Summary Worker exposure to airborne contaminants must be controlled and maintained below established regulatory limits. When the full range of administrative, engineering and collective control measures are not sufficient to achieve safe exposure levels, endangered workers must be provided with respiratory protective devices (RPDs). When there is no immediate danger to life or health (IDLH), air-purifying RPDs are used. In the case of gaseous contaminants, or vapours, these RPDs are equipped with adsorbent-filled cartridges. This raises the question of cartridge service life. For organic vapours, predictive models of breakthrough time are used, similar to that of Saturisk (http://www.irsst.qc.ca/saturisk/). For acid gases, including hydrogen chloride (HCl), hydrogen fluoride (HF), hydrogen sulfide (H2S), chlorine (Cl2), chlorine dioxide (ClO2) and sulfur dioxide (SO2), knowledge is so limited that no such tools exist at present, as far as we know. The goal of this project was therefore to determine the performance of respiratory protection cartridges against acid gases and the effects of environmental variables on service life. Sulfur dioxide was used as a tracer gas. The available literature on the treatment of SO2 in the air suggests a chemisorption mechanism when activated charcoal or activated carbon fibres are used. By this mechanism, in the presence of oxygen and water, the SO2 is oxidized into sulfuric acid (H2SO4) within the adsorbent. A number of cartridges were selected to be used in the experiments. On the basis of an examination of the cartridges approved for protection against acid gases on the National Institute for Occupational Safety and Health (NIOSH) Certified Equipment List (https://wwwn.cdc.gov/niosh-cel/), a sample of 10 cartridges that protect solely against acid gases was drawn up. Breakthrough testing of these cartridges under certification conditions highlighted a variety of different curves and dispersed breakthrough times. A characterization of the metal impregnants of charcoal, meant to catalyze the oxidation of SO2, served to produce a preliminary classification of the cartridges into three groups, based on their impregnation profile. One group stood out owing to high levels of copper and zinc, while another was found to have virtually none of the metals being searched for. This characterization, albeit preliminary, shed light on some of the different SO2 capture strategies used by manufacturers. With three cartridges chosen for their different impregnation profiles and breakthrough times, the concentration effect was measured. Overall, the acid gas cartridges seem, under our testing conditions, to be far more sensitive to changes in contaminant concentrations than the organic vapour cartridges do. Although it is generally acknowledged for organic vapours that a drop in concentration by a factor of 10 leads to an increase in service life by a factor of 5, a drop in the concentration of SO2 by a factor of 10 would lead to an increase in service life by a factor of around 15. Relative humidity has a major effect on SO2 capture. It greatly facilitates the retention of SO2, as it plays a role in the chemisorption reaction. At low relative humidity, the profiles of the cartridge breakthrough curves, with low or no measured metal impregnation, are atypical and show stepped adsorption beyond saturation. On the basis of similar phenomena observed with various adsorbent beds, a hypothesis specific to the study cartridges was proposed to explain this phenomenon. Experiments with different airflow rates confirmed that the kinetics of SO2 chemisorption may explain it in part. Tests of intermittent use of the cartridges showed that the storage of acid gas cartridges is not a problem, as it does not affect their service life. The chemisorption seems irreversible. The Wheeler-Jonas equation, useful for calculating service life in the case of organic vapours, cannot be directly applied in the case of SO2 or other acid gases, owing to the stepped curves obtained for some cartridges. It does, however, serve to describe the change in breakthrough time as a function of concentration. Before it can be used routinely to calculate service life, however, in-depth knowledge of the reactions occurring in the cartridges is required. This knowledge will give us a better understanding of changes in the reactional capability of the cartridges, depending on the ambient conditions of use. It will also be important to ensure that the SO2 adsorption rate can be calculated with sufficient accuracy on the basis of the available, established empirical expressions for organic vapours.