Quantifying the thermomechanical behaviour of carbon fibre reinforced polymer materials exposed to fire conditions
Aspinall, Timothy J.
Carbon fibre reinforced polymer (CFRP) materials are engineered materials consisting of carbon fibres arranged in layers of alternating directions and bonded with a resin matrix. At present, they are the most widely used material in the construction of new aircraft structures and are becoming increasingly popular across other industrial sectors. CFRP materials possess high specific strength and stiffness, excellent resistance to impact, and reduced maintenance costs compared to metallic alloys. However, due to the fire response behaviour of CFRP materials, which includes glass transition, a combustible polymer matrix, fibre oxidation and loss of load-bearing capacity, concerns have been raised with respect to their fire safety. Whilst a significant body of research exists for other common aviation materials, such as aluminium and titanium alloys, it is currently unknown if this knowledge can be applied directly to CFRP materials to help predict their thermomechanical behaviour. This knowledge is unknown because CFRP materials are often utilised differently from traditional aviation materials due to differences arising from the carbon fibre orientation. So far, quantifying the thermomechanical behaviour of CFRP materials has been mainly carried out using uncoupled (or post-fire) approaches, whereby the residual mechanical properties are used to assess the CFRP material’s performance during a fire. Because of this, current knowledge of CFRP material’s thermomechanical load-bearing behaviour when different fibre orientations are used in the manufacture of the composites during a fire remains poorly understood, with currently unknown implications for its load-bearing behaviour. Five separate experimental studies have therefore been undertaken to investigate and quantify the thermomechanical response of four CFRP materials containing a carbon fibre reinforcement and an epoxy resin matrix. Each CFRP material contains a unique fibre orientation and has been produced solely by the author of this thesis using carbon reinforcement and epoxy resin matrix sourced from different suppliers. The first study has been carried out to identify the main solid-phase thermal response characteristics of the four CFRP materials, neat epoxy resin and carbon fibre, in a kinetically-dominated heating regime. In this study, the glass transition, pyrolysis and oxidation temperatures of the CFRP materials have been quantified using different analysis techniques. The data from this chapter is critical for the following chapters as it identifies the temperatures that result in a loss of mechanical properties. The results from this study show that the solid-phase thermal response reactions of CFRP materials are complex, often consisting of several overlapping and competing physico-chemical processes. The second study investigates the burning behaviour of three CFRP materials in a heat-transfer-dominated heating regime. The motivation for this study is the lack of knowledge on the burning behaviour of CFRP materials containing common fibre orientations when different materials are placed adjacent to their unexposed rear surface boundary. Each CFRP material in this study has undergone cone calorimeter experiments to quantify the influence of fibre orientation and rear surface boundary conditions on the mass (loss) and heat release rate during separate flaming and non-flaming scenarios. The first rear surface boundary condition is a highly conductive aluminium heat sink, whilst the second contains low thermal conductivity ceramic insulation. These two rear surface boundary materials represent actual conditions that occur in aircraft structures where highly conductive and insulation materials are positioned in close proximity to fuel storage areas that have a risk of catching fire. The result from this study shows that CFRP materials exhibit distinct burning behaviours when the fibre orientation and rear surface boundary condition changes. The third study investigates and quantifies the influence of carbon fibre orientation on the post-fire (residual) three-point bending and tensile behaviour of three CFRP materials. In this study, the CFRP materials are first exposed to different thermal intensities using a cone calorimeter chosen to represent critical temperatures required to induce the physico-chemical processes most associated with the loss of mechanical properties in CFRP materials (i.e. glass transition of the epoxy resin, pyrolysis of the epoxy resin and the oxidation of the carbon fibre reinforcement). After this, the CFRP materials are left to cool and then mechanical tested to obtain post-fire mechanical data and compare results. The results of this study show that the post-fire mechanical response of the CFRP materials changes depending on the level of thermal intensity and the carbon fibre orientation. The fourth study presents a state-of-the-art approach for quantifying the thermomechanical bending behaviour of a CFRP material and an opportunity to investigate specific behaviours such as displacement, time-to-failure and failure modes. The motivation of this study is a lack