advanced crystallography: refinement of disordered structures november 13, 2012 1disorder webinar
TRANSCRIPT
Disorder Webinar 1
Advanced Crystallography: Refinement of Disordered Structures
November 13, 2012
This Webinar is being broadcast from our Madison, Wisconsin, USA, facility
November 13, 2012 2Disorder Webinar
3Disorder Webinar
Charles Campana, Ph.D.
November 13, 2012
Senior Applications Scientist at Bruker AXS, Madison, WI, USA• BS, Chemistry (1970), Montana
State University, Bozeman, Montana
• PhD (1975), Inorganic Chemistry (with L. F. Dahl), University of Wisconsin, Madison, Wisconsin
• Assistant Professor (1976 – 1980), University of New Mexico University, Albuquerque, New Mexico
• Senior Applications Scientist (1980 - present) Single-Crystal X-ray Diffraction, Nicolet, Siemens and Bruker;
• Crystallographic co-author on several hundred scientific papers.
Audience Poll
What is your experience level?
I have not done any crystal
structures
I have done only routine structures
I have done problem structures
I’m a crystallographer
Please use your mouse to answer the question on your screen:
4November 13, 2012
5Disorder WebinarNovember 13, 2012
Modern X-ray crystallographic systems with automatic features have made it possible for synthetic chemists with limited crystallographic training to obtain publication-quality crystal structures quickly and easily for routine structures.
While the automated structure routines perform well on straight-forward molecular compounds, these routines are not capable of automatically modeling various disorder problems.
In these cases, the user must utilize some of the advanced SHELXL/XL instructions involving the use of free variables, constraints and restraints to successfully refine and publish his / her results.
All of these techniques involve editing the SHELXL/XL output (*.res) file, with either a text editor or a specialized graphical editor, to produce a new, modified SHELXL/XL input (*.ins) file.
Introduction
6Disorder Webinar
George Sheldrick
November 13, 2012
Professor of Structural Chemistry and part-time programming technician at the University of Göttingen
• PhD (1966) University of Cambridge with E.A.V. Ebsworth; thesis entitled "NMR Studies of Inorganic Hydrides"
• 1966 - 1978: University Lecturer and Fellow of Jesus College, Cambridge
• Since 1978 Professor at the University of Göttingen
• Since 2011 Niedersachsen Professor at the University of Göttingen (emeritus)
• Author of about 800 scientific papers and of a computer program called SHELX (http://shelx.uni-ac.gwdg.de/SHELX/)
Sheldrick, GM (2008) A short history of SHELX. Acta Crystallogr., A64:112-122 (open access) This paper is currently the most highly cited scientific paper of the last five years in all subjects.
7Disorder WebinarNovember 13, 2012
• A short history of SHELX, Sheldrick, GM (2008) (http://journals.iucr.org/a/issues/2008/01/00/sc5010/sc5010.pdf)
• SHELX Manual (http://shelx.uni-ac.gwdg.de/SHELX/shelx.pdf)
• Crystal Structure Refinement: A Crystallographer's Guide to SHELXL (International Union of Crystallography Texts on Crystallography)
SHELX References
Disorder Webinar 8
SHELXTL vs. SHELX*http://shelx.uni-ac.gwdg.de/SHELX/index.html
SHELXTL (Bruker AXS)
• XPREP• XS• XM• XE• XL• XPRO• XWAT• XP• XSHELL• XCIF
SHELX (Public Domain)*
• None• SHELXS• SHELXD• SHELXE
• SHELXL• SHELXPRO• SHELXWAT• None• None• CIFTAB
November 13, 2012
9Disorder Webinar
Peter Müller, MIT
November 13, 2012
Director of the Diffraction FacilityMIT Department of ChemistryCambridge, Massachusetts
Objectives of this presentation
• Use of text editors or graphical editors to insert additional instructions into SHELXL / XL input instruction (*.ins) files.
• Introduction to the concepts of constraints and restraints in crystal structure refinement.
• Introduction to the internal atom connectivity concept and PART n and PART –n instructions.
• Introduction to the use of free variables in constraints and restraints.
