Thermomechanical analysis (TMA) is the industry-standard method for measuring the coefficient of linear thermal expansion (CTE) of printed circuit boards and copper-clad laminates. Accurate Z-axis CTE data is critical for predicting through-hole reliability, pad lifting, and delamination resistance during high-temperature processes such as lead-free reflow soldering. This article explains the TMA testing principle, applicable standards, sample requirements, test procedures, and typical applications for PCB CTE characterization.
Technical Overview
The coefficient of linear thermal expansion (CTE) describes how much a material elongates per unit length for each degree of temperature change, conventionally expressed in parts per million per degree Celsius (ppm/°C). For printed circuit boards, CTE is a strongly anisotropic property: in the X and Y directions, the woven glass fabric mechanically constrains expansion to roughly 10 to 20 ppm/°C, while in the Z (through-thickness) direction the polymer matrix dominates, producing expansion of approximately 40 to 70 ppm/°C below the glass transition temperature (Tg) and several hundred ppm/°C above it.
Thermomechanical analysis (TMA) is the recognized laboratory technique for measuring this property. In a TMA experiment, the specimen rests on a quartz platform beneath a probe that applies a small static force. A high-resolution displacement sensor continuously tracks the change in specimen dimension while a furnace ramps the temperature at a controlled rate, typically 5 °C/min. The resulting expansion curve yields the Z-axis CTE below Tg (α1), the CTE above Tg (α2), the extrapolated-onset Tg, and the total through-thickness expansion over a defined interval such as 50 °C to 260 °C.
Z-axis CTE is a first-order reliability parameter for PCBs. Because the resin system expands far more than the copper plated inside through-holes and vias, every thermal cycle imposes shear and tensile stress on the copper barrels. Excessive Z-axis expansion accelerates barrel cracking, innerlayer separation, and pad lifting, particularly during lead-free reflow excursions to 260 °C. The governing test methods include IPC-TM-650 method 2.4.24, ASTM E831, and ISO 11359-2, with equivalent national standards (such as GB/T and JIS series methods) following the same principle.
Testing Samples
Copper-Clad Laminate and Base Material Specimens
Coupons are cut from the laminate sheet, typically a square of approximately 6 mm x 6 mm (about 0.25 in) or larger, with a preferred thickness between 0.5 mm and 2.5 mm. When characterizing the resin system against a laminate specification such as IPC-4101 or IPC-4103, the copper foil is normally etched away before testing. Both faces must be machined flat, parallel, and free of cracks, voids, resin starvation, or other visible defects that could distort the displacement signal.
Finished PCB Coupons
Specimens may be taken from unpopulated regions of a fabricated board or from dedicated test coupons attached to the production panel. Testing the finished board captures the as-fabricated thermal expansion behavior, including the influence of copper distribution, hole patterns, and press cycles. Regions with heavy copper imbalance, severe warpage, or dense via fields should be avoided unless specifically under investigation.
Thin Cores and Cured Prepreg Plaques
Thin laminate cores thinner than about 0.5 mm can be stacked to reach an adequate measurement height, provided all layers are clean, dry, and aligned. Prepreg materials must first be fully cured into a solid plaque, since uncured B-stage resin does not exhibit representative expansion behavior.
Testing Items
A standard TMA characterization of PCB materials can include the following deliverables:
- Z-axis CTE below Tg (α1), expressed in ppm/°C
- Z-axis CTE above Tg (α2), expressed in ppm/°C
- Total Z-axis expansion from 50 °C to 260 °C, expressed in ppm or percent
- Glass transition temperature (Tg) by TMA, reported as the extrapolated onset of the expansion slope change
- X and Y in-plane CTE, as an optional measurement using an appropriate specimen orientation or fixture
- Time to delamination at 260 °C (T260) or 288 °C (T288), a related TMA test performed per IPC-TM-650 method 2.4.24.1
Testing Methods
Prior to testing, specimens are cleaned and conditioned. Common practice includes a bake at approximately 105 °C to remove absorbed moisture, followed by cooling in a desiccator, so that the measurement reflects the intrinsic material rather than drying shrinkage. Specimen thickness is measured at several points, and the average value is recorded.
