Multivariate Optimization of Column Liquid Chromatography for the Separation of Saturated, Aromatic, and Acidic Biomarkers in Petroleum and Rock Extracts

ABSTRACT Conventional geochemical biomarkers include saturated, aromatic, and carboxylic acid compounds. Saturated and aromatic biomarkers, however, form the foundation of many geochemical studies because of their abundance and intrinsic significance. Although carboxylic acids are also present in sediments and petroleum, they are often found at much lower concentrations, which complicates their isolation when using a single analytical method. For this reason, the application of a modified Brønsted‐based silica gel phase by classical liquid chromatography has the potential to separate and characterize these three classes. However, variations in the factors can compromise effective component separation. A well‐designed experiment can optimize the process by addressing multiple methodological aspects. The Brønsted bases tested to modify the silica gel were sodium and potassium hydroxides. The central composite design (CCD) showed that, besides the expected influence of the volume of Eluent 1 ( n‐ hexane), there is a difference between the base types in the separation of saturated and aromatic compounds. The SiO 2 /KOH phase appears to be more resistant to the influence of Eluent 1. Therefore, the trend in separation efficiency is K > Na. All two silica gel modifications yield a fraction rich in carboxylic acids. The modification of the silica gel with a Brønsted base had no significant effect on the molecular geochemical parameters in the samples. This separation method saves time and material, making it a potential approach for routine analysis of saturated and aromatic biomarkers and carboxylic acids in molecular organic geochemistry.