PCB Core Board Coefficient of Thermal Expansion (CTE) Testing and Measurement

PCB core board coefficient of thermal expansion (CTE) testing quantifies how much a laminate expands with rising temperature, a decisive factor in plated through-hole reliability, reflow survival, and long-term assembly durability. This technical guide explains the principle of CTE determination, the sample types, tested parameters, TMA-based measurement methods per IPC-TM-650, and practical applications for rigid, high-Tg, and specialty PCB core materials.

Technical Overview

The coefficient of thermal expansion (CTE) describes the fractional dimensional change of a material per degree of temperature change, conventionally expressed in parts per million per degree Celsius (ppm/°C). For PCB core boards, defined as cured copper-clad laminate sheets without circuit patterning, CTE is a critical thermomechanical property because the laminate, the copper foil, and the plated through-hole (PTH) barrels all expand at different rates during soldering and thermal cycling.

PCB cores are mechanically anisotropic. The woven glass fabric constrains expansion in the in-plane X and Y directions, producing CTE values typically between 12 and 16 ppm/°C for epoxy-glass systems. The resin-dominated Z axis expands far more aggressively, commonly 40 to 60 ppm/°C below the glass transition temperature (Tg) and 200 to 400 ppm/°C above it. This large Z-axis expansion stresses the copper barrels of vias and PTHs during reflow and power cycling, and is a leading root cause of barrel cracking, innerlayer separation, and interconnect failure.

Because the expansion slope changes sharply at Tg, CTE results are reported in two segments: alpha-1 (pre-Tg slope) and alpha-2 (post-Tg slope), together with the total Z-axis expansion percentage over a defined temperature span. Industry measurement practice follows IPC-TM-650 method 2.4.24 and its related revisions, with supporting references including ASTM E831, ASTM E228, and ISO 11359-2, while acceptance limits are defined in the IPC-4101 slash sheets for each laminate grade.

Testing Samples

Standard FR-4 Glass-Epoxy Cores

Woven E-glass fabric reinforced epoxy laminates in common thicknesses from 0.1 mm to 3.2 mm, with copper foil removed by chemical etching before testing so that the expansion of the neat laminate is measured. These represent the largest volume of samples submitted for routine CTE qualification.

High-Tg and Mid-Tg Laminate Grades

Multifunctional epoxy systems with Tg values of 150 °C to 180 °C or higher, including halogen-free formulations. Elevated Tg delays the transition to the steep alpha-2 expansion region, so CTE testing is essential to verify that high-Tg claims translate into reduced total expansion.

High-Speed Digital and RF Materials

Low-Dk and low-loss laminates based on modified epoxies, PPO/PPE, hydrocarbon-ceramic, or PTFE systems, often loaded with silica or ceramic fillers specifically to suppress CTE. In-plane CTE matching to silicon (approximately 3 ppm/°C) is frequently a design requirement for large flip-chip and BGA packages.

Flexible and Special Substrates

Polyimide film cores, adhesiveless clad films, metal-core (aluminum) boards, and heavy-copper constructions. These substrates may require adapted fixtures or orientation because their expansion behavior differs substantially from rigid glass-fabric laminates.

Testing Items

A complete CTE characterization of a PCB core board typically reports the following parameters:

• Z-axis CTE below Tg (alpha-1), in ppm/°C, calculated from the linear regression slope of the expansion curve before transition
• Z-axis CTE above Tg (alpha-2), in ppm/°C, calculated from the post-transition slope
• Total Z-axis thermal expansion percentage, usually integrated from 50 °C to 260 °C, which directly reflects via-barrel strain during lead-free reflow
• In-plane X and Y CTE, relevant to warpage, solder joint stress, and die-attach compatibility
• Glass transition temperature determined from the inflection or extrapolated onset point of the expansion curve
• Verification of cure state, since under-cured resin shifts Tg downward and inflates alpha-2 expansion

Testing Methods

Thermomechanical Analysis (TMA)

TMA is the reference technique for Z-axis CTE determination, performed in accordance with IPC-TM-650 2.4.24. A square specimen of approximately 6.4 mm by 6.4 mm, cut to represent the full laminate thickness with smooth, burr-free edges, is placed on a flat quartz stage. A silica expansion probe applies a minimal static force, typically no more than 0.1 N, to avoid indenting the resin surface, while the furnace is purged with nitrogen. The specimen is heated from ambient temperature to 260 °C at 5 °C/min, and the dimensional change is recorded continuously. Many protocols run two heating cycles, discarding the first run to relieve residual stress and machining history, and reporting data from the second run. The CTE is computed as the slope of the displacement-versus-temperature curve divided by the initial specimen length, expressed in ppm/°C.

