[bbrols]

Can anyone please help me with this problem.

I have had a single crystal x-ray diffraction conducted on a light atom crystal of unknown configuration, With only the least squares refinement, we acquired a flack parameter of -0.5 with an error of (10) when the data was merged and twin and basf instructions used the flack parameter was 0 with an error of (10).

This I have been told has given us the correct orientation but cannot give us the absolute configuration.

This makes no sense to me, I would have thought that they are one and the same.

At present we cannot add any heavy atoms or any known chiral centres so this is all the information we have to go by. we are awaiting results from the opposite enantiomer but if we could not tell which isomer is present from one enantiomer then I cant see how this is going to help.

Can anyone explain to me why, with a flack parameter of between -0.5 and 0 we cannot assign the absolute configuration. I have read many Acta Cryst. journals in an attempt to understand but I am still struggling.

Many Thanks

Rols

[AWK]

The standard deviattions (uncertainties) are important. Within 3 stds' Flack parameter should be less than 0.5.
eg 0.07(1) - 0.07+ 3x0.01=0.10 (far from 0.5), the absolute configuration is properly determined (with certainty over 99 %),
in your case 0(10) means 0.00(+-1.0), and configuration cannot be determined

[bbrols]

Thanks for that, I really appreciate your help.

Please forgive my ignorance, I think I have some issues with stds'
if an error of (1) means +/- 0.01 then surely an error of (10) would be +/- 0.1 not +/- 1.0? I did not study stats at school so this is all pretty difficult understand.

I am very new to x-ray crystallography but the shelxl program would not accept the other isomer as the correct configuration and forced us to keep inverting the structure to the same configuration.

[AWK]

0.2345(2) means 0.2345(+-0.0002)
0.2345(10) means 0.2345(0.0010)
0(10) means 0. (+-1.0)
su (std) concerns last or last two digits in your number
Shelxl  accepts a formal correct configuration which is rather impossible to determine from Flack parameter for structure containing only light atoms. Moreover, even for heavy atom structures, it strongly depends on the quality of crystal (quality of data), and precise correction of absorption.
In my practice, for sugar structures with known configuration of some chiral centers,  Flack parameter very often suggests the wrong (inverted) configuration

[bbrols]

Thank you very much, I still can't admit to understanding fully but that has been most helpful.

Back to the drawing board then.

[AWK]

http://lachlan.bluehaze.com.au/stxnews/stx/discuss/dis-fals.htm

Try to ask dr Flack personally:
http://www.flack.ch/howard/cristallo/Howard.Flack.html

[bbrols]

OK so I have not asked Prof. Flack yet as I don’t want to seem thick (though I am doing a grand job of that at present.

I have a flack parameter of -0.5 (10) for the natural product and a proposed configuration of RS.

I synthesised the opposite enatiomer and this gave the expected opposite configuration, SR, and a flack parameter of -0.7 (9)

I have been told that these flack parameters are meaningless, but how is the machine able to distinguish between the two isomers like this, if the structures are inverted the program will not accept the inverted structure.

I am being told that it is the correct orientation but not the correct absolute structure. No one seems able to explain what this means.

I also don’t understand the bit about standard deviation and the number in the brackets.

If anyone can point me in the direction of a text so that I can understand this a little better I would be most appreciative.

Thanks

Roland

[AWK]

Prof. Flack expalned this problem on Sci.techniques.xtallograph
http://www.ccp14.ac.uk/ccp/web-mirrors/hugorietveld/stxnews/stx/discuss/dis-fals.htm