Re: troublesome unit cell
Jacco I didn't actually use these cif files for my simulations, I entered the data by hand (including fractions where the software would take it). :-) The cif data is just there for other people's convenience - different powder software like different cif file formats anyhow. In any case, if there is a significant rounding error in the software, the TM 3a and 3b sites change to 6c. This space group is very finicky about the positions. The problem also occurred in a program that would take fractional coordinates, so that's not the only snag. Pam Dr Pamela Whitfield CChem MRSC Energy Materials Group Institute for Chemical Process and Environmental Technology Building M12 National Research Council Canada 1200 Montreal Road Ottawa ON K1A 0R6 CANADA Tel: (613) 998 8462 Fax: (613) 991 2384 Email: mailto:[EMAIL PROTECTED] ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca -Original Message- From: Jacco van de Streek [mailto:[EMAIL PROTECTED] Sent: October 18, 2004 1:32 PM To: [EMAIL PROTECTED] Whitfield, Pamela wrote: I've come across a unit cell that seems to give alot of software a headache when generating the powder patterns. I've tested 6 software packages with this (including commercial) and found only a 50% success rate in generating a good pattern with correct relative intensities. Not sure where the problem lies with the 'faulty' software, maybe some of the hkl multiplicities are off. These two structures (the file needs data_* inserted twice to generate a valid cif file, I believe) are in a trigonal or hexagonal space group, which has the following consequences: - There are problems with rounding errors when determining if two atoms overlap. Atoms in triclinic through tetragonal space groups do not have atoms on special positions with 0.333, but only special positions of the type 0.0, 0.25, 0.5 etc. These can be represented exactly in a text file, whereas 1/3 cannot. In the first crystal structure, they have even been truncated to 0.333, which gives errors of about 0.0003. Most computer programs will assume that floating point errors are 0.1. - The matrices representing the space-group operators are no longer diagonal, and the distinction between R*h and h*R (where R is the rotation matrix and h is the reciprocal space vector [h,k,l]) becomes important when calculating systematic absences. The current version of Mercury, the free crystal structure visualiser from the CCDC, exhibited both bugs in version 1.2, Mercury 1.2.1 which was released several weeks after Mercury 1.2 still contains the first bug: because the program that I had used to test Mercury against had the same bug... Mercury 1.3, due to be released before the end of this year should get the powder diffraction patterns for both crystal structures right without any problems (they are the same with version 1.3). The beta test of Mercury 1.3 is currently in progress. If you would like to participate in the beta test for Mercury 1.3, so that you can provide us with feedback for the simulated powder diffraction patterns of these and similar crystal structures before the final release, please contact mailto:[EMAIL PROTECTED] Best wishes, -- Jacco van de Streek Research Scientist Cambridge Crystallographic Data Centre Cambridge, United Kingdom
Re: troublesome unit cell
The cell is in space group P3112 There aren't even any structures with this space group in the ICSD (not even the other enantiomorph P3212) Hmm, I found 24 entries in ICSD for P3122, especially minerals such as Muscovite 3T but also things like V6C5 and Ag2HPO4 but not Li3MnNiCoO6 Cenzual, K.;Gelato, L.M.;Penzo, M.;Parthe, E. Inorganic structure types with revised space groups.I. (1991) Acta Crystallographica B 47, 433-439 Tordjman, I.;Boudjada, A.;Guitel, J.C.;Masse, R.Structure de l'Hydrogenophosphate D'Argent (1978) Acta Crystallographica B 34, 3723-3725 Alan.
Re: troublesome unit cell
Alan I agree - I found them as well, but this particular P3112/P3212 group is very rare. Given that software tends to deal with space group interpretation on a case-by-case basis it does mean that problems can be specific to a particular group - especially if it hasn't been tested before. Some software deals with it OK, others don't. You won't find the Li material in the database, it's a recent postulated superstructure for a layered battery material. Pam Dr Pamela Whitfield CChem MRSC Energy Materials Group Institute for Chemical Process and Environmental Technology Building M12 National Research Council Canada 1200 Montreal Road Ottawa ON K1A 0R6 CANADA Tel: (613) 998 8462 Fax: (613) 991 2384 Email: mailto:[EMAIL PROTECTED] ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca -Original Message- From: Hewat Alan [mailto:[EMAIL PROTECTED] Sent: October 18, 2004 2:42 PM To: [EMAIL PROTECTED] The cell is in space group P3112 There aren't even any structures with this space group in the ICSD (not even the other enantiomorph P3212) Hmm, I found 24 entries in ICSD for P3122, especially minerals such as Muscovite 3T but also things like V6C5 and Ag2HPO4 but not Li3MnNiCoO6 Cenzual, K.;Gelato, L.M.;Penzo, M.;Parthe, E. Inorganic structure types with revised space groups.I. (1991) Acta Crystallographica B 47, 433-439 Tordjman, I.;Boudjada, A.;Guitel, J.C.;Masse, R.Structure de l'Hydrogenophosphate D'Argent (1978) Acta Crystallographica B 34, 3723-3725 Alan.
