source: trunk/tutorial3/alumina11.html @ 930

Last change on this file since 930 was 930, checked in by toby, 11 years ago

rcs:* properties removed

  • Property svn:keywords set to Author Date Revision Id
File size: 6.7 KB
Line 
1<html>
2<head>
3   <title>EXPGUI Alumina Tutorial, part #11</title>
4</head>
5
6<style>
7A:link {text-decoration:none}
8A:vlink {text-decoration:none}
9</style>
10
11<BODY BGCOLOR="#FFFFFF"
12      topmargin="0" leftmargin="0" marginwidth="0" marginheight="0" 
13      text="#000000" link="#0033ff" vlink="#0033ff" alink="#0033ff">
14
15<?
16   include("/var/www/include/navigation.inc");
17   include("/var/www/include/utility.inc");
18?>
19<blockquote>
20
21<TABLE BORDER BGCOLOR="#FFFF40" ALIGN=RIGHT>
22<TR><TH><A  Href="index.html">Intro</A> 
23</TH><TH><A Href="alumina10.html">Previous page</A>
24</TH></TR></TABLE><BR CLEAR=ALL>
25
26
27<CENTER>
28<H1><HR noshade width="75%" size="2" align="center">
29GSAS/EXPGUI Alumina tutorial (part 11)<BR>
30Finishing Up
31<HR noshade width="75%" size="2" align="center">
32</H1>
33</CENTER>
34In the previous refinement steps we obtained quite good agreement
35between the computed pattern and the observed diffraction data,
36with a Chi<sup>2</sup> value of 2. CW neutron data have very regular peaks
37shapes and can often provide significantly better fits. Improving the fit
38allows perhaps slightly better values for the derived parameters, but more
39importantly offers smaller error estimates (standard uncertainties).
40<P>
41In this final step we will allow each atom to have independent U<sub>iso</sub>
42parameter, we will add more background terms,
43and we will switch the profile function to use the
44Finger-Cox-Jephcoat asymmetry parameter, which does a better job of modeling
45low-angle asymmetry.
46
47<P><HR noshade width="50%" size="4" align="center"><P>
48To delete the displacement factor constraint, go to the Constraints panel,
49and select the Delete check button to the right of the constraint, as shown
50below. Then press the "Delete" button below it, near the bottom of the window.
51<P>
52<img alt="tutorial screen dump" src="11a.gif" align=RIGHT>
53<BR clear=all>
54<P><HR noshade width="50%" size="4" align="center"><P>
55<img alt="tutorial screen dump" src="11b.gif" align=RIGHT>
56The program then prompts, as shown to the right, to confirm deleting the
57constraint.
58<BR clear=all>
59<P><HR noshade width="50%" size="4" align="center"><P>
60
61The constraint no longer appears in the panel, as shown below.
62<P>
63<img alt="tutorial screen dump" src="11c.gif" align=RIGHT>
64<BR clear=all>
65<P><HR noshade width="50%" size="4" align="center"><P>
66
67<img alt="tutorial screen dump" src="11d.gif" align=RIGHT>
68To increase the number of background terms, select the Histogram panel,
69and press the "Edit Background" button,
70as was done <a href="alumina4.html">in part #4</a>. The window to
71the right is then created. Click on the
72"Number of terms" pull-down menu and select 12 terms.
73<BR clear=all>
74<P><HR noshade width="50%" size="4" align="center"><P>
75
76To change the profile function, go to the Profile panel and press the
77"Change Type" button. This opens a window where the function can be selected
78and where the starting value for each profile parameter can be set.
79Set the function type to 3 and see how
80the number of terms expands to what is seen below.
81<img alt="tutorial screen dump" src="11e.gif" align=RIGHT>
82<BR clear=all>
83<blockquote><I>
84Note that the instrument parameter file usually contains default
85values for the various profile terms. The values constitute the left-hand
86columns of buttons. Where two terms are used the same way in
87different profile functions, the previous value is shown in the right-hand
88column.
89<P>
