Author: [Bruce Zhou]
Affiliation: [Jota Machinery Composites Material Prepreg Solution]
Corresponding Author: [jotamachinery@gmail.com]
Published : December 09 , 2025
Abstract
Carbon/phenolic laminates are established thermal protection materials for rocket nozzles and high-temperature aerospace components, but conventional micron-filled systems often require thick, heavy ablative layers and suffer from char erosion. In this study, carbon/phenolic prepregs based on a resol-type phenolic resin were developed with and without 5 wt% ammonium-modified bentonite nanoclay, and the influence of nanoclay on processing, mechanical behaviour and fire performance was systematically assessed. The bentonite was organophilized by cation exchange, as confirmed by an increase in basal spacing from 13.42 Å to 23.26 Å and by thermogravimetric signatures associated with organic degradation. FTIR-based cure monitoring of the phenolic resin showed that 4 h at 80 °C produced a B-stage conversion of ~0.35, while full crosslinking required 20 min at 190 °C. Gel time measurements indicated a modest acceleration of cure at elevated temperatures in the presence of nanoclay.
Unidirectional T700SC carbon-fibre prepregs were fabricated by vacuum-bag molding at 80 °C, followed by hot pressing of stacked plies using a multi-step cycle up to 190 °C. Nanoclay-modified prepregs exhibited comparable fibre volume fraction but higher density, slightly reduced flexural stiffness and, critically, measurable tack (~4.4 N/m) relative to virtually tack-free unmodified prepregs. In consolidated laminates, the nanoclay-containing system showed a reduced fibre volume fraction (82 → 73 vol%), leading to a decrease in flexural modulus from 130 ± 7 GPa to 104 ± 6 GPa and in flexural strength from 1.0 ± 0.7 GPa to 0.68 ± 0.09 GPa.
Cone calorimetry at 50 kW/m² revealed that both carbon/phenolic composites outperformed an epoxy/carbon reference in terms of peak heat release rate (pHRR) and total heat release (THR). Incorporation of nanoclay further reduced pHRR from 111.0 to 86.1 kW/m² and slightly lowered THR, while visual inspection and residue analysis indicated a more stable, thermally insulating char that prevented backside aluminium foil melting. However, total smoke production and smoke production rate increased by factors of approximately 2.5 and 2.4, respectively.
Overall, nanoclay-modified phenolic resin in carbon/phenolic prepregs improves tack and significantly enhances fire performance via char stabilization, at the expense of some reduction in flexural properties and higher smoke output. These results support the use of nanoclay-modified carbon/phenolic prepregs for unoccupied, high-temperature aerospace environments and motivate further optimisation of clay loading, consolidation parameters and hybrid filler strategies to balance mechanical integrity and ablative performance.
Keywords
Carbon/phenolic composite; nanoclay; prepreg; thermal protection; cone calorimetry; char formation; ablative materials; high-temperature composites
1. Introduction
Carbon-fibre/phenolic (CF/phenolic) composites are widely used as ablative thermal protection in rocket nozzles, re-entry vehicles and hot gas-path components owing to their ability to form a protective carbonaceous char when exposed to high heat fluxes. Phenolic resins provide inherent flame retardancy and char yield, while carbon fibres supply structural support and dimensional stability at elevated temperature. Classical CF/phenolic ablatives often employ micron-scale fillers such as glass, mineral oxides or carbon powders to stabilise the ablating surface and tailor thermal conductivity. Despite their success, these formulations typically require relatively thick layers to meet thermal and erosion requirements, leading to increased mass, and the char formed under severe heat flux can be mechanically fragile and susceptible to spallation.
