A super sandwich

The relative strengths of carbon fibre composites have been talked about for years. So by now, why are composites not used ubiquitously?

The relative strengths of carbon fibre composites have been talked about for years. So by now, why are composites not used ubiquitously?

One industry that is often used as an indicator of technology take-up is automotive. Composite components have been used for many years for low production sports and bespoke racing cars. But if concept cars indicate the intentions of the major car producers then they must be close to a breakthrough.

DaimlerChrysler estimates that its Dodge ESX3 concept car with composite body could be produced at a cost of 85% less than a conventional metal body and weigh 46% less. It would consist of 12 panels, injection moulded from glass-filled polyethelene. BMW’s Z22, which the company calls a ‘driveable feasibility study,’ could put carbon-fibre composites into production cars by the year 2004. The Z22’s body consists of 20 moulded parts, which could be produced at a speed acceptable for automotive assembly line production using the resin transfer molding process (RTM). BMW plans to to build a pilot plant to further develop composite processing technology. Afterwards there is technically nothing standing in the way, according to BMW.

The ‘Infrastructure Composites Report 2001’ compiled by Composites Worldwide Inc, indicates that under a calm commercial surface there are big changes to come. It forecasts that industrial markets for emerging fibre reinforced plastic applications will grow by over 500% between 2000 and 2010 worldwide. The demand is obviously there but what of the supply?

The problem is that carbon-fibre composites cost at least 20 times as much as steel, and the automobile industry is not interested in using them until the price of carbon fibre drops from $8 to $5 (and preferably $3) a pound. Production of carbon fibres is too expensive and slow. The raw material is typically pitch, or polyacrylonitrile (PAN) precursor. It is converted to carbon fibres using thermal pyrolysis, a slow, energy-consuming process that is combined with stressing to achieve the right properties. The precursor, the energy needed to heat it to make fibres, and the large ovens and other capital equipment required in the process contribute to the high cost.

According to carbon fibre producer Zoltek Corporation current precursor costs represent at least 60% of the expense of producing carbon fibre. Zoltek has been working to lower the cost of carbon fibre to make it more attractive to the auto industry. The company’s goal has been to establish a price of $5/lb for carbon fibre in high-volume applications this year.

Once the fibre has been purchased its integration into a composite part is also expensive. The immediate problem with mass production of composites is akin to mass producing timber or 12 year-old whisky – the curing process that defines the product’s quality takes a long time. It is therefore necessary to produce large numbers of the product concurrently over a large area to reach the required throughput. Traditionally automotive production has been characterised by production line techniques but this will have to reworked to allow for resin preparation and curing.

A technique to speed production that is showing great promise comes from Isle of Wight-based SP Systems. Already adopted by TVR Motorsport, SPRINT (SP Resin Infusion Technology) combines the advantages of other processes.

SPRINT materials consist of a layer of fibre reinforcement either side of a pre-cast, pre-catalysed resin film. The material therefore has the appearance of dry rein-forcement which has resin concealed at its centre. SPRINT materials differ from conventional prepreg (fabrics or fibres that are pre-impregnated with pre-catalysed resin) so the fibres in the reinforcements initially remain dry and unimpregnated by the resin.

Air extracted from dry fibre

SPRINT materials are layered in a mould and vacuum bagged like conventional prepreg by sealing the bagging film over the laminate and onto or around the mould tool. The air in the bag is extracted by a vacuum pump so up to 1 atmosphere of pressure can be applied.

However, when the vacuum is applied, the dry reinforcement enables air trapped in the fibre bundles and layers to be easily removed. The temperature is then raised for the resin to cure and the resin film softens and flows into the air-free reinforcement. The void content of the resultant laminate is low (typically 0 – 0.5%).

In any composite structure, the one factor that can greatly affect fatigue life is the void content of the laminate. Voids are stress concentration points and detract from the homogeneity of the laminate to accumulate stresses.

Traditionally low levels of voids have been reached by using auto-claves. Autoclaves are effectively high pressure ovens. Autoclaves need good quality tooling, relatively long cycle times and have a potentially constrained output. Autoclave processes are also widely recognised to be expensive – the preserve of F1 and aerospace parts. This has meant that composite parts have only been used in parts where cost is not limiting and factors such as weight saving and safety have been higher on the list.

And so designers and builders of moderate number (50-1000 per annum) production run cars have been excluded from fully optimised composite solutions because of cost.

One alternative process is liquid resin infusion which can either be a pressure moulding process e.g. RTM, or a vacuum moulding process e.g. SCRIMP and VARTM. These processes work by infusing a dry reinforcement laminate stack with a liquid resin. As the resin is moved through the laminate by the action of a vacuum pump or pressure, the air within the laminate is flushed out by the resin front. These processes yield very low void content structures with lower costs than the autoclave route.

However, liquid infusion is constrained by the need to use very low viscosity resin systems so that these resins can flow easily over and through the reinforcement fabrics, often a distance of many metres. Such low viscosity resins have a relatively low molecular weight.

High viscosity resins

From the chemist’s point of view the larger the molecule the greater the opportunity to add in features such as higher mechanical strength and increased fracture toughness. The low molecular weight resins that tend to be used for infusion processes therefore cannot offer the levels of mechanical and thermal performance that are attainable with the resin systems used in prepregs. This may have an effect on fatigue life, the very issue that having void-free laminate was looking to solve.

SPRINT uses higher molecular weight prepreg type resins which means that ultimate laminate properties have the potential to be near to autoclave cured laminates.

Although SPRINT is a form of infusion process, the resin systems in SPRINT only have to move the small distances through the thickness of its fabrics and therefore sophisticated, resin systems can be used.

But producing 5000 parts efficiently is a long way from mass production runs of 5 million parts. Many of these moulding process are scalable but not automated and still rely on skilled labour. So the composites revolution has not been cancelled, only postponed pending improvements in fibre and moulding production techniques.