Solving An X-Ray Crystal - Structure
Solving An X-Ray Crystal - Structure
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
II. XPrep
Screen shows: P A B C I ……
0.0 19.0 21.0 ……
These symbols are abbreviations for lattices (e.g. face centered, body centered, …)
You have to look for the one with the smallest value, which also determines the
lattice type for your crystal. If two columns are 0 chose the one with higher
symmetry.
Select [H] and press enter and go through commands
Before you type [F]: press [A] for absorption correction (only if crystal
dimensions have been measured before data collection)
This part of the program creates a new set of files: *.hkl (including the data)
*.ins (input file for next step)
*.prp (listing of what
happened in XPrep)
III. XS
Input for this program: *.ins and *.hkl
CFOM (Combined Figures of Merit): The lower this value, the better is your
collected data set (as low as 0.036)
RE: if this value is below 10% your data is good
The output from this program is: *.res (including the results)
*.lst (listing of what happened in XS
including an electron-density map)
IV. XP
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
help (gives you a list of all commands. Type help …. And receive a
description of this specific command)
info (lists all atoms and bonds of your current moiety after fmol has
been applied)
fmol (creates a list of atoms with according connectivities. If there is
still fmol occurring in the prompt, just press enter until the list is
complete. This command has to be applied everytime you enter
XP)
fmol/n (does the same as fmol but without showing on the screen. Later
on, if you are in unique mode, by typing fmol/n you can go back
to the entire molecule)
mpln (mean least square plane: calculates the best view of the electron-
density map to facilitate looking at the structure)
mpln/n (again, no visual output)
fmol/n $n (you will only see the nitrogen atoms)
fmol/n rest (you will see the rest of the atoms without the nitrogens)
proj (project: view structure: mono 1-dimensional
stereo 3-dimensional)
eyes 1 (always view structure in 1-D)
eyes 2 (always view structure in 3-D)
uniq ho1 (unique: view only the specified atom here Ho1 and what it is
connected to. If you then type proj you will only see Ho1 and its
immediate surroundings)
diag (like proj without a menu. If you now press enter, the picture of
the structure will stay in the top right corner of your screen even if
you are in prompt screen)
kill q40 q3 (deletes specified atoms here atoms q40 and q3 irreversibly. If you
now type proj, the deleted atoms will not show up any more)
read kv01 (if you accidentally deleted an atom and you want it back, this
command will re-read the *.res file until ‘END’ in the *.res file)
reap kv01 (reads complete *.res file. After read or reap, you type fmol and
proj and all the original atoms are back)
bang q22 q1 (calculates all bond angles and bond distances for the selected
atoms)
name q22 n1 (if you are sure about the identity of a specific electron density
here q22 you can name it here Nitrogen1 or n1. Note: you can’t
give two separate atoms the same name)
pick (in the window you can browse through atoms of a specific moiety
and label them)
sort (if you type sort and then info, you will see all the acquired
information so far sorted by numbers)
sort $o $n (sorts and lists all the oxygen atoms first and then all the nitrogen
atoms sorted by numbers)
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
V. XL
Run XL (this is the first step of the refinement. After computer is done, just press
any key)
VI. XP
info (after fmol and mpln type info and look at the column with the Ueq
values: All values for the assigned atoms should be roughly in a
range between 0.02x and 0.05x.
If you get 0.1xx you assigned too many electrons to that
peak. Kill, because it is a noise peak
If you get 0.000 Too little electrons assigned. Take an atom
with higher atomic number
If you then are missing an atom in your structure, take the q with
the highest value, which should be much higher than all the others,
and assign it an atom name)
proj less $q (projects everything except q’s)
kill $q (kill all the remaining q’s that aren’t atoms)
file kv01
VII. XL
VIII. Edit
copy .res .ins (takes the *.res file from XL and writes a new *.ins file for
further steps)
Edit .ins (opens window with the data of the *.ins file)
Below ‘PLAN 20’ make a new line and type in a new command:
‘ANIS’ for the atoms you wish to make anisotropic
Save and exit Editor
IX. XL
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
Run XL refinement cycle again (now XL takes sphere representation of the atoms
and calculates ellipsoid representation which is more accurate)
X. XP
Now you can also see some of the protons in the structure as q’s. Usually you
don’t need to refine protons except Hydroxide-protons or H-bonded protons. For
all the other protons you just add them later in optimized calculated positions. If
you want to refine the protons, just assign q’s as protons and go back and forth
between XL and XP to refine them. The Ueq values of protons are usually a little
bigger than the ones for other atoms.
