PET Ceramic Substrate Coefficient of Thermal Expansion Testing Standards: Methods, Parameters, and I

The coefficient of thermal expansion (CTE) is one of the most critical reliability parameters for PET ceramic substrates used in electronic packaging, power modules, and thermal management applications. This article presents a comprehensive review of internationally recognized CTE testing standards, including ASTM E228, ASTM E831, and ISO 11359-2, together with sample preparation requirements, measurement procedures, typical acceptance values, and common failure modes caused by thermal expansion mismatch. Engineers, quality managers, and materials scientists can use this guide to select the appropriate test method and interpret CTE data correctly for design and qualification purposes.

Technical Overview

The coefficient of thermal expansion (CTE) describes the dimensional change of a material per unit length per degree of temperature change, typically expressed in ppm/°C (10⁻⁶/K). For PET ceramic substrates, which combine a ceramic body or ceramic-filled composite structure with PET (polyethylene terephthalate) polymer layers, CTE control is essential because the material must remain dimensionally compatible with silicon chips, copper traces, solder joints, and encapsulants across the entire operating temperature range.

Thermal expansion mismatch is a leading cause of failure in electronic assemblies. When the CTE of the substrate differs significantly from that of the mounted components, repeated thermal cycling generates shear stresses at the interfaces, resulting in solder joint cracking, trace delamination, ceramic fracture, and warpage. Pure PET film exhibits a CTE of approximately 50 to 70 ppm/°C below its glass transition region, while common engineering ceramics fall between 3 and 8 ppm/°C. Ceramic-filled PET composite substrates are therefore engineered to bring the effective CTE into a window compatible with copper (approximately 17 ppm/°C) or silicon (approximately 2.6 ppm/°C), and accurate measurement is required to verify that the target value has been achieved.

CTE testing of PET ceramic substrates is governed by several internationally recognized standards, including ASTM E228 for push-rod dilatometry, ASTM E831 and ISO 11359-2 for thermomechanical analysis (TMA), IPC-TM-650 for interconnect substrate materials, and JIS R 1618 for fine ceramics. Selecting the correct standard depends on the sample geometry, the temperature range of interest, the required precision, and whether the PET constituent introduces polymer-specific transitions such as glass transition that must be reported alongside the CTE value.

Testing Samples

Bare Ceramic Substrates

Monolithic ceramic substrates such as alumina (Al₂O₃), aluminum nitride (AlN), and silicon nitride (Si₃N₄) plates are tested in the as-fired or polished condition. For push-rod dilatometry per ASTM E228, the recommended specimen is a rectangular bar with a length between 20 and 50 mm and a cross section of 3 to 10 mm. Both end faces must be flat, parallel, and perpendicular to the measurement axis to ensure uniform load transfer. Surface grinding with fine abrasive followed by cleaning in solvent and drying is typically required before testing.

PET-Film Laminated and Ceramic-Filled Composite Substrates

Substrates containing PET layers, including PET film laminated onto ceramic carriers and ceramic-filled PET composite sheets, require additional attention to specimen preparation. Because PET is a viscoelastic polymer, specimens must be conditioned at 23 ± 2 °C and 50 ± 5% relative humidity for at least 24 hours before measurement to stabilize the moisture content. The measurement direction must be documented, since rolled PET films and tape-cast ceramic sheets often exhibit anisotropic expansion between the machine direction (MD) and transverse direction (TD). Specimens containing copper cladding should either be etched to remove the metal or tested in the as-supplied configuration if the composite expansion is the property of interest.

Sample Quantity and Identification

A minimum of three specimens per material condition is recommended to assess statistical scatter. Each specimen must be uniquely marked on a non-critical surface, and the initial length must be measured at room temperature with a calibrated micrometer or optical measuring system to an uncertainty of no greater than ±2 µm, because this length is the denominator of the CTE calculation.

Testing Items

A complete CTE evaluation of PET ceramic substrates typically includes the following measurable parameters:

• Mean coefficient of linear thermal expansion between specified temperatures, for example 25 to 105 °C or 30 to 150 °C.
• Instantaneous (differential) CTE as a function of temperature across the full test range.
• Glass transition temperature of the PET constituent, detected as an inflection point in the expansion curve.
• Dimensional change ratio, expressed as total linear expansion percentage from the starting temperature to the maximum test temperature.
• Anisotropy ratio between machine direction and transverse direction for film-based and tape-cast substrates.
• Residual dimensional change after thermal cycling, indicating permanent shrinkage or warpage.

