Pitch-based carbon fibers, spun from petroleum or coal-tar pitch precursors, are high-performance reinforcement materials prized for their exceptionally high elastic modulus and thermal conductivity. Accurate determination of tensile strength is critical for material qualification, batch quality control, and structural design in aerospace, thermal management, and carbon–carbon composite applications. This article explains the specimen types, measured parameters, standardized test methods, and data evaluation procedures used to measure the tensile strength of pitch-based carbon fibers.
Technical Overview
Pitch-based carbon fibers are produced by melt-spinning isotropic or mesophase pitch, followed by oxidative stabilization, carbonization, and — for high-performance grades — graphitization at temperatures up to approximately 3000 °C. Two commercial families exist: isotropic pitch-based fibers (general-purpose grades) and mesophase pitch-based fibers (ultra-high-modulus, highly graphitic grades). Compared with PAN-based carbon fibers, pitch-based fibers deliver superior stiffness and thermal conductivity but are more flaw-sensitive, which makes standardized, reproducible tensile strength determination essential for both suppliers and users.
Carbon fibers are linear-elastic, brittle materials whose strength is governed by the statistical distribution of surface and internal flaws. Consequently, the measured tensile strength depends on the gauge length, filament diameter, and the uniformity of load transfer across the thousands of filaments in a tow. Two internationally recognized approaches address this: testing of resin-impregnated strands (tows) and testing of individual filaments. Typical property ranges for orientation are summarized below:
- Isotropic pitch-based fibers: tensile strength approximately 0.5–1.2 GPa, tensile modulus approximately 30–60 GPa.
- Mesophase pitch-based fibers: tensile strength typically 2.0–3.8 GPa, tensile modulus from roughly 350 GPa up to 900+ GPa for ultra-high-modulus grades.
- Failure strain is generally low, on the order of 0.4–1.5 %, reflecting the brittle, flaw-dominated failure mechanism.
Testing Samples
Single Filament Specimens
Individual filaments (typical diameters of 7–15 μm for pitch-based fibers) are carefully extracted from the tow and bonded with fast-curing epoxy onto slotted paper or cardboard tab frames that define a gauge length of approximately 20–25 mm, following ISO 11566 or ASTM C1557. Because calculated strength scales with the square of the diameter, each filament's diameter must be measured individually by laser diffraction, calibrated optical microscopy, or scanning electron microscopy. A large number of replicate specimens — typically 30 or more — is required because single-filament data exhibit substantial scatter.
Resin-Impregnated Tow (Strand) Specimens
Continuous tows (commonly 1K–12K filament counts) are impregnated with a low-viscosity, fiber-compatible epoxy resin, wound onto specimen frames, and cured into rigid, straight composite strands approximately 0.5–1.5 mm in diameter. The cured strands are cut to total lengths of about 250–350 mm and equipped with reinforced tabs, leaving a gauge length of 150–200 mm as specified in ASTM D4018, ISO 10618, JIS R 7608, or GB/T 3362. Resin impregnation is essential: it bonds the individual filaments into a unified load-bearing body, ensures uniform stress distribution, and protects filaments from grip-induced damage.
Specimen Conditioning and Quantity
- Condition specimens at 23 ± 2 °C and 50 ± 5 % relative humidity in accordance with ISO 291 prior to testing.
- Test a minimum of 5–10 valid strand specimens per condition; use 30 or more filaments when Weibull distribution parameters are required.
- Reject and replace specimens that fail within or immediately adjacent to the grips, as such failures do not represent the true material strength.
Testing Items
A complete tensile characterization of pitch-based carbon fiber typically includes the following reported quantities:
- Tensile strength (MPa or GPa), calculated as maximum force divided by fiber cross-sectional area.
- Tensile modulus (Young's modulus), determined from the linear portion of the stress–strain curve.
- Elongation at break (failure strain).
- Maximum tensile force (breaking load) of the impregnated strand.
- Auxiliary inputs: linear density (tex) per ISO 10120 and fiber density per ISO 10119, used to derive the nominal cross-sectional area.
- Statistical parameters: mean value, standard deviation, coefficient of variation, and — for filament data — Weibull characteristic strength and modulus.
Testing Methods
Resin-Impregnated Strand Method (ASTM D4018 / ISO 10618 / GB/T 3362 / JIS R 7608)
This is the primary industrial method for determining tow tensile strength. The workflow proceeds as follows:
- Determine the linear density (tex) and the fiber density, then calculate the nominal fiber cross-sectional area as A (mm²) = tex / (1000 × ρ), where ρ is the density in g/cm³.
- Prepare, cure, and tab the impregnated strand specimens; mount them in wedge or hydraulic grips with careful coaxial alignment.
- Load in tension at a constant crosshead speed (typically 2–10 mm/min) until failure, recording force and strain via a contact or non-contact extensometer.
