Simon Frost charts the history of the favoured materials in applications where strength, stiffness and low weight are key.
We know carbon fibre as the high-performance, weight-saving structural material of choice in motor sport, aeros
pace and, increasingly, renewable power and automotive components. Its origins, however, are far removed from the applications it serves today.
Often misattributed to Thomas Edison, it was the British physicist and chemist Sir Joseph Swan who created a carbon filament for a lightbulb, in 1860. His filament of carbonised paper in a glass bulb was, admittedly, impractical – the gases emitted from the burning paper quickly dulled and then blackened the glass. 19 years later, Edison created a practical lightbulb, using cellulose-based carbon filaments derived from heated cotton or bamboo and placed in a vacuum.
In the early 20th Century, carbon was widely superseded by the brighter tungsten filaments placed in an inert gas instead of a vacuum, which increased their luminosity and reduced bulb blacking. Swan’s lightbulb moment was not the application for which carbon filaments would become known, but his 1883 patent for extruding nitrocellulose through a fine die to form artificial fibres yielded a commercial process for creating the first synthetic fibres, which the textile industry would notably exploit – chemists at the American firm DuPont famously developed Nylon and Kevlar in the 1930s and ‘60s, respectively.
Rayon, PAN and pitch
It was not until the 1950s that R&D into carbon fibre began to gather momentum, catalysed again by the lighting industry. It was built on efforts by the Union Carbide Corporation (UCC), USA, to replace the tungsten in lightbulbs with carbonised rayon towards the end of the Second World War.
In 1958, UCC scientist Dr Roger Bacon was attempting to determine the triple point of graphite – where the liquid, solid, and gas are in thermal equilibrium – using a high-pressure carbon arc. As he decreased the pressure, the carbon turned from a vapour to solid, forming a stalagmite-like deposit on the lower electrode. When Bacon broke one of these open, he discovered whiskers embedded ‘like straws in brick’.
Just two or three centimetres long and a few micrometres in diameter, they could be bent into shapes and weren’t brittle. These were filaments of perfect graphite – the first high-performance carbon fibres discovered, boasting a tensile strength of 20GPa, around ten times that of steel, and a Young’s modulus of 700GPa. They were too small and expensive for any practical purpose, but the whiskers demonstrated the material properties that pure carbon fibres possessed.
The next year, Bacon’s colleagues patented a process to produce carbon fibres by heat-treating a rayon precursor up to 3,000°C. Rayon, made from purified cellulose, graphitised around this temperature when heated in a protective atmosphere of inert gases. The fibres created were relatively inexpensive and scalable, while much lighter than boron and fibreglass, which had dominated commercial composites. In 1964, Bacon and UCC colleague Wesley Schalamon developed a process that went some way towards replicating the mechanical properties of the whiskers. They found that when a carbon yarn from a rayon precursor was gradually stretched as it was heated, its modulus increased tenfold.
Meanwhile, Dr Akio Shindo at the Government Industrial Research Institute, Osaka (GIRIO) in Japan and William Watt at the Royal Aircraft Establishment, UK, were seperately developing carbon fibres using a different precursor – the polymer resin polyacrylonitrile (PAN). It offered a carbon yield of 50–60%, compared with rayon’s 20%, and a continuous string of carbon and nitrogen atoms in its polymeric backbone allowed for highly oriented graphitic-like layers, removing the need for hot stretching. It could also be adapted to substitute high modulus for extreme tensile strength – these low-modulus fibres would become especially popular in sporting equipment such as skis and tennis rackets and in the aerospace industry.
PAN displaced rayon in most applications, and still accounts for the majority of precursors used today, but in 1970, UCC chemist Leonard Singer introduced a third class of precursor. He found that liquid crystalline pitch could be graphitised to create carbon fibres with a lower tensile strength but an extremely high Young’s modulus – close to 1,000GPa – and high thermal conductivity. This made pitch-based precursors perfect for applications such as heat-facing aircraft components, where both stiffness and heat removal were essential.
