Thermal Interface Material Selection: Thermal Pad vs Grease vs Phase Change — Conductivity and Rth

Overview #

The specification parameter that most procurement teams get wrong when sourcing thermal interface materials (TIMs) from China is not bulk thermal conductivity — it is thermal resistance (Rth) at the actual bond line thickness in their assembly. A pad rated at 6 W/m·K on a datasheet can deliver worse junction-to-case performance than a 3 W/m·K grease if the pad’s contact compliance is insufficient for the surface finish of the heatsink and component. We have seen this failure mode repeatedly in qualification programs for power electronics assemblies, and it is almost never caught until thermal cycling reveals it.

Thermal Conductivity vs. Rth: The Specification That Actually Matters #

Bulk thermal conductivity (λ, W/m·K) is the number printed on every TIM datasheet. It is also the number most buyers use to compare products. The problem is that λ alone does not determine thermal performance in an assembly — Rth (thermal resistance, °C·W⁻¹ or cm²·K/W) does, and Rth is a function of both λ and bond line thickness (BLT), plus contact resistance at both interfaces.

The relationship is straightforward: Rth = BLT / (λ × A), where A is contact area. A phase change material with λ = 3.5 W/m·K that wets to a BLT of 0.05 mm will outperform a silicone pad with λ = 6 W/m·K compressed to a BLT of 0.3 mm on the same surface. In our qualification testing, we routinely measure Rth directly per ASTM International ASTM D5470 (Standard Test Method for Thermal Transmission Properties of Thermally Conductive Electrical Insulation Materials) rather than relying on supplier-reported λ values.

The test conditions matter: ASTM D5470 results at 50 psi contact pressure are not comparable to results at 200 psi. Most Chinese supplier datasheets report λ at a single pressure point — often 50 psi — which is not representative of clamped heatsink assemblies that operate at 150–300 psi. When we request COAs from Chinese TIM suppliers, we always ask for Rth data at three pressure points: 50, 100, and 200 psi. Fewer than 40% of suppliers we have evaluated can provide this.

Most procurement teams over-specify λ and under-specify BLT tolerance and Rth at operating pressure. The difference between a pad that compresses to 0.25 mm and one that compresses to 0.40 mm under the same clamp load is not marginal — at 100 W dissipation, it can mean a 4–8°C difference in junction temperature, which directly affects MTBF in power semiconductor applications.

For buyers sourcing sealing and thermal interface materials from China, the internal link between TIM selection and heatsink surface finish specification is critical. A Ra > 1.6 µm heatsink surface will not achieve rated BLT with a rigid pad — this is a system-level specification failure, not a material failure.

Performance Under Three Industrial Operating Conditions #

Condition 1: High-Temperature Power Electronics (Continuous 85–125°C) #

In IGBT modules, motor drives, and EV power inverters, the TIM must survive continuous junction temperatures up to 150°C with cyclic thermal loading. The critical failure mode is pump-out (for greases) and delamination (for pads).

Silicone-based thermal greases with λ ≥ 6 W/m·K are the dominant choice here, but pump-out under thermal cycling is a documented failure mode. In our qualification program for a motor drive OEM, we tested five Chinese-sourced greases through 1,000 thermal cycles (−40°C to +125°C, 15-minute dwell per ASTM International ASTM D5470 protocol). Three of the five showed measurable Rth increase (>20% degradation) by cycle 500. The two that passed used non-silicone carrier fluids — a specification detail absent from the standard datasheet.

Phase change materials (PCMs) eliminate pump-out risk because they transition from solid to liquid at 45–55°C and re-solidify on cooling, conforming to surface topography without migration. For assemblies with >500 thermal cycles per year, PCM is the specification we recommend over grease, even at a 30–50% unit cost premium.

Condition 2: High-Humidity and Chemical Exposure (Industrial Enclosures, IP54–IP65) #

In outdoor power electronics, variable frequency drives, and industrial control panels, TIMs are exposed to humidity cycling, condensation, and occasional solvent cleaning. The failure mode here is not thermal degradation — it is adhesion loss and ionic contamination.

Silicone pads with acrylic pressure-sensitive adhesive (PSA) backing are the standard choice, but the PSA peel strength after 96-hour humidity exposure at 85°C/85% RH (per IEC Standards IEC 60068-2-78) is the parameter to verify. In our testing, Chinese-sourced pads from three suppliers showed PSA peel strength dropping from 0.8 N/mm to below 0.3 N/mm after this exposure — insufficient to maintain pad position during vibration. The specification threshold we use for qualification is ≥0.5 N/mm peel strength after 96h/85°C/85% RH.

Non-silicone pads (polyurethane or acrylic matrix) are preferred in environments where silicone contamination is a concern — particularly in optical assemblies or conformal coating processes, where silicone outgassing causes adhesion failures downstream.

Condition 3: High-Vibration and Mechanical Cycling (Automotive, Rail, Industrial Machinery) #

In automotive ECUs, railway traction inverters, and CNC servo drives, the TIM must maintain contact under continuous vibration (typically 5–500 Hz, 5–20 g per SAE International SAE J1211 environmental profiles) and mechanical shock. Grease pump-out is accelerated by vibration; pad delamination is the equivalent failure for solid TIMs.

Graphite sheets and phase change materials outperform silicone pads in vibration environments because they have no adhesive layer to fatigue. In our fatigue testing, silicone pads with PSA backing showed 15% Rth increase after 200 hours of vibration at 10 g / 100 Hz. Graphite sheets on the same test fixture showed <3% Rth change over the same duration.

