This article provides a comparative analysis of the coefficient of thermal expansion (CTE) of PET-based PCB substrates against polyimide (PI), PEN, LCP, and FR-4 alternatives, covering typical CTE values, material behavior, and reliability implications. It also details the testing samples, test items, measurement methods, and application guidance engineers need when selecting a substrate for flexible and rigid printed circuit designs.
Technical Overview
Polyethylene terephthalate (PET) film is one of the most cost-effective dielectric substrates used for flexible printed circuit boards (FPCs). It offers good electrical insulation, high transparency, excellent chemical resistance, and strong mechanical toughness at a fraction of the price of polyimide. However, its thermal behavior, particularly its coefficient of thermal expansion (CTE), differs significantly from that of other common PCB materials, and this difference directly influences dimensional stability, copper adhesion, and long-term circuit reliability.
The CTE quantifies the fractional linear dimensional change of a material per degree of temperature change, expressed in ppm/°C. For PCB substrates, engineers distinguish between the in-plane CTE (X and Y directions, along the film surface) and the through-thickness CTE (Z-axis). In flexible circuits, the in-plane CTE is the dominant concern because it must remain compatible with the copper foil, whose CTE is approximately 17 ppm/°C. A large mismatch generates residual stress during thermal cycling, leading to curling, delamination, cracked traces, lifted pads, and misregistration during multilayer lamination or assembly.
PET is a biaxially oriented, semi-crystalline polyester. Its glass transition temperature (Tg) is approximately 75–80°C, below which the film expands at a relatively modest rate; above Tg and up to its softening range, expansion accelerates markedly. Heat-stabilized PET grades, which are annealed under tension by the film manufacturer, exhibit substantially lower and more stable in-plane CTE values than standard grades. Representative comparative values are summarized below:
- Heat-stabilized PET film: approximately 10–15 ppm/°C in-plane, below Tg
- Standard PET film: approximately 15–30 ppm/°C in-plane, below Tg
- Polyimide (PI) film: approximately 20–32 ppm/°C in-plane
- PEN film: approximately 13–20 ppm/°C in-plane
- LCP film: approximately 5–17 ppm/°C in-plane
- FR-4 laminate: approximately 12–18 ppm/°C in-plane and 45–70 ppm/°C in the Z-axis
- Electrolytic and rolled copper foil: approximately 17 ppm/°C
- Acrylic and epoxy adhesive layers: approximately 50–200 ppm/°C
Two conclusions follow from this comparison. First, high-quality heat-stabilized PET can approach copper and FR-4 in in-plane expansion, making it dimensionally predictable for moderate-temperature applications. Second, because PET loses mechanical strength near its Tg and softens well before polyimide does, its usable thermal window is limited to roughly −40°C to +105°C continuous (short peaks around 120–150°C), which means conventional lead-free reflow soldering is generally outside its capability. Any CTE comparison must therefore be interpreted together with the maximum process and operating temperature of the end product.
Testing Samples
PET-Based Flexible PCB Samples
Single-sided and double-sided flexible circuits were prepared using 35 µm electrodeposited copper foil laminated to 50 µm, 75 µm, and 125 µm PET base films with a 15–25 µm acrylic adhesive. Both standard and heat-stabilized PET grades were included so that the effect of film annealing on expansion behavior could be isolated. A subset of etched coupons with the copper fully removed was retained to measure the CTE of the bare substrate, while patterned coupons with grid arrays and via fiducials were used to evaluate the composite expansion and registration behavior of the finished circuit.
Polyimide Flexible PCB Samples
Control samples of equivalent construction were produced on 25 µm and 50 µm polyimide film, using both adhesive-based and adhesiveless (sputtered and plated) copper constructions. These samples establish the benchmark for high-temperature flexible circuits, since PI sustains continuous operation near 200°C and withstands multiple solder-reflow excursions at 260°C.
FR-4 Rigid PCB Reference Samples
Standard 1.6 mm FR-4 test coupons with 1 oz copper were included as a rigid-board reference. These samples allow direct comparison between the in-plane stability of flexible films and a woven-glass epoxy laminate, and they highlight the large Z-axis expansion typical of resin-dominated rigid materials, which is a primary driver of plated through-hole reliability.
PEN and LCP Comparison Samples
To complete the material map, flexible coupons based on PEN film and liquid crystal polymer (LCP) film were also evaluated. PEN occupies an intermediate position between PET and PI in both cost and thermal capability, while LCP represents the low-CTE, low-moisture-absorption option favored for high-frequency and fine-pitch applications.
Testing Items
The comparative evaluation program covered the following measurable characteristics, all of which are directly or indirectly governed by thermal expansion behavior:
- In-plane CTE (X and Y directions), reported separately for the machine direction (MD) and transverse direction (TD) of the base film
- Two-segment CTE values below and above the glass transition temperature, together with the Tg itself
- Through-thickness (Z-axis) CTE of laminated stacks and rigid reference boards
- Dimensional stability after thermal exposure, expressed as irreversible length change in ppm
- Registration shift of etched features and drilled or punched vias after thermal cycling
- Warpage and curl of finished circuits after soldering processes or thermal shock
- Copper peel strength before and after thermal cycling, as an indicator of interface stress caused by CTE mismatch
- Visual and microsectional inspection for trace cracking, adhesive flow, and delamination
Testing Methods
Thermomechanical Analysis (TMA)
Primary CTE measurement was performed by TMA in accordance with ASTM E831 and IPC-TM-650 method 2.4.24. Flat-etched specimens were clamped with a low-force expansion probe and heated at 3–5°C/min over a programmed range, typically 30–150°C for PET systems and 30–260°C for PI and FR-4 systems. The slope of the dimension-versus-temperature curve yields the CTE; a two-segment tangent fit provides separate values below and above Tg. Free-standing film specimens, adhesive layers, and fully laminated composites were each characterized so that the contribution of each stack component could be modeled.
Dynamic Mechanical Analysis (DMA) and Tg Verification
DMA was performed per ASTM D4065 and IPC-TM-650 method 2.4.24.2 to determine the glass transition and storage modulus of the PET, PI, and adhesive materials. Because CTE changes abruptly at Tg, confirming the transition temperature is essential for defining the temperature window over which a single CTE figure is valid.
Thermal Cycling and Dimensional Stability Testing
Finished coupons were subjected to thermal cycling per IPC-TM-650 method 2.6.7, using profiles such as −40°C to +85°C for PET assemblies and −55°C to +125°C for PI and FR-4 assemblies, typically for 100 to 500 cycles. Dimensional measurements of fiducials and grid patterns were taken before and after cycling using a calibrated video measurement system with a resolution better than 5 µm. The irreversible dimensional change, in-plane growth, and any out-of-plane curl were recorded and correlated with the CTE data.
Optical and Strain-Based Methods
For full-field expansion mapping, digital image correlation and moiré interferometry were applied to representative coupons during controlled heating. These methods reveal local anisotropy between machine and transverse film directions, expansion gradients near heavy copper planes, and edge effects that a single TMA specimen cannot capture. Strain gauges bonded to selected locations provided supplementary verification of composite expansion behavior.
All specimens were conditioned at 23 ± 2°C and 50 ± 5% relative humidity for at least 24 hours before testing, and results were reported as averaged values from a minimum of five specimens per material and construction, including measurement uncertainty.
Applications
The CTE comparison supports clear material selection decisions across the flexible electronics landscape. PET-based PCBs are the preferred choice where cost, transparency, and moderate thermal demands dominate:
- RFID antenna inlays and smart labels requiring high-volume, low-cost production
- Membrane switches, touch overlays, and decorative backlighting panels
- E-paper and display module backplanes benefiting from PET transparency and flatness
- Disposable and wearable medical sensors, monitoring patches, and diagnostic strips
- Automotive interior lighting, seat occupancy sensors, and cabin electronics isolated from engine-bay heat
- LED lighting strips and advertising displays assembled with low-temperature processes or conductive adhesives
- Printer head cables, battery protection circuits, and consumer device interconnects with limited thermal exposure
Polyimide remains the correct selection when the circuit must survive reflow soldering, sustained high temperature, or severe thermal cycling. PEN is an effective compromise in automotive cabin and industrial environments where temperatures exceed the PET comfort zone but do not justify PI pricing. LCP serves high-frequency and ultra-fine-pitch applications where the lowest CTE and lowest moisture absorption are decisive. Design mitigations for PET circuits include specifying heat-stabilized film grades, using symmetric copper distribution to balance stress, selecting low-CTE adhesives or adhesiveless constructions, applying artwork compensation factors derived from measured MD/TD expansion, and enlarging annular rings and bend-relief geometries to tolerate the residual strain field.
FAQ
What is the typical CTE of a PET PCB substrate?
Standard PET film exhibits an in-plane CTE of approximately 15–30 ppm/°C below its glass transition, while heat-stabilized grades achieve 10–15 ppm/°C. Above the Tg of roughly 75–80°C, the expansion rate increases significantly, so PET circuits should always be evaluated against their actual operating temperature range.
How does the CTE of PET compare with polyimide?
Heat-stabilized PET can match or undercut the in-plane CTE of polyimide film, which typically falls between 20 and 32 ppm/°C. The decisive difference is thermal capability rather than room-temperature expansion: polyimide retains its dimensions and mechanical strength at temperatures far beyond the point where PET softens, which makes PI the benchmark for solder-reflow-capable flexible circuits.
Can PET PCBs withstand reflow soldering?
Conventional PET laminates are generally not suitable for standard lead-free reflow profiles peaking near 245–260°C. Assembly of PET circuits is normally performed with low-temperature soldering, hot-bar or pulse heating on local areas, or conductive adhesives. High-temperature PET grades extend short-term tolerance to roughly 150°C but still do not replace polyimide for full reflow exposure.
Why does the in-plane CTE matter more than the Z-axis CTE for PET flex circuits?
Flexible circuits are thin, single- or double-layer structures in which expansion parallel to the surface controls trace stress, pad adhesion, and feature registration. Large Z-axis expansion is primarily a reliability concern in thick, plated-through-hole rigid boards such as FR-4, where barrel cracking and inner-layer interconnect failure are driven by resin expansion along the hole axis.
How is the CTE of a flexible PCB measured?
The standard method is thermomechanical analysis per ASTM E831 and IPC-TM-650 2.4.24, in which a specimen is heated at a controlled rate and its dimensional change is tracked to calculate ppm/°C. Complementary techniques include DMA for Tg determination, thermal cycling with optical dimensional measurement for stability assessment, and digital image correlation for full-field expansion mapping.
How can CTE mismatch problems be reduced in PET circuit design?
Effective countermeasures include using heat-stabilized PET film, balancing copper on both sides of the substrate, choosing low-expansion adhesives or adhesiveless copper cladding, scaling artwork to compensate for measured machine-direction and transverse-direction expansion, avoiding abrupt copper density transitions, and designing generous annular rings and strain-relief features at holes and bend zones.