PEEK (polyetheretherketone) is a high-performance semi-crystalline thermoplastic widely used in aerospace, semiconductor, and energy applications where dimensional stability at elevated temperatures is critical. Evaluating the high-temperature expansion behavior of PEEK—typically through coefficient of thermal expansion (CTE) measurement—provides essential data for design tolerances, material selection, and failure prevention. This article presents a comprehensive overview of PEEK high-temperature expansion performance evaluation, covering sample preparation, testing items, measurement methods, industry applications, and frequently asked questions.
Technical Overview
PEEK (polyetheretherketone) is a semi-crystalline engineering thermoplastic known for its exceptional combination of mechanical strength, chemical resistance, and thermal stability, with a glass transition temperature (Tg) of approximately 143°C and a melting point of approximately 343°C. Like all polymers, however, PEEK undergoes dimensional changes when heated, and these changes are not linear across the entire service temperature range. Accurate evaluation of PEEK's thermal expansion behavior is therefore essential for engineers designing precision components that must maintain tight tolerances under thermal cycling.
The key parameter in this evaluation is the coefficient of thermal expansion (CTE), commonly expressed in units of 10⁻⁶/K (ppm/K). Unfilled PEEK typically exhibits a linear CTE of approximately 45–55 × 10⁻⁶/K below its glass transition temperature, but this value can increase to 120–160 × 10⁻⁶/K above Tg, where the amorphous phase gains molecular mobility. Reinforced grades behave quite differently: carbon fiber reinforced PEEK (30% by weight) can display a CTE as low as 10–25 × 10⁻⁶/K, approaching that of metals such as aluminum, which makes filled PEEK particularly valuable in metal-replacement applications where thermal mismatch must be minimized.
High-temperature expansion evaluation of PEEK serves several engineering purposes: verifying that mating dimensions remain compatible during thermal excursions, confirming the uniformity and crystallinity of molded parts, detecting the glass transition and secondary relaxations, and qualifying materials against procurement specifications. Because PEEK parts frequently operate between room temperature and 260°C (and in some downhole applications up to 300°C), test programs must characterize expansion across the full anticipated range, including the Tg transition region where the CTE changes abruptly.
Testing Samples
Sample selection and preparation directly influence the reliability of expansion measurements. Because PEEK is anisotropic when reinforced or processed with flow-induced orientation, specimens must represent the actual part geometry and processing condition whenever possible. The following sample categories are commonly evaluated.
Unfilled Neat PEEK Resin
Specimens are prepared from unfilled PEEK granules by compression molding or injection molding into plaques or rods, from which rectangular prism specimens (typically 5–10 mm in length, 3–5 mm cross-section) are machined. Proper drying (3 hours at 150°C or per supplier recommendation) before molding and before testing is critical, because residual moisture can cause hygroscopic swelling that interferes with thermal expansion data. Molded samples should be examined for voids, sink marks, and uniform crystallinity, since crystallinity levels (typically 30–35% for standard processing) measurably affect the CTE.
Glass Fiber Reinforced PEEK
Grades containing 30% glass fiber are widely used for structural parts. Because glass fibers orient along the flow direction during injection molding, expansion is markedly anisotropic: the CTE in the flow direction may be 30–40% lower than in the transverse direction. Specimens are therefore cut in both the flow and cross-flow directions, and the orientation relative to the original part must be documented in the test report.
Carbon Fiber Reinforced PEEK
Carbon fiber reinforced PEEK (30% CF) offers the lowest CTE of the common PEEK grades and is a preferred choice for semiconductor handling and aerospace components requiring metal-like dimensional behavior. Test specimens follow the same preparation requirements as glass-filled grades, with additional attention paid to fiber orientation documentation. In extreme cases, highly oriented carbon fiber PEEK can exhibit near-zero or even slightly negative expansion along the fiber axis, which must be quantified for precision designs.
Bearing-Grade and Filled PEEK Compounds
Compounds containing PTFE, graphite, and carbon powder (typically 10% of each) are used for tribological components such as bushings and wear rings. The particulate fillers reduce but do not eliminate the increase in CTE above Tg, and the evaluation of these grades verifies whether the expansion remains within the design envelope for clearance-critical bearing assemblies.
