[RMCProfile-users] RMCPROFILE
helen.playford at stfc.ac.uk
helen.playford at stfc.ac.uk
Mon Apr 28 12:22:52 BST 2014
Dear H.Wu,
Thank you for your email.
The keyword XRAY_RECIPROCAL_SPACE_DATA should work (but requires the data to be in a very specific format, which we are working on an example for), but at the present time the keyword XRAY_REAL_SPACE_DATA is not implemented as there are questions over the accuracy of the Fourier transform. There are ways around this if you need to work with X-ray PDFs, which I can explain in more detail if necessary. Keywords with X_ at the start are not implemented.
I am a little confused by your second paragraph. Do you have data from SiO2? In SiO2 each Si is coordinated to four, not five O atoms. The neighbour list file you show is telling you that some atoms do not have the expected number of neighbours. If you were trying to run with the SiO2 polyhedral restraint this would have caused the errors. How did you generate the second file? It would not have produced errors as each Si atom has 4 neighbours, but it's important that the connectivity is correct (i.e. that Si atom #1 is really connected to O atoms #65, 74, 79 and 85).
Please let me know if you would like further help working with X-ray data, or if you have any other questions.
Best wishes,
Helen
From: rmcprofile-users [mailto:rmcprofile-users-bounces at rmcprofile.org] On Behalf Of ????
Sent: 22 April 2014 08:07
To: rmcprofile-users
Subject: [RMCProfile-users] RMCPROFILE
Dear author ,
I 'm working with local atomic arrangement in disorder materials. When I study the RMCProfile software, I have some questions so need your help.
In the SF6 example, I can get better results after studying the manual. But for the rmcsf6.dat file, I have some questions on the keyword of "NEUTRON_REAL(or RECIPROCAL)_SPACE_DATA". If I have X-ray diffraction data, I will modify "NEUTRON" to "XRAY", but there will be some errors; but if I modify "NEUTRON" to "X", then it will success, but there is no important "rmcsf6_PDF1.csv" and "rmcsf6_SQ1.csv" files in the makefiles. I don't know the reason.
Another question is restraint files. In the SF6 example, a S atom is adjacent with six O atoms and a O atom is adjacent with one S atom, because SF6 is unlinked octahedral material. When I research SiO2, SiO2 is linked tetrahedral material, so there exists boundary problem. In principle, a Si atom is adjacent with five O atoms and a O atom is adjacent with two Si atom. But when I use neighbour_list.exe to create restraint files, SiO2.sio is:
64
1 1 65
2 2 66 80
3 2 67 73
4 2 68 79
5 4 67 69 76 78
6 4 65 73 79 80
7 4 66 71 72 77
8 4 68 70 74 75
9 1 81
10 4 69 74 82 96
11 2 83 89
12 2 84 95
13 4 83 85 92 94
14 4 81 89 95 96
15 4 82 87 88 93
16 4 84 86 90 91
17 1 97
18 2 98 112
19 4 71 75 99 105
20 2 100 111
21 4 99 101 108 110
22 4 97 105 111 112
23 4 98 103 104 109
24 4 100 102 106 107
25 2 70 113
26 4 101 106 114 128
27 4 87 91 115 121
28 2 116 127
29 4 115 117 124 126
30 4 113 121 127 128
31 4 114 119 120 125
But there is error when I run the RMCProfile. I artificially change SiO2.sio file to SF6-like restraint files, there is no error. I don't know the reasons.This is modified file:
64
1 4 65 74 79 85
2 4 66 80 87 92
3 4 67 73 78 86
4 4 68 79 81 91
5 4 67 69 76 78
6 4 65 73 79 80
7 4 66 71 72 77
8 4 68 70 74 75
9 4 81 88 93 99
10 4 69 74 82 96
11 4 83 89 98 116
12 4 84 95 97 102
13 4 83 85 92 94
14 4 81 89 95 96
15 4 82 87 88 93
16 4 84 86 90 91
17 4 97 99 105 116
18 4 98 112 128 134
19 4 71 75 99 105
20 4 100 111 147 158
21 4 99 101 108 110
22 4 97 105 111 112
23 4 98 103 104 109
24 4 100 102 106 107
25 4 70 113 139 178
26 4 101 106 114 128
27 4 87 91 115 121
28 4 116 127 178 175
29 4 115 117 124 126
30 4 113 121 127 128
31 4 114 119 120 125
I know my expression have some questions, so I'm very sorry and hope you forgive me. Thank you very much!
Best wish!
H.Wu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rmcprofile.org/pipermail/rmcprofile-users/attachments/20140428/f1f44aa5/attachment.html>
More information about the rmcprofile-users
mailing list