Mapping applications of laser-laser isotopic measurement in carbonates

Abstract

Sedimentary geochemistry is very often associated with the measurement of isotopic composition of carbon and oxygen from carbonates. The usual technique combining acid digestion and mass spectrometry analysis is slow, costly and non-ideal for spatially resolved analyses. When carbonates are processed using laser calcination and the gas produced during calcination is analyzed by infrared spectrometry, the time required for isotopic analysis is reduced to around 15 min to analyze 30 mg of carbonate in situ. Although the time saved is significant, it is hardly reasonable to carry out a high-resolution isotopic mapping of large samples. A fully resolved isotopic mapping, for example, of a sample with 25 cm2 surface area at resolution of a tenth of a millimeter, would require a continuous measurement carried out for a month. The aim of this study is, therefore, to explore possible strategies for constructing an isotopic map with a minimum number of analyses. Two approaches are pro-posed: (i) a mathematical approach that seeks to establish a correlation between the position of the sample and the carbon or oxy-gen isotopes, and (ii) an approach that looks for a correlation between the color (spectral characteristics) of the sample surface subdomains and their isotopic compositions. The choice of the second approach stems from the assumption that color contains a priori information about geological or geochemical processes. Several algorithms were developed and tested, notably using artificial intelligence tools. To testify the isotopic maps produced by these algorithms, posteriori isotopic measurements are taken and compared with the predictions from computed isotopic maps.