Logo Individual Student Project Assessment
2001/2002

School of Crystallography, Birkbeck College, University of London
Welcome Page  Course Index  Student Index 

Zhanhui Yuan

General Comments

Despite the fact that you have obtained the correct solution, a rather poor understanding of crystallography is shown in places. In particular, there is a lack of understanding of when to use anisotropic displacement parameters. You made serious errors with the symmetry transformation for the 2nd refinement with data set B.

However, the report was written logically (despite some minor spelling mistakes and incorrect uses of English). It was easy to read and the use of separate figures helped the flow of text. Note that figure and table captions are important, especially for sub pages. It was also not necessary to include the details of every step in the refinement process: an overall impression of the steps taken would have been sufficient.

Detailed marks and comments are given below:

  Category Mark Comments
(a) General
Presentation		 4/5 The presentation is generally good though the figures of R-factor versus refinement cycle are not required: a figure of the structure would have been better. It would be better is more section headings had been given and in a consistent style (e.g. "Introduction" is bold but "Conclusion" is not).
  HTML &
Navigation		 5/5 Navigation is straightforward, the only sub pages being figures and CIF files. The HTML displays fine and links all work.
  Introduction 3/5 The introduction is a little weak: an explanation of the symmetry problem would have enhanced it considerably. There seems to be a lack of understanding of the problem set in the project and this is reflected in the poor conclusions given later.
(b) Data Set A
Strategy		 4/5 The strategy was fine until the temperature factors were refined. With such a limited range of data, the only reasonable option is to refine a single overall isotropic B value. At these low temperatures, it can reasonably be assumed that the vibrations of the atoms are isotropic, so anisotropic displacement parameters must not be used. The CIF output is not relevant since one does not normally publish wrong structures.
  Data Set A
Tables		 6/10 There are many faults (some minor, but others major) with the tables:
  • Table caption missing.
  • Fixed values should not be given an esd values of zero - no value is given.
  • Too many figures are given for esd values (e.g. 22 on S y coordinate, 82 on a bond angle, and worst of all 1.25 on a bond angle).
  • Units missing for B values and 2theta zero error.
  • When values are fixed as equal, this should be indicated (e.g. B O(2) = O(1)).
  • No peak shape parameters given. (This is fine if a pure Gaussian peak shape was used, but this was unclear in the text.)
  Data Set A
Figures		 3/5 An excessive number of figures have been given and figure captions are missing. They are particularly needed when figures are given as a sub-page.
(c-i) Data Set B
Strategy		 3/5 As for data set A, the strategy was fine until the temperature factors were refined. The logic of using anisotropic displacement parameters to account for the monoclinic distortion is plain nonsense. At these low temperatures, it can reasonably be assumed that the vibrations of the atoms are isotropic, so anisotropic displacement parameters must not be used. Secondly, although the choice of -0.4 for the z coordinate of O(4) works, it is a poor choice (due to application of the wrong mirror plane symmetry y=1/4 instead of y=3/4). Using 0.6 is better since O(4) then directly forms part of the SO42- group.
  Data Set B
Tables		 6/10 There are many faults (some minor, but others major) with the tables:
  • Table caption missing.
  • Fixed values should not be given an esd values of zero - no value is given.
  • Too many figures are given for esd values (e.g. 22 on S y coordinate, 82 on a bond angle, and worst of all 1.25 on a bond angle).
  • Units missing for B values and 2theta zero error.
  • When values are fixed as equal, this should be indicated (e.g. B O(2) = O(1)).
  • No peak shape parameters given. (This is fine if a pure Gaussian peak shape was used, but this was unclear in the text.)
  Data Set B
Figures		 4/5 Again, there are no figure captions, but the number of figures is more reasonable (though still large). One final figure with a good caption is all that is required here.
(c-ii) Data Set B
Strategy		 0/5 There is a major failure in your understanding here (or in your presentation) since you have failed to swop the x and y coordinates and a and b cell parameters. Effectively, what you have tried is a refinement is space group P 1 21/a 1 with coordinates based on Pbnm whereas you should have used coordinates based on the alternative setting Pnam for which P 1 21/a 1 is a sub-group.
  Data Set B
Tables &
Figures		 5/5 The tables and figures for the incorrect structure still share similar faults to those shown earlier, but no additional faults (so no further marks are deducted).
(d) General
Conclusions		 5/6 You stated the correct solution, although it was obtained with an element of luck. You also commented correctly on the relatively poorer profile R-factor for data set B.
  300K
versus
5K		 2/6 Apart from mentioning that atoms pack closer together at low temperature leading to loss of the mirror symmetry, you made no mention of the effect of temperature on lattice and atomic displacement parameters (and volume).
  Data Set A
versus B		 2/6 There was minimal discussion of this apart from noting that the wrong structure could be obtained with data set A. You should have provided a figure showing the effect of extrapolating the orthorhombic model to data set B.
  Unstable
Refinements		 1/6 Problems of refining with the wrong space group and how to overcome them were hardly discussed.
  X-rays
versus
Neutrons?		 3/6 Some useful points were made, but you failed to emphasise how Pb would dominate the scattering for X-rays and the problem of preferred orientation. Also Bragg-Brentano geometry is not the same as capillary geometry.
(e) CIF
Submission		 3/5 CIF outputs are OK, but there is no point generating/submitting CIF files for "wrong" structures. Also spaces were left in the esd values. There are also errors similar to errors made in the various tables also (e.g. to many digits for esd values).

Welcome Page  Course Index  Student Index 
© Copyright 1997-2002.  Birkbeck College, University of London.