Calibration of the temporal drift in absolute and relative Raman intensities in large Raman images using a mercury–argon lamp

Abstract

Raman imaging is a powerful tool for the analysis of astromaterials including meteorites, cosmic dust, and samples returned to Earth by spacecraft. Many astromaterials are heterogeneous on mm-or cm-sized scales requiring the collection of large Raman images with long accumulation times of days to weeks. It is known that the wavenumber and bandwidth calibration of a Raman instrument drifts with time and temperature resulting in significant changes in the observed Raman band over the course of Raman image collection. In this work, we examine the temporal and thermal drift of absolute and relative Raman intensities through the collection of a Raman image. We show that the spectrometer throughput and observed Raman intensities change over time and are correlated to small changes in lab temperature. We propose a method of calibrating for the drift in Raman intensities using a Hg-Ar (mercury-argon) calibration lamp integrated into the Raman instrument. We also show that relative Raman intensities drift with time as a function of temperature as a result of changes to the Raman instrument response function. We developed a method to calibrate the drifting instrument response function using only the calibration lines from the Hg-Ar lamp. Overall, our results elucidate the effects of temporal/thermal drift on Raman intensities and we propose calibration techniques to account for these effects.

Raman instrument calibration for astromaterials and analysis of Mars return samples

“…Thus, as discussed above for wavenumber calibration, one needs to know the temporal drift in bandwidth calibration to ensure that a calibration is valid over a given time period. It is important to note that the bandwidth calibration time dependence is independent of that of intensity and peak wavenumber as discussed in the original report of this work (Jakubek & Fries, 2020, 2021a, 2021b.…”

“…b) Dependence of the relative carbon D and G band intensities on graphite domain size. Panel (a) was reproduced with permission from Jakubek and Fries (2021b). Copyright 2021 Wiley.…”

“…To reduce the effect of temporal drift in relative Raman intensity for long time scale experiments, Jakubek and Fries (2021b) developed a method to calibrate relative Raman intensities using just the Hg‐Ar calibration lines from their integrated calibration lamp. To do this, they showed that every luminescence spectrum collected can be modeled as a linear sum of the luminescence spectra observed at the maximum and minimum lab temperature that occurred during the experiment.…”

“…Again, here we focus on the time dependence of relative intensities. Jakubek and Fries (2021b) utilized NIST luminescence glasses to measure the instrument response function of their Raman microscope. They measured the instrument response function as a function of time and temperature.…”

“…This is particularly important for continuous, long time scale data collection, such as Raman image collection, as the calibration will drift over the time of collection. In this case, the typical process of collecting calibration before and after a Raman image is insufficient for accounting for variable errors accrued during an image (Jakubek & Fries, 2020, 2021a, 2021b). As a result of these factors, variations in calibration quality between laboratories or even minor changes in methodologies employed by different instrument users in a laboratory may introduce significant errors, degrading the repeatability and understanding of measurements reported in the literature.…”