Obtaining a Crystal Structure

1
Obtaining a Crystal Structure

• A flowchart for the procedures used at Purdue

2
Mount The Crystal

• Look at the crystals under the polarizing
  microscope. Select the best quality crystal.
• Mount by attaching the crystal with either grease
  or epoxy glue to a fiber glued to a magnetic base.

3
Preliminary Examination

• Grab local mastership at the diffractometer
• Use acquisition? crystal mounting positions.
• Move to mountload magnetic base
• Move to \ and center every 90 degrees
• Use acquisition? quick scan and take a frame
• Use acquisition? acquire images to take a frame
• If ok give up mastership

4
Indexing with Denzo

• Type denzouse to select Denzo
• Start supergui
• Select Determine Unit Cell
• Use Defaults to determine the cell
• Programs useddenzo (kc)
• Output Files10 each .kcd .x

5
Indexing with Eval

• Type evaluse to set up for EvalCCD
• Start supergui
• Select Determine Cell
• Use Defaults to collect phi-chi frames and index
• Programs used --ndirax
• Output files 8 .kcd .kcd.pks and cell.drx
  collect.rmat

6
Face Indexing

• Very Optionalneeded for eval integration
• Use Orient/Measure Crystal in supergui
• Position faces along y
• Add faces as found and then save to file

7
Determine data collect exposure time

• Use Guess diffraction limit
• Note maximum resolution, suggested time, and time
  for overflows.
• Use appropriate values in data collection

8
Collect Data

• Use Collect images from supergui
• Use defaults to set up a strategy
• If data collection is not complete enough then
  rerun strategy and move back detector.
• Collect the data
• Files created-- .kcd

9
Automatic Integration with Denzo

• From supergui select Process images
  automatically
• In general use defaults
• Programs used nprocess
• Files created -- .x

10
Automatic Scaling with Denzo

• From supergui sellect scale raw data
• Use defaults except if sure centric then
• Enter more options and do Not keep Friedel
  pairs separate
• Programs usedmakescalein
• Ouput files--.sca .out

11
Other Integration/Scaling Techniques

• Manual with Denzo makesav, kc
• Scale with Scalepack in kc
• Must reindex with Denzo here if used
• Eval from supergui
• Scale with sadabs
• Need to know how to use these

12
Prelimanary Data Files

• Run solve.
• Select Create data files
• Program run dform
• Input .sca
• Output shelx.hkl xl.hkl scl.out
• Select Create XPREP Input
• Program run p4p
• Input scl.out and other files
• Output xl.p4p

13
Prepare Files for Solution/refinement

• Select XPREP from solve gui
• In general use defaults
• Determine cell symmetrycan check with absen
• If cell is rotated name new file xll and create
  new .hkl file
• Programs usedxprep xll, pcfix, datsum
• Input filesxl.p4p xl.hkl
• Outputxl.ins xl.org xl.pcf xl.prp

14
Solution by SIR

• Select SIR2002 from solve gui
• Run default run
• If needed change setup to add more data or use
  the relax procedure
• Programs usedautosir which includes xtlsir
  sir2002 sirxl shelxl
• Inputxl.sir (created by xtlsir using scl.out and
  xl.org) and xl.hkl
• Outputxl.ins and after Shelxl xl.res xl.lst

15
Solution by Shelxs

• Select Shelxs from solve gui
• If best cfom is greater than 0.7 than it probably
  failed
• Result run into SIR. If you let SIR recycle than
  it will try to assign atom type
• Programs used autoxs (xs xssir sir2002, sirxl
  shelxl)
• Input xl.org (or xl.ins) xl.hkl
• Output as per SIR

16
Solution by Dirdif Patty

• Select Patty from solve gui
• Run is automatic
• Results
• if R gt 45 no solution
• If Rgt20 solution with mis-assigned atom types
• If Rlt20 good solution
• Results into SIR2002
• Programs usedautopat (xtldir, dirdif, dirsir,
  sir2002, )
• Inputxl.org xl.hkl
• Outputas per SIR

17
Check the Structure Solution

• After solution there will be 5 cycles of SHELXL
  (xl) run
• Check statistics (r-factors, Goof) to make sure
  there is a solution
• Open the REFINE gui
• Use PLATON-ORTEP to look at atoms to ensure all
  are assigned correctly.

18
Isotropic Refinement

• Use UPDATE to rename xl.res to xl.ins and xl.ins
  to xl.inssav
• EDIT xl.lst to check if there are peaks that are
  atoms
• Run xl and update until isotropic refinement
  converges (max shift/esd lt 0.05) and no
  non-hydrogen atoms need to be added.

19
Rename Atoms

• Use the rename function to rename the atoms.
• This will start PLATON and select rename
• Rename the list and exit PLATON
• Use Sort Atoms to arrange the list
• Input xl.res(or .ins)
• Output xl.ins
• Programs used ARENAME and XSORT

20
Anisotropic Refinement

• Use Anisotropic to convert atoms to anisotropic
  refinement.
• Run Shelxl
• Run Platon-Ortep to check adps
• Use Split Atoms to create two new atoms from
  elongated adps.
• Run Shelx-Update to refine
• If atoms split check to add part numbers.

21
Add Hydrogens

• Use Add H
• When done this opens an edit window.
• Check HFIX cards as follows
• sp3 with 1H13 sp3 with 2H--23
• sp2 with 1H43 sp2 with 2H93
• sp163 CH3137 OH147
• Edit the mn number if not correct!!

22
Other Possible Commands

• Add part numbers for disordered atomsuse a text
  editor.
• Add SADI (same distance) command to restrain
  equivalent distances
• Add DFIX to restrain distance to a set value
• Use SQUEEZE to remove disordered solvents
• Change the enantiomorph if incorrect or the Flack
  parameter is about 1.0.

23
Final Refinement

• Continue with SHELX
• Upgrade by using WEIGHT to upgrade the weight.
• Run OUTLIER to remove bad data
• Add hydrogens not refined with AFIXlook for in
  xl.lst and add with Pks to Atoms
• Use ASSYM to move everything to the same
  assymetric unit.
• Refine until converged (max shift/esd lt 0.005)