Abstract Car lubricant additives are added to mineral or synthetic base stocks to improve viscosity and resistance to oxidation of the lubricant and to limit wear of engines. Their total amount in the commercial lubricant varies from a few percents to 20–25%. As they belong to various chemical classes and are added to a very complex medium, the base stock, their detailed chromatographic analysis is very difficult and time consuming as it should involve sample treatment and preparative scale separations in order to simplify the sample. The aim of this work is to determine the feasibility of the separation of low molecular weight lubricant additives using various packed columns with pure CO 2 as a mobile phase to enable implementation of flame ionisation detection as universal detector. This is part of a hypernated system including more sophisticated specific detectors, such as AED, FTIR or MS to obtain detailed structural information of compounds. This paper is devoted to the comparison of some stationary phases supposed to provide hydrocarbon group type separation (silica and normal phase) or separations on alkyl-bonded silica in non-aqueous mode of some selected classes of additives in test mixtures or in base stocks. Adsorption chromatography allows partial separation of additives from the base stocks while the direct elution of test additives can only be obtained on reversed phase supports having a very efficient silanol group protection so the interaction of the more polar compounds is much reduced. A two-dimensional scheme of analysis is also described. It combines adsorption chromatography to separate most of the polar additives from the base stock and alkyl-bonded silica for more detailed separation of the additives. However, overlapping between groups of compounds and the lack of resolution between some additives and the base stock should be addressed by the implementing of selective detectors.