• Examples of commonly used instructions for applying a variety of constraints and restraints to the refinement of disordered structures.
10November 13, 2012
11Disorder WebinarNovember 13, 2012
• In crystal structure refinement, there is an important distinction between a 'constraint' and a 'restraint'.
• A constraint enables one or more least-squares variables to be expressed exactly in terms of other variables or constants, and hence eliminated.
• A restraint takes the form of additional information which is not exact but is subject to a probability distribution; for example we could restrain two chemically but not crystallographically equivalent bonds to be approximately equal.
Constraints and Restraints
12Disorder WebinarNovember 13, 2012
The following general categories of constraints and restraints are available using SHELXL:
• Constraints for the coordinates and anisotropic displacement parameters for atoms on special positions: these are generated automatically by the program for ALL special positions in ALL space groups, in conventional settings or otherwise.
• Floating origin restraints: these are generated automatically by the program, so the user should not attempt to fix the origin in such cases by fixing the coordinates of a heavy atom.
• Two or more atoms sharing the same site: the xyz and Uij parameters may be equated using the EXYZ and EADP constraints respectively (or by using 'free variables'). The occupation factors may be expressed in terms of a 'free variable' so that their sum is constrained to be constant (e.g. 1.0). If more than two different chemical species share a site, a linear free variable restraint (SUMP) is required to restrain the sum of occupation factors.
• Geometrical constraints: these include rigid-group refinements (AFIX 6), variable-metric rigid-group refinements (AFIX 9) and various riding models (AFIX/HFIX) for hydrogen atom refinement, for example torsional refinement of a methyl group about the local threefold axis.
Constraints and Restraints in SHELXL
Disorder WebinarNovember 13, 2012 13
• Fragments of known geometry may be fitted to target atom, and the coordinates generated for any missing atoms. Four standard groups are available: regular pentagon, regular hexagon, naphthalene and pentamethylcyclopentadienyl. Other groups may be used simply by specifying orthogonal or fractional coordinates in a given cell (AFIX mn with m > 16 and FRAG...FEND).
• Geometrical restraints: a particularly useful restraint is to make chemically but not crystallographically equivalent distances equal without having to invent a value for this distance (SADI). The SAME instruction can be used to generate such restraints automatically, e.g. when chemically identical molecules or residues are present. This has the same effect as making equivalent bond lengths and angles but not torsion angles equal. The FLAT instruction restrains a group of atoms to lie in a plane (but the plane is free to move and rotate).
• Restraints on anisotropic displacement parameters: three different types of restraint may be applied to Uij values. DELU applies a 'rigid-bond' restraint to Uij of two bonded (or 1,3) atoms; the anisotropic displacement components of the two atoms along the line joining them are restrained to be equal. Isolated atoms may be restrained to be approximately isotropic (ISOR). Similarly, the assumption of 'similar' Uij values for spatially adjacent atoms (SIMU) is useful for partially overlapping atoms of disordered groups.
Constraints and Restraints in SHELXL
14Disorder WebinarNovember 13, 2012
• Constraints for special positions: the necessary constraints on co-ordinates, occupancies and Uij are derived automatically.
• Rigid groups (AFIX 6 … AFIX 0): the 3 positional parameters per atom are replaced by 3 rotations and 3 translations for the whole rigid group. Atoms may not be in more than one rigid group.
• Riding hydrogen atoms (AFIX mn): xH = xC + x – no extra positional parameters.
• Fixed parameters: add 10 to x, y, z, occ, U etc. Typically occupancies are fixed at 1.0 by adding 10, i.e. given as 11.0
• Free variables: can be used to add extra linear constraints to the usual refinement parameters and also be used instead of restraint target values. This provides a convenient way of getting target values with esd’s for use as restraints in other structures.
Types of constraints in SHELXL
15Disorder WebinarNovember 13, 2012
Example: Atom on twofold axis in space group C2. The two positions related by the twofold axis (x,y,z: -x,y,-z) coincide when x = 0 and z = 0. Since we still wish to sum over all symmetry operators in the structure factor calculation, the occupancy is fixed at 0.5. The probability ellipsoid used to describe the anisotropic motion should not be changed by the 180º rotation.
