| 1 | <html> |
|---|
| 2 | <head> |
|---|
| 3 | <title>EXPGUI Alumina Tutorial, part #6</title> |
|---|
| 4 | </head> |
|---|
| 5 | |
|---|
| 6 | <style> |
|---|
| 7 | A:link {text-decoration:none} |
|---|
| 8 | A: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="alumina7.html">Next page</A> |
|---|
| 24 | </TH><TH><A Href="alumina5.html">Previous page</A> |
|---|
| 25 | </TH></TR></TABLE><BR CLEAR=ALL> |
|---|
| 26 | |
|---|
| 27 | |
|---|
| 28 | <CENTER> |
|---|
| 29 | <H1><HR noshade width="75%" size="2" align="center"> |
|---|
| 30 | GSAS/EXPGUI Alumina tutorial (part 6)<BR> |
|---|
| 31 | Plotting the Initial Fit |
|---|
| 32 | <HR noshade width="75%" size="2" align="center"> |
|---|
| 33 | </H1> |
|---|
| 34 | </CENTER> |
|---|
| 35 | <I>While Chi-squared and R-values provide some measure of how a fit is |
|---|
| 36 | progressing, the only way to actually understand the quality of the |
|---|
| 37 | fit and what problems need to be corrected, is to look at agreement between |
|---|
| 38 | the observed diffraction data and the corresponding values computed |
|---|
| 39 | from the fit. This is sometimes called a Rietveld plot.</I> |
|---|
| 40 | <P> |
|---|
| 41 | The GSAS program POWPLOT and the EXPGUI program LIVEPLOT allow the |
|---|
| 42 | fit to be examined graphically. In this tutorial step, LIVEPLOT is used |
|---|
| 43 | to examine the results. |
|---|
| 44 | <P><HR noshade width="50%" size="4" align="center"><P> |
|---|
| 45 | Press the LIVEPLOT button on the button bar (or use a menu command) to |
|---|
| 46 | start LIVEPLOT. You should then see a plot like the one below. |
|---|
| 47 | <P><img alt="tutorial screen dump" src="6a.gif" align=RIGHT> |
|---|
| 48 | <BR clear=all> |
|---|
| 49 | <P><blockquote><I> |
|---|
| 50 | In this plot, the data are "X" characters, the calculated values are |
|---|
| 51 | a red line. The fitted background is shown as a green line. Offset below, |
|---|
| 52 | the observed pattern minus the computed pattern is shown in blue. |
|---|
| 53 | The observed and computed values do not agree very well, but seem to follow |
|---|
| 54 | the same trends. In the subsequent plots we will see in more detail what |
|---|
| 55 | some of the discrepancies are. |
|---|
| 56 | <P> |
|---|
| 57 | Note that the size of the plot can be changed by changing the size of the |
|---|
| 58 | window. |
|---|
| 59 | </I></blockquote> |
|---|
| 60 | <P><HR noshade width="50%" size="4" align="center"><P> |
|---|
| 61 | It will help to see the actual positions of the reflections. This display |
|---|
| 62 | can be turned on by pressing the "1" key (1 for phase 1, 2 for phase 2...) |
|---|
| 63 | in the LIVEPLOT window. (This can also be done using the Tickmarks |
|---|
| 64 | submenu in the File menu.) |
|---|
| 65 | <P><img alt="tutorial screen dump" src="6b.gif" align=RIGHT> |
|---|
| 66 | <BR clear=all> |
|---|
| 67 | <P><blockquote><I> |
|---|
| 68 | Note that the color, length, style, placement of the tick mark lines can be |
|---|
| 69 | changed in the Options menu. |
|---|
| 70 | <P> |
|---|
| 71 | Note also that the reflection indices for a tick mark can be |
|---|
| 72 | displayed by pressing the "h" key when the cursor is positioned |
|---|
| 73 | over a tick mark. |
|---|
| 74 | </I></blockquote> |
|---|
| 75 | <P><HR noshade width="50%" size="4" align="center"><P> |
|---|
| 76 | |
|---|
| 77 | With all the data displayed, at appears that the tick marks are in the |
|---|
| 78 | right places at lower angles, but are not well placed at 140 degrees and |
|---|
| 79 | higher. It would be good to see the plot at high magnification to see more |
|---|
| 80 | detail, however. "Zooming in" is accomplished by clicking the mouse in the |
|---|
| 81 | lower left and upper right corners of the region to be viewed |
|---|
| 82 | (lower right & upper left also works). A box is displayed, as below, after the |
|---|
| 83 | first mouse click. After the second the plot is redrawn. |
|---|
| 84 | <P><img alt="tutorial screen dump" src="6c.gif" align=RIGHT> |
|---|
| 85 | <BR clear=all> |
|---|
| 86 | <P><HR noshade width="50%" size="4" align="center"><P> |
|---|
| 87 | |
|---|
| 88 | Note with the plot zoomed in, it can now be seen that |
|---|
| 89 | the lattice parameters do not index the peak 42 degrees very well. |
|---|
| 90 | The computed peak widths agree reasonably well with the observed. |
|---|
| 91 | <P><img alt="tutorial screen dump" src="6d.gif" align=RIGHT> |
|---|
| 92 | <BR clear=all> |
|---|
| 93 | <P><HR noshade width="50%" size="4" align="center"><P> |
|---|
| 94 | |
|---|
| 95 | The fit is even worse at high angle. |
|---|
| 96 | Also, the computed peak widths are much narrower than the observed. |
|---|
| 97 | <P><img alt="tutorial screen dump" src="6e.gif" align=RIGHT> |
|---|
| 98 | <BR clear=all> |
|---|
| 99 | <HR> |
|---|
| 100 | Previous: <A HREF="alumina5.html"> |
|---|
| 101 | Initial Fitting: Refine Scale Factor and Background |
|---|
| 102 | </A> |
|---|
| 103 | <BR> |
|---|
| 104 | Next step: <A HREF="alumina7.html"> |
|---|
| 105 | Fitting the Unit Cell |
|---|
| 106 | </A> |
|---|
| 107 | |
|---|
| 108 | <HR> |
|---|
| 109 | </blockquote> |
|---|
| 110 | <P><font size=-1><A HREF="MAILTO:crystal@NIST.gov?subject=WWW page <?=$PHP_SELF?>">Comments, corrections or questions: crystal@NIST.gov</A></font><BR> |
|---|
| 111 | <font size=-1><? lastmod(); ?> |
|---|
| 112 | <BR>$Revision: 625 $ $Date: 2009-12-04 23:09:18 +0000 (Fri, 04 Dec 2009) $</font> |
|---|
| 113 | </body> |
|---|
| 114 | </html> |
|---|