The emergence of polymer nanocomposites has opened new routes to enhance ablative performance by exploiting nanoscale reinforcement in the resin matrix. Layered silicate nanoclays, in particular, are attractive due to their high aspect ratio, availability, and capacity to form tortuous diffusion paths and inorganic skeletons within the pyrolyzed matrix. Previous studies have reported that nanoclays and other nanoparticles such as nanosilica or carbon nanotubes can reduce ablation rate and heat release by promoting the formation of a more cohesive, ceramic-like char layer. However, most of these works have focused on bulk resins or moulded blocks, with limited attention to prepreg-based processing routes that are essential for fabricating high-quality laminates with controlled fibre volume fraction and low porosity.
In practical aerospace manufacturing, prepregs with adequate tack, drape and storage stability are critical for automated lay-up, autoclave curing and integration into complex architectures. Introducing nanoclays into phenolic resin modifies rheology and cure behaviour, which can affect prepreg handling as well as final composite properties. A comprehensive evaluation must therefore encompass both processability and in-service performance.
This work addresses that need by developing carbon/phenolic prepregs based on a resol-type phenolic resin, with and without 5 wt% ammonium-modified bentonite nanoclay, and by systematically evaluating their cure behaviour, prepreg handling characteristics, laminate mechanical properties and fire performance under cone calorimetry. The central hypothesis is that nanoclay will enhance char stability and reduce heat release during combustion, while its impact on mechanical properties will be governed primarily by changes in fibre volume fraction and matrix morphology.
2. Literature Review
CF/phenolic systems have a long history as ablative materials, exemplified by NASA’s MX-4926 and related carbon/phenolic formulations used in solid rocket motor nozzles. In such systems, phenolic resin decomposes to yield a porous char that insulates the underlying structure. Micron-sized inorganic fillers are traditionally added to stabilise the char, reduce thermal conductivity and adjust ablation rate. However, under high-enthalpy flows, the char can erode mechanically, exposing fresh material and requiring thicker ablative layers to ensure mission life.
Nanoparticle modification of ablatives has emerged as a promising approach to mitigate these drawbacks. Srikanth et al. reported that nanosilica-modified carbon/phenolic composites develop SiC-rich surface layers during ablation, significantly reducing erosion rates. Natali and co-workers compared carbon black and multi-walled carbon nanotubes (MWCNTs) in phenolic matrices and found that nanoscale carbon fillers promote thinner yet more effective protective zones. Koo and co-authors highlighted polymer nanostructured materials for propulsion applications, noting that nanoclays at loadings around 7.5 wt% can reduce surface recession and lower surface temperature in carbon/phenolic ablatives.
Layered silicate nanoclays, especially organomodified montmorillonite or bentonite, are particularly attractive because their plate-like morphology affords high barrier efficiency at relatively low volume fractions. Upon combustion, they may form an inorganic scaffold that reinforces the carbonaceous char and suppresses bubble coalescence and crack propagation. Several studies in epoxy and thermoplastic matrices have demonstrated reductions in peak heat release rate (pHRR) and improvements in limiting oxygen index with nanoclay additions, albeit with possible increases in smoke production and changes in mechanical performance.
In parallel, prepreg technology has become the dominant route for high-performance fibre-reinforced polymers in aerospace. Pre-impregnated fabrics or unidirectional tapes provide accurate resin content, consistent fibre distribution and low void content when processed under controlled pressure and temperature. For phenolic systems, prepregs must balance storage life with sufficient tack for lay-up and must be compatible with multi-step cure cycles that achieve high crosslink density without excessive exotherm.
Despite the obvious synergy between nanoclay modification and prepreg-based CF/phenolic systems, relatively few studies have examined nanoclay-filled phenolic prepregs. Most existing works on nanoclay-modified phenolic composites use bulk moulding or hand lay-up without a separate B-staged prepreg step. Consequently, the combined effects of nanoclay on resin cure kinetics, prepreg handling, fibre volume fraction, mechanical performance and fire behaviour remain incompletely understood.
