Author: [Bruce Zhou]
Affiliation: [Jota Machinery Composites Material Prepreg Solution]
Corresponding Author: [jotamachinery@gmail.com]
Published : December 08 , 2025

Abstract

Carbon–carbon (C/C) composites occupy a unique position among structural materials. By combining high-performance carbon fibres with a carbon matrix, they deliver outstanding specific stiffness, strength and thermal stability at temperatures where metals and polymer matrix composites have already failed. At the same time, they suffer from well-known drawbacks: high manufacturing cost, complex multi-cycle densification and a strong tendency to oxidise in air above about 500 °C. This paper revisits the foundations laid by Savage’s monograph Carbon-Carbon Composites and integrates later industrial practice to provide a concise, engineering-oriented overview.

After recalling the bonding and microstructural features that make carbon so versatile, the review summarises the main reinforcement families (PAN-, rayon- and pitch-based fibres) and matrix precursors (chemical vapour deposition/infiltration, thermosetting resins and pitch). Typical process routes for producing dense C/C and C/C–ceramic hybrids are outlined, together with their implications for porosity, anisotropy and cost. Oxidation mechanisms and protection concepts are then discussed across three temperature windows, from glass-forming sealants below 1500 °C to SiC-based and ultra-high-temperature ceramic coatings above 1800 °C.

Mechanical, thermal and electromagnetic properties are interpreted in terms of fibre architecture, matrix microstructure and interface design, emphasising the need for an intermediate interfacial bond strength to balance toughness and strength. Classic applications—aircraft and racing brakes, rocket nozzles, re-entry heatshields, aero-engine parts and high-temperature fixtures—are used to illustrate where C/C genuinely outperforms competing materials. Finally, the paper highlights two recurrent themes in both historical and current work: the need to treat C/C as a materials system (composite plus coatings plus environment) and the continuing pressure to reduce cost while improving oxidation resistance and quality control for wider adoption.

Keywords

carbon-carbon composites; C/C composites; carbon fibre; PAN-based fibres; pitch-based fibres; chemical vapour infiltration; thermosetting resin precursors; pitch matrix; oxidation protection; C/C–SiC; high-temperature materials; aerospace brakes; rocket nozzles

1. Introduction

Carbon–carbon composites were developed to solve a very specific problem: how to retain useful strength, stiffness and dimensional stability in environments where metallic alloys soften, creep or melt, and where conventional polymer matrix composites have long since decomposed. By embedding high-strength, high-modulus carbon fibres in a carbon or graphite matrix, engineers created materials that can survive excursions well above 2000 °C in inert atmospheres and short-term peaks even higher in re-entry or rocket exhausts.

Savage’s 1993 book Carbon-Carbon Composites remains one of the most complete treatments of this family of materials. It combines a rigorous discussion of carbon bonding and microstructure with practical processing know-how, laboratory-scale procedures and a sober assessment of applications and markets. Three conclusions from that work still frame current practice:

• C/C is a family, not a single recipe. Fibre type, architecture, matrix precursor, densification route and heat-treatment temperature can be combined in many ways.
• Performance is exceptional but expensive. High raw-material cost, repeated impregnation and heat-treatment cycles and complex quality control restrict volumes.
• Oxidation is the intrinsic weak point. Without adequate protection, even the best C/C will slowly burn away in hot oxidising environments.

This paper revisits those themes with an emphasis on process–structure–property relationships that are directly useful to designers and process engineers who work with high-temperature components.

2. Fundamentals of Carbon and Carbon–Carbon Composites

2.1 Carbon bonding and hybridisation

Carbon’s electronic structure (1s² 2s² 2p²) allows multiple hybridisation states:

• sp³: tetrahedral, as in diamond, giving a three-dimensional network with very high hardness and thermal conductivity.
• sp²: trigonal planar, as in graphite, with strong in-plane σ bonds and delocalised π bonding between layers, leading to electrical conductivity and anisotropy.
• sp: linear chains, less relevant to structural C/C but important during early pyrolysis stages.

Real engineering carbons are neither perfect diamond nor perfect graphite. They comprise stacked, misaligned and sometimes curved graphene layers, with varying degrees of order. X-ray diffraction parameters such as interlayer spacing and crystallite height, together with Raman spectroscopy (D and G bands), are commonly used to quantify this order and to follow the evolution from amorphous char towards more graphitic structures during heat treatment.

2.2 From carbon materials to carbon–carbon composites

Carbon–carbon composites are built by:

1. Selecting and shaping a suitable carbon fibre reinforcement.
2. Infiltrating that porous preform with a carbon-yielding matrix precursor.
3. Pyrolysing and often graphitising the matrix to convert it to carbon.
4. Repeating impregnation and heat-treatment cycles until density and properties meet the requirement.
5. Adding oxidation protection, where the service environment demands it.

