Reliability Testing of Kovar Alloy Glass-to-Metal Seals: Standards, Methods and Applications

Kovar alloy glass-to-metal seals are the backbone of hermetic packaging in electronics, aerospace, sensors, and medical devices, and their long-term performance depends directly on rigorous reliability testing. This article explains the test samples, testing items, standard methods, and acceptance criteria used by accredited third-party laboratories to qualify Kovar-to-glass seal assemblies against standards such as MIL-STD-883, MIL-STD-202, IEC 60068-2, ASTM, ISO, and GB/T. Read on for a complete technical reference covering hermeticity, thermal cycling, mechanical strength, corrosion endurance, and structural evaluation of these critical sealing systems.

Technical Overview

Kovar alloy is an iron-nickel-cobalt controlled-expansion alloy, specified in ASTM F15/F15M (UNS K94610), with a nominal composition of approximately 29% nickel, 17% cobalt, and the balance iron. Its thermal expansion coefficient of roughly 5.0 to 6.0 ppm/K between room temperature and 400°C closely matches that of borosilicate sealing glasses, which makes it the preferred metal for matched glass-to-metal seals.

During sealing, the Kovar surface is typically pre-oxidized at high temperature to form a thin, well-adherent oxide layer. The molten sealing glass wets and partially dissolves this oxide, producing a continuous chemical bond at the metal-glass interface. Because the seal is created at elevated temperature and cooled to room temperature, residual stresses are locked into the assembly; these stresses, together with interfacial quality, govern whether the seal remains hermetic and crack-free throughout its service life.

Reliability testing therefore focuses on three questions: whether the seal is leak-tight, whether it survives mechanical and thermal stresses without cracking or delamination, and whether electrical insulation performance is maintained over time. The dominant reference frameworks are MIL-STD-883 and MIL-STD-202 for microelectronic and passive components, MIL-STD-750 for discrete semiconductor devices, IEC 60068-2 and GB/T 2423 for general environmental testing, and ASTM F15 for material qualification.

Testing Samples

Matched Glass-to-Metal Seal Assemblies

These are production-representative assemblies in which Kovar pins, flanges, or housings are sealed to borosilicate glass under closely matched expansion coefficients. They represent the highest-risk seal type because both metal and glass deform together, and any coefficient mismatch translates directly into glass stress. Samples are usually drawn from finished production lots after all thermal processing steps.

Compression Seal Assemblies

Compression seals use an outer metal body, often a carbon or stainless steel with a higher expansion coefficient, that shrinks onto the glass insulator during cooling and places the glass in permanent compression. Pin materials may be Kovar or alloy steel with a glass-compatible expansion. Test samples must include the complete mechanical stack, since failure typically initiates at the housing-to-glass interface rather than at the pin.

Hermetic Packages and Feedthroughs

These include transistor-outline headers, ceramic-plus-Kovar hybrid packages, connector feedthroughs, pressure sensor headers, and battery terminal feedthroughs. Samples are tested as sealed units with known internal cavity volumes, because cavity volume determines the applicable fine leak acceptance limits.

Process Control Coupons

Button seals, bead seals, butt-joint specimens, and glass-to-metal wetting coupons are prepared alongside production runs. They are used for destructive metallographic verification of oxide layer thickness, wetting angle, interfacial continuity, and glass devitrification, providing process-level insight that finished-part testing cannot supply.

Testing Items

A comprehensive reliability program for Kovar glass-to-metal seals covers the following test items, typically organized into screening, qualification, and periodic lot acceptance groups:

• Fine leak and gross leak hermeticity testing
• Thermal cycling endurance and thermal shock resistance
• Mechanical strength: pin pull, pin push-out, shear, and torque integrity
• Insulation resistance and dielectric withstanding voltage
• Internal moisture content by residual gas analysis
• Corrosion endurance: salt fog, damp heat, and moisture resistance
• Mechanical environment survival: vibration, mechanical shock, and constant acceleration
• Constructional analysis: metallographic cross-section, oxide thickness, interfacial voids, and glass crack inspection
• Material verification: chemical composition, microstructure, and thermal expansion coefficient

Testing Methods

Hermeticity Testing

Fine leak testing is performed by helium bombing, in which samples are pressurized in helium, then measured with a helium mass spectrometer. Fixed reject limits per MIL-STD-883 Method 1014 range typically from 5×10⁻⁸ atm·cm³/s for small cavities to 1×10⁻⁶ atm·cm³/s for larger volumes. Gross leak detection uses perfluorocarbon liquid immersion and observation of bubble streams, or a weight-gain method. Equivalent sealing tests are defined in IEC 60068-2-17 (Test Q). Optical leak detection based on package-surface deflection is also accepted as an alternative fine leak method for certain cavity designs.

Thermal Cycling and Thermal Shock

Thermal cycling per MIL-STD-883 Method 1010 or IEC 60068-2-14 (Test Nb) and GB/T 2423.22 subjects seals to slow ramp transitions between extremes such as -65°C and +150°C, commonly for 10 cycles in screening and up to 100 cycles or more in qualification. Thermal shock per MIL-STD-883 Method 1011 transfers samples between liquid baths, for example 0°C to 100°C or -55°C to +125°C for 15 cycles, generating steep thermal gradients across the seal. After stressing, samples are re-examined for glass cracks, chipping, and leakage, since accumulated microdamage at the metal-glass interface is the primary crack initiation site.