kill $q
hadd $h $c (adds protons to all unsaturated carbons. Check if all protons are in
the right postion and if there is the right number of protons on all
atoms. Alternatively, protons can be added in the *.ins file. Under
‘PLAN 20’ write ‘HFIX 137 c1 c2’ to add protons to c1 and c2)
atyp 2 5 $c (set colors and style of all carbon atoms. First number encodes the
type, second number encodes the color of the chosen atoms. Type
help atyp to see a list of all the colors and styles)
file kv01 (writes new *.ins file with anisotropic refinement)
XI. XL
XII. Edit
Edit .ins
See ‘Afix 43’ (different for every atom. First number tells you the hybridization
of the atom. 4 is aromatic carbon, 3 is methyl carbon, 2 is
methylene carbon)
See ‘Afix 0’ (terminates the project for one atom and moves on to the next)
copy .res .ins
Edit .ins
Below ‘TEMP -100’ make a new line and type in a new command:
‘OMIT -3 55’ (this tells XL to only use reflections within 2 θ = 55o)
Save and exit Editor
XIII. XL
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
Run another XL refinement cycle (Before you exit XL look at the WR2 and R1
value)
Overwrite this line with the other line from the bottom (This tells XL to use a
weighting-scheme of the reflections including errors which is proposed by the
computer)
Save and exit Editor
XVI. XL
Run another XL refinement cycle (Now WR2 and R1 value should have gone
down)
Go back into *.ins file, copy the ‘WGHT …’ line again, run XL again… Go back
and forth until WR2 value doesn’t change significantly any more)
XVII. XP
kill $q
info less $h (you can see that the labels are not in order)
pick (rename all the atoms in an orderly fashion)
sort (now atoms are sorted in info-list)
sort $cl $n (sort in order of atomic numbers)
hadd $c (H’s are placed after the Carbon they are attached to)
save kv02 (Saves this session. Take a name different from the overall project
name. This is a good point the save because all the atoms are
labeled, refined, …)
next kv02 (To continue where you left from with the save command)
fmol (creates a list of atoms with according connectivities. If there is
still fmol occurring in the prompt, just press enter until the list is
complete)
file kv01
XVIII. Edit
Edit .ins
Below ‘L.S. 4’ make two new lines and type in the new commands:
‘ACTA’ and
‘BOND $H’
Save and exit Editor
XIX. XL
Run another XL refinement cycle (now XL creates a new set of files: *.cif
crystallographic information file and *.fcf)
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
XX. Edit
XXI. XL
XXII. XP Codes
matr 1 1 1 (makes the diagonal of the unit cell perpendicular to the screen)
↑ (up-arrow on keyboard brings back old commands)
cell (displays unit cell parameters. Put shortest axis perpendicular to
screen with matr command)
pack (packs molecules in unit cell. Dashed lines represent
intermolecular interactions.
‘Scan Mols.’ unit cell blinks (like pick command)
‘scan non-B’ scans dashed lines and delete what you don’t
need or want
‘sgen/fmol’ displays what you constructed in packing
diagram)
info (all those symmetry operations added new labels to the atoms)
name ???a ???(renames atoms that had an XXXA label back to XXX)
mpln (gives the best plane with deviations of atoms from that plane)
nopl (undoes all mpln commands and deletes all previous planes)
mpln n1 c1 … (creates best plane including specified atoms)
mpln n2 c6 … (creates this plane and shows you the dihedral angle between this
plane and the previous plane. The numbers on the left side of an
atoms show how far the atom is away from the plane. Good way
to check the planarity of a ring. If numbers are really small you
know that the specified atoms are in plane)
telp 0 -50 0.025(telp creates a drawing of the molecule. In the drawing, put labels
next to atoms. First number encodes type of drawing, second
number is %age of thermal ellipsoids, and third number is
thickness of the bonds. After you are done, press ‘Esc’ or ‘h’ to
exit without saving, press ‘Enter’ or ‘b’ to exit with saving)
exam *.plt (exam works like a directory command and shows you a list of all
files with .plt extension)
labl 1 400 (determines which atoms get labeled and in what style they get
labeled)
link 5 $h (makes all bonds to hydrogen atoms a line to distinguish them
from other bonds)
view drawing1 (to see the drawing on screen)
draw drawing1(chose ‘h’ for hpgl-file and then go through commands. Now you
can insert this *.hpgl file into word and see it as a picture)
cent n1 c1… (calculates the center of the specified ring)
cent/x n1 c1… (calculates the center of the specified ring and puts an imaginary
atom in that position. If you need to calculate the distance between
two rings, this is a good method)
XXIII. Edit
Since XP can’t calculate the error of a value (which you need in publications),
you have to do this in the *.ins file
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
Finalizing a project
I. Modify *.cif
Open modify.cif
Open *.cif
Replace the specified line with those from the modify.cif file
The ‘number of reflection used’ can be found in the *.p4p file
Save *.cif
II. XCIF
Table of choices:
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
IV. *.exp.doc
V. XShell
VI. Drawings
Make one black and white and one colored drawing of the molecule in XP
Insert the drawings into a word document and save it as *.drw.doc
Link 5 $h
The Center for X-Ray Crystallography
University of Florida
Written by Jürgen Koller, amended by Annaliese Thuijs
Labl 1 400
Telp 0 -50 0.025
II. Database
IV. Finish