Impact of Radiative Exposures on the Mechanical Properties of Fire-Resistant Fabrics

Firefighters’ protective garments are designed to protect them from elevated temperature and heat flux environments. Typically, these garments are made of three layers: an outer shell, a moisture barrier, and a thermal liner. At present, new firefighters’ garments must meet performance criteria specified in test standards, but these do not apply to in-use garments. As the performance of this clothing may degrade with use, quantitative methods for determining the useful life of these garments without destroying them is a need for the fire service. One area of such research has been the development of correlations between near infrared (NIR) spectral results and changes in fabric properties. Research at the University of Saskatchewan (U of S) has shown that NIR measurements can be correlated to deterioration in mechanical strength after thermal ageing. This study examined the performance of two Kevlar®/PBI fabrics after exposures to heat fluxes ranging from 10 to 70 kW/m2 for durations ranging from 15 to 1200 s. These exposures were conducted using a cone calorimeter and have been selected as representative of the wide range of conditions expected on the fireground. After examination using an NIR spectrometer, tensile testing was conducted on the specimens. The tensile strength values for thermally aged fabrics were then compared against criteria in standards for new clothing. Three correlations between exposure and duration were developed based on multi-variable linear regressions, multi-variable nonlinear regressions, and single-variable nonlinear regressions. These correlations were constructed from this research and past research datasets. These correlations aim to predict the degradation these types of fabrics experience, and could be used in future degradation. It was found that the multi-variable nonlinear regression correlation was the most successful across different exposures, while the single-variable nonlinear regression was able to predict degradation more accurately to an average standard error of 4%