Re: troublesome unit cell
Pam, I hope that when you used GSAS you entered (as per instructions) p 31 1 2 and not p3112. Bob R.B. Von Dreele IPNS Division Argonne National Laboratory Argonne, IL 60439-4814 -Original Message- From: Whitfield, Pamela [mailto:[EMAIL PROTECTED] Sent: Monday, October 18, 2004 12:50 PM To: '[EMAIL PROTECTED]' Alan I agree - I found them as well, but this particular P3112/P3212 group is very rare. Given that software tends to deal with space group interpretation on a case-by-case basis it does mean that problems can be specific to a particular group - especially if it hasn't been tested before. Some software deals with it OK, others don't. You won't find the Li material in the database, it's a recent postulated superstructure for a layered battery material. Pam Dr Pamela Whitfield CChem MRSC Energy Materials Group Institute for Chemical Process and Environmental Technology Building M12 National Research Council Canada 1200 Montreal Road Ottawa ON K1A 0R6 CANADA Tel: (613) 998 8462 Fax: (613) 991 2384 Email: mailto:[EMAIL PROTECTED] ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca -Original Message- From: Hewat Alan [mailto:[EMAIL PROTECTED] Sent: October 18, 2004 2:42 PM To: [EMAIL PROTECTED] The cell is in space group P3112 There aren't even any structures with this space group in the ICSD (not even the other enantiomorph P3212) Hmm, I found 24 entries in ICSD for P3122, especially minerals such as Muscovite 3T but also things like V6C5 and Ag2HPO4 but not Li3MnNiCoO6 Cenzual, K.;Gelato, L.M.;Penzo, M.;Parthe, E. Inorganic structure types with revised space groups.I. (1991) Acta Crystallographica B 47, 433-439 Tordjman, I.;Boudjada, A.;Guitel, J.C.;Masse, R.Structure de l'Hydrogenophosphate D'Argent (1978) Acta Crystallographica B 34, 3723-3725 Alan.
Re: troublesome unit cell
Bob I did indeed - I tried GSAS and it behaves itself. Pam Dr Pamela Whitfield CChem MRSC Energy Materials Group Institute for Chemical Process and Environmental Technology Building M12 National Research Council Canada 1200 Montreal Road Ottawa ON K1A 0R6 CANADA Tel: (613) 998 8462 Fax: (613) 991 2384 Email: mailto:[EMAIL PROTECTED] ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca -Original Message- From: Von Dreele, Robert B. [mailto:[EMAIL PROTECTED] Sent: October 18, 2004 3:08 PM To: [EMAIL PROTECTED] Pam, I hope that when you used GSAS you entered (as per instructions) p 31 1 2 and not p3112. Bob R.B. Von Dreele IPNS Division Argonne National Laboratory Argonne, IL 60439-4814 -Original Message- From: Whitfield, Pamela [mailto:[EMAIL PROTECTED] Sent: Monday, October 18, 2004 12:50 PM To: '[EMAIL PROTECTED]' Alan I agree - I found them as well, but this particular P3112/P3212 group is very rare. Given that software tends to deal with space group interpretation on a case-by-case basis it does mean that problems can be specific to a particular group - especially if it hasn't been tested before. Some software deals with it OK, others don't. You won't find the Li material in the database, it's a recent postulated superstructure for a layered battery material. Pam Dr Pamela Whitfield CChem MRSC Energy Materials Group Institute for Chemical Process and Environmental Technology Building M12 National Research Council Canada 1200 Montreal Road Ottawa ON K1A 0R6 CANADA Tel: (613) 998 8462 Fax: (613) 991 2384 Email: mailto:[EMAIL PROTECTED] ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca -Original Message- From: Hewat Alan [mailto:[EMAIL PROTECTED] Sent: October 18, 2004 2:42 PM To: [EMAIL PROTECTED] The cell is in space group P3112 There aren't even any structures with this space group in the ICSD (not even the other enantiomorph P3212) Hmm, I found 24 entries in ICSD for P3122, especially minerals such as Muscovite 3T but also things like V6C5 and Ag2HPO4 but not Li3MnNiCoO6 Cenzual, K.;Gelato, L.M.;Penzo, M.;Parthe, E. Inorganic structure types with revised space groups.I. (1991) Acta Crystallographica B 47, 433-439 Tordjman, I.;Boudjada, A.;Guitel, J.C.;Masse, R.Structure de l'Hydrogenophosphate D'Argent (1978) Acta Crystallographica B 34, 3723-3725 Alan.
