Hi Mangus, I am also using DICVOL extensively for indexing of the patterns from unknown compounds (mostly pharmaceuticals). It is a nice indexing software and gives you several possible solutions. Now you may use another indexing software (ITO, Treor) to see what it would give you, but the things may become more complicated. Another way that I am using routinely is to perform the whole-pattern-decomposition approach (peaks extraction from the pattern's profile) using either Le Bail or Pawley method. There are lot of software that can do this. I am using Topas. Put the lattice parameters from the best DICVOL's solutions and see which one would give the best overall fit. However, you need to assign a space group as well. Start with the lowest symmetry space group. For example, if the lattice is monoclinic - start with P2. In most of the cases you'll see that for one of the DICVOL's solutions the overall fit is quite better than for the others. This is the most probable lattice. Although I do not think that this is the ultimate and 100 % accurate solution, I found out that it works well and could be a quite effective and relatively simple approach to find the unit cell, or at least to rank the proposed solutions from any indexing software. Best regards, Srebri P.S. The indexing procedure is perfectly described with a lot of working examples and criteria for its reliability in "Fundamentals of Powder Diffraction and Structural Characterization of the Materials (Pecharsky V. & Zavalij P.), 2003.
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Dr. Srebri Petrov, PXRD Lab Dept. of Chemistry, University of Toronto, Tel/Fax: 416-978-1389 ----- Original Message ----- From: "Magnus Helgerud Sørby" <[EMAIL PROTECTED]> To: <[email protected]> Sent: Thursday, June 22, 2006 5:00 AM > Dear Leonid, > > If the structure of bethanechol chloride is not known, I agree that we > cannot know its unit cell for certain. Still, it seems very likely that the > unit cell proposed by all the programs in the test based on synchrotron data > (no peaks excluded among the first 20), is in fact correct. > > Anyway, my question still stands (in a slightly modified form): Why is the > presumably correct unit cell considered as the best solution in case A(1) > when Dicvol propose another unit cell with better FoMs and smaller volume? > > Best regards, > Magnus > > > -----Opprinnelig melding----- > > Fra: Leonid Solovyov [mailto:[EMAIL PROTECTED] > > Sendt: 22. juni 2006 10:29 > > Til: [email protected] > > Emne: Re: Ranking of solutions in DICVOL04 > > > > > > > > > The benchmarking tests were done on various > > > XRD data for bethanechol chloride. I assume > > > that its crystal structure is determine > > > (since knowing its unit cell with certainty > > > should be a prerequisite for choosing it > > > as a benchmarking compound), but I cannot > > > check it since I don't have access to > > > any database for organic crystal structures. > > > > Dear Magnus, > > > > Neither CSD nor COD contain structural information on this compound. > > The two ICDD entries for this substance indicate that it was not > > obtained in a pure single-phase form. > > > > Best regards, > > Leonid > > > > > > > > __________________________________________________ > > Do You Yahoo!? > > Tired of spam? Yahoo! Mail has the best spam protection around > > http://mail.yahoo.com > > > > > >