The resulting composite microstructure depends not only on the fibre architecture (unidirectional, 2D woven, 3D braided or needled) but also on the matrix route—CVD/CVI, thermosetting resin or pitch—and the thermal history. This tunability is one of C/C’s strengths, but it also explains why data from one system cannot be transferred blindly to another.

3. Reinforcements: Carbon Fibre Families and Their Role in C/C

3.1 Rayon-based fibres

Historically, rayon was the first widely used precursor for carbon fibres. After stabilisation and carbonisation, it produced low-modulus, relatively low-strength fibres with a distinctive “crenulated” surface. Carbon yields are modest (around 20–25 %), and the microstructure is dominated by fine, disordered graphite-like domains.

In the context of C/C, rayon-derived fibres still find use in ablative components where through-thickness thermal conductivity must be kept low and where interlaminar shear strength is important. However, they have largely been displaced by PAN- and pitch-derived fibres in structural applications.

3.2 PAN-based fibres

Polyacrylonitrile (PAN) has become the workhorse precursor for structural carbon fibres and, by extension, for many C/C composites. Processing typically involves:

1. Wet or dry-jet spinning of oriented PAN filaments.
2. Stabilisation in air at 200–300 °C to form a ladder polymer and prevent melting.
3. Carbonisation in inert atmosphere at roughly 1000–1500 °C.
4. Graphitisation up to 2500–3000 °C, depending on the targeted modulus.

During stabilisation and carbonisation, the PAN chains cyclise, dehydrogenate and aromatise, eventually forming turbostratic graphite-like regions aligned along the fibre axis. Higher tension and higher graphitisation temperatures promote orientation and increase modulus, but they can also reduce strain to failure.

Commercial PAN fibres are usually grouped as low modulus (LM), intermediate modulus (IM) and high modulus (HM). Intermediate modulus grades, with tensile strengths of 4–5 GPa and moduli around 250–280 GPa, dominate aerospace and motorsport structures and are widely used in C/C brake disks and heatshields. Their relatively robust processing window and established supply base support consistent quality—an important part of the “trust” factor around C/C technology.

3.3 Pitch-based fibres

Pitch-derived fibres, particularly those based on mesophase pitch, address a different part of the design space. Petroleum or coal-tar pitches can be heat-treated to form discotic liquid-crystalline (mesophase) domains, which, when melt-spun, produce fibres with highly aligned graphene layers.

Key characteristics include:

• High carbon yield (often 75–85 %), improving material utilisation.
• Very high modulus (400 GPa and above) and, in some grades, exceptional in-plane thermal conductivity.
• Sensitive processing window, especially in melt spinning and stabilisation.

Although the underlying cost of pitch is low, the combination of strict purity requirements and narrow process tolerances has historically kept high-performance pitch fibres in the premium price bracket. In C/C, pitch fibres are selected where ultra-high modulus or high thermal conductivity is crucial, for example in some brake and heat-spreading components.

4. Matrix Precursors and Densification Routes

4.1 Gas-phase impregnation: CVI and CVD

Gas-phase densification by chemical vapour infiltration (CVI) is a central route for producing high-quality C/C and C/C–ceramic composite matrices. In a typical process, a porous fibre preform is exposed to a hydrocarbon gas at elevated temperature and reduced pressure. Decomposition on internal surfaces deposits carbon, gradually filling pores.

Advantages include:

• Low matrix processing stress, reducing fibre damage.
• High purity and control over matrix texture.
• Capability to deposit not only carbon but also ceramic phases such as SiC, TiC, B₄C or BN, enabling hybrid C/C–ceramic matrices and protective coatings.

The main practical challenge is achieving uniform densification. Excessive deposition at the outer surfaces can seal the preform and trap internal porosity (“overcrusting”). Reactor design, gas flow, pressure and temperature must therefore be tuned carefully, often aided by microprobe mapping and tomography to check density gradients.

4.2 Thermosetting resin precursors

An alternative route uses thermosetting resins—phenolics, furans and related systems—as matrix precursors. The basic steps are:

1. Impregnation of the fibre preform with liquid or solution resin.
2. Curing to a crosslinked polymer.
3. Pyrolysis under inert atmosphere to convert the polymer to an amorphous carbon char.
4. Re-impregnation and further pyrolysis cycles to raise density.

Resin selection is guided by:

• Carbon yield, which should be as high as practicable to minimise shrinkage and porosity.
• The relationship between glass transition temperature and decomposition temperature: a matrix that spends a long time in a rubbery state while gases evolve tends to crack and foam.
• Compatibility with fibre sizing and wet-out behaviour.

