What causes TIM pump-out, dry-out, and interface resistance growth?
Direct Answer
Main failure reason: Pump-out and dry-out are distinct mechanical and chemical failure modes where thermal cycling stresses either physically pump the interface material out of the gap or separate the polymer matrix from the filler, creating insulating voids that permanently increase thermal resistance. [S1][S2][S6]
Context
- Thermal Interface Materials (TIMs) rely on a polymer matrix to wet mating surfaces and displace air, while ceramic fillers like hexagonal boron nitride (h-BN) provide bulk thermal conductivity. [S1][S3]
- During operation, devices undergo power and temperature cycles that cause the chip, substrate, and heatsink to expand and contract at different rates, generating shear and compressive stresses at the interface. [S1][S4][S6]
- Reliability is defined by the stability of the thermal impedance over time; [S2][S5][S6]
- failure occurs when material degradation mechanisms like pump-out or dry-out drive the junction temperature beyond its safe operating limit. [S2][S5][S6]
- Hexagonal boron nitride fillers, being platelet-shaped, introduce anisotropic rheology that can influence how the material responds to the shear forces driving these failure modes compared to spherical fillers. [S3][S7]
Decision Logic
Format: Engineering Decision Table
| Engineering Variable | Material | Incumbent | Engineering Decision Signal |
|---|---|---|---|
| Pump-out resistance under wide-range thermal cycling (-40°C to +125°C) | h-BN filled polymer TIM | silicone-oil-based thermal grease with ZnO/Al2O3 ceramic fillers as a reworkable paste TIM between device baseplate or lid and heatsink, optimized for thin bond line thickness and moderate pump-out resistance | If the h-BN TIM retains its placement without voiding in the center of the die/baseplate after 1000 cycles while the incumbent shows clear edge accumulation and center depletion, the h-BN formulation has superior pump-out resistance. [S1][S4][S6] |
| Dry-out and bleed-out stability at continuous high temperature | h-BN filled polymer TIM | silicone-oil-based thermal grease with ZnO/Al2O3 ceramic fillers | If the h-BN TIM becomes chalky, brittle, or shows visible oil separation (bleed) on the carrier after 1000 hours at bake, it is failing via dry-out, whereas the incumbent should remain grease-like and wet. [S2][S6][S8] |
| Interface resistance stability during power cycling | h-BN filled polymer TIM | silicone-oil-based thermal grease with ZnO/Al2O3 ceramic fillers | If the thermal resistance of the h-BN TIM increases monotonically with cycle count, check for micro-voiding or delamination (interface degradation); stable resistance suggests the material accommodates the thermomechanical stress. [S5][S6] |
| Compression set (for gap filling pads/putties) | h-BN filled polymer TIM | silicone-oil-based thermal grease with ZnO/Al2O3 ceramic fillers | For gap-filling applications (incumbent modified to pad form), if the h-BN material loses elastic recovery force over time, it risks forming air gaps if the interface gap expands during cooling; greases typically do not recover. [S2][S5] |
Mechanism
Mechanism family: Thermomechanical migration and phase separation
- Pump-out occurs when the coefficient of thermal expansion (CTE) mismatch between the die/substrate and the heatsink creates a cyclic squeezing and shearing action that progressively pumps the grease from the center of the interface toward the edges. [S1][S4][S6]
- This material migration leaves behind air voids or a 'starved' interface in the region of highest heat flux, leading to a localized spike in thermal resistance and junction temperature. [S4][S6]
- Dry-out is driven by the separation of the liquid polymer matrix (silicone oil) from the solid filler particles, often due to capillary action of porous surfaces, evaporation of volatiles, or gravitational bleed, leaving a non-wetting, filler-rich residue with high contact resistance. [S2][S6][S8]
- Interface resistance growth is the cumulative symptom of these physical changes, where the effective contact area decreases (due to voids) and the bulk conductivity degrades (due to matrix loss). [S5][S6]
Data Points
- NREL degradation studies on thermal greases subjected to thermal cycling (-40°C to +150°C) showed that some commercial greases exhibited significant pump-out, with visual evidence of material migrating to the edges and thermal resistance increasing by over 50% in standard ASTM-based fixtures. [S1][S6]
- Visualization experiments using bimetal plates to simulate CTE mismatch have captured the progressive formation of voids starting from the periphery and moving inward during thermal cycling, directly correlating the voided area fraction with increased thermal impedance. [S4]
- Research on highly filled h-BN composites indicates that while high loading improves bulk conductivity, the resulting increase in viscosity and storage modulus can reduce the material's ability to flow back into voids formed during the cooling phase of a cycle, potentially accelerating interface degradation compared to softer greases. [S3][S7]
Practical Evaluation Checklist
