This work is devoted to reduce spontaneous sublimation of ammonium azide at ambient and elevated temperatures by means of two microencapsulation techniques involving solvent/non-solvent and solvent evaporation methods in which stearic acid, Viton and nitrocellulose (NC) have been tested as coating agents. Scanning electron microscopy (SEM) was employed to examine the coating morphology. The thermal behavior of pure and coated ammonium azide samples have been studied by using simultaneous thermogravimetery-differential thermal analysis (TG-DTA) and differential scanning calorimetry (DSC). The results showed that ammonium azide microparticle could be effectively coated with nitrocellulose through a solvent/non-solvent experiment in which the coating quality depends on some experimental factors such as coating agent to NH4N3 weight ratio and volume and addition time of non-solvent. The effect of these factors on coating quality and thermal properties of NH4N3 has been revealed by results of SEM and thermal experiments. The most stabilized coated ammonium azide was achieved by using 4.5% (w/w) of NC as stabilizer, and by addition of 30 ml n-hexane as non-solvent within 75 min, where the maximum sublimation temperature increases about 30 degrees C with respect to uncoated sample and reaches to 145.5 degrees C. The influence of the heating rate (5, 10, 15 and 20 degrees C/min) on the DSC behavior of the pure and coated ammonium azide particles at the optimum condition was verified, whereas sublimation temperature of the NH4N3 was increased as the heating rate was increased. Also, the kinetic parameters such as the activation energy and frequency factor of the sublimation processes for pure and coated ammonium azide were obtained from the DSC data by non-isothermal methods proposed by Kissinger and Ozawa. The results showed that, activation energy for sublimation of completely coated NH4N3 particles is considerably higher than (nearly 1.5 times) that of the pure one. Also, the first order rate constant of sublimation of completely coated sample is noticeably lower than that of pure sample (k(coat)/k(pure); = 3 x 10(-4)). The kinetic results led us to conclude that the applied microencapsulation technique caused efficient stabilization of volatile NH4N3. (C) 2010 Elsevier BM. All rights reserved.