Platon, New Options Ton Spek, National Single Crystal Structure Facility

PLATON, New Options

• Ton Spek,

• National Single Crystal Structure Facility,

• Utrecht, The Netherlands.

• Delft, Sept. 18, 2006.

Overview of this Presentation

• PLATON development started more than 25 years ago.

• Start of the talk with some historical notes.

• Recent additions to PLATON such as the analysis for TWINNING and the analysis of Bijvoet pairs will be introduced.

• Preliminary results of the new Charge Flipping algorithm for ab-initio structure determination will be shown.

• Some life Demo’s

PLATON BACKGROUND

• I started as a student in crystallography in the second half of the 1960th.

• As it happens, that was also the period of the first publication of the FFT algorithm by Cooley & Tukey.

• The relevance of FFT in the context of this talk will become clear later on.

PLATON BACKGROUND

• As a student, I was given a small batch of colorless crystals of a compound X of unknown composition.

• The assignment was to find out what was in that sample using X-ray diffraction only.

• It took at that time about half a year to produce an interpretable Fourier map and solve the mystery structure.

Puzzle Solved !

PLATON BACKGROUND

• In order to arrive at a solution, a new Symbolic Addition Method program had to be created, AUDICE, written in ALGOL-60.

• It subsequently also solved other difficult structures in the lab.

PLATON BACKGROUND

• AUDICE was superseded by MULTAN and later by SHELXS, SIR and DIRDIF with the change of the central university computer to a CDC6400 + FORTRAN.

• Structure refinement moved via XRAY-72, XRAY-76, SHELX76 to the current SHELXL-97.

• Since 1980, virtually everything else needed for our national service was developed or incorporated as part of the PLATON package since around 1980.

What is PLATON

• A Multipurpose Crystallographic Tool.

• Compatible with and Complementary to the SHELX & Bruker-AXS SHELXTL Software. (.res,.hkl,.cif,.fcf)

• ‘Semi Public domain’ (I.e. free-of-charge for academics, but with a License Fee for For Profit Organizations).

• Developed on UNIX/LINUX and available on MS-Windows & MAC-OSX Platforms.

Multipurpose Crystallographic Tool

• Absorption Correction Tools

• (e.g. MULABS,TOMPA)

• Space Group Determination from Extinctions.

• ADDSYM - Check for Missed Symmetry.

• SQUEEZE – Disordered Solvent Handling..

• Geometry (Bonds,Angles, Rings etc.).

• Molecular Graphics (ORTEP, PLUTON, Contour)

EXAMPLE ORTEP

• Input Shelx Style: sucrose.res

• (Alternatively: .cif,.pdb,.dat,.spf style)

• Automatic ORTEP style PLOT 

Multipurpose Crystallographic Tool – (continued)

• Generation of Powder Patterns (Coordinates, hkl)

• Structure Validation (part of IUCr CHECKCIF).

• Analysis of Fo/Fc data including Bijvoet Pairs for absolute structure determination.

• Analysis of (Pseudo) Merohedral Twinning.

• System-S, Automated Structure Determination.

System S

• Automatic structure determination

• (Space group determination, structure solution, refinement and analysis)

• Build-in in PLATON (Unix version only)

• Calls external programs (including itself) for various functions (e.g. SHELXS, SIR).

• Program runs in either guided or no-questions-asked mode.

• LIFE DEMO for ‘mistery structure’ X

Twinning

• Twinning results in overlap of reflections with different hkl indices.

• Twinning can be detected in some cases during the data collection experiment

• Cases of (Pseudo) Merohedral twinning are generally detected during the structure determination

(Pseudo)Merohedral Twinning

• Options to handle twinning in L.S. refinement available in SHELXL, CRYSTALS etc.

• Problem: Determination of the Twin Law that is in effect.

• Partial solution: coset decomposition, try all possibilities

• (I.e. all symmetry operations of the lattice but not of the structure)

• ROTAX (S.Parson et al. (2002) J. Appl. Cryst., 35, 168.

• (Based on the analysis of poorly fitting reflections of the type F(obs) >> F(calc) )

• TwinRotMat Automatic Twinning Analysis as implemented in PLATON (Based on a similar analysis but implemented differently)

Example

• Structure refined to R= 20% in P-3

• Run TwinRotMat on CIF/FCF

• Result: Twinlaw with estimate of the twinning fraction and drop in R-value

Absolute Structure

• The absolute structure of a compound is normally determined with the refinement of the Flack parameter.

• The value of the Flack parameter can be inconclusive in view of a high su.

• A detailed scatter-plot may be more conclusive.

New: Charge Flipping

• Ab Initio structure solution by charge flipping

• See G.Oszlanyi & A. Suto, (2004). Acta Cryst. A60,134.

• The approach does not use Direct or Patterson methods to solve structures, just number crunching via FFT transforms.

Procedure

• Expand the reflection set to P1 from the assumed Laue symmetry.

• Assign random starting phases to all reflections.

• Calculate a Fourier map. Change sign of all density below a given level. Do a back-Fourier transformation to get calculated structure factors. Assign the calculated phases to the observed data. Recycle to convergence.

• After convergence: Interprete the map in terms of atom types and determine the correct symmetry.

Procedure

Applications

• Structure determination of powders.

• Structure determination of incommensurate structures.

• Routine structure determinations. (Preliminary implementation of the algorithm in PLATON with a routine named FLIPPER).

• LIFE DEMO for Stucture X 

Conclusions

• Charge Flipping solves structures ab-initio !

• Very limited a-priori info on the structure is needed and no a-priori choise of space-group is required.

• Theoretical background is limited.

• Room for improvement and extension of the algorithm.

• Comparison of the applicability with alternative techniques still limited.

Thanks

• Info: www.cryst.chem.uu/platon

• Thanks to users for:

• Complaints

• Bug reports (‘undocumented features ..)

• Suggestions for extensions

• And you for your attention