The effect of atmospheric particulate matter (PM) on health, visibility, and regional climate has become a major concern worldwide, and control of PM is now the major challenge in air pollution abatement. Many of the adverse effects are directly associated with the aerosol products of incomplete combustion, which include in large part the refractory carbon component know as elemental carbon (EC). Unfortunately, the complex organic and inorganic nature of PM can lead to severe measurement inconsistencies, and this is particularly the case for EC. Numerous methods have been developed over the last few decades to measure refractory carbon in PM, including methods based on chemical oxidation, thermal oxidation, optical behavior alone, photoacoustic behavior, and thermal oxidation combined with optical behavior. A number of intercomparison studies have revealed that while measurements of total PM carbon are fairly consistent among different methods, EC measurements are not. These inconsistencies in EC are attributed mainly to the belief that different methods measure different mixtures of substances as EC. Thermal oxidation and wet oxidation methods in particular often disagree substantially with optical methods. However, owing to the chemical and physical complexity of refractory carbon in PM, it is not reasonable to expect that the mass of refractory carbon by non-optical oxidative methods should be equivalent to the mass of light-absorbing carbon by optical methods. The goal here was to derive optimal temperatures and durations for the critical steps in the thermal protocol.