Matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry of kerogen extracts: effect of interactions between sample structure and matrix

This paper describes experiments carried out within the framework of examining capabilities of MALDI‐TOF‐MS in the characterisation of fossil fuels and related materials. The work has focused on the effect of interactions between sample structure and matrix composition on the quality of MALDI spectra. Solvent extracts from six kerogens have been examined by MALDI‐MS in the presence of three matrices of different composition and in the absence of matrix (laser ablation). The kerogen extracts have also been characterised by size exclusion chromatography (SEC) and UV‐fluorescence spectroscopy (UV‐F). In the presence of the three matrices (sinapinic acid, 2,5‐dihydroxybenzoic acid and α‐cyano‐3‐hydroxycinnamic acid), spectra of extracts from the three geologically ‘younger’ kerogens showed larger molecular mass distributions compared to spectra of extracts from the corresponding ‘older’ kerogens. This result directly contradicted trends observed in the absence of matrix, where extracts from ‘older’ kerogens showed larger molecular mass distributions. Results from SEC and UV‐F agreed with trends observed in MALDI‐spectra acquired in the presence of matrix, particularly sinapinic acid. By contrast, we recently reported that high mass signal in laser ablation mass spectra of the pyridine‐insoluble fraction of a coal tar pitch, obtained in the absence of matrix, closely matched signals obtained in the presence of sinapinic acid. Compared to the highly condensed aromatic structure of the coal tar pitch, the kerogen extracts (of considerbaly higher oxygen content) appear less able to readily absorb incident laser radiation (337 nm) in the absence of externally added matrix. The argument is consistent with apparently more efficient absorption of incident laser radiation by the geologically ‘older’ samples, when no matrix was used. Structural characteristics suggest that the geologically ‘older’ samples would be better able to act as ‘self‐matrix’. Thus, differences in structure and composition of the samples appear to go some way towards explaining differences between trends observed in the presence/absence of matrix. Dependence of MALDI spectra on molecular structure has also been found for two sets of biomass derived samples. Considerably more work appears warranted to determine MALDI‐MS parameters appropriate to the analysis of the more oxygenated, lower rank, samples. © 1997 John Wiley & Sons, Ltd.