PEEK Films and Additively Manufactured Parts
Thin PEEK films used in flexible electronics and insulation are tested with adapted fixtures to prevent buckling under the TMA probe load. For 3D-printed (FDM or SLS) PEEK parts, the layer-by-layer construction produces pronounced anisotropy and variable porosity, so specimens are prepared in both the XY plane and Z (build) direction to capture the full expansion behavior.
Testing Items
A complete high-temperature expansion evaluation of PEEK typically includes the following measurement items:
- Mean (average) coefficient of linear thermal expansion over specified temperature intervals, such as 23–100°C, 23–150°C, 23–200°C, and 23–260°C.
- Instantaneous (differential) CTE as a continuous function of temperature, obtained from the slope of the dimension-versus-temperature curve.
- Glass transition temperature determined from the inflection point in the expansion curve, providing a cross-check against DSC results.
- CTE values below and above Tg, reported separately because of the large discontinuity at the transition.
- Anisotropy ratios for reinforced grades, comparing flow-direction and transverse-direction expansion.
- Dimensional stability across repeated thermal cycles, revealing irreversible shrinkage from post-crystallization or residual stress relaxation.
- Volumetric expansion estimation for isotropic grades, calculated from linear measurements in multiple axes where required.
- Qualification against specification limits, such as maximum permissible CTE in customer drawings or material datasheets.
Testing Methods
The two principal techniques for evaluating PEEK expansion behavior at high temperature are thermomechanical analysis (TMA) and push-rod dilatometry. Both measure the change in specimen length as a function of temperature under controlled heating, but they differ in sensitivity, temperature capability, and applicable standards.
Thermomechanical Analysis (TMA)
TMA is the most widely used method and is governed by ASTM E831 and ISO 11359-2. A specimen of 2–10 mm length is placed on a quartz stage, and a low-force expansion probe (typically 0.05 N or less) rests on its upper surface. The instrument is heated at a controlled rate—commonly 2–5°C/min—under nitrogen purge, while a linear variable displacement transducer records dimensional change with sub-micrometer resolution. Temperature calibration is performed using indium (156.6°C) and zinc (419.5°C) melting standards, and length calibration is verified with a certified quartz or platinum reference. Typical test ranges for PEEK extend from room temperature or −30°C up to 250–300°C, remaining below the melting point to avoid gross specimen deformation.
Push-Rod Dilatometry
A horizontal or vertical push-rod dilatometer (ASTM E228) accommodates larger specimens, usually 25–50 mm in length, and offers excellent accuracy for mean CTE over broad temperature intervals. The longer gauge length improves the relative precision of the expansion measurement, making dilatometry a strong choice for production lot acceptance testing and for calibrating reference values used to validate TMA results.
Test Procedure and Data Treatment
The general procedure comprises: (1) conditioning specimens in a desiccator after drying to eliminate moisture effects; (2) measuring initial length at 23°C with a calibrated micrometer; (3) mounting the specimen with flat, parallel, polished end faces perpendicular to the measurement axis; (4) applying the probe force and running an initial conditioning cycle to remove residual stresses if required by the protocol; (5) heating at the specified rate under inert atmosphere while recording dimension and temperature continuously; and (6) computing the mean CTE using the formula alpha = (L2 − L1) / (L0 × (T2 − T1)), where L0 is the initial length. Instantaneous CTE is obtained by differentiating the length curve, and Tg is identified at the inflection point where the slope changes. Results are reported with temperature intervals, heating rate, specimen orientation, and the number of replicate specimens—normally a minimum of three.
Quality Control Considerations
Key sources of error include probe indentation into the specimen surface under excessive force, thermal gradients along the specimen, moisture-related swelling, and specimen end-face non-parallelism. Laboratory quality programs address these through certified reference materials, instrument-level verification runs, replicate testing, and documented deviation limits. For reinforced PEEK, results are only comparable when fiber orientation and processing history are matched between specimens.