90The profile functions are described in detail in the
91<A HREF="http://www.nist.gov/cgi-bin/exit_nist.cgi?url=http://www.ccp14.ac.uk/ccp/ccp14/ftp-mirror/gsas/public/gsas/manual/GSASManual.pdf">
92GSAS documentation</A>.
93For the CW neutron and x-ray functions, the functions can be
94summarized as follows:
95<UL>
96<LI>Type 1: Simple Gaussian peak shapes, poor asymmetry correction;
97appropriate for CW neutrons only.
98<LI>Type 2: Pseudo-Voight function, poor asymmetry correction; good for
99refinements where low-angle peaks are not significant
100<LI>Type 3: Similar to type 2, except this includes the Finger-Cox-Jephcoat
101asymmetry correction. Good even with significant low angle peaks.
102<LI>Type 4: Similar to type 3, except this includes the Stephens model for
103anisotropic strain broadening (where different classes of reflections
104have different widths).
105</UL>
106
107</I></blockquote>
108<P><HR noshade width="50%" size="4" align="center"><P>
109As is shown below,
110press on the "Current" button for GU, GV and GW, to change the starting
111values for those parameters to what was obtained previously, rather than the
112values in the instrument parameter file.
113<P><img alt="tutorial screen dump" src="11f.gif" align=RIGHT>
114<BR clear=all>
115<P><HR noshade width="50%" size="4" align="center"><P>
116
117Again select GU, GV and GW for refinement in the Profile panel, as below.
118<P>
119<img alt="tutorial screen dump" src="11g.gif" align=RIGHT>
120<BR clear=all>
121<P><HR noshade width="50%" size="4" align="center"><P>
122
123With these extra parameters, the refinement converges with still better
124values for the agreement factors, as seen below.
125<P>
126<img alt="tutorial screen dump" src="11h.gif" align=RIGHT>
127<BR clear=all>
128<P><HR noshade width="50%" size="4" align="center"><P>
129
130Since the diffractometer constants and profile terms have been changed,
131POWPREF should be run again.
132<P>
133<img alt="tutorial screen dump" src="11i.gif" align=RIGHT>
134<BR clear=all>
135<P><HR noshade width="50%" size="4" align="center"><P>
136
137The Chi<sup>2</sup> and Rwp improve even more when GENLES is run, indicating
138that the previous run of POWPREF was needed.
139<P>
140<img alt="tutorial screen dump" src="11j.gif" align=RIGHT>
141<BR clear=all>
142<P><HR noshade width="50%" size="4" align="center"><P>
143
144The resulting fit is quite good. By turning on "Cumulative
145Chi-squared" in the Options menu, the purple line diagonal line
146is shown. This highlights the worst fit areas of the pattern in terms
147of their impact on the weighted profile r-factor (Rwp) or in Chi<sup>2</sup>.
148<P>
149<img alt="tutorial screen dump" src="11k.gif" align=RIGHT>
150<BR clear=all>
151<blockquote><I>
152The Cumulative Chi-squared plot was first demonstrated by W. I. F. David
153at the Accuracy in Powder Diffraction Meeting-III (2001) (See the
154<a href="http://www.ncnr.nist.gov/xtal/software/expgui/liveplot.html#Cchi2">LIVEPLOT documentation</a> for more information.
155The line would
156have a constant slope (slope=1), if all regions of the data are fit at the
157statistically expected level. The areas where the Cumulative Chi-squared
158has a much greater slope are regions that have poorer fits.
159</I></blockquote>
160<HR>
161Previous: <A HREF="alumina10.html">
162Group U<sub>iso</sub> parameters &
163Refine coordinates and Overall U<sub>iso</sub>.
164
165</A>
166
167<HR>
168</blockquote>
169<P><font size=-1><A HREF="MAILTO:crystal@NIST.gov?subject=WWW page <?=$PHP_SELF?>">Comments, corrections or questions: crystal@NIST.gov</A></font><BR>
170<font size=-1><? lastmod(); ?> 
171<BR>$Revision: 930 $ $Date: 2009-12-04 23:14:35 +0000 (Fri, 04 Dec 2009) $</font>
172</body>
173</html>
Note: See TracBrowser for help on using the repository browser.