The coefficient of thermal expansion (CTE) is a critical parameter that determines the dimensional stability and long-term reliability of PET-based printed circuit boards. This article provides a comprehensive guide to PET PCB board CTE testing methods, covering thermomechanical analysis (TMA), strain gauge techniques, and dilatometry in accordance with IPC-TM-650 and ASTM standards. Engineers, quality managers, and materials scientists will find practical guidance on sample preparation, test conditions, data interpretation, and acceptance criteria for PET flexible substrates.
Technical Overview
PET (polyethylene terephthalate) films are widely used as base dielectric substrates for flexible printed circuits, membrane switches, and low-cost flexible PCBs due to their excellent electrical insulation, mechanical strength, moisture resistance, and favorable cost compared to polyimide. However, PET is a semi-crystalline thermoplastic with a glass transition temperature of approximately 78 to 80 degrees Celsius, which means its thermomechanical behavior changes significantly across the operating temperature range of many electronic assemblies.
The coefficient of thermal expansion (CTE), commonly expressed in parts per million per degree Celsius (ppm/degree Celsius), quantifies the fractional dimensional change of the material per unit temperature change. For biaxially oriented PET films used in PCB applications, the in-plane CTE below Tg typically ranges from 15 to 35 ppm/degree Celsius, which is reasonably close to the 17 ppm/degree Celsius of copper foil. Above Tg, however, the CTE of PET increases dramatically and can exceed 100 ppm/degree Celsius, while oriented films may also exhibit irreversible thermal shrinkage caused by relaxation of molecular orientation.
Accurate CTE characterization is therefore essential for predicting warpage, delamination, copper trace cracking, via failure, and registration drift during soldering, lamination, and thermal cycling. Because PET substrates cannot withstand standard lead-free reflow temperatures, CTE data also defines the process windows for low-temperature assembly and hot-melt lamination used in flexible circuit manufacturing.
Testing Samples
Bare PET Film Substrates
Unclad PET film is tested to establish the baseline thermomechanical behavior of the dielectric material. Specimens are typically cut from both the machine direction (MD) and transverse direction (TD) of the roll, since biaxial orientation produces measurable anisotropy between the two axes. Film thicknesses commonly evaluated range from 25 to 250 micrometers, and multiple plies may be stacked where a minimum specimen thickness is required by the test method.
Copper-Clad PET Laminates
Copper-clad PET laminates, usually manufactured with acrylic or modified epoxy adhesive layers, are tested to evaluate the composite CTE of the finished construction. Specimens may be tested as received, or the copper may be chemically etched away to isolate the adhesive-plus-substrate system. These samples reveal the influence of the adhesive layer, whose CTE is often an order of magnitude higher than that of the PET base film.
Finished Flexible PET PCBs
Processed circuit samples containing etched traces, coverlayers, stiffeners, and through-features are tested to verify end-product dimensional stability. Testing finished circuits captures the cumulative effect of all processing steps, including any thermal history introduced during lamination or curing that alters the residual stress state of the oriented PET film.
Adhesive, Coverlayer, and Coating Materials
Freestanding films or plaques of adhesive, covercoat, and protective coating materials are tested separately to build complete material models. These auxiliary layers frequently dominate Z-axis expansion in multilayer constructions and are critical inputs for finite element thermal stress simulations.
Testing Items
A complete thermomechanical characterization of PET PCB materials typically includes the following measured parameters:
- In-plane coefficient of thermal expansion (X and Y axes, alpha 1 below Tg and alpha 2 above Tg)
- Out-of-plane Z-axis coefficient of thermal expansion for laminates and multilayer stacks
- Glass transition temperature (Tg) of the PET base film and of adhesive systems
- Percent thermal expansion (PTE) over a defined temperature interval
- Irreversible thermal shrinkage of oriented PET film upon first heat excursion
- Anisotropy ratio between machine direction and transverse direction expansion
- Time to delamination and thermal degradation onset for qualification purposes
Reporting should distinguish clearly between reversible expansion and one-time shrinkage, since the first heating run of oriented PET often differs substantially from the second run after thermal history has been erased.
Testing Methods
Thermomechanical Analysis (TMA)
TMA is the primary method for CTE determination of PET PCB materials and is performed in accordance with ASTM E831 and IPC-TM-650 method 2.4.24. A specimen with flat, parallel faces is placed under a low-mass quartz or fused-silica probe applying a small static force, typically around 0.05 newtons. The furnace is heated at a controlled rate of 3 to 5 degrees Celsius per minute under nitrogen purge, and dimensional change is recorded with a linear variable differential transformer capable of submicrometer resolution.
For Z-axis testing of laminates, specimens are usually cut into squares of several millimeters on a side, stacked to reach the recommended thickness, and tested through the glass transition. The resulting expansion curve yields alpha 1 from the slope below Tg, alpha 2 from the slope above Tg, the Tg inflection point, and the percent thermal expansion between specified limits such as 50 to 260 degrees Celsius as called out in procurement specifications. For PET substrates, the maximum test temperature must remain well below the melting region of approximately 250 degrees Celsius, and test plans should be capped near 150 to 200 degrees Celsius unless degradation behavior is specifically under investigation.
