Image Registration in Multi-Modal Scanning Microscopy: A Solar Cell Case Study

Abstract

Scanning probe measurements are an indispensable tool of solar cell research today, and the compatibility with simultaneous acquisition of complementary measurement modes is a particular strength. However, multi-modal data acquisition is often limited by different scan-parameter requirements. As a consequence, the modalities may be assessed subsequently rather than simultaneously. In this instance, image registration serves as a tool to align two-dimensional datasets at nanoscale. Here, we showcase an example of two subsequent scanning Xray microscopy measurements of solar cells with a Cu(In,Ga)Se 2 absorber, the first measurement being optimized for X-ray beam induced current and the second for X-ray fluorescence. We discuss different approaches and pitfalls of image registration and its potential combination with Gaussian filtering. This finally allows us to proceed with the investigation of point-by-point correlations.

Effects of X-rays on Perovskite Solar Cells

“…Because of drift and lack of reproducibility of scan positions that are inevitable at the nanoscale, the XBIC and XRF measurements will require alignment, for example, by image registration. 34 Decrease Measurement Time. Given that the signal-tonoise ratio in XBIC measurements is rather limited by currents induced by the environment than by statistics, the integration time can be even below 1 ms if the response chain of the solar cell/amplifier/data acquisition system is fast enough.…”

“…A second scan covering the same area can then be optimized for high sensitivity and spatial resolution of XRF measurements. Because of drift and lack of reproducibility of scan positions that are inevitable at the nanoscale, the XBIC and XRF measurements will require alignment, for example, by image registration …”

X-ray vision of Cu(In,Ga)Se2: from the Ga/In ratio to solar-cell performance

“…to obtain XRF count rates Φ Element and stoichiometric fractions ν Element (see SI of [33]) for every element, taking self-absorption, photon flux and dwell time into account. The resulting images from two subsequent scans optimized for XBIC and XRF signal were registered and aligned as described in [43] based on the distribution of Se.…”