Underfill Adhesive Martens Heat Resistance Temperature Testing: Principles, Methods and Applications

The Martens heat resistance temperature test is a classical thermal-mechanical characterization method used to determine the heat endurance limit of cured underfill adhesives employed in flip-chip, BGA, and CSP electronic packaging. This article explains the test principle, specimen requirements, testing procedure, applicable standards, and practical applications, providing packaging engineers and quality assurance professionals with a complete technical reference for evaluating the thermal reliability of underfill materials.

Technical Overview

Underfill adhesives are epoxy-based thermosetting materials dispensed into the gap between a silicon die and its substrate in advanced electronic assemblies. Once cured, they redistribute thermomechanical stress away from fragile solder interconnects, dramatically improving resistance to thermal cycling, mechanical shock, and board-level drop events. Because underfills must maintain mechanical strength during high-temperature operation and subsequent assembly processes, their heat resistance is a fundamental qualification parameter.

The Martens heat resistance temperature test measures the temperature at which a cured, free-standing specimen, loaded in flexure under a defined bending stress, deflects by a specified amount while being heated at a constant rate. The method is standardized in GB/T 1699 and traces its origin to the Martens method described in DIN 53458 and DIN 53462. For highly cross-linked thermosets such as underfill adhesives, the Martens temperature sits slightly below or near the glass transition region and serves as a practical indicator of the upper service temperature limit of the cured material.

A high Martens heat resistance temperature confirms that the cured underfill network will retain stiffness and load-transfer capability under thermal load, protecting solder joints in automotive, industrial, and high-power computing environments. Accredited third-party laboratories routinely perform this test as part of incoming material inspection, formulation benchmarking, and packaging qualification programs.

Testing Samples

Cured Standard Molding Specimens

The Martens test requires free-standing rectangular bars, typically 120 mm in length, 15 mm in width, and 10 mm in thickness. Because underfill adhesives are supplied as liquids, the test material is first cast into a release-treated mold, degassed to remove entrapped air, and cured according to the supplier-recommended cure profile, including any required post-cure schedule. Only fully cured, void-free specimens yield representative results.

Representative Production Material

Samples should be drawn from the actual production batch under evaluation. The adhesive must be within its shelf life, stored at the specified temperature, and handled according to the manufacturer's thawing and dispensing guidelines to ensure the test reflects real-world material behavior.

Specimen Conditioning

Before testing, specimens are conditioned in a standard atmosphere of 23 ± 2 °C and 50 ± 5% relative humidity in accordance with ISO 291 for a minimum of 40 hours. At least three valid specimens are tested for each material condition, and the arithmetic mean is reported.

Testing Items

The core deliverable of the test is a single, well-defined parameter:

• Martens heat resistance temperature of the cured underfill adhesive, expressed in degrees Celsius (°C)

To build a complete thermal reliability profile, the Martens result is frequently evaluated alongside complementary measurements performed in accredited third-party laboratories:

• Glass transition temperature (Tg) by DSC per ISO 11357, DMA, or TMA methods
• Heat deflection temperature (HDT) under flexural stress per ISO 75
• Vicat softening point per ISO 306 for comparative assessment
• Coefficient of linear thermal expansion (CTE) below and above Tg by TMA
• Thermal degradation onset by thermogravimetric analysis (TGA)

Testing Methods

Test Principle

A cured specimen is supported horizontally on two supports set 100 mm apart and loaded at mid-span through a lever system so that it sustains a nominal bending stress of 5 MPa on the outer fibers. The surrounding medium, typically a heat-transfer oil bath or forced-air oven, is heated at a constant rate of 50 °C per hour. The temperature at which the deflection indicator registers the specified endpoint deflection of 6 mm is recorded as the Martens heat resistance temperature.