The present study contributes to this gap by (i) organomodifying bentonite for compatibility with phenolic resin, (ii) producing nanoclay-modified carbon/phenolic prepregs via vacuum-bag molding, and (iii) comparing their behaviour against unmodified CF/phenolic systems using a consistent processing route and test matrix.
3. Methodology
3.1 Materials
A resol-type phenolic resin was synthesised in-house from phenol and 37% formaldehyde at a mole ratio of F:P = 3:1 under basic conditions (pH ≈ 9, NaOH catalyst) and reacted at 90 °C for 2 h. The resin was subsequently neutralised to pH ~7 using boric acid and dehydrated under vacuum at 75 °C. The resulting resin was stored at −10 °C to preserve reactivity.
Commercial bentonite clay was organomodified by ammonium cation exchange following a previously established procedure. X-ray diffraction confirmed an increase in basal spacing indicative of successful intercalation of organic cations. To prepare the nanoclay-modified resin, the phenolic resol was pre-concentrated to a viscosity of approximately 1100 cP, after which 5 wt% of the modified bentonite was dispersed using mechanical stirring and sonication for 30 min.
Unidirectional stitched carbon-fibre fabric based on Toray T700SC-12000 tow was used as reinforcement.
3.2 Prepreg fabrication
Prepregs were produced by vacuum-bag molding at 80 °C. A flat mould was coated with release agent, and carbon fabric plies were hand-impregnated with either the neat phenolic resin or the nanoclay-modified phenolic resin. The lay-up was completed with a release film, bleeder layer and vacuum bag. Cure was conducted for 4 h at 80 °C, with vacuum applied during the first 2 h. This step partially advanced the resin to a B-stage suitable for storage and subsequent consolidation.
Two prepreg types were obtained:
- PC: carbon/phenolic prepreg without nanoclay,
- PBC: carbon/phenolic prepreg with 5 wt% ammonium-modified bentonite.
Prepreg sheets of approximately 45 × 35 cm were produced for each formulation.
3.3 Laminate consolidation
Prepregs were cut into 10 × 12 cm plies and stacked unidirectionally. Consolidation was performed in a hot press under a nominal load of 1 ton, using a stepped cure cycle designed to achieve full crosslinking:
- 45 °C for 15 min
- 60 °C for 1 h
- 80 °C for 4 h
- 110 °C for 1 h
- 150 °C for 1 h
- 190 °C for 2 h
After cooling to room temperature under pressure, panels were demoulded and machined into specimens for mechanical and fire testing.
3.4 Characterisation
Clay modification: X-ray diffraction (XRD) was used to measure basal spacing d001 before and after organomodification. Thermogravimetric analysis (TGA/DTG) in nitrogen, from ambient to ~800 °C, characterised water loss, organic decomposition and dehydroxylation.
Resin cure behaviour: FTIR spectroscopy was employed to monitor the evolution of methylene bridge vibrations associated with ortho–para and para–para linkages at 1473 cm⁻¹ and 1456 cm⁻¹. Dynamic and isothermal cure experiments were conducted to evaluate conversion versus time and temperature. Gel times at 80, 100 and 120 °C were measured by standard tube methods for both neat and nanoclay-modified resins.
Prepreg properties: Prepreg density and fibre content were determined from mass and volume measurements and fibre areal weight. Flexural rigidity was measured according to ASTM D1388 (method A), and tack was assessed via a peel-type test akin to ASTM D3167, expressed as force per unit width.
Composite mechanical properties: Laminate density and fibre volume fraction were obtained using burn-off or acid digestion methods. Flexural modulus and strength in the fibre direction were measured following ASTM D790 under three-point bending.
Fire performance: A cone calorimeter (ASTM E1354) was used to evaluate fire behaviour at an external heat flux of 50 kW/m². Square specimens of 100 × 100 mm and thickness ~4 mm were tested in horizontal configuration. Time to ignition, time to peak heat release rate, flaming duration, peak heat release rate (pHRR), total heat release (THR) and residue mass fraction were recorded. Backside aluminium foil was inspected post-test for evidence of melting.
