The purpose of this study was to examine how CO2 affects the burning behavior of two coal chars, char 1 and char 2. The work consisted of experiments and numerical modeling. The experiments were conducted under high heating rates in a laboratory-scale drop-tube reactor (DTR). The char samples were produced by pyrolyzing coal particles in the DTR at 850 degrees C in pure N-2. Before pyrolysis, the coal particles were ground and sieved to a particle size fraction of 100-125 mu m. The mass loss of the char particles was determined after the DTR combustion process. The surface temperature of the char particles was measured with a two-color pyrometer during combustion. The diameter evolution and the falling velocity of the particles were studied optically with a CCD high-speed camera. The oxygen concentrations used in the measurements were 2-12 vol.% in either N-2 or CO2. The combustion was assumed to take place within the Zone I and Zone II regimes. Zone I describes the conditions where the combustion process is controlled by chemical kinetics. In Zone II both chemical kinetics and intraparticle diffusion control the combustion. With char 2 the effect of replacing N-2 gradually with CO2 was also tested. This was done for the purpose of examining the interactions of the oxidation and CO2 gasification reactions. When the N-2 was entirely replaced with CO2 from the reactor atmosphere, the mass loss rate of both chars decreased slightly compared to the N-2 setting. A more drastic decrease was observed in the particle surface temperature. This study also presents the numerical modeling results of combusting the two coal chars in the DTR in N-2/O-2 and CO2/O-2 atmospheres. The apparent chemical kinetic parameters of the oxidation reactions were calculated based on the measurement results in the N-2/O-2 atmosphere. The apparent chemical kinetic parameters of the CO2 gasification reaction were also calculated for char 2. In the modeling calculations the internal heat transfer of the char particles, oxygen diffusion in the boundary layer, Stefan flow, and the size distribution of the particles were taken into consideration. The modeling results indicated the importance of determining the initial size distribution of the sample particles. An average diameter model could not explain the large variation in the measured particle surface temperatures. As a result, a comparison between the modeling results and the measurement results suggested that high CO2 partial pressure in the combustion atmosphere can affect the combustion process in other ways than merely through the differences in the gas properties. (C) 2014 Elsevier Ltd. All rights reserved.