[CCP14 Home: (Frames | No Frames)]
CCP14 Mirrors: [UK] | [CA] | [US] | [AU]

(This Webpage Page in No Frames Mode)


Methods, Problems and Solutions - and Tutorials

GSAS (General Structure Analysis System) Rietveld powder diffraction and Single Crystal software

Doing Le Bail (whole pattern structureless) fitting in GSAS via Brian Toby's EXPGUI interface as part of setting up for a combined X-ray / Neutron Rietveld refinement

The CCP14 Homepage is at http://www.ccp14.ac.uk

[Back to Tutorials] [Back to Problems and Solutions] [Back to GSAS Hints/Resources]
[Back to Le Bailing in GSAS via EXPEDT]

[The reference to use for GSAS in any resulting publications is: A.C. Larson and R.B. Von Dreele, "General Structure Analysis System (GSAS)", Los Alamos National Laboratory Report LAUR 86-748 (1994).]

[The reference to cite in any resulting publications for using EXPGUI is: B. H. Toby, EXPGUI, a graphical user interface for GSAS, J. Appl. Cryst. (2001). 34, 210-213]

This example is Orthorhombic Lead Sulphate used in the first Rietveld Round Robin organised by Rod Hill under the auspices of the IUCr Commission on Powder Diffraction.

Click here to download a zipped file of the PbSO4 GSAS XRD and Neutron data - as well as the Parameter files

The main strategy here is to a) add the phase of interest (PbSO4), then b) add both XRD and Neutron histograms then c) followed by Le Bail fitting both datasets together.

As noted below, Le Bail fitting in GSAS can go very easily if you use the EXPGUI "Fit Background Graphically" feature.

Information on Brian Toby's EXPGUI, with installation instructions for UNIX and MS-Windows can be obtained via the following webpages. Make sure you have the latest Beta version for the best results (personal opinion).

Run EXPGUI and goto the directory with the diffraction data and PRM files for the neutron and powder diffractometers. Enter a new file name and press READ to create a new EXP file. (of course you can open and edit an existing GSAS EXP file)

When prompted, create the new EXP file, then provide the title information leading to the 2nd EXPGUI screen.

Creating a new EXP file

EXPGUI ready for input

Click on the Phases Tab to present the following screen image, then click on the Add Phase Tab

Adding a new phase

Then input the Spacegroup and Cell information then click Add then continue to lead back to the Phase Tab with the cell and spacegroup information (but no atom information) :

S.G., a, b, c, al, be, ga := P n m a 8.472940 5.393338 6.953579 90.0000 90.0000 90.0000

(if you have the structure information in PowderCell or CIF format, you can easily importn all the structure information).

Inputting the Spacegroup and Cell information

Phases GUI with the Spacegroup and Cell Information

Now (if convenient) add the starting structure (or at least a dummy atom) using the Add Atoms GUI interface. Leading to the second screen image below.

PB     PB1   .187427   .250000   .167072  1.00  .025
S      S2    .066713   .250000   .684284  1.00  .025
O      O3   -.092390   .250000   .595179  1.00  .025
O      O4    .193956   .250000   .543152  1.00  .025
O      O5    .081057   .027005   .808874  1.00  .025

Adding the atom information

Atom information is added

Now click on the Histogram GUI interface and the Add New Histogram Icon for in this case to starting adding the first X-ray diffractogram. Then browse and select the Data file and PRM file and click Add to continue.

Note: It is quite common for GSAS instrument parameter files to have incomplete or incorrect information. Thus always check that the parameters displayed by EXPGUI make sense. In this case, you will have to correct the Ratio (of Alpha 1 and 2) to 0.5 and reset the Zero (offset) to 0. Also check that the wavelength(s) are correct.

Histrogram Interface in EXPGUI

Selecting the data and PRM file

XRD Histogram added into GSAS

Now repeat the same sequence to add in the Neutron dataset and parameter file via the Add New Histogram Icon. As there are two Banks/Sets defined by this particular instrument parameter file, select the one that is appropriate (in this case Number 1).

Adding the Neutron Histogram into EXPGUI / GSAS

Neutron Histogram has been added into EXPGUI / GSAS

To enable Le Bail mode fitting for the two datasets, select the LS Controls tab, click on the XC histogram (XRD dataset 1) and then click on Equally Weighted Le Bail Method. Now select the NC histogram (Neutron dataset 2) and then click on Equally Weighted Le Bail Method.

Enable Le Bail fitting for the XRD data

Enable Le Bail fitting for the Neutron data

Now run POWPREF and then GENLES and slowly release parameters as per normal to fit the data (remember to turn off the Scale Factor as by default this may be left on). This can also be very easily done in EXPGUI using point and click.

From Brian Toby: Personally, I would also suggest setting the cycles to zero & running GENLES 2-4 times.