RE: troublesome unit cell
Pam, As a developer and someone who uses other software to test against, I have been following the discussion with interest. This is a complex area and it goes to show the importance of having a few programs if only for testing purposes. I am puzzled by your cif data however as I cant get past verifying the equivalence of the number of equivalent positions for the Li site. In R-3m you have as fractional atomic coordinates for Li1: x = 0y = 0.5 and z = 0 For P3112 you have three Li sites with x, y, z's as 0.77780 0.0 0.83330 0.0 0.88890 0.83330 0.0 0.55560 0.83330 which translates to: 0.77780 0.0 5/6 0.0 0.88890 5/6 0.0 0.55560 5/6 Listed below are equivalent positions for a general site. What I dont understand is that for R-3m I get 9 non-overlapping fully occupied equivalent positions within the unit cell with a volume of 101.2975. For P3112 I get 18 non-overlapping equivalent positions with a volume three times as large of 303.8968. I was expecting to get 27 if the two structure were to be equivalent. Thus, from your cif data I am left to conclude that I am generating the wrong number of equivalent positions, or, if I am generating the correct number of equivalent positions then the two cif data sets should indeed give different powder patterns. Have you checked what the number of non-overlapping fully occupied equivalent positions for Li should be for the two structures. alan P3112 xyzs { -x+y, -x, z-1/3 -x+y, y, -z+1/3 -y, -x, -z-1/3 -y, x-y, z+1/3 x, y, z x, x-y, -z } 6 R-3m xyzs { -x, -x+y, -z -x, -y, -z -x+y, -x, z -x+y, y, z -y, -x, z -y, x-y, z y, -x+y, -z y, x, -z x-y, -y, -z x-y, x, -z x, y, z x, x-y, z ' +(-1/3, 1/3, 1/3) - -x-1/3, -x+y+1/3, -z+1/3 -x-1/3, -y+1/3, -z+1/3 -x+y-1/3, -x+1/3, z+1/3 -x+y-1/3, y+1/3, z+1/3 -y-1/3, -x+1/3, z+1/3 -y-1/3, x-y+1/3, z+1/3 y-1/3, -x+y+1/3, -z+1/3 y-1/3, x+1/3, -z+1/3 x-y-1/3, -y+1/3, -z+1/3 x-y-1/3, x+1/3, -z+1/3 x-1/3, y+1/3, z+1/3 x-1/3, x-y+1/3, z+1/3 ' +(1/3, -1/3, -1/3) - -x+1/3, -x+y-1/3, -z-1/3 -x+1/3, -y-1/3, -z-1/3 -x+y+1/3, -x-1/3, z-1/3 -x+y+1/3, y-1/3, z-1/3 -y+1/3, -x-1/3, z-1/3 -y+1/3, x-y-1/3, z-1/3 y+1/3, -x+y-1/3, -z-1/3 y+1/3, x-1/3, -z-1/3 x-y+1/3, -y-1/3, -z-1/3 x-y+1/3, x-1/3, -z-1/3 x+1/3, y-1/3, z-1/3 x+1/3, x-y-1/3, z-1/3 } 36 /* Fractional atomic coordinates for for R-3m Li1 0.0 0.5 0.0 0.5 0.0 0.0 0.5 0.5 0.0 0.7 0.8 0.3 0.16667 0.3 0.3 0.16667 0.8 0.3 0.3 0.16667 0.7 0.8 0.7 0.7 0.8 0.16667 0.7 */ /* Fractional atomic coordinates for P3112 Li 0.0 0.0 0.5 0.77780 0.0 0.8 0.77780 0.55560 0.16667 Li 0.77780 0.88890 0.5 0.0 0.88890 0.8 0.0 0.0 0.16667 Li 0.11120 0.55560 0.5 0.0 0.55560 0.8 0.0 0.0 0.16667 */
RE: troublesome unit cell
AlanI'm afraid that the cif files weren't overly good ones - I don't normally use them. For the R-3m the Li 1/2 fractional coordinate is for the z, not y. The structures are related by a Ö3a x Ö3a relationship with Z = 3 for the R-3m and Z = 9 for the P3112.Here are the STR files. You can refine the 3a and 3b positions for the P3112, but there is a strict relationship that has to be maintained between the x and y coordinates - I have to do it in launch mode to make it work. It's a very picky space group to work with.PamDr Pamela Whitfield CChem MRSCEnergy Materials GroupInstitute for Chemical Process and Environmental TechnologyBuilding M12National Research Council Canada1200 Montreal RoadOttawa ON K1A 