Strain Gauge Method
The bonded strain gauge technique described in IPC-TM-650 method 2.4.41.2 measures in-plane CTE directly on actual circuit materials with high sensitivity. A precision foil strain gauge with self-temperature compensation matched to the expected CTE range is bonded to the specimen using a high-temperature adhesive that is cured before testing. The specimen is placed in an environmental chamber and subjected to a programmed temperature ramp, while the gauge output is captured through a Wheatstone bridge circuit.
Data reduction corrects for the apparent strain contribution of the gauge itself, determined by testing a reference material such as titanium silicate or fused quartz with a known, near-zero CTE. The method allows simultaneous monitoring of the X and Y axes on a single specimen and is particularly valuable for evaluating finished PET circuits, where it reproduces the constraint conditions of real hardware. The achievable accuracy is on the order of plus or minus 1 ppm/degree Celsius when gauge selection, bonding, and apparent strain calibration are performed rigorously.
Push-Rod Dilatometry
Quartz push-rod dilatometry per ASTM E228 and ASTM D696 provides bulk linear expansion data over moderate temperature spans. A specimen is placed inside a quartz tube and contacted by a quartz rod, and the length change is measured with a dial gauge, capacitive sensor, or LVDT as the assembly is heated in a bath or furnace. Dilatometry is well suited to thicker molded or laminated plaques and serves as an independent cross-check of TMA results, though its sensitivity is lower and it is not well suited to very thin PET films.
Dynamic Mechanical Analysis (DMA)
DMA per ASTM E1640 is used primarily to determine Tg and to evaluate the storage modulus transition that accompanies the change from alpha 1 to alpha 2 behavior. Some DMA instruments also operate in a stress-controlled thermomechanical mode that permits expansion measurement. Combining DMA modulus data with TMA CTE data enables accurate prediction of thermal stress at copper-PET interfaces during thermal cycling.
Sample Preparation and Conditioning
Regardless of the chosen method, the following practices are essential for valid PET CTE data:
- Condition specimens at 23 plus or minus 2 degrees Celsius and 50 plus or minus 5 percent relative humidity for a minimum of 24 hours before testing
- Cut specimens with clean, parallel edges and verify thickness at multiple points with a calibrated micrometer
- Mark and preserve the machine direction versus transverse direction orientation of all film samples
- Remove copper by chemical etching rather than mechanical peeling when isolating the dielectric, to avoid introducing mechanical damage
- Perform a duplicate heating run when separating reversible expansion from first-cycle thermal shrinkage
- Test a minimum of three specimens per material, direction, and condition to establish statistical confidence
Applications
CTE test data for PET PCB materials directly supports design and manufacturing decisions across a broad range of applications, including flexible printed circuits for printers, scanners, and camera modules; membrane switch keyboards and control panels; automotive instrument clusters and body electronics; flexible heaters and sensor arrays; medical monitoring devices and wearable electronics; RFID antenna substrates; and display interconnects. In each case, matching the substrate CTE to copper foil and adjacent layers within an acceptable window preserves trace integrity, adhesion, and circuit registration throughout the product life.
Test results also feed qualification programs under IPC-4204 for flexible base dielectrics and downstream IPC-6013 performance requirements for flexible printed boards. Purchasing specifications increasingly cite alpha 1, alpha 2, Tg, and thermal shrinkage limits verified by the methods described above, making standardized CTE testing a contractual element of the PET PCB supply chain.
FAQ
What is the typical CTE of PET film used for printed circuits? Biaxially oriented PET film generally exhibits an in-plane CTE of roughly 15 to 35 ppm/degree Celsius below its glass transition temperature, with values differing slightly between machine and transverse directions. Above Tg the expansion rate rises sharply, and first-time heating may also produce measurable shrinkage from relaxation of the oriented molecular structure.
Which standards govern PET PCB CTE testing? The most commonly applied standards are IPC-TM-650 methods 2.4.24 and 2.4.41.2 for laminate and strain gauge testing, ASTM E831 for CTE by TMA, ASTM E228 for dilatometry, ASTM D696 for plastics in the minus 30 to plus 30 degree Celsius range, and ASTM E1545 and E1640 for Tg by TMA and DMA respectively.
What temperature range should be used for testing? For application-oriented data, a range such as minus 40 to plus 150 degrees Celsius reflects most PET circuit service environments. Testing above approximately 200 degrees Celsius should be avoided for routine CTE work because PET approaches its melting region and irreversible degradation begins to distort the measurement.
Why do first and second heating runs give different results? Oriented PET films contain residual stresses from the biaxial stretching process. The first heating run releases these stresses, producing apparent shrinkage that overlaps the true expansion signal, so a second run yields the intrinsic reversible CTE of the material.
How is Z-axis expansion different from in-plane expansion? Z-axis CTE, measured normal to the plane of the laminate, is dominated by the adhesive and dielectric layers and is often far higher than in-plane values. It is the key parameter for assessing plated hole reliability and lamination pressure requirements in multilayer PET constructions.
How many specimens are required for a valid test? A minimum of three specimens per direction and material condition is standard practice, and five or more replicates are recommended when results will be used for qualification, supplier comparison, or specification acceptance decisions.