of fundamental knowledge on the mechanical response of a loaded CFRP material as it undergoes heating. The data produced from this study is important to aircraft manufacturers and aerospace and defence contractors who use CFRP materials in hazardous areas of aircraft (adjoining or in close proximity to fuel tanks) or in military applications where a chance of fire is always possible due to munitions fragments during combat operations. Drawing on material flammability, the thermal irradiance is induced using an electric coil heater allowing a systematic evaluation of the material response. By manipulating the applied heat flux, the process causing failure is shown to vary. At low heat fluxes, the failure is elastic and is dominated by a large proportion of the specimen reaching the glass transition temperature. Whereas, at higher heat fluxes, the failure is dominated by the pyrolysis of the epoxy resin at the locally exposed surface, resulting in a more brittle failure. Because the developed apparatus allows the systematic variation of the thermal and mechanical load, it is possible to utilise it to replace conventional uncoupled approaches where residual mechanical properties are often used to assess the performance of materials exposed to thermal loads. The final study describes a series of experiments using the approach described in the fourth study and describes the setup, execution, results, and analysis of thermomechanical experiments performed on the four CFRP materials using a novel ’rig’. The proposed test rig allows the thermomechanical behaviour, relating the mechanical performance degradation with particular surface temperature and temperature gradient inside the CFRP materials to be investigated. The motivation of this study is to understand the mechanical response of loaded CFRP materials containing unique fibre orientations as they undergo heating. Experiments have been performed on specimens produced from four unique CFRP materials to study their behaviour under three-point bending when exposed to different heat fluxes. Failure times, displacement and temperature distribution data are recorded from specimens produced from each unique CFRP material, whilst failure modes and degradation mechanisms have also been investigated using high-definition videography. The data produced in this study has shown that the carbon fibre orientation and heat flux influence the thermomechanical load-bearing response of CFRP laminates. It is generally observed that laminates containing unidirectional [90°] fibres demonstrate the worst overall load-bearing response to thermomechanical loading conditions. In contrast, woven bi-directional [0°, 90°] fibres demonstrate the best. Unidirectional [0°] fibres and unwoven multidirectional [0°, 45°, 90°] fibres present a modest overall load-bearing response to thermomechanical loading conditions. It should also be added that specimen thickness and boundary conditions also govern the thermal response of the CFRP materials, as shown in Chapters 4 and 5. Overall, the contribution to knowledge that the work in this thesis presents are original and have significant potential for engineers and designers to understand the fire safety of CFRP materials in load-bearing applications. The originality of this work is hence twofold. The first novelty of this work is that the apparatus developed in this thesis can quantify a CFRP material’s thermomechanical behaviour by transiently coupling a well-defined thermal condition and a mechanical load to allow the material’s mechanical response to be explicitly linked to the thermal (fire) environment. Quantifying a material’s thermomechanical behaviour using this approach has not been carried out until now due to practical difficulties in accurately controlling a heat flux and mechanical load simultaneously. At the same time, most of the approaches used in existing literature assess the thermomechanical behaviour of CFRP materials when they have cooled, which is unrealistic in practical terms. The importance of this first novelty is that the method and apparatus developed can be used to measure the load-bearing behaviour of CFRP materials whilst exposed to the heat from a fire to reflect real-world hazards encountered in aircraft. The second novelty of this thesis is that the majority of existing studies have examined CFRP materials containing a single carbon fibre orientation without assessing the behaviour (and comparing their performance) to other commonly used carbon fibre orientations whilst coupling the fire and load. Whereas the work in the thesis does this by creating four CFRP materials, each containing a unique, albeit common carbon fibre orientation used in specific areas of aircraft structures. The practical significance of this work is that the knowledge identified, collated and created can be used to improve the fire safety of aircraft and make the engineers and designers of them aware of the limits of their own knowledge and the limits of the state-of-the-art when CFRP materials are used in primary load-bearing components that can often contain materials with significantly different thermomechanical responses adjacent to or in close proximity to fuel storage areas in order to create a workable solution.