Determining sample alignment in X-ray Reflectometry using thickness and density from GaAs/AlAs multilayer certified reference materials

TL;DR

This study investigates how misalignment in X-ray reflectometry affects parameter estimation and demonstrates that using known properties of buried layers can effectively calibrate instrument alignment, enhancing measurement accuracy.

Contribution

The paper develops calibration curves linking angle misalignment to layer parameters and validates the use of buried layer properties for robust instrument calibration in XRR.

Findings

Buried layers show highest sensitivity to misalignment.

Calibration using known layer properties improves alignment accuracy.

Buried layer density provides a robust calibration method.

Abstract

X-ray reflectometry (XRR) provides researchers and manufacturers with a non-destructive way to determine thickness, roughness, and density of thin films deposited on smooth substrates. Due to the nested nature of equations modeling this phenomenon, the inter-relation between instrument alignment and parameter estimation accuracy is somewhat opaque. In this study, we intentionally shift incident angle information contained in a high-quality XRR data set and refine this shifted data using an identical structural model to assess the effect angle misalignment has on parameter estimation. We develop a series of calibration curves relating data misalignment to variation with layer thickness and density for a multilayer GaAs/AlAs Certified Reference Material on a GaAs substrate. We then test the validity and robustness of several approaches of using known thickness and density parameters from this structure to calibrate instrument alignment. We find the highest sensitivity to and linearity with measurement misalignment from buried AlAs and GaAs layers, in contrast to the surface layers, which show the most instability. This is a fortuitous result, as buried AlAs and GaAs exhibit the highest long-term stability in thickness. Therefore, it is indeed possible to use reference thickness estimates to validate XRR angle alignment accuracy. Buried layer mass density information also shows promise as a robust calibration approach. This is suprising, as electron density of buried layers is both a highly-correlated phenomenon, and a subtle component in the XRR model.