Smoke production: Total smoke production (TSP) and smoke production rate (SPR) were obtained from the cone calorimeter’s optical system for both composite types.
4. Results
4.1 Clay modification and resin cure
XRD results showed that the basal spacing of bentonite increased from 13.42 Å in the unmodified clay to 23.26 Å after ammonium cation exchange, indicating significant interlayer expansion and successful organomodification. TGA/DTG revealed additional mass-loss events in the 150–550 °C range associated with decomposition of the organic modifier, along with a reduction in low-temperature water loss, consistent with increased hydrophobicity.
FTIR-based cure monitoring demonstrated that the phenolic resin reached a maximum methylene-bridge conversion after approximately 20 min at 190 °C under isothermal conditions. At 80 °C, a 4 h hold produced a conversion of about 0.35, confirming that the prepregs were B-staged but far from fully crosslinked. Gel-time measurements showed that the presence of nanoclay slightly accelerated cure at higher temperatures: at 100 °C, gel time decreased from 2 h 57 min (neat resin) to 2 h 5 min (modified), and at 120 °C from 41 min to 34 min. At 80 °C both systems retained gel times beyond 4 h.
4.2 Prepreg characteristics
Prepreg properties are summarised in Table 1.
Table 1. Prepreg properties (mean ± standard deviation).
| Property | PC (neat phenolic) | PBC (nanoclay-modified) |
|---|---|---|
| Density (g/mL) | 1.01 ± 0.19 | 1.26 ± 0.10 |
| Fibre content (vol%) | 43 ± 8 | 45 ± 4 |
| Flexural stiffness (mJ/m) | 18.5 ± 6.0 | 13.8 ± 2.8 |
| Tack (N/m) | not measurable | 4.4 ± 0.79 |
Both prepregs achieved comparable fibre contents within experimental scatter. The nanoclay-modified prepregs exhibited a higher density, reflecting the incorporation of inorganic filler. Flexural stiffness decreased moderately, indicating slightly improved drape. Most notably, tack, which was effectively negligible for the neat phenolic prepregs, reached ~4.4 N/m for the nanoclay-containing prepregs, enabling more reliable ply adhesion during lay-up.
4.3 Laminate mechanical properties
Laminate properties are listed in Table 2.
Table 2. Flexural properties of consolidated laminates.
| Property | PC | PBC |
|---|---|---|
| Fibre volume fraction (vol%) | 82 ± 1 | 73 ± 4 |
| Density (g/mL) | 1.60 ± 0.04 | 1.70 ± 0.02 |
| Flexural modulus (GPa) | 130 ± 7 | 104 ± 6 |
| Flexural strength (GPa*) | 1.0 ± 0.7 | 0.68 ± 0.09 |
*Units for strength are nominally reported as GPa in the source data; in practice, these values likely correspond to hundreds of MPa.
The nanoclay-modified laminates (PBC) exhibited a lower fibre volume fraction than the unmodified PC laminates (73 vs 82 vol%), accompanied by a reduction in flexural modulus (~20%) and strength (~30–35%). The higher composite density of PBC reflects both the greater matrix content and the inorganic clay fraction.
4.4 Fire performance
Cone calorimetry results are summarised in Table 3.
Table 3. Cone calorimeter results at 50 kW/m².
| Property | PC | PBC |
|---|---|---|
| Peak HRR (kW/m²) | 111.0 | 86.1 |
| THR (MJ/m²) | 16.7 | 15.9 |
| Residue (%) | 88.2 | 83.8 |
Both carbon/phenolic systems exhibited low pHRR and THR relative to typical epoxy/carbon laminates reported in the literature. The incorporation of nanoclay reduced pHRR by approximately 22–23% (111.0 → 86.1 kW/m²) and marginally decreased THR.
