Production and use of carbon-carbon composites

TWI Bulletin, November - December 1996

Wendy Hanson, Senior Research Engineer in the Advanced Materials and Process Department, joined TWI in 1992. Her work involves polymer-derived ceramics and composites, joining of ceramics to metals, glass-ceramics, ceramic and metallic brazing, and high temperature inorganic adhesives

Carbon-carbon composites consist of carbon fibres in a carbon matrix. Their high specific strength at high temperatures, excellent fracture toughness and thermal shock resistance have resulted in these composites finding use in applications such as aircraft and racing car brakes, heatshields for re-entry vehicles and rocket nozzles, as Wendy Hanson explains.

Carbon has many characteristics which make it a material with potential for high temperature applications. Most importantly, carbon retains its strength at elevated temperatures (up to 2500 deg.C), eventually subliming at approximately 3500 deg.C. Its other desirable properties include: good thermal shock resistance, high thermal conductivity, high specific heat, and low wear rate. However, carbon also has two major property drawbacks, its low strain to failure (brittleness) and its susceptibility to oxidation at temperatures above about 400 deg.C. These can both be overcome, but they have contributed to the limited application of carbon in industry.

Availability of carbon fibres in the late 1950s led to development of improved materials now known as carbon-carbon (C/C) composites. These composites are a family of materials which consist of a carbon (or graphite) matrix reinforced with carbon (or graphite) fibres. Thus, the attractive properties of carbon are combined with the high strength, versatility and toughness of composites. The C/C family is unique in that it is the only elemental composite.

This material was originally developed in the early 1960s for elevated temperature aerospace applications such as nose cones and rocket nozzles. It can be tailored to higher strengths and stiffness than other engineering metallic alloys and, unlike metals, can maintain these properties to high temperatures. Figure 1 shows the strength of a number of engineering materials as a function of temperature.

Carbon-carbon composites range from simple unidirectional fibre reinforced structures to complex woven 3-dimensional structures. The variety of carbon fibres and multidirectional weaving techniques now available, allow tailoring of C/C composites to meet complex design requirements. By selection of fibre-type, lay-up (or fibre-weave), matrix and composite heat treatment, the properties can be suited to different applications.

Processing of C/C composites

The approach for producing structural C/C composites is to orientate the required volume fraction of fibres into a preform, to accommodate design loads of the final component in service. All of the C/C preform structures must then be densified by a process that fills the open volume of the preform with a well-bonded carbon or graphite matrix. Selection of a densification process is dictated by the characteristics of the preform structure (fibre type and weave) and the required properties of the composite (matrix). There are two main routes to producing C/C composites, carbonisation of organic liquids (polymers) impregnated into the fibre preform, and chemical vapour infiltration of carbon from gaseous hydrocarbon precursors. These methods are shown in Fig.2. ^[1] Solid state processes are unfavourable because carbon is very difficult to sinter.

Carbonisation of organics

This requires use of a polymer precursor, which when heated, is stripped of its hydrogen and oxygen atoms, to leave a carbon residue. There are two main types of precursor used:

Thermosetting resins

(eg phenolics and furans) - these are used because they are processible by current industrial forming routes and readily impregnate preform structures. Thermosetting resins polymerise at low temperatures (<250 deg.C), to form a highly-crosslinked, glassy solid which does not graphitise, even at temperatures up to 3000 deg.C. The carbon yield from these precursors is up to 60%, and hence density of the composite may be adversely affected by the low density of the matrix formed.

Thermosetting resins are used to impregnate fibrous substrates, which are then taken through a carbonisation cycle. To build up density, it is frequently necessary to repeat the impregnation/carbonisation stage.

Pitches

(eg coal-tar and petroleum pitch) - these are used because they have a low melt viscosity, high carbon yield and a tendency to form graphitic structures. Impregnation under high pressure also forces the pitch into the pores, thereby increasing overall density; however, carbonisation must be performed slowly, so as to allow the gaseous products to evolve without blowing the component apart.

The most widely used approach for introducing a carbon matrix into the fibrous preform is through impregnation with a liquid pitch or resin precursor. This is followed by carbonisation in an inert atmosphere to convert the impregnant to coke. This conventional technique is usually repeated several times to improve density. Use of high isostatic pressure during the impregnation and coking stages of densification results in a more efficient process. Coke yields for pitch increase from 50% at ambient pressure to 85% at 70MPa.

Both thermosetting and pitch matrix precursors are compared in Fig.3. ^[2]

Successful impregnation processes require a low weight loss during carbonisation and, although a large range of organic compounds could be used, when all of the process and property requirements are considered, the choice becomes more limited. The following characteristics are required:

• low viscosity to help impregnation
• high carbon yield to provide a dense matrix
• low porosity microstructure and high crystallinity (directly affecting mechanical properties)

The mechanical properties of composites prepared using different precursors is given in Table 1. ^[3]

Table 1: Comparison of mechanical properties of composites produced from different precursors ^[3]

*PEEK - polyetheretherketone **PEI - polyetherimide

Infiltration/deposition processes

In simple terms chemical vapour infiltration (CVI) involves a solid material being deposited from gaseous reactants via thermal decomposition on to a heated substrate, in this case the fibrous preform. The preform is placed in a high temperature furnace and gaseous reagents such as methane are introduced. Thermal degradation of the gas occurs at ~1000 deg.C on the heated fibre surfaces forming a pyrolytic carbon deposit. This continues until the deposited material becomes the matrix. Advantages include relatively low processing temperatures and potential for near-net-shaping. Difficulties include a tendency to deposit preferentially near the component surface, leading to pore blockage or crusting, so inhibiting further densification and long processing times. Intermittent machining operations are needed to remove the crusting which adds to the long time needed for product manufacture.