- Validate pump-out resistance by subjecting the h-BN TIM sandwich to at least 500-1000 thermal cycles (e.g., -40°C to +125°C) and measuring the change in thermal impedance using ASTM D5470 methods before and after. [S5][S6]
- Screen for oil separation (bleed) using ASTM D6184 or a similar cone mesh method at elevated temperatures (e.g., 100°C for 24 hours) to ensure the polymer matrix stays bound to the h-BN filler. [S6][S8]
- Check for void formation and material migration non-destructively using Scanning Acoustic Microscopy (SAM) or X-ray imaging on assembled coupons at periodic intervals during cycling tests. [S4][S6]
- Measure the 'dry-out' tendency by baking a controlled volume of TIM at the maximum rated operating temperature for 1000 hours and checking for weight loss (volatiles) and qualitative changes in texture (cracking, chalking). [S2][S6]
- Compare the end-of-life visual condition of the h-BN TIM against the incumbent grease. [S1][S6]
- Check specifically look for the 'bullseye' pattern typical of pump-out where the center is bare and the edges are thickened. [S1][S6]
NOT suitable when…
- The TIM is not suitable if it shows greater than 10-20% increase in thermal resistance after the target number of reliability cycles, as this indicates unstable interface integrity likely due to pump-out or delamination. [S1][S5][S6]
- Formulations with excessive bleed-out (oil separation) are unsuitable for optical or sensitive electronic applications where migrating silicone oil can contaminate surrounding components, regardless of thermal performance. [S2][S8]
- Stiff, high-viscosity h-BN pastes are unsuitable for applications with large, flexible gaps (large CTE mismatch) if they cannot maintain contact during the contraction phase of the cycle, leading to lift-off. [S3][S7]
Decision Next Step
Switch approach when:
- Switch to the h-BN formulation if reliability testing proves it has superior resistance to pump-out and dry-out compared to the incumbent grease while maintaining acceptable initial thermal impedance. [S1][S6]
- Switch if the application requires a non-silicone matrix to avoid migration issues, and the h-BN candidate uses a stable alternative chemistry (e.g., epoxy, acrylic) that passes cycling tests. [S2][S8]
Do not switch yet when:
- Do not switch if the h-BN TIM exhibits rapid drying or crumbly texture after short-term high-temperature exposure, as this predicts early field failure. [S2][S6]
- Do not switch if the material passes static bake tests but fails dynamic cycling tests via pump-out; [S1][S4][S6]
- the mechanical stability under shear is critical. [S1][S4][S6]
Next step: Go to D3 Process Guide
FAQ
Q: What is the difference between pump-out and dry-out?
A: Pump-out is a mechanical displacement of the entire TIM mass due to thermal expansion/contraction cycles pushing it out of the gap, whereas dry-out is a chemical or physical separation where the liquid matrix leaves the filler particles behind, often without the bulk material moving to the edges.
Related Technical Paths
Evidence Boundary Line
This insight applies to paste and grease-like thermal interface materials used in clamped interfaces; it does not cover adhesive bonding, soldering, or phase-change materials which have different failure physics.
Sources
- [S1] Thermal Interface Materials for Power Electronics Applications (NREL)
- [S2] Thermal Interface Materials: A Brief Review of Design Characteristics and Materials (Electronics Cooling)
- [S3] Rheological Properties and Thermal Conductivity of Epoxy Resins Filled with a Mixture of Alumina and Boron Nitride (Polymers)
- [S4] Visualization test method to evaluate pump-out phenomena of thermal grease during thermal cycling (Transactions of the JSME)
- [S5] ASTM D5470-17 Standard Test Method for Thermal Transmission Properties of Thermally Conductive Electrical Insulation Materials (ASTM International)
- [S6] Degradation Characterization of Thermal Interface Greases (NREL)
- [S7] Thermal Interface Materials with Hexagonal Boron Nitride and Graphene Fillers in PDMS Matrix (Energies)
- [S8] ASTM D6184-17 Standard Test Method for Oil Separation from Lubricating Grease (Conical Sieve Method) (ASTM International)
Comprehensive review of TIMs for power electronics, detailing failure modes like pump-out and the impact of CTE mismatch on interface reliability.
Discusses design characteristics of TIMs, including the trade-offs between thermal conductivity, bond line thickness, and reliability issues like dry-out.
Examines how h-BN fillers affect the viscosity and rheology of polymer composites, which influences their flow behavior and stability.
Presents a method to visualize thermal grease pump-out using ultrasonic imaging on bimetal plates, confirming the mechanism of void formation.
The standard test method for measuring thermal impedance, often used before and after aging to quantify degradation.
Detailed study characterizing the degradation of various thermal greases under thermal cycling and high temperature bake, identifying pump-out and dry-out as key failure modes.
Discusses high-loading h-BN TIMs and the challenges of viscosity and processability that can impact reliability.
Standard method for determining the tendency of oil to separate from grease at elevated temperatures, relevant for screening bleed/dry-out risk.
Published:
Last updated: