The coefficient of thermal expansion (CTE) of PET PCB boards is a critical parameter that determines dimensional stability, solder joint reliability, and long-term performance of flexible circuits. This technical article explains the complete CTE measurement experiment for PET-based printed circuit boards, covering sample preparation, thermomechanical analysis (TMA) procedures, applicable IPC and ASTM standards, and interpretation of test results. Engineers involved in flexible PCB design, materials qualification, and quality control will find practical guidance for accurate and repeatable CTE characterization.
Technical Overview
Polyethylene terephthalate (PET) is one of the most widely used substrate materials for flexible printed circuit boards, offering an excellent balance of cost, dielectric performance, chemical resistance, and mechanical flexibility. However, like all polymeric materials, PET exhibits significant dimensional change with temperature. The coefficient of thermal expansion (CTE), typically expressed in parts per million per degree Celsius (ppm/°C), quantifies this dimensional change and directly affects the reliability of the finished circuit board.
For PET PCB boards, CTE measurement is particularly important because mismatched expansion between the PET base film, copper conductor layers, adhesives, and solder joints generates internal stress during thermal cycling. Excessive CTE mismatch can cause conductor cracking, delamination, copper trace fatigue, and solder joint failure. The measurement experiment described in this article uses thermomechanical analysis (TMA) to determine the linear CTE of PET PCB materials in both the in-plane (X/Y) and through-thickness (Z-axis) directions, along with the glass transition temperature (Tg) that marks the boundary between the two expansion regimes.
PET films typically exhibit an in-plane CTE of approximately 15 to 25 ppm/°C below the glass transition temperature, which occurs around 75 to 80°C for amorphous or partially crystalline grades. Above Tg, the CTE increases significantly, often exceeding 60 ppm/°C. In the Z-axis direction, the expansion can be substantially higher. When PET is laminated with copper foil and adhesive to form a finished PCB board, the effective CTE changes due to the constraint imposed by the copper layers, which is why measurement on the actual board construction is essential rather than relying on raw film data.
Testing Samples
Bare PET Base Film
Raw PET film samples with thicknesses of 25 µm, 50 µm, 75 µm, or 125 µm are tested to establish baseline CTE values. Thin films are often stacked in multiple layers to reach the minimum measurable thickness for stable TMA probe contact, typically requiring a total stack thickness of at least 0.5 mm.
Copper-Clad PET Laminate
Samples of PET film laminated with electrolytic or rolled-annealed copper foil (commonly 12 µm, 18 µm, or 35 µm copper) and acrylic or polyester adhesive represent the actual board construction. These samples may be tested with copper intact to determine the effective composite CTE, or with copper etched away to isolate the adhesive system contribution.
Finished PET Flexible PCB Boards
Production-representative circuit boards with patterned copper traces, coverlay, and any stiffeners are tested to verify real-world dimensional behavior. Samples should be cut from areas free of traces, vias, and dense circuitry to ensure uniform material under the probe.
Sample Preparation Requirements
- Cut samples into squares or discs of 6 to 10 mm, with surfaces flat and parallel within close tolerance to ensure uniform load distribution.
- Condition samples prior to testing, typically 24 hours at 23 ± 2°C and 50 ± 5% relative humidity, or dry to constant mass at low temperature when moisture effects must be excluded.
- Measure and record initial dimensions (length, width, thickness) using a calibrated micrometer or optical measuring system with accuracy better than 2 µm.
- Remove burrs, debris, and surface irregularities that could cause probe slippage or artificial displacement readings.
- Mark the orientation of samples relative to the machine direction (MD) and transverse direction (TD) of the PET film, since CTE is anisotropic.
Testing Items
The CTE measurement experiment for PET PCB boards encompasses the following quantifiable parameters:
- Alpha 1 (α1) – In-plane or Z-axis CTE below Tg: the linear expansion coefficient measured in the temperature region below the glass transition, typically from 30°C to just below Tg onset.
- Alpha 2 (α2) – CTE above Tg: the expansion coefficient in the rubbery or softened region above the glass transition.
- Glass transition temperature (Tg): determined from the inflection point of the dimension-versus-temperature curve, marking the transition between α1 and α2 regimes.
- X/Y-axis CTE: expansion behavior in the plane of the board, critical for dimensional stability, registration accuracy, and trace-to-substrate stress.
- Z-axis CTE: through-thickness expansion, critical for plated through-hole reliability and via integrity, usually significantly higher than in-plane values.
- Total dimensional change: cumulative expansion or shrinkage over the full test temperature range, expressed as a percentage of original dimension.
- Anisotropy ratio: the ratio of TD to MD expansion, indicating directional bias introduced by film manufacturing.
Testing Methods
Thermomechanical Analysis (TMA) – Primary Method
TMA is the industry-standard technique for PET PCB CTE measurement, referenced by IPC-TM-650 method 2.4.24 (CTE and Tg by TMA) and ASTM E831. The procedure is as follows:
- Calibrate the TMA instrument for temperature using indium and zinc melting point standards, and for length displacement using a certified calibration standard.
- Place the sample on the flat quartz platform and apply the expansion probe with a static force of 5 to 10 mN, low enough to avoid penetrating the PET surface but sufficient to maintain contact.
- Purge the furnace with nitrogen at approximately 45 to 55 mL/min to prevent oxidative degradation during heating.
- Apply a preload temperature hold at 23°C for 5 minutes to establish thermal equilibrium and a stable baseline.
- Heat the sample at a controlled rate of 5°C/min (per IPC-TM-650) from ambient temperature to 150 to 180°C, staying below the PET melting range of approximately 250°C to avoid irreversible damage.
- Record the dimension-versus-temperature curve continuously, then compute CTE as the slope of dimensional change divided by initial dimension: CTE = (ΔL / L₀) / ΔT.
- For finished boards exhibiting stress relaxation, perform a second heating run and report values from the second scan where the first-run artifacts such as cure shrinkage or moisture-driven transitions have been eliminated.
Dilatometry Method
For larger samples or verification purposes, a horizontal or vertical push-rod dilatometer per ASTM D696 or ISO 11359-2 can be used. This method employs a fused silica dilatometer with the specimen heated in a controlled bath or furnace, and it is particularly suitable for in-plane measurements on larger coupons cut directly from PET circuit boards.
Strain Gauge and Optical Methods
Foil strain gauges bonded to PET board surfaces allow in-situ thermal expansion measurement during thermal cycling, while non-contact optical techniques such as digital image correlation (DIC) enable full-field mapping of in-plane expansion across patterned circuit areas, revealing local CTE variations near copper features.
Measurement Direction Considerations
In-plane CTE requires either a TMA configured for tensile or lateral measurement of thin strips, or calculation from dilatometer data on oriented coupons. Z-axis CTE uses the standard compression configuration on stacked discs. Because PET film is biaxially oriented during manufacture, in-plane values must be reported separately for machine direction and transverse direction.
Quality Control of Test Results
- Verify baseline flatness: the pre-transition region of the curve must be linear with a correlation coefficient of at least 0.999 for a valid regression slope.
- Confirm repeatability: test a minimum of three specimens per condition and report mean values with standard deviation.
- Document thermal history: note whether values derive from first or second heating scans, as this materially affects reported numbers.
- Cross-check Tg values from TMA against DSC results when discrepancies exceed 5°C.
Applications
Accurate CTE data from PET PCB measurement experiments supports numerous engineering and manufacturing activities:
- Material selection and qualification: comparing PET against alternative flexible substrates such as polyimide, PEN, and LCP during new product development.
- Solder joint reliability prediction: CTE mismatch inputs for finite element simulation and Coffin-Manson fatigue life modeling of surface-mount assemblies on PET boards.
- Dimensional stability control: predicting layer-to-layer registration error and pattern shift during lamination, etching, and multilayer registration processes.
- Thermal cycling qualification: establishing acceptance criteria for reliability testing such as thermal shock between -40°C and 85°C common in automotive and consumer electronics specifications.
- Through-hole and via integrity: using Z-axis CTE to assess the risk of copper barrel cracking in mechanically drilled or laser-formed interconnects.
- Process window definition: determining safe reflow and lamination temperatures that remain below the region of rapid PET expansion above Tg.
- Supplier incoming inspection: verifying that delivered copper-clad PET laminate lots conform to contractual CTE specifications.
FAQ
What is the typical CTE value of a PET PCB board?
Bare PET film generally shows an in-plane CTE of 15 to 25 ppm/°C below Tg and 60 to 120 ppm/°C above Tg. Z-axis expansion of laminated constructions commonly ranges from 50 to 150 ppm/°C depending on adhesive content. Copper-clad boards exhibit effective in-plane CTE values between those of copper (~17 ppm/°C) and the polymer constituents, typically 20 to 40 ppm/°C.
Why is TMA preferred over other CTE measurement techniques for PET PCBs?
TMA offers microlevel displacement sensitivity, small specimen requirements, direct simultaneous determination of both CTE and Tg from a single run, and full alignment with IPC-TM-650 and ASTM E831 standards recognized across the electronics industry.
Does the copper layer affect the measured CTE?
Yes. Copper constrains in-plane expansion of the composite, pulling the effective CTE closer to that of copper. For this reason, tests should be performed on the exact board stackup, and results must specify whether copper was present, etched away, or present only on one side.
What temperature range should be used for PET CTE testing?
A practical range is 23°C to 150°C for the first scan and 23°C to 120 to 150°C for the second scan. This captures the α1 region, the Tg inflection near 75 to 80°C, and the α2 region while remaining safely below the PET melt, preventing sample destruction.
How thick should a PET sample be for reliable TMA measurement?
A minimum stack thickness of approximately 0.5 mm is recommended for Z-axis measurements. Thin PET films below this threshold are stacked in multiple aligned layers; the resulting curve represents through-thickness behavior of the film laminate.
What CTE values are considered acceptable for PET flexible PCBs?
Acceptance limits depend on the application, but typical procurement specifications call for in-plane CTE below 30 ppm/°C (α1) with Tg above 70°C for standard flexible circuit use. Automotive and high-reliability applications impose tighter limits and often mandate polyimide substrates with lower expansion characteristics.
How does moisture influence PET CTE test results?
Absorbed moisture plasticizes PET, shifts the apparent Tg, and can introduce non-linear dimensional artifacts during the first heating scan. Drying specimens at 60°C under vacuum or dry nitrogen before testing, or reporting second-scan values, mitigates this effect and improves reproducibility.