With the existing analytical pyrolysis systems (e.g., thermogravimetric analysis (TGA), pyroprobe, drop tube furnace, etc.) that are available on the market, it is not possible to monitor the real temperature of a sample or visualize the transformation of the solids. Furthermore, typical sample preparation methods (e.g., using a powder) result in poor mass and heat transfer, because of large sample thickness and the explosion of particles during the heating regime. To study and address these problems, (i) a new experimental setup was designed to allow temperature measurement and pyrolysis sample visualization using a CDS Pyroprobe 5000; (ii) the system was completely characterized with cellulose (85.7 mu m thick) as the solid sample; and (iii) preliminary results showed the propulsion of cellulose particles out of the quartz sample tube, which could be a major source of error. As a result, a thin sample film was used for further experimentation. Pyrolysis experiments using cellulose were performed under near-vacuum conditions (3.5 mbar) and at near-atmospheric pressure (950 mbar) to gain insight into the effect of pressure on pyrolysis. The temperature of the samples and the yield of solid residue (char) were measured. Results showed that the heating rate of the sample was much slower than the heating rate of the platinum coil. Furthermore, endothermic reactions suppressed the heating rate as the temperature rose above 400 degrees C. Data was extracted from videos of pyrolysis at both pressures to track the darkening of the solid samples. A computer simulation of these experiments including heat, mass, and momentum transport at 3.5 and 950 mbar shows how operation at different pressures and heating rates affect the generation and distribution of pyrolysis products. These simulations indicate that convective heat transfer dominates over radiative heat transfer to the cellulose sample at high and low pressures if the heating coil is wrapped tightly around the quartz sample tube. However, radiative heat transfer is somewhat more important at lower pressures where the gas-phase thermal conductivity is smaller, especially if the heating coil is loosely wrapped around the sample tube.