The role of intensity and instrument sensitivity in Raman mineral identification

Abstract

Mineral identification is a basic and common element of investigations in a wide variety of fields that deal with natural crystalline materials and their synthetic analogues. Raman spectroscopy has been shown to be effective for mineral identification through spectral pattern matching and, in addition, offers a number of practical advantages over established mineral‐identification technologies. In order to make Raman technology accessible to the user community that routinely requires a mineral identification tool, the systematics of Raman spectra of minerals must be investigated and incorporated into instrument design, software design, and methodology documentation. This paper quantifies the range of Raman intensities that can be expected from natural minerals and investigates the incorporation of this information into instrument standards and analytical methodology. For this study, 7978 Raman spectra representing 2159 mineral species were extracted from the RRUFF database (http://rruff.info) and separated into four instrument‐specific groups. Pseudo‐logarithmic histograms show that Raman intensities of minerals can span four orders of magnitude, and opaque minerals tend to have one order of magnitude lower intensities than non‐opaque minerals. Examples of inter‐instrument comparisons show that instrument sensitivity can vary by a factor of 35. An experiment investigating the data quality required for mineral identification indicates that a functional minimum signal‐to‐noise ratio is 15. The results of this study can be used to design data collection strategies. The minerals quartz and pyrite are shown to represent the high‐frequency center of intensity distributions for non‐opaque and opaque minerals, respectively. The requirements for a relevant instrument‐sensitivity standard are also discussed. Copyright © 2015 John Wiley & Sons, Ltd.