Also from Brian Toby:
My recommendation for a LeBail refinement scheme would be (1) to fit 2
background terms/histogram, (2) then unit cell, (3) then zero, (4) then
neutron wavelength (5) then more background terms until the fit is good
(6) then profile terms. On occasion I need to fix the bkg terms while I
refine the profile terms. BTW, I never refine the FCJ S/L & H/L terms. I
will probably need to rerun POWPREF once or twice through this scheme
and after I do this, I rerun GENLES a few times with zero cycles until
the Rwp's settle down. 

The value of BKGEDIT comes in if one of two things happen: the Le Bail
fit does weird things with intensities, rather than draw the background
where it should be, or the background is sufficiently complex that one
cannot start the LeBail w/o a reasonable background. 

Set background, wavelength and two-theta offset in the Histogram tab; Profile parameters in the Profile Tab; UNIT Cell in the Phase tab.

Hints that you can ignore if you have better ideas:

  • Refine Unitcell and zero offset for the XRD data and refine Wavelength and zero offset for the neutron data.

  • For the profile function: try using Type 3 profile - Pseudo Voight with FCJ assymmetry correction for both the XRD and Neutron data.

  • Running Powpref resets the Le Bail intensities. Normally you would not want to reset the Le Bail intensities as then you would have to fix most parameters to allow the Le Bail intensities to refit in a stable manner.

Fitting Background while Le Bailing in GSAS ---- Use the EXPGUI "Fit Background Graphically" feature - it is totally brilliant!

(This will normally fix most of the difficulties Le Bailing in GSAS)

If having a problem fitting the background, you may like to use the very nifty and almost life saving EXPGUI manual Background Point selection and editing feature. (you then fit a Chebyshev to this and save back into the GSAS EXP file where you have the option of refining or fixing the background). This "Fit Background Graphically" feature is accessed via Edit Background in the Histogram Tab. (if you have not done this before, refer to tutorial runthrough on Using EXPGUI, BKGEDIT and GSAS to manually define a background using Chebyshev Polynomial coefficients.

Manually defining the background in EXPGUI

Fit the profile until you are happy. Then turn on the Rietveld option for both Histograms to start refining the structure. Following is the Le Bail fit for the Neutron pattern. If you have difficulties with Le Bail fitting Alpha_1,2 XRD data, you may want to turn the Rietveld option back on(?). Refer below to possible reasons why you may be experiencing difficulty here with GSAS.

If you have trouble fitting a powder XRD profile, refer to Difficulty in profiling Powder XRD data in GSAS, Peak Cutoff values and the influence of Powpref

Fit to the Neutron pattern after Le Bail fitting

From the Structure Solution from Powder Diffractometry List:
At 12:44 07/09/99 -0500, Jaap Vente wrote:
>On Tue, 7 Sep 1999, Peter Y. Zavalij wrote:
>> > Lachlan wrote :
>> > A belated reply on the topic of Le Bail extraction
>> > in GSAS:
>> > 
>> There is small problem with LeBail extraction in GSAS:
>> alpha1 and alpha2 components are refined independently...
>Yes I found that as well, but thought that I had made a mistake somewhere.
>I think it is a major problem though. Bob could you please fix that?

Not sure if Bob has subscribed to this list. The way this problem
was fixed in ARIT was to consider the two structure factor amplitudes
independently derived from the alpha 1 and 2 components, then
to make them both equal to the mean value, after each new
iteration. I did not find a better way. This is the reason why
Rp is more slowly decreasing when dealing with alpha1+2 than
with alpha 1 alone.

In FORTRAN, this operation corresponds to :
( see in ARIT source at http://sdpd.univ-lemans.fr/arit.html )

In ITER subroutine :
make both structure factors equal to the mean

       DO 1 I=1,NN
       IF(ICH(I).EQ.0)GO TO 1

In the MAIN program, IHC is the order
of the alpha2 component in the hkl list :

       DO 4 I=1,NN-1
       IF(ICC(I).EQ.5)GO TO 4
       DO 3 J=I+1,NN
       IF(ICC(J).NE.5)GO TO 3
       IF(ICO(I).NE.ICO(J))GO TO 3
     1HHH(3,I).NE.HHH(3,J))GO TO 3
       GO TO 4



PS : To subscribe, send an e-mail to sdpd-subscribe@egroups.com 

[Back to Tutorials] [Back to Problems and Solutions] [Back to GSAS Hints/Resources]
[Back to Le Bailing in GSAS via EXPEDT]

[CCP14 Home: (Frames | No Frames)]
CCP14 Mirrors: [UK] | [CA] | [US] | [AU]

(This Webpage Page in No Frames Mode)

If you have any queries or comments, please feel free to contact the CCP14