Characteristic times derived from the HRR curves indicated that both PC and PBC composites had longer ignition times than an epoxy reference laminate. The nanoclay-modified composite displayed a somewhat longer flame duration but a delayed time to peak HRR compared with the unmodified system. Visual inspection after testing revealed that the backside aluminium foil in PC specimens had melted or significantly deformed, while the foil remained intact in PBC specimens, suggesting improved thermal insulation by the nanoclay-stabilised char.
4.5 Smoke production
Smoke parameters are given in Table 4.
Table 4. Smoke production metrics.
| Material | TSP (m²) | SPR (m²/s) |
|---|---|---|
| PC | 0.9 | 0.0017 |
| PBC | 2.4 | 0.0040 |
Nanoclay addition led to a near tripling of total smoke production and more than doubling of smoke production rate.
5. Discussion
The results demonstrate that introducing 5 wt% ammonium-modified bentonite into a resol-type phenolic resin has multiple effects on CF/phenolic prepregs and laminates.
From a processing standpoint, the organomodified clay increased resin viscosity and slightly accelerated cure at elevated temperatures, yet the prepregs remained B-staged after 4 h at 80 °C, retaining adequate processing latitude. The higher viscosity appears to be beneficial at the prepreg stage: the nanoclay-containing prepregs showed measurable tack within the range of commercial structural prepregs, whereas the neat phenolic prepregs were essentially non-tacky. Improved tack is advantageous for ply positioning, lay-up on contoured tools and automated handling. The modest reduction in flexural stiffness suggests that drape remains acceptable.
During laminate consolidation, however, the increased resin viscosity reduced resin bleed into the bleeder layers, leading to a lower fibre volume fraction in PBC laminates compared with PC laminates. Since axial flexural modulus and strength in unidirectional composites scale primarily with fibre volume fraction and fibre properties, a significant portion of the observed mechanical degradation can be rationalised in terms of increased matrix content. Simple rule-of-mixtures estimates show that a drop in FVF from 82 to 73 vol% would be expected to reduce modulus by approximately 10–20%, consistent with the measured 20% decrease. The somewhat larger reduction in flexural strength suggests additional contributions from potential changes in interfacial adhesion, matrix brittleness or void content, which were not explicitly measured.
In contrast, the fire performance clearly benefits from the presence of nanoclay. The reduction in pHRR, despite higher matrix content, indicates that the nanoclay substantially alters the degradation pathway and char architecture of the phenolic matrix. The increased inorganic content and the layered morphology of bentonite likely promote the formation of a more continuous, mechanically robust char layer with reduced permeability to volatile products and oxygen. This stabilised char acts as an effective heat shield, as evidenced by lower heat release, slightly reduced THR and the absence of backside aluminium melting in PBC specimens. Adjusting for fibre volume fraction, the effective char yield attributable to the matrix is higher in PBC than in PC, further supporting this interpretation.
The increase in smoke production is a notable trade-off. The organic component of the clay modifier and altered degradation chemistry can generate additional aromatic fragments and soot precursors, leading to higher TSP and SPR. For external propulsion components, rocket nozzles and uninhabited engine compartments, smoke generation is generally of secondary concern relative to thermal protection and structural integrity. However, for interior or cabin-adjacent structures subject to stringent flame, smoke and toxicity regulations, the observed smoke penalty would be unacceptable without further mitigation.
In the broader context of nanoparticle-modified ablatives, the behaviour observed here is consistent with previous findings for nanoclay- or nanosilica-filled phenolic systems: enhanced char stability and reduced ablation or heat release are typically accompanied by increased viscosity and potential changes in smoke and mechanical properties. The present study is distinctive in that it implements a prepreg route and explicitly quantifies prepreg tack and stiffness, connecting nanoscale modification to industrially relevant processing metrics.
