1. Introduction

Thermally conductive adhesives combine structural bonding with thermal interface material (TIM) functions in a single bond line, and they have become standard in electric-vehicle (EV) power electronics, where junction temperatures of silicon-carbide (SiC) devices can exceed 150 deg C under continuous load. This article presents three production case studies - an inverter power module, an onboard charger (OBC), and a battery cold-plate interface - with measured thermal and mechanical data. Thermal conductivity is tested per ASTM D5470 and lap-shear strength per GB/T 7124-2008. Dongguan DENSON Functional Materials Co., Ltd. (DENSON) supplies carbon-black and functional color pastes compatible with thermally conductive adhesive systems.

2. Technical Features and Mechanism

A thermally conductive adhesive is defined as a filled polymer adhesive that conducts heat through a percolation network of ceramic or metallic particles while maintaining mechanical adhesion. Typical matrices are epoxies, silicones, and polyurethanes; fillers include alumina (Al2O3, 60-80 wt%), boron nitride (BN), aluminum nitride (AlN), and sometimes silver or graphite for high-performance grades. Thermal conductivity rises sharply once filler loading crosses a percolation threshold, typically 55-65 vol%; a well-formulated product reaches 1.0-3.0 W/(m K) while retaining enough flow to wet the interface. Silane coupling agents such as KH-560 reduce filler-matrix interfacial thermal resistance (Kapitza resistance), and bond-line thickness is usually controlled at 50-200 um with glass beads or spacer particles.

Measurement conditions matter when comparing datasheets: ASTM D5470 uses a guarded heat-flow meter under controlled clamping pressure, and bulk conductivity values measured on cast coupons can overstate in-joint performance because real bond lines contain micro-voids, thickness variation, and surface roughness. Engineers should therefore request both bulk conductivity and thermal resistance at the specified bond-line thickness, typically 50-200 um, together with data after thermal aging (for example 1000 h at 150 deg C), since epoxy post-cure, silicone reversion, and filler sedimentation can all shift performance over service life.

3. Case Study 1: Inverter IGBT/SiC Module to Pin-Fin Cooler

An 800 V traction inverter replaced thermal grease and mounting clamps with a two-part silicone thermally conductive adhesive between the SiC module baseplate and an aluminum pin-fin cooler. The adhesive measured 2.2 W/(m K) (ASTM D5470), bond-line 150 um, and lap-shear strength 2.1 MPa (GB/T 7124). In a 120 kW continuous-load dyno test, the SiC junction-to-coolant thermal resistance Rth(j-w) dropped by approximately 16% versus the greased assembly, and peak junction temperature fell by 9-11 deg C. After 1000 thermal cycles (-40 to 125 deg C, IEC 60068-2-14), X-ray inspection showed no delamination, and the assembly passed IEC 60664-1 creepage requirements with UL 94 V-0 rating.

4. Case Study 2: Onboard Charger (OBC) Magnetic Component Potting

A 22 kW OBC potted its transformer and PFC inductors with a two-part polyurethane thermally conductive potting compound (1.1 W/(m K), Shore A 60, dielectric strength above 20 kV/mm per IEC 60243-1). The low-modulus compound absorbed CTE mismatch between copper windings and ferrite cores: winding hot-spot temperature at 45 deg C ambient and full load decreased from 138 deg C to 124 deg C, and audible transformer noise fell by 4-6 dB(A). After 85 deg C/85% RH damp heat for 1000 h, insulation resistance remained above 500 MOhm and no cracks were observed on sectioning.

5. Case Study 3: Battery Pack Cold-Plate Gap Filling

A blade-cell battery pack compared three TIMs between the cell carrier and liquid cold plate: a 2.0 mm silicone gap pad (3.0 W/(m K)), a dispensable silicone gap filler (3.5 W/(m K)), and a structural epoxy thermally conductive adhesive (1.8 W/(m K), 0.8 mm bond line). Although the pad had the highest bulk conductivity, total interface resistance was lowest for the adhesive because of its thin, void-free wet contact: measured pack thermal resistance per cell was 0.31 K/W (adhesive) versus 0.38 K/W (gap filler) and 0.45 K/W (pad). The adhesive also eliminated 14 fasteners per module and passed GB 38031-2020 thermal propagation, crush, and random-vibration (ISO 16750-3) tests; shear strength remained above 7 MPa after 500 temperature cycles.

Process control was decisive in all three cases. Meter-mix equipment with gear pumps and static mixers with at least 18 elements kept mix-ratio drift within +-1.5%; bead patterns were designed to push air toward a single escape edge; and in-line pressure monitoring on the dispensing robot flagged blocked nozzles before starved bond lines occurred. First-article inspection combined cross-sectioning, X-ray void analysis (acceptance below 2% void fraction), and infrared thermography under power, giving a repeatable acceptance baseline before ramp-up.

6. Key Selection Parameters and Recommendations

Recommendations: specify total interface resistance rather than bulk conductivity alone; control metering ratio within +-2% for two-part materials; run X-ray or acoustic-microscopy checks on first articles; and validate color pastes for cure compatibility. See the DENSON website and the Literature section for compatible color paste data.

7. Conclusion

In EV power electronics, thermally conductive adhesives reduce total thermal resistance primarily through thin, void-free, permanent interfaces rather than through headline conductivity values. Selection must balance conductivity, modulus for CTE stress, temperature rating, flame requirements, and dispensing process windows, with final validation on thermal-cycle and vibration tests at pack level.

8. FAQ

Q1: Is a higher W/(m K) adhesive always better for EV electronics?
A1: No. Total thermal resistance includes contact resistance and bond-line thickness; a 1.8 W/(m K) adhesive at 0.8 mm with full wetting can outperform a 3.0 W/(m K) pad at 2 mm with trapped air, as the battery cold-plate case demonstrated.

Q2: What causes voids in dispensed thermally conductive adhesive?
A2: Voids come from air entrained during mixing, excessive dispense speed, moisture on the substrate, or filler settling. Use static mixers, controlled bead patterns, degassed cartridges, and clean dry surfaces, then verify with X-ray inspection.

Q3: Silicone or epoxy thermally conductive adhesive for inverter modules?
A3: Silicones tolerate higher temperatures and lower stress but have low shear strength and risk siloxane contamination; epoxies give structural strength and chemical resistance but are rigid. Choose by stress analysis, and use primers where CTE mismatch is large.

Q4: Can thermally conductive adhesives be tinted black?
A4: Yes; carbon black pastes are common and also aid UV stability, but loading must be controlled so as not to disturb the ceramic filler network or poison catalyst cure. DENSON provides heat-resistant black pastes with compatibility test reports.