[U11, U22, U33, U23, U13, U12] [U11, U22, U33, -U23, U13, -U12] which is only true if U23 = 0 and U12 = 0. All these constraints are generated automatically by SHELXL for all special positions in all space groups.
Special position constraints
16Disorder WebinarNovember 13, 2012
In SHELXL, rigid groups are defined by three rotations about the first atom in the group and by three translations of the group as a whole. Special position constraints may be applied to the first atom and restraints and riding hydrogen atoms are allowed on all atoms in the group. Note that the esd’s of bond lengths and angles but not of co-ordinates within a rigid group come out as zero from the L.S. matrix algebra.
AFIX 6 rigid group – all AFIX 9 variable metric … bond lengths and … rigid group - anglesatoms angles fixed atoms fixed, bond lengths … … multiplied by theAFIX 0 AFIX 0 same factor
Rigid group constraints
17Disorder WebinarNovember 13, 2012
The connectivity list is used for the automatic generation of hydrogen atoms and some restraints. Non-hydrogen atoms i and j are considered to be ‘bonded’ if:
dij < ri + rj + 0.5 Å
The CONN instruction may be used to modify r and to set a maximum connectivity for an atom (e.g. 0 for water). A shell of symmetry equivalents is generated automatically around the unique atoms. Bonds may be added with BIND or deleted with FREE.
PART N controls the generation of bonds for disordered groups. Most atoms have N = 0; multiple conformations have N = 1, 2 etc. Bonds are generated only when the N are equal or one N is zero. If N is negative, bonds are not made to symmetry equivalents.
The connectivity list
18Disorder WebinarNovember 13, 2012
Free variables are an extremely concise but effective way of applying linear constraints to atom parameters (especially occupancies), restraint targets etc. The parameter x is given as (10m+p), which is interpreted as follows:
m = 0: refine normally, starting at value p
m = 1: fix at value p
m > 1: x = p* fv(m)
m <-1: x = p* [fv(–m) – 1]
e.g., 30.25 (m = 3, p = 0.25) means 0.25*[fv(3)] and –30.25 (m = –3, p = –0.25) means 0.25*[1 – fv(3)], which could be used to constrain two occupancies to add up to 0.25 (only one parameter, free variable #3, is refined). The starting values for the free variables are given on the FVAR instruction (but free variable #1 is the overall scale factor).
Free variables
19Disorder WebinarNovember 13, 2012
TITL YLID in P2(1)2(1)2(1)CELL 0.71073 5.9651 9.0437 18.4047 90.000 90.000 90.000ZERR 4.00 0.0002 0.0003 0.0006 0.000 0.000 0.000LATT -1SYMM 0.5-X, -Y, 0.5+ZSYMM -X, 0.5+Y, 0.5-ZSYMM 0.5+X, 0.5-Y, -ZSFAC C H O SUNIT 44 40 8 4TEMP 23.000SIZE 0.32 0.34 0.34 L.S. 4BONDFMAP 2PLAN 20 FVAR 1.00000S1 4 0.19050 0.68120 0.74160 11.00000 0.05000C1 1 0.36850 0.62830 0.67080 11.00000 0.05000C2 1 0.31150 0.50160 0.62620 11.00000 0.05000O1 3 0.16360 0.40830 0.63170 11.00000 0.05000C3 1 0.49610 0.49860 0.56590 11.00000 0.05000C4 1 0.52540 0.41640 0.51090 11.00000 0.05000C5 1 0.70740 0.44010 0.46220 11.00000 0.05000C6 1 0.84660 0.54890 0.47510 11.00000 0.05000C7 1 0.82480 0.64010 0.53680 11.00000 0.05000C8 1 0.65310 0.61620 0.58110 11.00000 0.05000O2 3 0.66890 0.80130 0.67590 11.00000 0.05000C9 1 0.56180 0.69740 0.64820 11.00000 0.05000C10 1 0.16570 0.88170 0.72670 11.00000 0.05000C11 1 0.34520 0.68260 0.82310 11.00000 0.05000
HKLF 4
Standard SHELX Instructions
20Disorder WebinarNovember 13, 2012
• XP (Bruker AXS Inc.)