Applications
High-temperature expansion data for PEEK directly supports engineering decisions across demanding industries:
- Aerospace: bushings, seals, and structural brackets experience thermal cycling from sub-zero cruise altitudes to engine-bay heat; CTE data ensures interference fits and clearances are maintained across the envelope.
- Semiconductor manufacturing: wafer carriers, CMP retaining rings, and test sockets made from carbon fiber reinforced PEEK must match or undershoot silicon and aluminum expansion to preserve alignment and prevent wafer damage.
- Oil and gas: downhole seals, bearing rings, and electrical connector housings operate continuously at 150–260°C under pressure; reliable above-Tg expansion data prevents seal bypass and assembly loosening.
- Automotive: transmission components, thrust washers, and under-hood sensors benefit from PEEK grades whose expansion is matched to steel and aluminum housings.
- Medical devices: sterilization-tray components and instrument handles undergo repeated autoclave cycles at 134°C, and long-term dimensional stability must be confirmed through cyclic expansion testing.
- Food and chemical processing: pump wear rings and valve seats require verified clearance behavior at elevated service temperatures in aggressive media.
- Metal replacement programs: CTE data allows direct substitution of aluminum or brass components, provided the expansion mismatch with adjacent metal structures is quantified.
FAQ
Why does the CTE of PEEK increase significantly above the glass transition temperature?
Below Tg (approximately 143°C), the amorphous fraction of PEEK is in a glassy state with restricted molecular motion, so expansion is modest. Above Tg, the amorphous regions gain segmental mobility and free volume increases more rapidly with temperature, roughly doubling to tripling the linear CTE. The crystalline fraction continues to expand slowly, which is why highly crystalline or fiber-filled PEEK shows a smaller step change at Tg than amorphous-dominant polymers.
Which standards govern PEEK expansion testing?
The most commonly applied standards are ASTM E831 and ISO 11359-2 for thermomechanical analysis of CTE, and ASTM E228 for push-rod dilatometry. ASTM D696 is occasionally used for the narrow −30°C to 30°C range. Procurement specifications may additionally invoke internal OEM standards that define temperature intervals, heating rates, and acceptance limits.
What temperature range should be used for testing?
The range must bracket the intended service environment. For general qualification, 23°C to 260°C is typical because 260°C represents the common continuous-use limit of PEEK. Test programs that need to capture the Tg transition should include at least 100°C to 200°C, and cryogenic or downhole applications may extend the range from −70°C to 300°C.
What is the difference between mean and instantaneous CTE?
The mean CTE is a single averaged value over a defined temperature interval and is convenient for tolerance stack-up calculations. The instantaneous CTE is the point-by-point slope of the expansion curve and reveals how expansion changes continuously with temperature, including the abrupt increase at Tg. Both values should be reported for a complete evaluation.
How do fillers such as carbon fiber reduce the expansion of PEEK?
Carbon and glass fibers possess CTE values that are far lower than the polymer matrix—in the case of carbon fiber, even slightly negative along the fiber axis. When embedded in PEEK, the fibers mechanically constrain matrix expansion in the direction of orientation, reducing the composite CTE substantially while introducing anisotropy that must be characterized in multiple directions.
Can repeated thermal cycling permanently change PEEK dimensions?
Yes. The first heating cycle may release residual molding stresses or allow limited post-crystallization, producing a small irreversible shrinkage. Subsequent cycles are generally reproducible. Standard practice is to run one conditioning cycle before recording the reported expansion data, or to explicitly document first-cycle behavior when as-molded stability is the property of interest.
How many specimens are required for a valid evaluation?
A minimum of three replicate specimens per material and orientation is standard, with five recommended for qualification or lot-release testing. Reported values should include the mean, standard deviation, and the specific temperature intervals, heating rate, and applicable standard used.
Does moisture affect PEEK expansion measurements?
PEEK absorbs only a small amount of moisture (typically below 0.5% at saturation), but even this level can introduce hygroscopic swelling that biases early-stage expansion data. Specimens are therefore dried and stored in a desiccator prior to testing so that the measured dimensional change reflects thermal expansion alone.