Discussion Session 3 Abstract
Challenges in integration, scaling and filtering of peak intensity data in crystallographic multigrain analysis
Chairpersons: Przemek Dera (pdera@hawaii.edu ; dera@cars.uchicago.edu) and Michael Ruf (Michael.Ruf@bruker.com)
In crystallography scaling is the process by which the measured integrated peak intensities are corrected, normalized, and converted to structure factor amplitudes, usable in structure solution and refinement. Scaling process applies corrections that account for Lorenz and polarization effects, absorption by sample and its environment (e.g. diamond anvil cell components), changes in incident beam intensity, and sample movement within the X-ray beam. Scaling also allows to merge together peak intensity data coming from different scans (rotations about different axes or data collected at different detector positions). The stage of scaling often also includes merging of symmetry equivalents and high redundancy of data is utilized to refine several important data corrections. In conventional crystallography scaling is typically performed for one crystal at a time, with merging of data from multiple crystals possible as a separate step in the data analysis workflow. Scaling solutions for crystallographic one-crystal-analysis available in variety of single crystal software packages are well established and robust.
In high pressure experiments, additional phenomena such as peak overlaps with diamond or pressure medium signal, diamond extinction effects, etc. add intrinsic challenges to the scaling process, require additional peak filtering, and typically result in noticeably lower data quality (lower number of observations completeness and redundancy, higher internal consistency factors and refinement FOMs). Availability of specialized corrections and features accounting for high-pressure specific data artifacts varies noticeably between different crystallographic software suites. For high-pressure multigrain analysis problems such as pressure gradients and anisotropy (each grain may have different strain state), defects and high grain mosaicity, and presence of multiple crystalline phases add even more complexity.
In multigrain analysis we are dealing with data composed of multiple (few to few hundreds) individual crystal contributions corresponding to the same crystalline phase. Individual crystallites are typically very small (micrometer or less) and not positioned ideally at the center of the goniometer. As a result the peak intensity datasets from each crystal are limited (relatively few independent observations, due to the crystallite moving out of the beam during rotation), heavily affected by the changes in the illuminated volume and incident intensity (as the crystallites move across the focused beam profile) and limited redundancy. In addition, with many crystallites belonging to the same phase, peak overlaps between different grains are also common. All these factors make deriving proper scaling corrections for each grain independently very challenging. On the other hand, the large number of grains available, and their shared crystal structure offer possibility to perform simultaneous scaling. Currently there are only few crystallographic programs allowing this kind of data treatment, and they do not include high-pressure specific features.
The objective of the discussion session-2 on integration and scaling will be to review the most important factors and corrections that need to be applied during the multigrain scaling process of high-pressure data, discuss available and prospective approaches and algorithms, and review the current software suites that could handle such analysis or incorporate the necessary functionality in the future.
Useful links and resources:
Advanced Macromolecular Structure Determination: Integration & Scaling by Tim Grüne:
http://shelx.uni-ac.gwdg.de/~tg/teaching/ggnb/A106/pdfs/A106-day1.pdf
Integration and scaling by Harry Powell:
Multigrain crystallography by Henning Sorensen et al. (2012):
http://www.degruyter.com/downloadpdf/j/zkri.2012.227.issue-1/zkri.2012.1438/zkri.2012.1438.xml
<