One of Savage’s important practical observations is the existence of an optimum fibre–matrix bond strength. If the interface is too weak, the matrix shrinks away during pyrolysis, leading to gaps and poor load transfer. If it is too strong, the matrix is dragged with the fibres, generating high internal stresses and extensive cracking. Controlled surface treatment of fibres and careful choice of precursor chemistry are therefore essential to obtain both acceptable shrinkage and attractive mechanical properties.

4.3 Thermoplastic/pitch precursors

Pitch-based matrices occupy a middle ground between CVI and resin routes. High carbon yields and the possibility of developing graphitic microstructures make pitch matrices attractive where high thermal conductivity is needed. However, infiltration viscosity, phase separation and volatile evolution during pyrolysis all require close control.

Processing options include:

• Low-pressure infiltration with subsequent carbonisation.
• High-pressure or autoclave routes, which can improve penetration but impose additional equipment costs.
• Use of additives to tailor mesophase formation and final microstructure.

In practice, many industrial C/C systems combine multiple routes: for example, a resin-derived first matrix, followed by CVI or pitch impregnation to close residual porosity and tune properties.

5. Oxidation Behaviour and Protection Concepts

5.1 Oxidation mechanisms

In inert atmospheres, C/C can withstand temperatures far beyond those tolerable by most structural materials. In oxygen-containing atmospheres, however, oxidation becomes appreciable above about 400–500 °C and accelerates rapidly as temperature increases.

Key factors include:

• Degree of graphitisation and crystallite orientation.
• Open porosity and microcrack network providing gas access.
• Presence of catalytic impurities (alkali metals, transition metals).
• Local stress state and thermal gradients.

Without protection, a component may gradually lose surface material, develop internal “wormholes”, or suffer non-uniform thinning that undermines mechanical integrity.

5.2 Temperature-window approach to protection

Savage recommends viewing oxidation protection in three temperature bands:

1. Below about 1500 °C
   Glass-forming sealants and borosilicate or phosphate coatings can provide effective barriers, flowing to seal cracks and pores. Silica-rich systems that form viscous glassy layers around 1000–1200 °C are widely used.
2. Approximately 1500–1800 °C
   Simple glassy sealants soften excessively or volatilise. Multi-layer systems combining carbides (SiC, TiC), nitrides and refractory oxides become necessary. Hybrid C/C–SiC matrices produced by CVI often provide both structural and protective functions in this range.
3. Above about 1800 °C
   Protection becomes increasingly challenging, particularly under high heat flux. Sacrificial layers, ultra-high-temperature ceramic coatings and controlled atmospheres are often combined. In some re-entry or rocket applications, designers accept controlled ablation as part of the mission profile.

An important practical lesson is that oxidation protection cannot be treated as a separate “paint-on” step. Coating choice interacts with substrate microstructure, thermal expansion mismatch, residual stresses and service environment. Successful designs treat C/C plus its protective system as a single materials architecture.

6. Properties of Carbon–Carbon Composites

6.1 Microstructure–property relationships

The macroscopic behaviour of C/C is governed by several coupled microstructural features:

• Fibre architecture: unidirectional laminates offer maximum in-plane stiffness and strength but modest interlaminar properties. 2D weaves, 3D braids and needled preforms trade some in-plane performance for better through-thickness toughness.
• Matrix texture: isotropic resin-derived carbons tend to be relatively tough but thermally less conductive; graphitised pitch matrices provide high in-plane conductivity but can be more brittle.
• Interface characteristics: interfacial debonding and crack deflection contribute to toughness. Too strong an interface encourages straight-through cracking; too weak an interface yields low shear strength.

6.2 Mechanical properties

Typical ranges for structural C/C include:

• Tensile strengths from several hundred MPa up to around 1–2 GPa in the fibre direction, depending on fibre grade, volume fraction and lay-up.
• In-plane elastic moduli from roughly 50 GPa (low-modulus, highly porous systems) to beyond 200 GPa for dense, high-modulus designs.
• Modest interlaminar shear and through-thickness tensile strengths, particularly in layered 2D architectures without through-thickness reinforcement.

The combination of high specific strength, retention of properties at elevated temperature and gradual, damage-tolerant failure in well-designed microstructures underpins the strong safety record of C/C brake and nozzle components when correctly engineered.

6.3 Thermal and electrical properties

Thermal behaviour is one of C/C’s defining attractions:

• In-plane thermal conductivity can range from modest (tens of W m⁻¹ K⁻¹) in resin-derived matrices up to several hundred W m⁻¹ K⁻¹ in graphitised pitch systems.
• Through-thickness conductivity is typically much lower and sensitive to porosity and fibre orientation.