0R6CANADATel: (613) 998 8462 Fax: (613) 991 2384Email: mailto:[EMAIL PROTECTED]ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca-Original Message-From: alan coelho [mailto:[EMAIL PROTECTED]]Sent: October 18, 2004 4:16 PMTo: [EMAIL PROTECTED]Subject: RE: troublesome unit cellPam,As a developer and someone who uses other software to test against, Ihave been following the discussion with interest. This is a complex areaand it goes to show the importance of having a few programs if only fortesting purposes.I am puzzled by your cif data however as I cant get past verifying theequivalence of the number of equivalent positions for the Li site.In R-3m you have as fractional atomic coordinates for Li1: x = 0 y = 0.5 and z = 0For P3112 you have three Li sites with x, y, z's as0.77780 0.0 0.833300.0 0.88890 0.833300.0 0.55560 0.83330which translates to:0.77780 0.0 5/60.0 0.88890 5/60.0 0.55560 5/6Listed below are equivalent positions for a general site.What I dont understand is that for R-3m I get 9 non-overlapping fullyoccupied equivalent positions within the unit cell with a volume of101.2975.For P3112 I get 18 non-overlapping equivalent positions with a volumethree times as large of 303.8968. I was expecting to get 27 if the twostructure were to be equivalent.Thus, from your cif data I am left to conclude that I am generating thewrong number of equivalent positions, or, if I am generating the correctnumber of equivalent positions then the two cif data sets should indeedgive different powder patterns.Have you checked what the number of non-overlapping fully occupiedequivalent positions for Li should be for the two structures.alanP3112xyzs{ -x+y, -x, z-1/3 -x+y, y, -z+1/3 -y, -x, -z-1/3 -y, x-y, z+1/3 x, y, z x, x-y, -z} 6R-3mxyzs{ -x, -x+y, -z -x, -y, -z -x+y, -x, z -x+y, y, z -y, -x, z -y, x-y, z y, -x+y, -z y, x, -z x-y, -y, -z x-y, x, -z x, y, z x, x-y, z ' +(-1/3, 1/3, 1/3) - -x-1/3, -x+y+1/3, -z+1/3 -x-1/3, -y+1/3, -z+1/3 -x+y-1/3, -x+1/3, z+1/3 -x+y-1/3, y+1/3, z+1/3 -y-1/3, -x+1/3, z+1/3 -y-1/3, x-y+1/3, z+1/3 y-1/3, -x+y+1/3, -z+1/3 y-1/3, x+1/3, -z+1/3 x-y-1/3, -y+1/3, -z+1/3 x-y-1/3, x+1/3, -z+1/3 x-1/3, y+1/3, z+1/3 x-1/3, x-y+1/3, z+1/3 ' +(1/3, -1/3, -1/3) - -x+1/3, -x+y-1/3, -z-1/3 -x+1/3, -y-1/3, -z-1/3 -x+y+1/3, -x-1/3, z-1/3 -x+y+1/3, y-1/3, z-1/3 -y+1/3, -x-1/3, z-1/3 -y+1/3, x-y-1/3, z-1/3 y+1/3, -x+y-1/3, -z-1/3 y+1/3, x-1/3, -z-1/3 x-y+1/3, -y-1/3, -z-1/3 x-y+1/3, x-1/3, -z-1/3 x+1/3, y-1/3, z-1/3 x+1/3, x-y-1/3, z-1/3} 36/* Fractional atomic coordinates for for R-3mLi1 0.0 0.5 0.0 0.5 0.0 0.0 0.5 0.5 0.0 0.7 0.8 0.3 0.16667 0.3 0.3 0.16667 0.8 0.3 0.3 0.16667 0.7 0.8 0.7 0.7 0.8 0.16667 0.7*//* Fractional atomic coordinates for P3112Li 0.0 0.0 0.5 0.77780 0.0 0.8 0.77780 0.55560 0.16667Li 0.77780 0.88890 0.5 0.0 0.88890 0.8 0.0 0.0 0.16667Li 0.11120 0.55560 0.5 0.0 0.55560 0.8 0.0 0.0 0.16667*/ LiMnNiCoO2 R-3m.str Description: Binary data Li3MnCoNiO6.str Description: Binary data
RE: troublesome unit cell
