystals
188. to avoid them in diffraction experiments if something like that is possible of course But whatever your choice it is first of all necessary to recognize twins which could be done by the so called warning signs for twinning They are as follows Under an optical microscope twins show re entrant angles between faces in other words faces appear to leave grooves or depressions on the crystal surface If the crystal is not opaque or uniaxial cubic under the optical microscope with polarized light the sample is dark in some parts and transmit light in other parts In the diffraction pattern some reflections are sharp others split There are problems with cell indexing or refinement and many reflections do not fit An unusually long axis or a large cell volume with too big Z value appears Systematic absences are not consistent with any known space group Apparent trigonal or hexagonal space group Metric symmetry is higher than the Laue symmetry Rint for the higher symmetry Laue group is only slightly higher than for the lower symmetry one Rint is often lower than that for non twinned crystal Exceptionally low value of E 1 The structure cannot be solved or if solved R factors cannot be reduced although data look good Physically impossible Patterson function for heavy atom structures K mean F mean F is very high say 4 or above for groups of low intensity reflections
189. tomatically or manually a new set of medium to strong reflections and attempt to index them The remaining two cases are in fact obvious but we wish to leave them for the end of the chapter Why discourage beginners immediately If you have been following our suggestions you have already checked the collected reflections using RLATT and got an idea about their usability If it all looks just like a mess you could probably try autoindexing as described under option no 1 by changing some parameters blindly Maybe the program sees better than you and you really get something but need to check to see if it is reasonable anyway If you fail a few times we suggest you do not waste your time anymore proceed to the next options If you have seen the light a real lattice then you are probably in a hurry to try approach number 1 right away Good luck you impatient soul Oh you took your time and already measured some lattice parameters For your bureaucratic type there would be no problem to enter the expected parameters and small tolerances in procedure no 1 and check Assume known cell see later Maybe you have seen too much light Several lattices in a beautiful confusion Skip immediately to option no 2 Again you could be without any quantitative data or with quite a detailed premeasurement in RLATT but it depends on how you do other things in your life and you will be rewarded accordingly If you decide to use collected ref
190. uccessful and you are still confused about what to do in future work keep in mind the following temporary rules 1 In the great majority of cases absorption has no significant effect on the final atomic coordinates i e they agree within 30o limits with and without absorption correction Therefore bond distances and angles as the main results of the crystal structure analysis will be reliable perhaps only slightly less precise even with un corrected data 2 When the crystal is low absorbing say with u lt 2 mm maybe even more but the upper limit still needs to be determined in most cases absorption has no significant effect even on a d p s Here the crystal shape is not important the sample could be spherical regular irregular plate like or rod like 3 Although this could be time consuming use face indexed absorption correction whenever possible 4 One of the Bruker fans since our documentation is lost do not ask for his her name has sent the following comment to the users list In limited testing found that for a well defined moderately absorbing u about 10 mm crystal of relatively anisotropic shape a SADABS correction did rather well a face indexed correction did a bit better and a face indexed SADABS correction did slightly better yet Except for the first part our experience is very similar even with crystals having u 70 mm Therefore sequentially use XPREP for face indexed and SADABS f
191. undancy is always highly overestimated sometimes even by a factor of 1 5 Have a nice day playing with ASTRO Tip Itis very useful to include several identical single frames so called control frames at the beginning somewhere in the middle and at the end of data collection As shown in the following example suitable for a hemisphere routine data collection these frames can be labeled Run 9 with sequential frame numbers and only one frame in each run 9 000 28 00 00 00 200 3 300 1 20 00 0 000 28 00 14 00 00 30 00 3 300 1210 20 00 9 001 28 00 00 00 00 3 300 1 20 00 a 000 28 00 28 00 00 60 00 3 300 610 20 00 9 002 28 00 00 lt 00 00 3 300 L 20 00 2 000 28 00 2 00 00 60 00 2 300 166 20 00 3 000 28 00 2 00 90 00 60 00 2 300 166 20 00 9 003 28 00 00 00 00 3 300 20 00 After the data collection step you should compare control frames all on one screen displaying them in different quadrants as described in chapter How to analyze crystal data FILE gt Display In this way you will immediately see if and when the crystal has moved or decayed Control frames should not be integrated see later COSMO COSMO is a new program for planning data collection strategy distributed for the first time as far as we know in 2002 Although the graphical interfaces of ASTRO and COSMO are quite different COSMO can be considered a subset of ASTRO options with jus