Mechanical and Structural Evaluation

Pin and terminal integrity is evaluated through axial pull, push-out, shear, and torque tests, with acceptance values defined by the product detail specification. Lead integrity testing follows MIL-STD-883 Method 2004. Structural quality is confirmed by metallographic cross-sectioning prepared per ASTM E3, verifying an oxide interlayer of typically 0.3 to 2 micrometers, continuous wetting without continuous interfacial voids, and the absence of radial or ring cracks in the glass. Dye penetrant inspection is used to reveal through-cracks not detectable by leak testing alone.

Electrical Characterization

Insulation resistance and dielectric withstanding voltage are measured pin-to-pin and pin-to-case, before and after environmental stress, per MIL-STD-202 Methods 302 and 301 or the equivalent MIL-STD-883 electrical requirements. Typical requirements include insulation resistance of 1×10¹⁰ ohms or greater at 100 VDC and dielectric withstand of several hundred volts without breakdown or excessive leakage. Degradation after humidity exposure frequently indicates surface contamination or alkali migration in the glass.

Environmental and Corrosion Endurance

Corrosion resistance is assessed by salt fog exposure per ASTM B117, ISO 9227, or GB/T 10125, with durations of 24 to 48 hours or longer for severe environments. Damp heat steady-state testing follows IEC 60068-2-78, for example 85°C and 85% relative humidity for 1000 hours, while moisture resistance cycling follows MIL-STD-883 Method 1002 and salt atmosphere exposure follows Method 1009. Internal water vapor content of sealed packages is measured by residual gas analysis per MIL-STD-883 Method 1018, with a common limit of 5000 ppmv, because trapped moisture drives internal corrosion and fogging over service life.

Material-Level Verification

Kovar chemistry and expansion behavior are confirmed against ASTM F15/F15M, with thermal expansion measured by dilatometry per ASTM E228 or thermomechanical analysis per ASTM E831. Glass expansion is measured on the same basis, and the metal-glass expansion match over the 25°C to 400°C range is documented, since drift in expansion behavior is a root cause of field cracking when raw material lots vary.

Applications

Reliability-tested Kovar glass-to-metal seals are essential wherever hermetic isolation must be maintained for decades:

• Hermetic microelectronic packages, including transistor-outline cans, flat packs, and multichip module housings for aerospace and defense electronics
• Pressure and inertial sensor headers for automotive engine, transmission, and exhaust systems
• High-voltage connector feedthroughs and battery terminal feedthroughs for electric vehicles and energy storage systems
• Implantable medical device feedthroughs, where seal integrity is combined with additional biocompatibility requirements under the ISO 10993 series
• Vacuum devices, infrared detector Dewars, fiber-optic feedthroughs, and optoelectronic packages requiring long-term internal atmosphere stability
• Down-hole and industrial instrumentation exposed to pressure, vibration, and corrosive media

FAQ

What leak rate is considered acceptable for a Kovar glass-to-metal seal?

Acceptance depends on internal cavity volume. Under the fixed reject levels of MIL-STD-883 Method 1014, a fine leak limit of 5×10⁻⁸ atm·cm³/s helium applies to small cavities, with relaxed limits up to about 1×10⁻⁶ atm·cm³/s for larger volumes, and any observable gross leak indication constitutes failure. Many high-reliability programs impose tighter internal limits.

Why is the Kovar pre-oxidation step critical to seal reliability?

The pre-oxidation firing controls the interfacial oxide that the molten glass dissolves to form a chemical bond. Insufficient oxidation causes poor wetting and interfacial gaps, while excessive oxidation produces a porous, weakly adherent layer that creates leak paths. Metallographic verification of oxide thickness and continuity is therefore a routine reliability control.

How do matched and compression seals differ in testing emphasis?

Matched seals are most sensitive to thermal expansion mismatch and are tested aggressively for glass cracking under thermal cycling and shock. Compression seals rely on mechanical pre-compression from the outer housing, so testing emphasizes housing yield strength, assembly integrity under thermal and mechanical stress, and leakage after temperature excursions.

How many thermal cycles are required for qualification?

Typical qualification profiles use 100 cycles at a condition such as -65°C to +150°C, while mission-profile-driven programs may extend to 500 or 1000 cycles. Screening applications commonly use 10 cycles. The required profile should be selected against the end-use environment and the governing specification.

Can helium leak testing detect all glass cracks?

No. Leak testing only detects through-going leak paths. Subsurface and non-through cracks, as well as interfacial delamination, require dye penetrant inspection, ultrasonic or acoustic micro-imaging as a screening tool, and destructive metallographic cross-sectioning for confirmation.

What causes insulation resistance degradation in glass seals?

Common causes include conductive surface contamination, moisture films on the glass, alkali ion migration within the glass under bias and humidity, and silver or copper migration from adjacent metallization. Measuring insulation resistance before and after damp heat exposure is the standard method for detecting these mechanisms.

Which standards should a supplier reference for acceptance testing?

MIL-STD-883 and MIL-STD-202 form the baseline for electronic component applications, IEC 60068-2 and GB/T 2423 cover general environmental testing, ASTM F15/F15M governs the Kovar material itself, and ISO 9227 or ASTM B117 apply to corrosion testing. Accredited third-party laboratories can consolidate these requirements into a single qualification and lot acceptance test plan.