Abstract
This study applies Design of Experiments and Response Surface Methodology (RSM) to the ternary Al/ KClO4/binder system, using the adiabatic flame temperature as the target response. A three factor Box–Behnken design was constructed in the Al content (26-38 wt.%), binder content (2-10 wt.%) and initial temperature of the reactants (298-398 K), and the flame temperature of each of the fifteen resulting formulations was computed from a fully transparent Hess law energy balance based on tabulated NIST–JANAF data, rather than from a proprietary equilibrium code. The thermochemical calculation yielded a reaction enthalpy of -10.6 kJ·g-1. The effective dissociation coefficient was calibrated so that the stoichiometric flame temperature fell within the literature range of 2900-3200 K; this agreement therefore represents consistency with the calibration target rather than an independent validation of the model. A full quadratic response surface model fitted the computed data with a coefficient of determination of 0.9990 and a maximum relative error below 0.13%, and passed all residual adequacy checks. The temperature was governed primarily by the quadratic effect of the aluminum content and the linear cooling effect of the binder, whereas the initial temperature showed a comparatively weak contribution within the investigated range, acting primarily additively and independently of composition. Numerical optimization located a maximum adiabatic temperature of about 3089 K at 32.7 wt.% aluminum and the lowest binder loading.