Brian O’Rourke, Chief Composites Engineer at Williams Martini Racing, UK, has been working with carbon fibre since 1973, when he first encountered pre-impregnated composites working for the British Aircraft Corporation in Weybridge, UK. Williams recruited him in 1982 to help transfer the composite materials technology that had recently boomed in the aerospace industry to Formula One.
‘Carbon fibres were originally of two types – high strength (HS, with a modulus of around 230GPa), and high modulus (HM – more than 400GPa), with the former being used for most applications. Beginning in the mid-1980s, we began to see a new category – intermediate modulus (IM) fibres,’ O’Rourke explains. When the PAN precursor is heated to higher levels, a fibre with greater modulus can be achieved, but its strength reaches a peak at a relatively low temperature, so a compromise in favour of one or the other is made.
‘The IM fibres instead attempted to find the best of both, giving a moderate increase in modulus to 280GPa, but with an enhanced failure strain, up from 1.5% to 2%. The IM types are mostly chosen today for all-round structural performance. Some F1 parts, such as a wing, are designed mostly for stiffness, so HM fibres can be used for these.’
A similar trade-off between toughness and temperature capability was a key challenge in the formulation of resin matrices. ‘There were some epoxies that cured at 120°C and had good toughness but little temperature capability, and others that cured at 180°C that performed at higher temperatures but were rather brittle […] the results of efforts to provide the best of both were seen in the late 1980s, with the third-generation toughened epoxies that form the bulk of those used in aerospace today.’
As well as epoxies, other specialised resins have been developed – cyan-esters, for example, offer structural performance above 200°C, while ablative-type phenolic resins can survive beyond 400°C, making them useful in heat-shielding applications where strength is not as important.
In a crash, carbon fibre composites can be ‘exceptionally useful and completely useless at the same time’, O’Rourke says. ‘It depends on the nature of the prevailing loading conditions. These materials fail in a quasi-brittle manner via a range of different fracture modes. Importantly, they exhibit no plastic deformation when under load, so the manner in which equivalent metallic structures absorb energy during an impact – via panel buckling and the formation of a network of plastic “hinges” – does not happen. Instead, a vast number of small fracture processes occur which, if controlled to progress in the right order, are extremely efficient in absorbing energy.’
Impact attenuators built into the components of a racing car are designed to fracture into millions of small particles, and while the obliteration of carbon fibre may appear more dramatic than buckling metals, this spectacular failure mode is also far safer.
Not all composite parts behave in the same way, though. ‘By definition, this fracture process assumes that the loads encountered in an impact will be, largely, oriented in the plane of the absorber skin panels and that those laminates are stabilised to resist buckling – by stiffening, for example, with a honeycomb core. The opposite,
however, applies when fully out-of-plane loading is encountered. This has been witnessed with other composite components, such as suspension elements, which nowadays contribute little in an impact because their buckling is rapidly followed by fracture in bending.’
Racing into the mainstream
Fittingly, the introduction of new carbon fibre composites happens the fastest in Formula One, which benefits from huge investment and a short qualification process for materials compared with the aerospace industry, where production runs can last decades. Despite progression in both fibre and resin formulations, O’Rourke notes that there are still aerospace components in production that are formed of the same materials used in the mid-1980s. He describes Formula One as ‘a global shop window for all of the advanced engineering used, including what we do with composites.’
Carbon fibre composites, however, are gradually becoming more accessible for mainstream applications – the unit price is the same today as it was 30 years ago, making the materials less expensive in real terms. In terms of volume, wind energy has become a major user of composites, and the mainstream automotive industry is moving into composite use on a large scale.
‘The future looks very exciting,’ says O’Rourke, who has had the fortune to work with the most up-to-date composites available for the past 35 years. We won’t be seeing F1 cars on the M25, but this shift away from exclusivity means that much lighter, and therefore more fuel-efficient, vehicles could well become the norm.