Alan, One of your operators is a typo. -x+y, y, -z+1/3 should be Y-X Y 1/3-Z Bob From: alan coelho [mailto:[EMAIL PROTECTED] Sent: Mon 10/18/2004 3:16 PM To: [EMAIL PROTECTED] Pam, As a developer and someone who uses other software to test against, I have been following the discussion with interest. This is a complex area and it goes to show the importance of having a few programs if only for testing purposes. I am puzzled by your cif data however as I cant get past verifying the equivalence of the number of equivalent positions for the Li site. In R-3m you have as fractional atomic coordinates for Li1: x = 0y = 0.5 and z = 0 For P3112 you have three Li sites with x, y, z's as 0.77780 0.0 0.83330 0.0 0.88890 0.83330 0.0 0.55560 0.83330 which translates to: 0.77780 0.0 5/6 0.0 0.88890 5/6 0.0 0.55560 5/6 Listed below are equivalent positions for a general site. What I dont understand is that for R-3m I get 9 non-overlapping fully occupied equivalent positions within the unit cell with a volume of 101.2975. For P3112 I get 18 non-overlapping equivalent positions with a volume three times as large of 303.8968. I was expecting to get 27 if the two structure were to be equivalent. Thus, from your cif data I am left to conclude that I am generating the wrong number of equivalent positions, or, if I am generating the correct number of equivalent positions then the two cif data sets should indeed give different powder patterns. Have you checked what the number of non-overlapping fully occupied equivalent positions for Li should be for the two structures. alan P3112 xyzs { -x+y, -x, z-1/3 -x+y, y, -z+1/3 -y, -x, -z-1/3 -y, x-y, z+1/3 x, y, z x, x-y, -z } 6 R-3m xyzs { -x, -x+y, -z -x, -y, -z -x+y, -x, z -x+y, y, z -y, -x, z -y, x-y, z y, -x+y, -z y, x, -z x-y, -y, -z x-y, x, -z x, y, z x, x-y, z ' +(-1/3, 1/3, 1/3) - -x-1/3, -x+y+1/3, -z+1/3 -x-1/3, -y+1/3, -z+1/3 -x+y-1/3, -x+1/3, z+1/3 -x+y-1/3, y+1/3, z+1/3 -y-1/3, -x+1/3, z+1/3 -y-1/3, x-y+1/3, z+1/3 y-1/3, -x+y+1/3, -z+1/3 y-1/3, x+1/3, -z+1/3 x-y-1/3, -y+1/3, -z+1/3 x-y-1/3, x+1/3, -z+1/3 x-1/3, y+1/3, z+1/3 x-1/3, x-y+1/3, z+1/3 ' +(1/3, -1/3, -1/3) - -x+1/3, -x+y-1/3, -z-1/3 -x+1/3, -y-1/3, -z-1/3 -x+y+1/3, -x-1/3, z-1/3 -x+y+1/3, y-1/3, z-1/3 -y+1/3, -x-1/3, z-1/3 -y+1/3, x-y-1/3, z-1/3 y+1/3, -x+y-1/3, -z-1/3 y+1/3, x-1/3, -z-1/3 x-y+1/3, -y-1/3, -z-1/3 x-y+1/3, x-1/3, -z-1/3 x+1/3, y-1/3, z-1/3 x+1/3, x-y-1/3, z-1/3 } 36 /* Fractional atomic coordinates for for R-3m Li1 0.0 0.5 0.0 0.5 0.0 0.0 0.5 0.5 0.0 0.7 0.8 0.3 0.16667 0.3 0.3 0.16667 0.8 0.3 0.3 0.16667 0.7 0.8 0.7 0.7 0.8 0.16667 0.7 */ /* Fractional atomic coordinates for P3112 Li 0.0 0.0 0.5 0.77780 0.0 0.8 0.77780 0.55560 0.16667 Li 0.77780 0.88890 0.5 0.0 0.88890 0.8 0.0 0.0 0.16667 Li 0.11120 0.55560 0.5 0.0 0.55560 0.8 0.0 0.0 0.16667 */
RE: troublesome unit cell
Title: Message Pam, If the other programs are in error at the hkl stage then my suggestion is to send the hkls from the apparently working programs to the authors of the apparently not working programs. Or better yet those authors should contact you as I for one am always on the look out for "apparent" discrepencies. alan -Original Message-From: Whitfield, Pamela [mailto:[EMAIL PROTECTED] Sent: Tuesday, October 19, 2004 12:05 AMTo: '[EMAIL PROTECTED]'Subject: RE: troublesome unit cell Alan In the original simulations I used them exactly as you describe :-) I modified that file to be able to start refining the TM x and y positions for some data I have. Topas is still seeing them as 3a and 3b positions, so internally they are still 1/9, 5/9 etc, as they started out. In any case it doesn't make a massive difference. Theproblems I refer to in some softwareare not subtle, they're huge, and seem to reflect a problem in hkl-related multiplicities. Pam Dr Pamela Whitfield CChem MRSCEnergy