Various techniques have been devised to speed up the production process, including plasma enhancement, directing the gas through the component, and thermal gradient techniques, (where the sample is heated from underneath, whilst facing a cold wall, hence setting up a localised thermal gradient). ^[4]

A general rule when deciding which processing route to opt for is:

• the gas phase route (although slow and expensive) produces high quality, thin-walled (up to a few centimetres) parts.
• the high pressure, pitch impregnation route produces high density components, particularly suited for ablative applications.

Applications of C/C

Carbon/carbon composites have been developed to combine high specific strength, toughness and stiffness of fibre-reinforced composites with the refractory properties of carbon. Their properties include: thermal shock resistance, high thermal conductivity, low thermal expansion, high specific strength, chemical inertness and biocompatibility.

Carbon/carbon is used in applications such as the nose cone and leading wing edges of the NASA Space Shuttle, the brakes of large commercial aircraft, the clutches and brakes of Formula 1 racing cars, high temperature furnace furniture, and as prosthetic devices and heating elements. Table 2 outlines the typical markets of carbon- and ceramic-matrix composites. ^[5]

Table 2: Markets for C/C and ceramic matrix composites in US$m

Currently, use of C/C composites is concentrated on components for missiles, re-entry vehicles and aircraft brakes.

Heat shields

Although the space shuttle, Fig.4, is probably the best known example of a C/C re-entry heat shield, the greatest numbers of parts used are in the nose cones of ballistic missiles. On a weight for weight basis, C/C can endure higher temperatures for longer periods of time than any other ablative material. Excellent thermal shock resistance permits rapid transition from -160 deg.C in the cold of space to approximately 1700 deg.C during re-entry, without fracture.

Brakes

Almost two-thirds by volume of C/C produced is used in aircraft braking systems. Use of C/C as a brake material is a result of its low weight and excellent friction and wear resistance. 40-60% weight saving is made when compared to steel, for Concorde this translates to approximately 600kg. The number of landings also increases by two to four times compared to steel/cermet brakes. Due to the prohibitive price of current C/C production, these components are usually only found on military aircraft, although Concorde and the Airbus (eg the A320, Fig.5) are notable exceptions. A number of land-based applications are being investigated including racing cars and high-speed trains.

Biomaterials

A further promising area of application is for biomedical materials, such as heart valves and artificial hips, Fig.6. This is due to the good compatibility of C/C with living tissue and also, since the modulus of the composite is close to that of bone, a C/C prosthesis embedded in a bone gives much less risk of damage/breakage than a conventional metal one. Thus, although a C/C prosthetic is more expensive than a metal one, it is less likely to need replacement, leading to less patient trauma and cost. C/C can be engineered to mimic bone, for example a femur is not uniform in its properties, but with suitable fibre orientation a very close match can be made.

Miscellaneous

The relatively high price of C/C materials has tended to restrict their use to aerospace and military application. Although there are other suitable areas, including glass-making equipment, high temperature mechanical fasteners, hot press dies and furnace furniture and heating elements, Fig.7.

The nuclear industry is interested in C/C for fusion reactors, since it resists ultra high temperatures and thermal shock. If used in automotive pistons, the low friction of C/C reduces drag and, in combination with low density, leads to greater efficiency and performance while reducing engine size and weight.

Conclusion

Carbon-carbon composites consist of carbon fibres in a carbon matrix. High specific strength at high temperatures, excellent fracture toughness and thermal shock resistance have resulted in C/C composites finding use in aircraft and racing car brakes, heatshields for re-entry vehicles and rocket nozzles. When C/C materials do break they generally undergo a non-brittle, graceful, type of fracture, rather than catastrophic failure.

Two factors are limiting widespread application of these materials: cost of production and oxidation of the matrix above 400 deg.C.

The oxidation factor is accommodated by using oxidation-resistant coatings. For example, piston-head crowns and the Space Shuttle use very similar technology. ^[6] The C/C parts are packed in silicon powder and heated to several thousand degrees. This vaporises the silicon allowing it to react with the carbon to form a tough SiC barrier coating ~0.0125mm thick, the parts are machined to final size by diamond tooling.

The cost of C/C composites is controlled, not by the material costs, but by the cost of fabrication; therefore, reduction in costs will be achieved by an improvement in the carbon yield. An increase in pressure from ambient, where a 50% carbon yield is achieved, to 100MPa, improves the carbon yield to 90%. This is the basis of high pressure carbonisation and pyrolysis, a technique being investigated by TWI and others.

References