It is better to spent time learning how to collect without ice… :-)
FF
Dr Felix Frolow   
Professor of Structural Biology and Biotechnology
Department of Molecular Microbiology
and Biotechnology
Tel Aviv University 69978, Israel

Acta Crystallographica F, co-editor

e-mail: [email protected]
Tel:  ++972-3640-8723
Fax: ++972-3640-9407
Cellular: 0547 459 608

On Jun 27, 2012, at 21:00 , JORGE IULEK wrote:

>     I thank for the references and the comprehensive discussion from Dr. 
> Holton.  Also, for the reference indicated by Dr. Berry. I think I will get 
> what I am looking for, now I need to "process" all this information. 
>     Partially answering Dr. Holton, my aim is to have a side guide for 
> improving parameters to process images with ice diffraction rings. In 
> general, the idea is to exclude the minimum area from the detector, but large 
> enough to avoid bad regions. In this, ice ring widths have a role, so, 
> besides the amount of ice, exposure time and beam intensity, relative 
> diffraction intensities contribute, and this way one could decide better what 
> the "best" width might be for each individual ring. Sure, looking at the 
> images are the base here, but it is interesting to be seconded by the ring 
> expected positions and "relative intensities". 
>     Yours, 
> 
> Jorge 
> 
> 
> On Tuesday 26 June 2012 09:16 PM, James Holton wrote: 
>> I think an appropriate reference is Bragg (1921) 
>> http://dx.doi.org/10.1088/1478-7814/34/1/322 who compared various 
>> possible crystal structures to the relative "strength" of the 
>> reflections from "ice powder" measured by Dennison (1921) 
>> http://dx.doi.org/10.1103/PhysRev.17.20. 
>> 
>> However, as Bragg noted Dennison's intensities don't agree all that 
>> well with those you would expect from what we now know is the 
>> structure of hexagonal ice (Ih).   It is possible that Dennison's 
>> preparation (plunge-cooling pure water in a capillary) actually 
>> created some cubic ice (ice Ic) along with his hexagonal ice (ice Ih). 
>>   The high intensity he saw for the "middle" line of the triplet we 
>> normally see for true hexagonal ice is consistent with this.  Cubic 
>> ice is actually more commonly seen in MX diffraction patterns than 
>> hexagonal ice (in my experience). 
>> 
>> However, you do have to be very careful about what you mean by 
>> "intensity".  Are you talking about photons/pixel?  Photons/spot? 
>> Photons integrated over a powder_ring?  All these will be different 
>> relative numbers.  I'm not sure if they knew about Lorentz factors yet 
>> in 1921  There is no mention of correcting for them in either paper. 
>> Anyway, if you are after the "true" hexagonal ice ring intensities, I 
>> would advise taking the following PDB file: 
>> 
>> CRYST1    4.511    4.511    7.346  90.00  90.00 120.00 P 63/m m c 
>> SCALE1      0.221680  0.127987 -0.000000       -0.00000 
>> SCALE2     -0.000000  0.255974 -0.000000        0.00000 
>> SCALE3      0.000000 -0.000000  0.136129       -0.00000 
>> ATOM      1  O   WAT A   1       0.000   2.604   0.457  1.00  0.00           
>> O 
>> ANISOU    1  O   WAT A   1      603    630    172    302      0      0       
>> O 
>> ATOM      2  H   WAT A   1       0.000   2.604   1.308  0.50  0.00           
>> H 
>> ANISOU    2  H   WAT A   1      510    510     56    255      0      0       
>> H 
>> ATOM      3  H   WAT A   1       0.000   3.432   0.148  0.50  0.00           
>> H 
>> ANISOU    3  H   WAT A   1      487    361    163    185    199     95       
>> H 
>> 
>> Calculate structure factors from it, add up F2 of same-resolution 
>> indicies and plot them out that way.  remember, the square of a 
>> structure factor is proportional to the integrated intensity of a 
>> single-crystal spot, which is not the same thing as a powder ring 
>> intensity.  The relationship was described most recently by Norby 
>> (1997) http://dx.doi.org/10.1107/S0021889896009995 
>> Which I paraphrase as: 
>> for a flat detector, the average intensity of a pixel in a powder ring 
>> is given by: 
>> 
>> Ipix = k*p*sum(F2)*omega/sin(theta)/sin(2*theta) 
>> 
>> where Ipix is the recorded value of one pixel, 'k" is a 
>> resolution-independent scale factor, "p" is the polarization factor 
>> (see Holton&  Frankel 2010), "F" is the structure factor of an hkl 
>> index that falls on the ring (there can be more than one, hence the 
>> "sum"), "omega" is the solid angle subtended by the pixel and "theta" 
>> is the Bragg angle. 
>> 
>> For the above PDB, I get: 
>> d  sum(F2) 
>> 3.907 121 
>> 3.673 156 
>> 3.449 111 
>> 2.676 88.5 
>> 2.255 111 
>> 2.075 153 
>> 1.953 62.9 
>> 1.922 84.2 
>> 1.888 62.5 
>> 1.836 4.33 
>> 1.725 47.8 
>> 1.662 1.84 
>> 1.527 96.7 
>> 1.477 42.8 
>> 1.448 39.5 
>> 1.375 78.9 
>> 1.370 34.6 
>> 
>> HTH, 
>> 
>> -James Holton 
>> MAD Scientist 
>> 
>> 
>> On Tue, Jun 26, 2012 at 2:34 PM, Edward A. Berry<[email protected]>  wrote: 
>>> Maybe figure 4 in 
>>> Viatcheslav Berejnov et al.   Vitrification of aqueous solutions 
>>> J. Appl. Cryst. (2006). 39, 244–251 ? 
>>> 
>>> 
>>> JORGE IULEK wrote: 
>>>> Hi, all, 
>>>> 
>>>>      I thought I could easily find a reference to comment upon the 
>>>> relative 
>>>> intensity of 
>>>> rings in an image due to diffraction by polycrystal ice, but no luck 
>>>> googling for that. A 
>>>> reference that would contain a picture (with visual relative intensities) 
>>>> would be even 
>>>> better. Of course absolute intensity depends on the amount of ice, but a 
>>>> relative "scale" 
>>>> is what I am looking for now. 
>>>>       Yours, 
>>>> 
>>>> Jorge 
>>>> 
> 

Reply via email to