Thermal analysis and pyrolysis modeling of NOMEX IIIA fabric. Issue 9 (2nd September 2018)

• Record Type: Journal Article
• Authors: Li, Yanjun; Liao, Ya-Ting
• Is Part Of: Combustion science and technology. Volume 190:Issue 9(2018)
• Page Start: 1580
• Page End: 1593
• Publication Date: 2018-09-02
• Subjects: NOMEX -- pyrolysis model -- TGA, MCC -- fire modeling; Combustion -- Periodicals; Combustion engineering -- Periodicals; 541.36105
• DOI: 10.1080/00102202.2018.1459587
• Languages: English
• ISSNs: 0010-2202

Abstract

ABSTRACT: NOMEX® fabric is widely used in fire safety and military applications for its advantages in thermal and chemical resistance. In this research, thermogravimetric analysis (TGA) and microscale combustion calorimetry (MCC) tests of a NOMEX blend, NOMEX IIIA, were performed in nitrogen flow between 20°C and 800°C. Four heating rates, ranging from 10°C/min to 50°C/min, were used. At all heating rates, mass loss rate data from TGA exhibit peaks at 350–500°C and at 500–650°C. Heat release rate data from MCC also exhibit peaks at the same temperature ranges. This suggests that the sample went through thermal decomposition and yielded combustible gas products at these temperature ranges. As heating rate increased, these peaks shifted to higher temperatures, became wider, and had smaller magnitudes. Based on the test results, three pyrolysis models consisting of multiple consecutive and parallel reaction steps were proposed with different complexities. An Arrhenius expression was used to describe reaction rate as a function of density and temperature of the sample. Pyrolysis kinetic parameters were first theoretically estimated by correlating TGA data with Arrhenius assumptions employing Kissinger, Friedman, and Flynn–Wall–Ozawa methods. The proposed pyrolysis models and associated kinetic parameters were then implemented into a solid combustion code (GPYRO) to simulate the TGA and MCC heating processes. The pyrolysis parameters were fine-tuned to optimize the match between TGA data and modeling results. Combustion heat release rates of gaseous products were estimated using MCC data. All three proposed models were able to reasonably predict TGA and MCC data at all heating rates while the models with higher complexities showed slightly better performance.