The TMA instrument is calibrated before use: temperature calibration is verified with the melting points of certified metals such as indium (156.6 °C) and zinc, while length (expansion) calibration is verified with a certified reference material such as fused quartz or an aluminum oxide standard. A flat-tipped expansion probe is used with a minimal static force, typically no more than about 0.05 N, so that the probe load does not compress or indent the specimen. The furnace is purged with nitrogen at roughly 20 to 50 mL/min to prevent oxidative degradation of the resin at elevated temperature.
During the run, the specimen is equilibrated at the starting temperature (commonly 25 °C to 30 °C) and then heated at 5 °C/min to an endpoint high enough to define α2 clearly, generally at least Tg plus 40 °C; for total-expansion reporting the run typically extends to 260 °C to simulate a lead-free reflow peak. Data reduction is performed by linear regression on the two linear segments of the expansion curve, below and above the transition, with the Tg reported as the extrapolated onset of the slope change. Testing is normally repeated on two or three replicate specimens, and the average values with scatter are reported together with the complete expansion curve.
Applications
Z-axis CTE data generated by TMA supports a wide range of engineering and quality decisions across the electronics supply chain:
- Qualification of copper-clad laminates and prepreg against slash-sheet limits in IPC-4101 and IPC-4103 material specifications
- Verification of high-Tg, low-CTE, and filled resin systems used in high layer count boards, thick backplanes, and automotive or aerospace electronics
- Assessment of lead-free reflow compatibility, where total Z-axis expansion from 50 °C to 260 °C is frequently limited to approximately 3.5 to 4.0 percent
- Reliability screening of IC package substrates and HDI structures, where plated via integrity is tightly coupled to thermal expansion
- Failure analysis of barrel cracking, pad lifting, corner cracking, and resin delamination, by comparing the failed material with control data
- Incoming material inspection, supplier comparison, and input data for warpage and thermo-mechanical stress simulation
FAQ
What is a typical Z-axis CTE value for standard FR-4?
Standard FR-4 laminates generally exhibit α1 between about 40 and 70 ppm/°C and α2 between about 200 and 400 ppm/°C. High-Tg, ceramic-filled, or low-CTE epoxy systems can reduce α1 to roughly 20 to 40 ppm/°C, which is one reason they are preferred for lead-free assembly and thick multilayer constructions.
Why is the Z-axis CTE so much higher than the X and Y CTE?
In the plane of the board, the woven glass fabric carries most of the load and its low expansion (about 3 to 5 ppm/°C) dominates the composite response. Through the thickness, the resin expands freely without fiber restraint, so the Z-axis CTE approaches that of the polymer matrix itself.
What is the difference between α1, α2, and total expansion?
α1 is the CTE measured below Tg, where the resin is glassy and relatively stiff; α2 is the CTE measured above Tg, where the softened resin expands far faster; total expansion is the cumulative dimensional change over a specified interval, such as 50 °C to 260 °C, regardless of Tg. All three are commonly specified together in laminate datasheets.
Can TMA measure in-plane CTE?
Yes. With a suitable expansion fixture and flat specimens oriented so the probe tracks lateral displacement, TMA can provide X and Y CTE values, although dilatometry is sometimes used as a complementary technique. In-plane data is mainly relevant for chip packaging, where silicon expansion must be matched by the substrate.
How does Tg measured by TMA compare with DSC and DMA values?
TMA Tg is typically the lowest reported value, DSC Tg is intermediate, and DMA Tg is the highest, because each technique senses the transition through a different physical response. When comparing datasheets or specifications, always confirm which method and which parameter definition apply.
Which standards govern PCB CTE testing by TMA?
The primary methods are IPC-TM-650 method 2.4.24 for Z-axis expansion and Tg of printed boards and laminates, ASTM E831 for linear expansion of solid materials by TMA, and ISO 11359-2 for thermomechanical analysis of plastics. Accredited third-party laboratories routinely perform these tests in accordance with the applicable IPC, ASTM, ISO, GB, and JIS procedures.
Why Choose ZKGX?
- State-of-the-art analytical equipment
- Highly qualified scientific team
- Fast turnaround time
- Competitive pricing