Testing Methods

Push-Rod Dilatometry (ASTM E228)

Push-rod dilatometry is the reference method for rigid ceramic specimens requiring the highest accuracy. The specimen is placed in a fused silica or alumina sample holder with a spring-loaded push rod transmitting length changes to a linear variable differential transformer (LVDT). The furnace heats the assembly at a controlled rate of 3 to 5 °C/min under a nitrogen or argon purge of approximately 50 mL/min. The system must be calibrated with a certified reference material such as NIST standard reference material 732 or a sapphire standard over the same temperature range used for the specimen. ASTM E228 reports repeatability within approximately ±2% and reproducibility within approximately ±5% for CTE values above 5 ppm/°C.

Thermomechanical Analysis (ASTM E831 / ISO 11359-2)

TMA uses a small specimen, typically 2 to 10 mm in thickness, resting on a quartz platform with a floating or low-force expansion probe. The method is preferred for PET-containing substrates because it requires small specimens, applies minimal force that will not deform the polymer layer, and simultaneously yields the glass transition temperature of PET from the same scan. A heating rate of 5 °C/min under nitrogen is standard, and the mean CTE is calculated from the slope of the length-versus-temperature segment between two user-defined temperatures.

Test Conditions and Temperature Range Selection

For PET ceramic substrates, the maximum test temperature must remain safely below the PET melting point of approximately 250 °C and below the onset of polymer degradation; a practical upper limit of 200 to 220 °C is commonly specified. A typical test program is:

• Temperature range: room temperature to 200 °C for PET-containing samples; room temperature to 400 °C or higher for bare ceramic bodies.
• Heating rate: 5 °C/min, constant.
• Atmosphere: high-purity nitrogen at 40 to 60 mL/min to prevent oxidative degradation of the PET layer.
• Pre-test hold: 5 minutes at the starting temperature to establish thermal equilibrium.
• Optional second heating cycle to remove the effect of residual stress and moisture from the first scan.

Calculation and Reporting

The mean coefficient of linear thermal expansion between temperatures T₁ and T₂ is calculated as the change in specimen length divided by the initial length L₀ and by the temperature interval: α = (L₂ − L₁) / [L₀ × (T₂ − T₁)]. The test report must state the applicable standard, specimen geometry and orientation, temperature interval, heating rate, atmosphere, calibration reference, and the mean CTE with measurement uncertainty. Any inflection attributed to the glass transition of PET must be reported separately so that CTE values are not quoted across the transition region without qualification.

Applications

CTE testing of PET ceramic substrates supports qualification and process control across a broad range of industries:

• Power electronics modules, where substrate CTE must be matched to silicon carbide and silicon dies to survive thousands of power cycles.
• LED packaging and ceramic LED substrates, where CTE compatibility with phosphor layers and encapsulants prevents delamination and light output degradation.
• Flexible and rigid-flex circuits using ceramic-filled PET laminates, where dimensional stability during reflow soldering is critical.
• Thick-film and thin-film circuit substrates, where conductor and dielectric pastes are selected based on the measured substrate CTE.
• Automotive electronics, including engine control units and onboard chargers, governed by AEC-Q100 and AEC-Q200 qualification requirements that reference CTE stability.
• 5G and RF modules, where low-loss ceramic substrates require predictable expansion behavior to maintain impedance tolerances.
• Supplier incoming inspection and comparative material selection studies during new product introduction.

FAQ

What is the typical CTE range for PET ceramic substrates? Ceramic-filled PET composite substrates generally exhibit a CTE between 10 and 30 ppm/°C depending on filler loading, while bare engineering ceramics range from approximately 2.8 ppm/°C for silicon nitride to 7.5 ppm/°C for 96% alumina. The exact value must be confirmed by testing because composition, orientation, and processing history all influence the result.

Should I choose ASTM E228 or ASTM E831 for my samples? Use ASTM E228 dilatometry for rigid, dimensionally stable ceramic bodies where maximum accuracy is required and sufficient specimen length is available. Use ASTM E831 or ISO 11359-2 TMA when specimens are small, thin, or contain PET polymer layers whose glass transition must also be characterized.

Why does the first heating scan sometimes differ from the second? The first scan may include residual stress relaxation, moisture loss, and physical aging effects from the PET constituent. Running a second heating cycle after controlled cooling isolates the intrinsic reversible expansion behavior and is recommended for lot-release data.

What sample size and quantity are required? Dilatometry requires bars of approximately 25 mm in length, while TMA accepts specimens from 2 to 10 mm thick. A minimum of three specimens per condition is standard practice to evaluate variability.

How is CTE mismatch risk assessed? Design engineers compare the substrate CTE with the CTE of the die, copper metallization, and solder using finite element thermal cycling simulation. As a rule of thumb, solder joint fatigue life improves significantly when the substrate-to-silicon CTE difference is kept below a few ppm/°C for direct die attach configurations.

What is a typical turnaround time for CTE testing? Including specimen preparation, conditioning, calibration verification, and a dual heating cycle, most accredited laboratories report results within three to seven working days after sample receipt.