Before using DIFABS, it is suggested that you read the following journal article:
A web page to refer to is "Should DIFABS be Banned?": Within the limitations described in the original DIFABS paper above, it does seem(?) to be a method worth keeping in your crystallographic bag of tricks to get the job done under some circumstances. Anyway, one of the brilliant advantages of WinGX is that it does let you easily test out a variety of absorption correction methods for yourself. (It should be repeated that for many of these methods, WinGX uses the facilities within Ton Spek's Platon [CCP14 Web Mirror]. This is available for UNIX, DOS and Windows). It should also be noted that for any other crystallographic software to access this wide variety of absorption correction methods, one easy way to do this is just make your software "Platon Friendly". This also gives you the ability to access a wide variety of analytical algorithms inside Platon. |
This example uses the raw hkl data used to determine the crystal
structure of Cs2TiSi6O15
as published in I.E. Grey, R.S. Roth, M.L. Balmer,
Journal of Solid State Chemistry, 131, 38-42 (1997).
The starting information that is normally known before starting DIFABS is:
While the face information is available, DIFABS would normally be used if this information did not exist. Though it is possible that DIFABS could be used on top of face index absorption corrected data to tweak the absorption(?) A solved crystal structure where major atoms are refined isotropically and no extinction or similar corrections have been applied. The following structure solved with DIRDIF PATTY and refined on Shelx93 are given.
CS1 2 0.11731 0.24153 0.41000 11.00000 0.03096 TI1 4 0.25000 0.25000 0.00000 10.50000 0.01342 SI1 3 -0.00258 0.23084 -0.14685 11.00000 0.01192 SI2 3 0.32391 0.42925 -0.16631 11.00000 0.01826 SI3 3 0.34547 0.04062 -0.14463 11.00000 0.02011 O1 1 0.38138 0.23869 -0.17005 11.00000 0.03024 O2 1 0.30201 0.05003 -0.05699 11.00000 0.02313 O3 1 0.10762 0.18887 -0.07442 11.00000 0.02694 O4 1 0.24671 0.45812 -0.27011 11.00000 0.04433 O5 1 0.00000 0.23843 -0.25000 10.50000 0.07003 O6 1 0.26365 0.42388 -0.09048 11.00000 0.02687 O7 1 -0.08560 0.07390 -0.14348 11.00000 0.05764 O8 1 -0.04847 0.41752 -0.12658 11.00000 0.04985 |
At this point we have solved the structure, refined the atoms
isotropically and wish to apply DIFABS to generate the
absorption corrected data. This assumes you have
an Unmerged HKL file with direction cosines. Run WinGX to
bring up the following Menu Bar
Click on Absorb, RefDelF, DIFABS, General menu option to enter the Difabs absorption correction option.
WinGX will pass this over to Platon, where it will generate a output window (as shown below). WinGX may also quickly bring up then quit a Platon absorption surface plot.
After removing the Platon summary screen, WinGX will leave a confirmation dialog box stating .
The thorough Platon output is in the absorb.lst file where you can check the exact corrections Platon has performed on your HKL data. This can be accessed easily via the top menu under Analyse, Open List File, ABSORP
1975 Reflections corrected (Corrections on F**2) Min. absorption corr. = 0.378 for -16 0 -4 at Phi(P)= 176.4 Mu(P)= 4.2 Phi(S)= 298.9 Mu(S)= 4.2 Max. absorption corr. = 2.103 for 0 0 -2 at PHI(P)= 89.8 MU(P)= 0.0 PHI(S)= 264.1 MU(S)= 0.0 Average absorption correction = 1.017 Minimum, Maximum Virtual Transmission: 0.101 0.564 ***VALUES FOR COEFFICIENTS*** 0.6365 -0.0138 0.0010 0.2830 0.0040 0.0010 -0.2128 -0.0685 0.2126 0.0774 -0.0081 -0.0134 0.4841 0.1398 -0.0280 -0.1501 -0.0016 -0.0183 -0.0097 0.1744 -0.1732 -0.0187 -0.0024 -0.3789 0.0258 -0.0014 0.0599 0.4957 0.0355 -0.2192 0.0344 -0.0450 0.0860 -0.0889 -0.7614 -0.3010 -0.1200 -0.0308 -0.0602 0.1350 0.2162 0.3040 0.1385 -0.6506 Absorption Surface for PHI = 0 ---> 170 WITH MU RANGE = 0 ---> 30 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 30 1.49 1.63 1.75 1.83 1.85 1.84 1.87 1.94 2.01 2.03 1.95 1.78 1.60 1.43 1.30 1.17 1.03 0.90 25 1.53 1.68 1.80 1.87 1.87 1.85 1.88 1.96 2.06 2.10 2.02 1.85 1.63 1.43 1.26 1.11 0.97 0.84 20 1.58 1.74 1.87 1.93 1.91 1.89 1.91 2.00 2.12 2.17 2.11 1.92 1.68 1.45 1.25 1.07 0.92 0.79 15 1.65 1.82 1.95 2.00 1.98 1.95 1.97 2.06 2.20 2.27 2.22 2.02 1.76 1.49 1.26 1.05 0.88 0.74 10 1.73 1.92 2.06 2.11 2.09 2.05 2.07 2.17 2.31 2.40 2.35 2.15 1.86 1.56 1.29 1.06 0.86 0.71 5 1.84 2.03 2.18 2.25 2.24 2.21 2.24 2.35 2.49 2.58 2.51 2.30 2.00 1.67 1.37 1.10 0.87 0.69 0 2.05 2.20 2.32 2.41 2.48 2.56 2.69 2.86 2.99 2.99 2.82 2.51 2.17 1.85 1.55 1.26 0.98 0.72 |
The new Shelx ready absorption corrected
hkl file is titled difabs_g.hkl. You can rename you cosine
based hkl file with direction cosines to a backup file then copy the
difabs_g.hkl file to be the project hkl file that will be used for
solution and/or refinement. Though for refinement, WinGX can
detect what types of HKL files you have allowing you to vary
the type of absorption corrected HKL files you are using so you
can explore the effect of absorption on the structure.
|