Apparatus

Testing is conducted on a dedicated Martens heat resistance apparatus comprising:

• A uniformly heated oil bath or air oven with a temperature control accuracy of ±2 °C and a verified heating rate of 50 °C/h
• A rigid specimen frame with supports spaced at 100 mm and a central loading lever
• Dead-weight loading system calibrated to produce the standard 5 MPa bending stress
• A dial gauge or optical sensor capable of reading deflection to 0.1 mm resolution
• Calibrated thermocouples or mercury thermometers positioned immediately adjacent to the specimen surface

Procedure

• Measure and record the width and thickness of each conditioned specimen at three points along its length; reject specimens with visible voids, cracks, or warpage greater than acceptable limits
• Mount the specimen on the supports with its as-cast surface in the defined orientation, engage the loading lever, and apply the calibrated load without shock
• Set the deflection indicator to zero or its reference position once the initial creep stabilizes
• Begin heating at 50 ± 5 °C/h, starting the test at a bath temperature at least 50 °C below the expected Martens temperature
• Continuously monitor deflection and record the bath temperature at the exact moment the indicator reaches 6 mm deflection
• Repeat for at least three specimens and report the arithmetic mean, individual values, and any observations such as discoloration, cracking, or delamination

Reporting

The test report identifies the material and lot, curing schedule, specimen dimensions and number, conditioning conditions, heating rate, applied bending stress, individual and mean Martens temperatures, and the testing standard referenced, ensuring full traceability for qualification and audit purposes.

Applications

Martens heat resistance temperature data for underfill adhesives directly support critical decisions across the electronics manufacturing value chain:

• Flip-chip packaging: verifying that the cured underfill preserves joint reinforcement through solder reflow of adjacent components and extended high-temperature operation
• Ball grid array (BGA) and chip-scale package (CSP) corner and edge bonding: selecting adhesives that remain rigid above worst-case board operating temperatures
• Automotive electronics: qualifying underfills for powertrain and domain-controller modules where sustained ambient temperatures exceed 125 °C, complementing AEC-oriented reliability testing
• Stacked-die and memory packaging: confirming that cap and mold underfills withstand molding and burn-in thermal excursions without softening
• Incoming material inspection: comparing Martens temperature between production lots to detect formulation drift or improper cure
• Research and development: benchmarking new epoxy, cyanate ester, or filled formulations against incumbent materials and against Tg and HDT data for balanced thermal design

FAQ

What is the Martens heat resistance temperature of an underfill adhesive?

It is the temperature at which a cured underfill bar, held under a 5 MPa bending stress and heated at 50 °C/h, deflects by 6 mm. It represents a practical heat endurance limit closely associated with the onset of glass transition-related softening.

How does the Martens test differ from Vicat and HDT tests?

The Martens method applies a 5 MPa bending stress with a 6 mm deflection endpoint, whereas the Vicat method per ISO 306 uses a penetrating needle under a small load, and HDT per ISO 75 measures deflection under 1.80 MPa, 0.45 MPa, or 8.00 MPa flexural stress. Martens temperature is typically the most conservative, demanding indicator among the three for rigid thermosets.

Why do underfill adhesives need a high Martens temperature?

If the cured underfill softens below the operating or processing temperature, its modulus drops sharply, stress transfer to solder joints is lost, and thermomechanical reliability degrades. A Martens temperature comfortably above the maximum service temperature ensures the adhesive network continues reinforcing the interconnect structure.

What Martens temperatures do high-performance underfills typically achieve?

Well-cured, highly cross-linked epoxy underfills generally exhibit Martens heat resistance temperatures in the range of approximately 110 °C to 170 °C, depending on resin chemistry, filler loading, and cure schedule; specific values must always be verified by laboratory testing of the actual formulation.

Can an underfill be tested in its liquid dispensed state?

No. The Martens test is a solid-state flexural test performed on fully cured, free-standing bars. Liquid adhesives are therefore cast, degassed, and cured in molds before testing.

How many specimens are required for a valid result?

A minimum of three conditioned, defect-free specimens should be tested, and the mean value reported; additional specimens improve statistical confidence, particularly for qualification submissions.