The use of lightweight carbon fibre-reinforced polymer (CFRP) composites in transportation vehicles has necessitated the need to guarantee that these new materials and their structures are able to deliver a sufficient level of crashworthiness to ensure passenger safety. Unlike their metallic counterparts, which absorb energy primarily through plastic deformation, CFRPs absorb energy through a complex interaction of damage mechanisms involving matrix (polymer) cracking, fibre/matrix debonding, fibre pull-out/kinking/fracture, delamination and inter/intralaminar friction. CFRP is primarily deployed as a laminate and can potentially deliver a higher specific energy absorption than metals. Translating this capability to a structural scale requires careful design and is dependent on geometry, fibre architecture, laminate stacking sequence and damage initiation strategies for optimal uniform crushing. Consequently, the design of crashworthy CFRP structures currently entails extensive physical testing which is expensive and time consuming. This paper reports on progress and challenges in the development of a finite-element computational capability for simulating the crushing of composites for crashworthiness assessments, with the aim of reducing the burden of physical testing. It addresses the 'tyranny of scales' in modelling structures constructed of CFRP composites. Intrinsic to this capability is the acquisition of reliable material data for the damage model, in particular interlaminar and intralaminar fracture toughness values. While quasi-static values can be obtained with a reasonable level of confidence, results achieved through dynamic testing are still the subject of debate and the relationship between fracture toughness and strain rate has yet to be satisfactorily resolved. This article is part of the theme issue 'Nanocracks in nature and industry'.
Keywords: composite materials; computational simulation; crashworthiness; finite-element modelling.