Several limitations should be acknowledged. First, only a single nanoclay loading (5 wt%) was investigated. Prior work suggests there may be an optimal clay content balancing mechanical properties and fire performance; exploring a broader range (e.g., 2–10 wt%) would enable more informed trade-offs. Second, the mechanical characterisation was limited to room-temperature flexural behaviour of unidirectional laminates. For real ablative structures, through-thickness properties, interlaminar shear and high-temperature performance are critical; these were not addressed. Third, microstructural analysis of the cured matrix and post-fire char, for instance by SEM or TEM, would be valuable to directly connect nanoclay dispersion and char morphology to the observed cone calorimeter results. Finally, the cone calorimeter provides a controlled, relatively low-flow heat flux; more aggressive tests such as oxy-acetylene torch or arc-jet ablation would be required to fully assess erosion resistance under realistic rocket or re-entry conditions.
Despite these limitations, the study provides a useful foundation for future optimisation of nanoclay-modified CF/phenolic prepregs. Adjusting consolidation parameters (pressure, bleeder design, dwell times) to recover higher fibre volume fractions in nanoclay systems could mitigate the observed mechanical losses. Combining nanoclay with other nanoscale fillers—such as nanosilica, carbon black or graphene oxide—may further refine char microstructure and compensate for matrix brittleness.
6. Conclusion
Carbon/phenolic prepregs incorporating 5 wt% ammonium-modified bentonite nanoclay have been developed and evaluated in terms of cure behaviour, prepreg handling, mechanical performance and fire response. Organomodification of bentonite increased interlayer spacing and hydrophobicity, enabling its dispersion in a resol-type phenolic resin. FTIR and gel-time measurements confirmed that B-staged prepregs could be obtained at 80 °C while full cure required temperatures near 190 °C, with nanoclay causing mild acceleration of cure at elevated temperatures.
Nanoclay addition produced prepregs with comparable fibre content, slightly enhanced drape and, importantly, measurable tack suitable for practical lay-up. In consolidated unidirectional laminates, the higher viscosity of the nanoclay-modified resin reduced resin bleed, leading to a lower fibre volume fraction and corresponding reductions in flexural modulus and strength.
Cone calorimetry demonstrated that both carbon/phenolic systems exhibit excellent fire performance relative to typical epoxy composites, and that nanoclay further decreases peak heat release rate and slightly reduces total heat release while improving backside thermal insulation. The enhanced fire behaviour is attributed to the formation of a more stable, inorganic-reinforced char layer. This benefit is counterbalanced by increased smoke production and a modest mechanical penalty.
These findings indicate that nanoclay-modified carbon/phenolic prepregs are promising candidates for high-temperature aerospace applications where mass efficiency, fire resistance and manufacturability are critical, and where smoke is not a primary design constraint. Further optimisation of nanoclay content, consolidation parameters and hybrid filler strategies, coupled with more extensive mechanical and ablation testing, is recommended to fully exploit the potential of this material system.
References
- Asaro, L.; Manfredi, L. B.; Alvarez, V.; Rodríguez, E. S. Development of high temperature resistant materials using carbon/phenolic prepregs with nanoclays. Proceedings of ICCM19 – 19th International Conference on Composite Materials, 2013.
- Srikanth, I.; et al. Nano silica modified carbon–phenolic composites for enhanced ablation resistance. Scripta Materialia 2010, 63, 200–203.
- Natali, M.; et al. Ablative properties of carbon black and MWCNT/phenolic composites. Composites Part A 2012, 43, 174–182.
- Koo, J. H.; et al. Polymer nanostructured materials for propulsion systems. Journal of Spacecraft and Rockets 2007, 44(6), 1250–1262.
- Sarath, P. S.; et al. State-of-the-art on advancements in carbon–phenolic and carbon–elastomeric ablatives. Polymers 2024, 16, 1461.
- Adem, M. A.; et al. The effect of nanocarbon inclusion on mechanical, tribological, and thermal properties of phenolic resin-based composites: An overview. Engineering Reports 2024, 6, e12861.