Individual Student Project Assessment

Individual Student Project Assessment
2001/2002

School of Crystallography, Birkbeck College, University of London

Welcome Page

Course Index

Student Index

Josef Bursik

General Comment:

The first part of the project was done excellently. However, you failed to realize that two possible solutions for the low temperature phase are not equivalent, i.e. the fact that the distortion could take place in one of two directions. Since the correct solution is the one in which the sub-group is a semi-standard setting of space group 14 (i.e. b axis unique, then you managed to achieve the correct solution. I suspect you may have missed it had P2(1)/b11 been the correct solution. It was pleasing to see that you knew when to stop refining the model both for data sets A and B: well done in this respect.

	Category	Mark	Comments
(a)	General Presentation	4/5	Overall presentation was good with the project presented logically. Only criticism is that you went into too much detail in places: for example, more detail is required for successful results than for unsuccessful ones. The additional pictures of the structures enhanced the report, though the ones of the asymmetric unit were less useful.
	HTML & Navigation	4/5	Navigation was very easy with the pictures done as sub-pages so as to improve the flow of text. Some file names seem to be a mixture of upper and lower-case: this is bad practice and potentially gives problems when files are transferred from one operating system to another. (It was not a problem on the server here which is case insensitive.) However, the non-standard characters in some of the filenames gave problems: the marks given here are therefore a little generous. A minor point: it would be nice if the background colour was set to white using the BGCOLOR attribute of the <BODY> tag.
	Introduction	4/5	An excellent introduction, though slightly missing the point concerning the physical meaning of the different settings of space group 14.
(b)	Data Set A Strategy	5/5	Your strategy for data set A was perfect. You finished the refinement at the correct point, i.e. with a single B value for all atoms.
	Data Set A Tables	9/10	Generally very good, though too many figures quoted for the esd values in places (e.g. lattice parameter and bond angle esd values). One of the bond angles has a typographical error?
	Data Set A Figures	4/5	Figures were fine, though slightly excessive in number. The captions could have been improved by using individual figure numbers (e.g. Figure 1.1, Figure 1.2, etc.).
(c-i)	Data Set B Strategy	4/5	The strategy was fine except for the creation of O(4). Space group Pbnm has three types of plane, but only one of these is a mirror plane. Generating O(4) using the glide planes is not a sensible option since O(3)' is related to O(3) in the orthorhombic model by mirror symmetry.
	Data Set B Tables	8/10	The tables were fine, though the extra ones are not required. You seem to have an esd missing on the B value for Pb and no units are given for B. Units are missing from the cell angle (beta) also.
	Data Set B Figures	4/5	Figures generally fine, but with the same comment as that made earlier for the figures of data set A.
(c-ii)	Data Set B Strategy	0/5	You fell down in a bad way here. The point is that the orthorhombic structure could distort such that either of the angles alpha or beta are no longer equal to 90°. The idea was not to use different unit cells and thus different settings of space group 14, but to determine in which direction the distortion occurs.
	Data Set B Tables & Figures	1/5	You needed to produce a figure of the "failed" model here plus tables showing the "failed" model. The latter would have high esd values and the alpha angle (if using P2(1)/b11) would refine back close to 90°, but not to 90° exactly. Without these you cannot make the correct conclusions.
(d)	General Conclusions	2/6	Given the misunderstanding over the different distortions of the orthorhombic structure, the final conclusions were inevitably a little weak.
	300K versus 5K	2/6	The comments on the decrease in the atomic displacement parameters was excellent (i.e. anisotropic and large at 300K and isotropic and small at 5K). It would have been good to mention changes in cell parameters, and a brief discussion of the loss of mirror symmetry as a result of decreasing molecular libration.
	Data Set A versus B	3/6	You made some excellent points here, especially about the danger of extrapolating beyond the range of ones measurements. However, the other key point to be made is that data set A correspond to low resolution in real space, hence the difficulty in determining the correct symmetry from a limited data set range. Measuring out to smaller d spacings results in better real space resolution.
	Unstable Refinements	0/6	Since you failed to try out the alternative model correctly, you were unable to make comments about the instability of the refinement. A key point here is the use of restraints that would give stability to the refinement: the danger though is that one has a stable false minimum as a result.
	X-rays versus Neutrons?	3/6	This was reasonable well discussed. However, you needed to point out that scattering from lead would dominate the powder X-ray diffraction pattern leading to large uncertainties for the refined parameters of the light atoms and thus poor geometric parameters for the sulphate anion.
(e)	CIF Submission	4/5	The main point here is that one only requires CIF files for the correct model: one does not submit CIF files for wrong models. A minor point is that the symbols for the atom types would be better as, say, Pb rather than Pb2+ - the concept of oxidation state is less relevant in neutron diffraction since oxidation state is not related to the scattering factor, b, in contrast to the X-ray case.

Welcome Page

Course Index

Student Index