The compressive load specification is critical in vibration applications: a pad that requires 150 psi to achieve rated BLT will relax under vibration if the fastener torque is not controlled. We specify fastener torque to ±10% and require pad compression set data per ASTM International ASTM D395 Method B — a test that fewer than 30% of Chinese TIM suppliers include in their standard documentation package.

Compliance, Certification, and What Chinese Suppliers Actually Provide #

Most Western buyers do not realize that Chinese TIM suppliers operate under SAC China Standards GB/T standards that do not require Rth measurement at multiple pressure points — only a single-point λ value is mandated for product certification. This means a GB/T-compliant Chinese TIM product may have a datasheet that is technically accurate but operationally incomplete for a Western engineering drawing that specifies Rth at 100 psi.

REACH compliance is the most commonly requested certification for TIMs entering the EU market. Silicone-based TIMs are generally REACH-compliant, but the filler materials — typically aluminum oxide, boron nitride, or aluminum nitride — must be verified for SVHC (Substances of Very High Concern) status. Boron nitride is not currently an SVHC, but aluminum compounds require verification against the current SVHC candidate list. We require a full REACH declaration of conformity, not just a checkbox on a supplier questionnaire.

EU RoHS Directive compliance is relevant for TIMs used in electronic assemblies sold in the EU. The key restricted substances to verify in TIM formulations are lead (in some legacy indium-lead alloy TIMs) and certain flame retardants used in pad matrix materials. Request the full RoHS test report, not just a supplier declaration.

For food processing and pharmaceutical equipment applications, NSF International NSF/ANSI 51 (food equipment materials) or NSF/ANSI 61 (drinking water system components) certification may be required. Very few Chinese TIM suppliers hold NSF certification — this is a genuine supply chain gap, and buyers in these sectors should plan for longer qualification timelines or consider Western-certified alternatives for direct food-contact applications.

In our supplier qualification program, we have seen suppliers pass initial sample approval with full documentation and then deliver out-of-spec material at production volume. The trigger is almost always a filler particle size change at the compounder level — something that a standard COA will not catch without incoming λ spot-testing and BLT measurement. We require incoming inspection at AQL 2.5 (per ISO Standards ISO 2859-1) for all TIM shipments above 500 units.

For buyers also managing cleanroom consumables procurement, note that TIM outgassing specifications (total mass loss <1.0% and collected volatile condensable material <0.1% per ASTM E595) are relevant for any assembly that enters a cleanroom environment — and this data is almost never included in a standard Chinese supplier datasheet without explicit request.

Practical Guidance for Buyers #

When sourcing thermal interface materials from China, the first specification to request from suppliers is not λ — it is Rth at your actual operating pressure, measured per ASTM D5470. Most buyers ask for λ because it is the headline number on every datasheet. The parameter that determines whether your assembly passes thermal qualification is Rth at bond line thickness, and that requires a pressure-specific test that most Chinese suppliers do not run by default.

The most common sourcing mistake we see is qualifying a TIM at 50 psi contact pressure when the production assembly operates at 150–200 psi. At higher pressure, BLT decreases and Rth improves — but if the pad’s compressive modulus is too high, the clamping force transfers to the component package rather than compressing the pad, causing mechanical damage. This failure mode appears at production volume, not at sample approval.

Before committing to volume order, require three consecutive batch COAs showing λ, BLT at 100 psi, and Shore A hardness (for pads). Require a thermal cycling test report (minimum 500 cycles, −40°C to +125°C) showing Rth stability within ±15% of initial value. If the supplier cannot provide this data, treat it as a disqualifying gap — not a negotiating point.

Frequently Asked Questions #

Q1: What is the most important specification to verify on a thermal pad COA from a Chinese supplier?
A: Rth at your operating contact pressure, not bulk λ. A pad with λ = 6 W/m·K compressed to 0.4 mm BLT delivers worse performance than a 3.5 W/m·K phase change material at 0.08 mm BLT.

Q2: When should I specify phase change material instead of thermal grease?
A: For assemblies exceeding 500 thermal cycles per year or operating above 100°C continuously, PCM eliminates the pump-out failure mode that affects silicone greases. In our thermal cycling tests (1,000 cycles, −40°C to +125°C per ASTM D5470 protocol), PCM showed <10% Rth degradation versus >20% for three out of five greases evaluated. The unit cost premium of 30–50% is justified by the reduction in field failure rate.

Q3: What is the most common quality failure when sourcing TIMs from Chinese suppliers at production volume?
A: Filler particle size drift at the compounder level. This is where most sourcing decisions go wrong — it does not appear on a standard COA, but it shifts λ by 15–25% and changes BLT behavior. The threshold we use is incoming λ spot-testing on every lot above 500 units, with a rejection criterion of ±10% from the qualified value.

Q4: What compliance documentation should I require for TIMs entering the EU market?
A: A full REACH declaration of conformity (not a checkbox) and a EU RoHS Directive test report from an accredited third-party lab. Supplier self-declarations are not sufficient for EU customs compliance in most member states. Verify the SVHC candidate list status of the filler material — aluminum compounds require specific verification.

Q5: Is higher thermal conductivity always better when selecting a TIM?
A: No. At λ above 6–8 W/m·K, the limiting factor in most assemblies shifts from bulk conductivity to contact resistance at the interfaces. Chasing λ = 12 W/m·K when your heatsink surface is Ra 1.6 µm is a specification error, not an upgrade.

Published by sinoraw.com Technical Team | Request a sourcing consultation