Materials GroupInstitute for Chemical Process and Environmental TechnologyBuilding M12National Research Council Canada1200 Montreal RoadOttawa ON K1A 0R6CANADATel: (613) 998 8462 Fax: (613) 991 2384Email: mailto:[EMAIL PROTECTED]ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca/ -Original Message-From: alan coelho [mailto:[EMAIL PROTECTED]Sent: October 18, 2004 5:36 PMTo: [EMAIL PROTECTED]Subject: RE: troublesome unit cell Pam, I looked at your INP file and there was a problem in regards to how atomic coordinates were defined. As you know TOPAS uses computer algebra with a tollerance of 1.0e-15 for checking the equivalence of equivalent positions. If I use x = 1/9; y = 5/9; etc,,, then the structures are similar. It is not often obvious these equations but it is easy to check using the number of equivalent positions or by inspecting the bond lengths. My guess is that the other programs may have conventions that are also not obvious. it is always good to do some checking cheers alan -Original Message-From: Whitfield, Pamela [mailto:[EMAIL PROTECTED] Sent: Monday, October 18, 2004 10:32 PMTo: '[EMAIL PROTECTED]'Subject: RE: troublesome unit cell AlanI'm afraid that the cif files weren't overly good ones - I don't normally use them. For the R-3m the Li 1/2 fractional coordinate is for the z, not y. The structures are related by a 3a x 3a relationship with Z = 3 for the R-3m and Z = 9 for the P3112.Here are the STR files. You can refine the 3a and 3b positions for the P3112, but there is a strict relationship that has to be maintained between the x and y coordinates - I have to do it in launch mode to make it work. It's a very picky space group to work with.PamDr Pamela Whitfield CChem MRSCEnergy Materials GroupInstitute for Chemical Process and Environmental TechnologyBuilding M12National Research Council Canada1200 Montreal RoadOttawa ON K1A 0R6CANADATel: (613) 998 8462 Fax: (613) 991 2384Email: mailto:[EMAIL PROTECTED]ICPET WWW: http://icpet-itpce.nrc-cnrc.gc.ca-Original Message-From: alan coelho [mailto:[EMAIL PROTECTED]]Sent: October 18, 2004 4:16 PMTo: [EMAIL PROTECTED]Subject: RE: troublesome unit cellPam,As a developer and someone who uses other software to test against, Ihave been following the discussion with interest. This is a complex areaand it goes to show the importance of having a few programs if only fortesting purposes.I am puzzled by your cif data however as I cant get past verifying theequivalence of the number of equivalent positions for the Li site.In R-3m you have as fractional atomic coordinates for Li1: x = 0 y = 0.5 and z = 0For P3112 you have three Li sites with x, y, z's as0.77780 0.0 0.833300.0 0.88890 0.833300.0 0.55560 0.83330which translates to:0.77780 0.0 5/60.0 0.88890 5/60.0 0.55560 5/6Listed below are equivalent positions for a general site.What I dont understand is that for R-3m I get 9 non-overlapping fullyoccupied equivalent positions within the unit cell with a volume of101.2975.For P3112 I get 18 non-overlapping equivalent positions with a volumethree times as large of 303.8968. I was expecting to get 27 if the twostructure were to be equivalent.Thus, from your cif data I am left to conclude that I am generating thewrong number of equivalent positions, or, if I am generating the correctnumber of equivalent positions then the two cif data sets should indeedgive different powder patterns.Have