Electrical conductivity is similarly anisotropic but generally high, enabling electromagnetic shielding and, in some designs, resistive heating or sensing functions.

7. Applications and Market Experience

7.1 Brakes and clutches

Perhaps the most visible success story for carbon-carbon composites lies in aerospace and motorsport brakes. Aircraft main-wheel brake stacks and Formula 1 brake disks exploit:

• High friction coefficients and stable braking performance over a wide temperature range.
• Low density and high specific heat, improving dynamic performance and reducing unsprung mass.
• Controlled oxidation and wear behaviour when combined with suitable protective systems and operating envelopes.

Decades of service experience have built a strong evidence base for reliability under demanding cyclic loading and thermal shock, provided that manufacturing and inspection are tightly controlled.

7.2 Rocket motors, nozzles and heatshields

In rocket nozzles and re-entry heatshields, C/C provides:

• High thermal shock resistance.
• Dimensional stability at extreme temperatures in inert or reducing environments.
• Tailorable ablation behaviour where controlled material removal protects underlying structure.

Hybrid C/C–SiC and related C/C–ceramic systems extend usefulness in oxidising exhaust streams and long-duration missions.

7.3 Aero-engine and industrial components

Aero-engine components exposed to hot gas paths, as well as industrial furnace fixtures, crucibles and linings, benefit from:

• Reduced mass vs refractory metals and ceramics.
• Resistance to thermal cycling.
• Non-reactivity with many process atmospheres, provided oxidation is mitigated.

7.4 Biomedical and niche applications

C/C’s biocompatibility and chemical inertness have led to exploratory and limited use in orthopaedic implants and related devices. Here, microstructure must be tailored for both mechanical performance and surface interactions with tissue and fluids.

7.5 Market constraints

Savage’s market analysis in the early 1990s already highlighted key barriers that remain relevant:

• High part cost driven by fibre price, multiple densification cycles and energy-intensive heat treatments.
• Oxidation protection systems that add further cost and complexity.
• Limited supply base and specialist know-how, concentrating production in a small number of companies.

Although fibre and process costs have evolved since, the underlying economic picture has not fundamentally changed: C/C remains a high-value, niche solution reserved for situations where no other material meets the combined thermal and mechanical requirements.

8. Challenges and Future Directions

Drawing together the historical record and current practice, several themes stand out:

1. Cost reduction and process efficiency
   Higher carbon yield precursors, optimised CVI cycles, improved impregnation schemes and automation all contribute to lower part cost. However, any reduction must preserve reproducibility and safety margins in critical components.
2. Improved oxidation protection
   Multi-layer coating systems, C/C–ceramic hybrid matrices and smart sealants continue to evolve. A promising direction is integrated design of substrate microstructure and coating so that thermal expansion, crack paths and degradation modes are managed as a whole.
3. More robust design methods
   Physics-based modelling of pyrolysis shrinkage, crack evolution, oxidation kinetics and coating behaviour, combined with probabilistic design, would allow engineers to exploit C/C more confidently in variable and long-life service conditions.
4. Lifecycle and sustainability considerations
   Although C/C is used in relatively low volumes, there is growing interest in recyclability, reduced energy consumption during processing and the use of alternative or bio-derived precursors where feasible.
5. Expanded application domains
   Advanced propulsion systems, hypersonic vehicles, high-temperature process equipment and energy technologies continue to pose demands that play to the strengths of carbon-carbon composites. Each new application must, however, be matched carefully to a suitable C/C variant and protection system.

9. Conclusion

Carbon–carbon composites remain one of the most capable materials families for extreme temperature environments. Savage’s monograph laid out the foundations: a clear description of carbon science, fibre and matrix processing, microstructure–property relationships and the realities of commercialisation. Many of the core insights still guide practice today: the need to balance fibre type, matrix route and interfacial chemistry; the central importance of oxidation protection; and the recognition that C/C earns its place where performance, not initial cost, is the primary driver.

For engineers and researchers working on high-temperature structures, C/C offers an instructive case study in how deep understanding of atomic bonding, microstructure and process integration can be translated into reliable, if specialised, industrial products. Future progress will likely come from incremental improvements in cost and protection, better modelling tools and careful expansion into applications where its unique combination of properties justifies the investment.

References

1. Savage, G. Carbon-Carbon Composites. Chapman & Hall, London, 1993.
2. Windhorst, T.; Blount, G. Carbon-carbon composites: a summary of recent developments and applications. Materials & Design 1997, 18(1), 11–15.
3. Quenisset, J.M. Carbon combinations. Advanced Materials 1994, 6(2), 178.