In-Plane (X/Y) CTE Measurement

In-plane expansion is measured by orienting a rectangular specimen horizontally under a side-acting TMA fixture, or by supplementary techniques such as bonded foil strain gauges and digital image correlation (DIC), which track two-dimensional strain across the panel surface. In-plane values are direction-dependent because the woven fabric geometry differs along warp and fill, so both axes are reported separately when required by specification.

Push-Rod Dilatometry

For larger specimens or comparative bulk studies, a horizontal or vertical push-rod dilatometer per ASTM E228 provides CTE over extended temperature ranges with excellent baseline stability, and serves as a cross-check for TMA results.

Sample Preconditioning

Prior to testing, specimens are baked at 105 °C to 110 °C for one to two hours to remove absorbed moisture, then kept in a desiccator until measurement. Moisture plasticizes the resin, depresses the measured Tg, and can distort both alpha slopes, making preconditioning mandatory for reproducible results.

Applications

CTE data from PCB core boards support nearly every stage of the electronics reliability chain:

• Incoming material qualification against IPC-4101 slash sheets, where many laminate grades limit total 50 °C to 260 °C expansion to approximately 3.5 to 4.0 percent and impose grade-specific alpha-1 and alpha-2 maxima
• Prediction of plated through-hole and via reliability under thermal cycling, reflow excursions, and IST testing, since Z-axis strain drives barrel fatigue
• Material selection for high-density interconnect (HDI) designs, package substrates, and chip-scale assemblies where low in-plane CTE reduces solder joint and die stress
• Warpage and coplanarity control in large, thin boards intended for BGA mounting and double-sided reflow
• Supplier comparison, lot-to-lot statistical process control, and evaluation of formulation or cure changes
• Failure analysis, where abnormally high expansion confirms resin degradation, under-cure, or moisture absorption as the mechanism behind delamination or open vias

FAQ

What are typical CTE values for an FR-4 core board?

Standard FR-4 typically exhibits a Z-axis alpha-1 of 40 to 60 ppm/°C, an alpha-2 of 200 to 350 ppm/°C above Tg, and total 50 °C to 260 °C expansion of roughly 2.5 to 4.0 percent. In-plane CTE usually falls between 12 and 16 ppm/°C. Filled, high-Tg, and low-CTE laminate grades can reduce total Z expansion below 2.5 percent.

Why is Z-axis CTE so much higher than X and Y CTE?

The woven glass fabric carries the mechanical load in the plane and expands only slightly, while the resin phase dominates the thickness direction. Once the resin passes Tg and softens, the fabric no longer restrains through-thickness expansion, so alpha-2 increases several-fold.

What do alpha-1 and alpha-2 mean in a CTE report?

They are the linear expansion slopes below and above the glass transition temperature. Alpha-1 applies to the rigid, glassy state and alpha-2 to the rubbery state. The jump between them marks Tg, which is why TMA yields Tg and CTE from a single run.

What specimen size and preparation are required?

The standard TMA specimen is a 6.4 mm square representing the full core thickness, with copper etched away, cleanly cut edges, and a bake at 105 °C to 110 °C to eliminate moisture. Specimens must be flat and free of chips so the probe seats evenly.

What temperature range and heating rate are used?

The most common program runs from room temperature or 50 °C up to 260 °C at 5 °C/min under nitrogen. The upper limit corresponds to lead-free reflow peak temperatures and ensures alpha-2 is fully captured. Some specifications extend the scan or apply multiple cycles to evaluate aged behavior.

How does CTE affect PCB reliability in practice?

Each reflow or thermal cycle expands the laminate through thickness more than the copper barrel can accommodate, imposing cyclic plastic strain. Over repeated cycles this produces barrel cracks, corner cracks in annular rings, and latent interconnect opens. Lower total Z-axis expansion directly extends interconnect fatigue life, which is why CTE is a primary acceptance criterion for core laminates.