When infused with an oxidizer, on-chip porous silicon (PS) shows tremendous potential as an energetic material. However, applications such as fuzing and propulsion require long-term material stability. The present work utilizes accelerated lifetime testing for PS samples with sodium perchlorate (NaCIO4) oxidizer. Devices were exposed to elevated temperatures for various time intervals, then subjected to an electrical pulse through an integrated bridgewire to test for ignition. The pass/fail data from ignition testing was then analyzed to determine the mean and median failure point at each temperature using an Arrhenius aging model and a lognormal probability density function (PDF). Samples were heated over a temperature range of 185-300 degrees C for up to 46 h with median failure times ranging from 0.4-20 h. Differential scanning calorimetry (DSC) and Fourier transform infrared (FTIR) spectroscopy were used to evaluate chemical changes of PS/NaCIO4 during aging. Comparison of DSC traces for aged samples showed differences for heating above and below 250 degrees C, suggesting two different aging mechanisms. FTIR results suggest that for low temperature aging (below similar to 250 degrees C) the hydrogen termination layer of the porous silicon was still intact, but backbond oxidation did occur. Above degrees 250 degrees C the hydrogen termination layer was no longer evident in FFIR and considerable oxidation occurred, suggesting reaction or dissociation of the hydrogen layer. This corresponds with an exothermic peak shown during DSC at 300 degrees C. Thermogravimetric analysis/mass spectroscopy (TGAIMS) was also performed, and confirmed hydrogen gas release beginning at similar to 280 degrees C. A DSC peak at similar to 400 degrees C is evident for fresh samples, but is greatly reduced or not evident for all aged samples. We attribute this peak to backbond oxidation, which is shown to occur in FFIR for all aged samples. Variable heating rate DSC experiments were also performed to determine activation energy of fresh and aged samples. Activation energy was calculated using the Kissinger Method for the first two exothermic peaks in DSC experiments. For the first peak, corresponding to hydrogen desorption (similar to 300 degrees C), activation energy for fresh and aged samples was 132.1 kJ/mol and 132.7 kJ/mol, respectively. The activation energy of the second exothermic peak, corresponding to backbond oxidation, for fresh and aged samples was 183.8 kJ/mol and 159.6 kJ/mol, respectively. Device failure during accelerated aging tests was also used to predict the mean lifetime of porous silicon/sodium perchlorate at room temperature (25 degrees C) as 100 years with a 90% confidence interval of 31 to 328 years. (C) 2017 Published by Elsevier Inc. on behalf of The Combustion Institute.