In this study, two types of carbon fiber reinforced polymer (CFRP) flame retardant composites, CFRP@MatEG‐GF and CFRP@MatEG‐GF‐MW (EG, GF, and MW are the acronyms of expandable graphite, glass fiber and mineral wool, respectively), were prepared via compression molding, with expandable graphite‐based mat incorporated at the surface of the CFRP. The results showed that compared with the control CFRP, both CFRP@MatEG‐GF and CFRP@MatEG‐GF‐MW passed vertical burner test and 45° burner test according to FAR25.853. Moreover, the limiting oxygen index (LOI) values of both CFRP@MatEG‐GF and CFRP@MatEG‐GF‐MW were increased, and the peak of heat release rate (PHRR) and total heat release (THR) were reduced. Besides, CFRP@MatEG‐GF‐MW had higher flame retardancy than CFRP@MatEG‐GF. This was attributed to the formation of a more stable and dense carbon layer structure. The mechanical test results showed that the tensile strength of CFRP@MatEG‐GF‐MW was higher than that of CFRP@MatEG‐GF. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2019, 136, 47801.

No abstract

This study investigated the effect of intumescent mats (M1 and M2) with different compositions on the post-fire performance of carbon fibre reinforced composites (CFRP). The sandwich structure was designed for composites where M1 (CFRP-M1) or M2 (CFRP-M2) mats were covered on the composite surface. A significant reduction in the peak heat release rate (PHRR) and total heat release (THR) was observed from the cone calorimetric data and CFRP-M1 composite showed the lowest value of 148 KW/m 2 and 29 MJ/m 2 for PHRR and THR, respectively. Additionally, a minor influence on mechanical properties was observed due to the variation of composite thickness and resin volume in the composite. The post-fire properties of composite were characterised and the M1 mat presented better retention of flexural strength and modulus. The feasibility of two-layer model was confirmed to predict the post-fire performance of composites and reduce the reliance on the large amounts of empirical data.