• XShell (Bruker AXS Inc.)
• APEX2 (Bruker AXS Inc.)
• WinGX (L. Farrugia)
• Crystals (D. Watkin et al.)
• X-Seed (L. Barbour)
• shelXle (C. Hubschle et al.)
• OLEX2 (O. Dolomanov et al.)
Graphical Editors for SHELXL/XL files
21Disorder WebinarNovember 13, 2012
The best strategy is to constrain the positions and displacement parameters to be the same, and refine the occupancies so that their sum is constrained to be unity:
EXYZ MG CA EADP MG CAFVAR 1.0 0.6..PART 1MG 6 0.37041 0.34874 0.03824 21.0 0.20936PART 2CA 7 0.37041 0.34874 0.03824 -21.0 0.20936PART 0
If the cations were sharing a special position on a twofold axis, their occupancies would be specified as 20.5 and –20.5. For three atoms (or molecules) sharing a site, it is better to tether each occupancy to a free variable (e.g. by 31, 41 and 51) and to restrain the sum of these free variables to unity:
SUMP 1.0 0.001 1.0 1 1.0 2 1.0 3
Two cations sharing the same site
22Disorder WebinarNovember 13, 2012
The DFIX restraint is able to restrain bond lengths to target values but sometimes the target is uncertain. For example the P―O distance in a phosphate may vary with the pH and the extent of libration. SADI can be very useful in such cases, e.g.
SADI P O1 P O2 P O3 P O4SADI O1 O2 O1 O3 O1 O4 O2 O3 O2 O4 O3 O4
ensures that the phosphate will be a regular tetrahedron, but allow the bond length to refine.
The same can however be achieved by an AFIX 9 constraint or by using DFIX with a free variable, e.g.
FVAR …… …… 1.55
DFIX 31 P O1 P O2 P O3 P O4DFIX 31.6330 O1 O2 O1 O3 O1 O4 O2 O3 O2 O4 O3 O4
DFIX or SADI?
23Disorder WebinarNovember 13, 2012
• In SHELXL, the DELU restraint is a strict rigid-bond restraint, i.e. the components of the anisotropic motion of two atoms along the line joining them are restrained to be equal.
• Although DELU is a reliable restraint and so can be given a small esd, there are not as many DELU restraints as Uij, so it may be necessary to supplement them with the less accurate but more numerous SIMU and ISOR restraints with larger esd’s.
The rigid-bond restraint DELU
24Disorder Webinar
Restraints on ADP's
November 13, 2012
25Disorder WebinarNovember 13, 2012
• Toluene is a good solvent for growing crystals because of its long liquid range, but it simply cannot resist inversion centers:
• This can be handled with one complete toluene molecule with occupancies of 10.5 (fixed at 0.5) and PART -1. Equivalent 1,2- and 1,3-distances can be restrained to be equal with SADI and a FLAT restraint applied to all 7 carbons, or a rigid hexagon can be used for the 6-membered ring (plus two SADI and one CHIV for the CH3). SIMU and DELU are recommended. The hydrogen atoms should be set with HFIX in a later job:
• HFIX 43 C1 > C5 (generates 5H with occupancies of 0.5)• HFIX 123 C7 (generates 6H with occupancies of 0.25)
Toluene on an inversion center
26Disorder WebinarNovember 13, 2012
ISORDELUWGHT 0.045500 FVAR 0.22187PART 1C1A 1 0.42827 0.23894 0.85590 10.50000 0.04467 0.03563 = 0.04915 0.00377 0.01463 -0.00748..C17A 1 0.42023 0.25662 1.14925 10.50000 0.05042 0.04274 = 0.05014 -0.00305 -0.00756 0.00391AFIX 43H17A 2 0.53968 0.31799 1.15025 10.50000 0.05761AFIX 0PART 2SAME C1A > C17AC1B 1 0.47550 0.26048 1.14359 10.50000 0.04169 0.03667 = 0.04839 0.00438 -0.00897 0.00686..C17D 1 0.23382 -0.11832 0.35155 10.50000 0.03704 0.02547 = 0.04522 0.00395 0.00013 -0.00093AFIX 43H17D 2 0.16654 -0.00313 0.35229 10.50000 0.02412AFIX 0PART 0
Use of PART 1, PART 2 and SAME
27Disorder WebinarNovember 13, 2012
Part 1 Part 2 Parts 1 & 2
Use of PART 1, PART 2 and SAME
November 13, 2012 28Disorder Webinar
FVAR 0.07920 0.57624C1 1 1.056270 0.672587 0.015161 11.00000 0.07754 0.17320 = 0.19702 -0.03184 0.00034 0.03626..O6 3 0.793446 0.479760 0.011891 11.00000 0.07630 0.08652 = 0.06854 -0.00308 -0.00415 0.05024SAME C1 > O6PART -1C1' 1 0.122354 -0.092142 0.016919 20.50000 0.16721 0.13875 = 0.20344 0.02675 -0.02623 0.07727..AFIX 0O6' 3 0.167969 0.159420 0.007654 20.50000 0.08280 0.06135 = 0.12535 0.00968 -0.00772 0.03937PART 0
Ethyl acetate disordered about a 2-fold axis
29
Disordered t-butyl group
November 13, 2012
30
Disorder between Cl and C
November 13, 2012
31
Disorder of a cyclopentadienyl ring
November 13, 2012
32Disorder WebinarNovember 13, 2012
33Disorder Webinar
Disorder of a chloroform molecule on a mirror plane
November 13, 2012
34Disorder Webinar
DMSO disordered over three positions
November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
• Background• Sample from UCSD Summer School• Prof. Michael Richmond et al. (U. of North Texas)• NMR indicated dynamic equilibrium between two isomers
• Structure was easily ‘solved’, but could not be refined• R1 = 16%• Many NPD atoms• Three very large difference peaks (‘Star of David’)
35November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
preliminary structure – chelating phosphine ligand
36November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
L.S. 4BONDFMAP 2PLAN 99WGHT 0.040000FVAR 0.15477 0.8500ANIS 3OS1 5 1.38180 0.36327 0.23664 21.00000
0.01306OS2 5 1.16037 0.44356 0.29321 21.00000
0.01737OS3 5 1.22802 0.46498 0.16310 21.00000
0.01676C1A 1 1.22737 0.29544 0.22637 21.00000
0.01424O1A 3 1.14724 0.25371 0.21696 21.00000
0.01327C2A 1 1.53166 0.43473 0.24515 21.00000
0.02705O2A 3 1.61483 0.47245 0.24881 21.00000
0.01637C3A 1 1.00907 0.38885 0.25481 21.00000
0.02644O3A 3 0.91212 0.35992 0.23556 21.00000
0.03082C4A 1 1.33420 0.49045 0.32785 21.00000
0.02520O4A 3 1.42200 0.51620 0.35349 21.00000
0.02519C5A 1 1.13429 0.40796 0.37402 21.00000
0.01968O5A 3 1.11507 0.38537 0.42076 21.00000