192. use the crystal system constraint only once for the post integration global refinement You of course expect to get the best intensities this way Well be prepared to pay for this pleasure Very likely the procedure will be performed only if very high quality data are required for example in charge density studies In a limited test using only one crystal this procedure resulted in slightly better statistics especially for strong reflections but the final results did not justify expectations and effort But if you really are an incurable During this procedure you will have to repeat indexing and if necessary also Bravais operations for the reflections of each run and we guarantee that SMART will choose different orientations of the axes for different runs plus instead of minus direction Thus you will have to reorient the orientation matrices for some p4p files but whatever you do do not try to do it via the Crystal gt Transform operation This procedure has such unbelievable bugs that it can easily drive you to or beyond the edge of a nervous breakdown Good transformation matrices usually produce the wrong lattice parameters Instead use the Crystal gt Orientation Matrix operation This one in contrast works correctly It also reindexes your reflections after a new orientation matrix It is just that you have to introduce its components one by one AND THIS IS NOT AN EASY TASK You should print out the matrix you want t
193. ut files Advice keep these two pages separate and use them for interpretation of SAINT outputs Tables containing overall reflections statistics Pairs Unig Merg lt 2s lt gt lt Sigma gt lt Bg gt A Rsym dl l dl s R Ranom Canom Erx ErY Erz RmX RmY Rmz Total number of reflections Total number of anomalous pairs Total number of unique reflections Total number of mergeable reflections Percentage of reflections less than 2 sigma l Mean integrated intensity Mean I sigma l Mean background intensity Average of the merging R factor computed from the integrated intensity and mean lt I gt of all observed reflections and their symmetry equivalents Rsym abs lt I gt abs lt I gt Average signed Rsym Average of dl sig abs I lt I gt sigma where sig is standard uncertainty This is a goodness of fit indicating the agreement of each reflection with the average of its group of equivalents relative to that expected from the standard uncertainty Same as Rsym but computed only for reflections with the same Bijvoet parity Average of R factor of mean intensity of positive Bijvoets lt I gt in each group of equivalents and mean negative Bijvoets lt l gt in the group Ranom abs lt gt lt I gt 0 5 abs lt gt lt gt Anomalous clustering coefficients as a measure of strength of the anomalous
194. ve note about centered lattices Percentage of the hash table occupied by 3D profiles of reflections close to the diffracting conditions If Ful is greater than 80 performance is probably suffering due to hash collisions Spot profile tables SUM MAX Intensity of the total normalized profile Maximum value in profile divided by integrated intensity 98 Examples of MultiRun scans See chapters 6 and 7 for necessary fundamentals For four circle diffractometers only several characteristic examples suitable for different crystal to detector distances and coded according to the key described in chapter 7 are given here The lists are prepared as SMART input while lists for ASTRO have one additional column to the right The settings in this section should cover at least a hemisphere For a three circle platform diffractometer the table lists parameters for different total measurement times We hope the listed run parameters will be helpful in planning your own data collection strategy Four circle diffractometer Columns are Run Frame 2 Theta Omega Phi Chi Axis Width Frames Time r30_30 0 001 43 00 43 00 00 27 00 3 300 600 10 00 1 001 43 00 43 00 180 00 27 00 3 300 600 10 00 2 001 43 00 53 00 00 80 00 2 300 150 10 00 3 001 43 00 53 00 90 00 80 00 2 300 150 10 00 4 001 43 00 53 00 180 00 80 00 2 300 150 10 00 5 001 43 00 53 00 270 00 80 00 2 300 150 10 00 r40_25 0 001 38
195. y CRYSTAL gt Redtn Cell gt Edit or Ctrl E shortcut will allow you to view and possibly edit the contents of the Reflection Array An ACHS flag indicates the status of reflections A flag means that the reflection record includes goniometer setting angles and X Y coordinates of the reflection relative to the detector origin this is always the case for reflections determined experimentally S flag indicates that the values of the X Y spot positions of this reflection have been corrected with a spatial correction this condition is also always satisfied for experimentally determined reflections C flag denotes centered reflections that is there was enough information to determine the centroid inside the scan angle the necessary condition is that the reflection is visible on or spanned by more than one frame H flag means that the reflection has HKL indices of course this is possible only after indexing The other data in Reflection Array are HKL indices if determined 2 theta omega phi and chi angles X and Y detector coordinates of the centroid normalized intensity 1 and I sigma ratio It is possible to modify any entry to change flags to delete some for example very weak or very strong reflections to type in some already known data or to add some reflections by an interactive means provided by CRYSTAL gt Redtn Cell gt Pick which will be described below I Be very careful one of the SMART bugs flie
