Optimal estimated standard uncertainties of reflection intensities for kinematical refinement from 3D electron...

Area of Science:

• Crystallography
• Materials Science
• Data Analysis

Background:

• Estimating errors in merged reflection intensities is vital for accurate crystallographic analysis.
• Current methods often focus solely on counting statistics, potentially underestimating total error.
• A comprehensive understanding of all error sources in electron diffraction data is needed.

Purpose of the Study:

• To investigate and compare three methods for correcting estimated errors in electron diffraction reflection intensities.
• To identify a model that accurately accounts for all contributions to reflection intensity errors.
• To improve the quality and reliability of electron diffraction data.

Main Methods:

• Three distinct methods for error correction in electron diffraction data were examined.
• These methods were applied to both organic and inorganic test datasets for direct comparison.
• A specific error model incorporating original uncertainty and symmetry-related reflection intensities was evaluated.

Main Results:

• The application of a specific error correction model demonstrated significant improvements.
• This model, including terms for original uncertainty and largest symmetry-related intensities, enhanced overall structure quality.
• A notable improvement in the final R_all factor was achieved using the proposed error model.

Conclusions:

• The developed error model provides a more complete and adequate estimation of reflection intensity uncertainties.
• Implementing this model in data reduction software like PETS2 leads to superior crystallographic structure determination.
• This advancement is critical for enhancing the precision and trustworthiness of electron diffraction studies.