- Calculate tensile strength σ = F_max / A; determine chord modulus between two defined strain points within the linear elastic region (for example, between about 0.1 % and 0.6 % strain, as specified by the governing standard) and report failure strain at maximum force.
Single Filament Method (ISO 11566 / ASTM C1557)
Single-filament testing measures the intrinsic strength of the individual filament. The mounted paper frame is clamped in the grips, the frame shoulders are cut so that only the filament is loaded, and the specimen is pulled to failure at a controlled displacement rate. The cross-sectional area is derived from the individually measured filament diameter, assuming a circular cross-section. This method provides fundamental flaw-statistics data but involves higher scatter and greater sensitivity to diameter measurement error than strand testing.
Equipment and Measurement Traceability
- A universal testing machine verified in accordance with ISO 7500-1 or ASTM E4, fitted with a low-capacity, high-resolution load cell appropriate for fiber-level forces.
- Strain measurement by clip-on, video, or laser extensometer, with accuracy meeting ISO 9513 or ASTM E83 requirements.
- Laser micrometry, optical microscopy, or SEM for filament diameter determination; a calibrated balance for linear density; density-gradient or Archimedes methods for fiber density.
Data Reduction and Statistical Evaluation
Because brittle fiber strength follows a weakest-link statistical model, results are conventionally evaluated with the two-parameter Weibull distribution, yielding a characteristic strength and a Weibull modulus (shape parameter) typically between about 4 and 10 for pitch-based fibers. The Weibull modulus also quantifies the well-known gauge length effect — shorter gauge lengths produce higher measured strengths — which must be considered when comparing data generated under different standards. A typical coefficient of variation is 5–15 % for impregnated strand tests and 15–30 % for single-filament tests. All raw values, the mean, the standard deviation, and the number of valid failures should be reported together with the test standard, gauge length, and conditioning state.
Applications
Reliable tensile strength data for pitch-based carbon fibers underpin material selection and product qualification across many advanced industries:
- Aerospace and satellite structures: optical benches, antenna reflectors, and truss members that exploit the near-zero thermal expansion and extreme stiffness of mesophase pitch-based fiber composites.
- Thermal management: high-thermal-conductivity heat spreaders, thermal straps, and electronic packaging materials based on highly graphitic pitch fibers.
- Carbon–carbon composites: aircraft brake discs, clutch components, rocket nozzle throats, and re-entry heat-shield structures.
- Friction and industrial materials: chopped isotropic pitch fibers for brake pads, gaskets, insulation felts, and conductive fillers.
- Sporting goods: high-modulus golf shafts, fishing rods, and bicycle components.
- Research and quality control: precursor and process development, graphitization optimization, incoming material inspection, and batch release testing performed by manufacturers or accredited third-party laboratories.
FAQ
Why must the tow be resin-impregnated before tensile testing?
Without impregnation, the load is transferred unevenly across thousands of individual filaments, causing progressive, non-simultaneous failure and severely underestimating the true tow strength. The cured resin bonds the filaments into a unified composite strand, ensures equal strain distribution, and prevents premature fiber damage at the grips, which is why ASTM D4018, ISO 10618, GB/T 3362, and JIS R 7608 all mandate impregnated specimens.
Why does the measured tensile strength depend on gauge length?
Carbon fibers fail from the most severe flaw within the stressed volume. A longer gauge length contains a larger population of potential flaws, increasing the probability of encountering a critical defect and lowering the measured strength. This size effect is quantified by the Weibull modulus, so gauge length must always be stated when comparing results.
Should I choose the single-filament method or the impregnated strand method?
Strand testing is preferred for routine quality control, datasheet values, and composite design inputs because it is faster, less scattered, and reflects collective tow behavior. Single-filament testing is preferred for fundamental research, flaw characterization, Weibull statistics, and cases where resin interaction must be excluded.
How is the fiber cross-sectional area determined without cutting and imaging every specimen?
For strand tests, the nominal area is calculated from the measured linear density (tex) and the fiber density using the relationship A = tex / (1000 × ρ). For single filaments, the area is computed from the individually measured diameter, which must be determined accurately because strength scales with the square of the diameter.
What level of scatter is normal in the results?
Coefficients of variation of 5–15 % are typical for resin-impregnated strand tests, while single-filament results commonly show 15–30 %. Larger scatter usually indicates problems with impregnation quality, filament alignment, grip damage, diameter measurement error, or insufficient specimen counts.
Which standards govern this testing, and how should a testing provider be selected?
The principal standards are ASTM D4018, ISO 10618, ISO 11566, ASTM C1557, GB/T 3362, and JIS R 7608, supported by accessory standards for linear density (ISO 10120), density (ISO 10119), machine verification (ISO 7500-1 / ASTM E4), and conditioning (ISO 291). Users should select accredited third-party laboratories with ISO/IEC 17025-accredited mechanical testing scopes, calibrated force and strain instrumentation, and documented experience with brittle high-modulus fibers.