196. y repeated rotations Set the graduated scale reticle in the eyepiece vertical and finely adjust the vertical position of the crystal Now depress C or D and activate AXIS PRINT While the crystal rotates to the new position you can check it through the microscope it rotates almost through all possible positions It can well happen that SMART will choose to rotate around chi so that the goniometer head runs into the beam stop mounted on the collimator according to Murphy s law it almost certainly will see also footnote 4 although it could as well go the other way and avoid it Do not loose YOUR head and with it your crystal collimator or centering of the microscope by making some stupid panic move Just gently push the beam stop around ahead of the racing goniometer head It is easily moved and the goniometer head could do it by itself without any damage but we suggest you do it with a gentle touch of the human hand here robots still cannot cope Readjust the vertical position if necessary by switching between A or B and C or D and that s it If you prefer it is possible to center the crystal without default positions In that case use GONIOM gt Manual Ctrl M shortcut Then press the desired axis button on the manual box and use the right hand buttons fast slow forward or reverse to get suitable angle positions Be careful think about collision danger Tip It is easier to measure a crystal in Manual than in Opt
197. y vs data collection time Fig 7 2 Main COSMO window If COSMO is properly installed on start up it loads default hardware diffractometer profile characteristics for your site If two or more different diffractometers are in use select the correct profile from the Hardware menu However there is an auxiliary program COSMATIC that allows you to create your own hardware profile to take care of specific details The main COSMO window shown in Fig 7 2 is relatively simple and mostly self explanatory When you start COSMO in a new project the first thing you have to do is to load the crystal parameter file File gt Import Crystal Data which very probably will be a SMART p4p file Then check carefully and adjust if necessary all parameters listed on the right side of the main window When everything is properly set press the button at the lower right side and Refine Strategy to start the refinement After that take three coffee breaks if you do not drink too fast or have lunch Hopefully the refinement will be finished just before you come back Inspect the results first in the main window and then by invoking View gt Detailed Strategy You will see several series of active 55 colored runs with detector 2Theta angle very probably cycling as a merry go round Definitely something like that cannot be recommended for a four circle diffractometer the detector is too heavy and the detector arm is too weak although
198. zoom menu A Crosshair cursor is obtained by invoking ANALYZE gt Cursor or by pressing the F5 shortcut At the same time several useful quantities coordinates of crosshair center counts at current pixel 2 theta value at current pixel resolution in A of current pixel hkl values if orientation matrix is in effect etc are displayed to the right of the image Press the left mouse button or Enter key to finish e e e e 21 Valid for all kinds of cursor Pressing Ctrl can speed up cursor movement It is possible to drag the cursor with the arrow keys This is much more precise A Box rectangular cursor appears if you call ANALYZE gt Cursor gt Box or press F6 shortcut This cursor is very useful if you are interested in integrated intensities and the coordinates of the corresponding centroid Pressing Esc or clicking the right mouse button toggles between position change mode and size change mode so you can adjust box dimensions Move the box to the approximate position press the right mouse button and tune the box dimensions then press the right mouse button again and position the box properly Finally press the left mouse button or the Enter key As above to the right of the image you will see X Y coordinates of the box center height and width of the box sum of the pixel values maximum pixel value average pixel value integrated intensity X Y coordinate of the centroid of intensity etc

BRUKER AXS PACK - X-ray Diffraction Texas A & M University

Contents

Download Pdf Manuals

Related Search

Related Contents

BRUKER AXS PACK - X-ray Diffraction Texas A &amp; M University

Contents

Download Pdf Manuals

Related Search

Related Contents

BRUKER AXS PACK - X-ray Diffraction Texas A & M University