The effect of sodium dihydrogenphosphate, trisodium pyrophosphate, and sodium aluminocarbonate on the thermal decomposition of rigid polyurethane (PUR) foams, based on diphenylmethane-4,4-diisocyanate, diphenyl-2,2-propane-4,4-dioxyoligo(ethylene oxide), and oxyalkylenated toluene-2,6-diamine, blown with pentane, was studied. Thermogravimetric (TG) data have shown that there is a stabilization effect of additives in the initial stage of degradation, both in nitrogen and air atmosphere, and the decomposition proceeded in two steps up to 600degreesC. Results of the kinetic analysis by the isoconversional methods of Ozawa-Flynn-Wall and Friedman yielded values of (apparent) activation energy (E-a) and preexponential factor (A). For phosphate-stabilized PUR samples, E-a remained stable over a broad area of the degree of conversion, while for carbonate-containing sample two regions of E-a were observed. Further advanced kinetic analysis by a nonlinear regression method revealed the form of kinetic function that was the best approximation for experimental data-for a two-stage consecutive reaction the first step was the Avrami-Erofeev nucleation-dependent model, and the second step was a chemical reaction (1st or nth order) model. The integrated thermogravimetric (TG)/Fourier transform infrared (FTIR) technique probed the thermal degradation of modified PURs by analyzing the evolved gases. The solid residue remaining at different temperatures was identified by diffuse reflection FTIR (Kubelka-Munk format). The complex thermal behavior was discussed on the basis of the obtained results-it can be shown that the global stabilization effect is a multistage process whose initial conditions are of critical importance in governing the nature of the entire process. (C) 2003 Wiley Periodicals, Inc.