0.04168C6A 1 1.03019 0.51972 0.29063 21.00000
0.02206O6A 3 0.94876 0.56097 0.28462 21.00000
0.04682C7A 1 1.12667 0.38230 0.13189 21.00000
0.02768O7A 3 1.06313 0.34129 0.11120 21.00000
0.02261C8A 1 1.33629 0.53964 0.20167 21.00000
0.02088O8A 3 1.39946 0.58535 0.21995 21.00000
0.02206C9A 1 1.34144 0.46478 0.08771 21.00000
0.02607O9A 3 1.39311 0.47459 0.04379 21.00000
0.02768C10A 1 1.07098 0.52231 0.13931 21.00000
0.02470O10A 3 0.98114 0.55458 0.12305 21.00000
0.04529ANIS 2P1 4 1.49269 0.30401 0.31810 11.00000
0.01685P2 4 1.51235 0.30162 0.16667 11.00000
0.02253C1 1 1.64848 0.25992 0.21803 11.00000
0.02086C2 1 1.64668 0.26068 0.28099 11.00000
0.02963C3 1 1.77319 0.22528 0.30686 11.00000
0.03008•
37November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
three large difference peaks
38November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
L.S. 4BONDFMAP 2PLAN 99WGHT 0.040000FVAR 0.14765 0.84986PART 1ANIS 3OS1 5 1.381809 0.363269 0.236644 21.00000 0.01398OS2 5 1.160375 0.443563 0.293216 21.00000 0.01832OS3 5 1.228021 0.464972 0.163111 21.00000 0.01769PART 2ANIS 3OS4 5 1.279201 0.372329 0.290337 -21.00000 0.01897OS5 5 1.356916 0.404339 0.165486 -21.00000 0.01516OS6 5 1.132255 0.478575 0.221611 -21.00000 0.02121PART 1C1A 1 1.227149 0.295498 0.226379 21.00000 0.01463O1A 3 1.147208 0.253766 0.216857 21.00000 0.01437C2A 1 1.531363 0.434503 0.245167 21.00000 0.02786O2A 3 1.614757 0.472385 0.248824 21.00000 0.01766...
C9A 1 1.341290 0.464934 0.087821 21.00000 0.02657O9A 3 1.392650 0.474727 0.043766 21.00000 0.02908C10A 1 1.071073 0.522226 0.139298 21.00000 0.02516O10A 3 0.981357 0.554645 0.123065 21.00000 0.04658PART 0ANIS 2P1 4 1.492683 0.304015 0.318087 11.00000 0.01777P2 4 1.512382 0.301606 0.166718 11.00000 0.02357C1 1 1.648401 0.259913 0.218106 11.00000 0.02206C2 1 1.646876 0.260442 0.280954 11.00000 0.03039
39November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
remaining carbonyl atoms revealed in difference map
40November 13, 2012
Example 1 - Os3(CO)10(PPh2~PPh2)
superposition of both isomers
41November 13, 2012
Disorder Webinar 42
Example 1 - Os3(CO)10(PPh2RPPh2)
chelating phosphine ligand (85%)bridging phosphine ligand (15%)
November 13, 2012
Disorder Webinar 43
Example 1 - Os3(CO)10(PPh2~PPh2)
Crystallographic restraints (SADI, ISOR, SIMU, EADP) were used in initial refinementFinal refinement converged at R1 = 5.2%
Kandala, S.; Yang, L.; Campana, C. F.; Nesterov, V.; Wang, X.; Richmond, M. G. (2010) Isomerization of the diphosphine ligand 3,4-bis(diphenylphosphino)-5-methoxy-2(5H)-furanone (bmf) at a triosmium cluster and P-C bond cleavage in the unsaturated cluster 1,1-Os3(CO)9(bmf): Synthesis and x-ray diffraction structures of the isomeric Os3(CO)10(bmf) clusters and HOs3(CO)8(μ-C6H4)[μ-PhPC:C(Ph2P)CH(OMe)OC(O)]. Polyhedron, 29, 2814-2821.
November 13, 2012
Example 2 – Urea Host : Guest Complex
• Background• Prof. Mark Hollingsworth et al. (Kansas State U.)• Urea host lattice with long-chain carboxylic acid in
channels
• Structure of urea lattice was ‘solved’, but carboxylic acid molecules could not be located
• Apparent unit cell• Orthorhombic P212121
• a = 8.3096 Å , b = 10.9591 Å , c = 13.6330 Å • 12 Urea molecules, 4 carboxylic acid molecules per cell
44November 13, 2012
Example 2 – Urea Host : Guest Complex
projection down b-axis
45November 13, 2012
Example 2 – Urea Host : Guest Complex
Projection down a-axis
46November 13, 2012
Example 2 – Urea Host : Guest Complex
analysis of Q-peaks
47November 13, 2012
Example 2 – Urea Host : Guest Complex
analysis of Q-peaks
48November 13, 2012
Example 2 – Urea Host : Guest Complex
N2C 3 0.662814 0.937759 0.024542 11.00000 0.02028AFIX 93H2CA 2 0.626060 1.006875 0.048581 11.00000 0.05118H2CB 2 0.753628 0.936746 -0.008654 11.00000 0.05692AFIX 0PART -1C1S 1 0.047758 0.093743 0.256429 10.33333 0.02221O1S 4 -0.065938 0.059457 0.208123 10.33333 0.03123O2S 4 0.147213 0.019479 0.303494 10.33333 0.03285AFIX 3H2S 2 0.117603 -0.055411 0.292184 10.33333 0.05883AFIX 0C2S 1 0.090446 0.226023 0.271096 10.33333 0.02369AFIX 23H2SA 2 0.196427 0.240424 0.240008 10.33333 0.03582H2SB 2 0.103898 0.240028 0.342366 10.33333 0.05245AFIX 0
....
C10S 1 0.080032 1.122110 0.284898 10.33333 0.02266AFIX 23H10A 2 0.095393 1.108073 0.356020 10.33333 0.03019H10B 2 0.185537 1.108980 0.252859 10.33333 0.07844AFIX 0C11S 1 0.034726 1.252421 0.270929 10.33333 0.02477O3S 4 -0.087391 1.287204 0.233376 10.33333 0.03143O4S 4 0.147522 1.327757 0.305089 10.33333 0.03101AFIX 3H4S 2 0.123082 1.402847 0.301909 10.33333 0.66676PART 0
49November 13, 2012
Example 2 – Urea Host : Guest Complex
• Anisotropic refinement of dicarboxylic acid
• Final refinement• R1 = 3.83%• Temperature factors on hydrogen atoms refined
50November 13, 2012
Example 3 - Fe3(CO)12 Crystallographic Problem
• Space group is P21/n with Z = 2
• Each molecule must lie on a crystallographic center of symmetry
• Successful refinement is difficult because it requires two half-weighted molecules superimposed on the center of symmetry
Example 3 - Fe3(CO)12 Refinement Results
100K Data• Resolution – 0.40Å • R1 = 3.82%• wR2 = 9.14%• 12633 Ind. Refl.
298K Data• Resolution – 0.65Å • R1 = 5.11%• wR2 = 13.45%• 2989 Ind. Refl.
53Disorder WebinarNovember 13, 2012
PART -1FE1 3 -0.055374 -0.020549 -0.170035 10.50000 0.01726 0.01813 = 0.01192 -0.00304 0.00029 0.00180FE2 3 0.178487 -0.013488 0.072272 10.50000 0.01366 0.03072 = 0.01623 -0.00211 0.00057 0.00456FE3 3 -0.113197 0.044515 0.114740 10.50000 0.01190 0.02706 = 0.01485 -0.00692 0.00156 0.00141C1 1 0.081260 0.147371 0.115557 10.50000 0.01956 0.02010 = 0.01724 -0.00283 0.00109 -0.00231O1 2 0.126014 0.245819 0.134149 10.50000 0.04116 0.01783 = 0.02873 -0.00442 0.00366 -0.00946C2 1 0.021315 -0.103724 0.184318 10.50000 0.02054 0.01953 = 0.01435 -0.00137 0.00099 0.00160O2 2 0.008896 -0.191357 0.252741 10.50000 0.04893 0.02094 = 0.02342 0.00376 0.00923 0.00349C3 1 -0.081841 -0.177419 -0.115438 10.50000 0.01853 0.01948 = 0.01636 -0.00508 0.00216 -0.00150O3 2 -0.103377 -0.276377 -0.093423 10.50000 0.03593 0.02098 = 0.03320 -0.00557 0.00597 -0.00602 . . . .C11 1 -0.287792 -0.062646 0.088306 10.50000 0.02252 0.03041 = 0.02577 -0.00620 0.00431 -0.00036O11 2 -0.394246 -0.123109 0.074063 10.50000 0.02464 0.04749 = 0.05354 -0.01977 0.00761 -0.00933C12 1 -0.225303 0.177277 0.034806 10.50000 0.01673 0.03692 = 0.02324 -0.00162 0.00748 -0.00024O12 2 -0.293440 0.264375 0.000260 10.50000 0.02529 0.04527 = 0.03472 -0.00706 0.01154 -0.00290PART 0
Example 3 - Fe3(CO)12
Disorder Webinar 54
Example 3 - Fe3(CO)12 - shelXle diagrams
November 13, 2012
50% Thermal Ellipsoids
100K 298K
Example 3 - Fe3(CO)12 Bond Lengths
100K• Fe(1) -Fe(2) 2.6932(3) Å• Fe(2)- Fe(3) 2.6993(3) Å• Fe(2)-Fe(3) 2.5591(4) Å
• Fe(2)-C(1) 2.013(4) Å• Fe(2)-C(2) 1.985(4) Å• Fe(3)-C(1) 1.989(4) Å• Fe(3)-C(2) 2.020(4) Å
298K• Fe(1) -Fe(2) 2.6766 (11) Å• Fe(2)- Fe(3) 2.6806 (11) Å• Fe(2)-Fe(3) 2.5547 (12) Å
• Fe(2)-C(1) 2.095(8) Å• Fe(2)-C(2) 2.042(8) Å• Fe(3)-C(1) 2.063(8) Å• Fe(3)-C(2) 2.129(8) Å
Example 3 - Fe3(CO)12 Bond Lengths
100K• Fe(1) -Fe(2) 2.6932(3) Å• Fe(2)- Fe(3) 2.6993(3) Å• Fe(2)-Fe(3) 2.5591(4) Å
• Fe(2)-C(1) 2.013(4) Å• Fe(2)-C(2) 1.985(4) Å• Fe(3)-C(1) 1.989(4) Å• Fe(3)-C(2) 2.020(4) Å• C(1)-O1) 1.161(3) Å• C(2)-O(2) 1.163(3) Å
298K• Fe(1) -Fe(2) 2.6766 (11) Å• Fe(2)- Fe(3) 2.6806 (11) Å• Fe(2)-Fe(3) 2.5547 (12) Å
• Fe(2)-C(1) 2.095(8) Å• Fe(2)-C(2) 2.042(8) Å• Fe(3)-C(1) 2.063(8) Å• Fe(3)-C(2) 2.129(8) Å• C(1)-O1) 1.088(6) Å• C(2)-O(2) 1.112(6) Å
Example 3 - Fe3(CO)12 Comparison of 100K and 298K Structures
Example 3 - Fe3(CO)12 Conclusions
Collection of low temperature (100K) data to high resolution (0.40 Å) facilitates the separation of overlapped peaks in the disordered structure.
The refinement of the structure using the PART -1 instruction in SHELX(TL) allows a stable refinement of the Fe3(CO)12
structure with no restraints.The final structure a 100K exhibits only minor deviations from
idealized C2v molecular symmetry with symmetrically bridging carbonyl ligands.
Although the results of the lower resolution (0.65 Å) dataset collected at room temperature are not as precise, both structures are nearly superimposable with the 100K structure.
Review and Conclusions
• Use of text editors or graphical editors to insert additional instructions into SHELXL / XL input instruction (*.ins) files.
• Introduction to the concepts of constraints and restraints in crystal structure refinement.
• Introduction to the internal atom connectivity concept and PART n and PART –n instructions.
• Introduction to the use of free variables in constraints and restraints.
• Examples on commonly used instructions for applying a variety of constraints and restraints to the refinement of disordered structures.
60November 13, 2012
61Disorder Webinar
Questions and Answers
Any questions?
Please type any questions you may have for our speakers in the Q&A panel and click Send.
How did we do?
When you exit the webinar, please fill out our evaluation survey to let us know. We appreciate your feedback.
Thank you!
November 13, 2012
62
Available at www.bruker.com/service/education-training/web
inars/sc-xrd.html
November 13, 2012
Innovation with Integrity
© Copyright Bruker Corporation. All rights reserved.