Below are the questions asked during the presentation, along with their respective answers.
Q: How do metal thermal interface materials compare with graphite products?
A: Metals are thermally isotropic which means they conduct heat equally in all directions. Graphite is much more of a heat spreader. It will have very high conductivity in the X/Y direction but relatively poor in the Z direction. Both products are good options for very different applications.
Q: What industries are likely to need and use metal thermal interface materials?
A: Metal based thermal interface materials are often used as a TIM1 or TIM2 when high heat dissipation is required and the required lifetime is long.
Q: How much of electrical isolation can we have in terms of voltage difference between die and heat sink?
A: Because metals are electrically conductive, you cannot expect to have electrical isolation when using a metal TIM. If electrical isolation is required, a different type of thermal interface material will be required.
Q: If the metal is used for TIM, there are any problem about the electrical conductivity?
A: Metal based TIM materials are electrically conductive. If an application require electrical isolation, these are not going to be the appropriate materials.
Q: Is it possible to mix or embed a liquid metal material into a typical silicone gap filler matrix?
A: This scenerio is unlikely. It may be possible to combine liquid metal with a silicone gap filler but only if the gap filler is already cured and in a very porous state. Then, the liquid metal could be infiltrated into the pores. If this would work, the benefits will likely be limited due to the relatively poor performance of silicone materials.
Q: How do the compressible metal foil and Solid Liquid Hybrid metal TIM compare to Silicone Based Gap pad type TIMs?
A: In general metal based TIMs are going to have much higher thermal conductivity than silicone based materials. Silicone based materials have two features that make them viable in a number of applications. First, they are very cheap compared to most other TIMs. If you don’t have a high thermal demand, silicone materials can be just fine. The second attribute is that they generally exhibit low interfacial resistance. If you are able to apply a thin bondline, the silicone material performance can be pretty good at time zero. Over time, these materials will outgas and dry out so you can expect some deterioration as time goes on.
Q: Which part is a patterned TIM most useful?
A: Patterned TIMs are most often used in TIM2 applications. The patterning dramatically improves the compressibility of the material. For example, a flat foil of pure indium might require 100 PSI to get compression. By patterning the foil, you can reduce the required pressure by more than 50%.
Q: Is there a way to simulate interfacial thermal resistances?
A: The standard for measure the overall thermal resistance of a system is the ASTM D5470. The thermal resistance obtained from this test will take into account both the thermal conductivity of the TIM material itself as well as the interfacial resistances. Because this is a standardized test, it is very useful for comparing materials. It is possible to take the results from this test method and estimate the impact of the interfacial resistance using a little math.
Q: What is the thermal conductivity of such metallic TIM (in x ,y and z direction)?
A: Metals are thermally isotropic which means they conduct heat equally in all directions. For metals that are commonly used as thermal interface materials, they generally have a thermal conductivity between 35 and 86 W/mK. The actual conductivity is dependent on the specific alloy used.
Q: Whether the TIM 1 and TIM 2 materials need to be same?
A: TIM1 and TIM2 materials don’t have to be the same. Most of the time, they are not the same material. In most cases, the performance requirements for the TIM1 are more stringent. As this is in direct contact with the die, it is critical that this material is effective at drawing the heat away. For the TIM2 applications, the heat is already away from the die and you are typically trying to pull the heat from a heat spreader to the heat sink. Here, the surface area is typically larger, so the demands on the TIM2 are not quite as high.
Q: Are Ga based TIMs compatible with copper heat sinks as opposed to aluminum?
A: Copper and nickel both have very low dissolution rates in gallium and are suitable metals to use. Because of copper oxidation, it is common to use nickel plated copper.
Q: What is the volumetric change at phase change of some common liquid interface materials?
A: Most of these alloys shrink slightly upon solidification and expand on melting. However, you should expect these changes to be very small (<1%). If trying to use a metal as a phase change material, you may need to take this into account if the spacing is tight.
Q: How would you compare metallic TIM to polymer based TIM’s?
A: There are many thermal interface materials that are polymer based. However, when people refer to a polymer TIM, it is usually a material that cures at elevated temperature to make a permenent bond. A polymer TIM has relatively low thermal conductivity that is usually less than 6 W/mK. A metal TIM will be much higher. Polymer TIMs also tend to be somewhat brittle after curing. This could cause failures during power or thermal cycling.
Q: How do you rework such applications?
A: Liquid metals can be reworked. The liquid metal will easily wipe off of a surface with soapy water. If you are re-applying liquid metal, you don’t even need to worry about complete removal.
Q: What are the suitable materials for TIM 2?
A: There are many options that can be considered for a TIM2 application. Before selecting a TIM2 material, it is important to lay out the critical material properties prior to material down selection. In addition to thermal conductivity, you will want to know if it needs to fill a large vs small gap, how much pressure can be applied, surface planarity, and potential needs for electrical isolation. Identifying these requirements can help you select the most appropriate materials.
Q: Can you discuss the effect of density change in PCM’s? Are material that expand upon phase change preferable? Can voids or high pressure / leakage happen after phase change?
A: As noted in the question, materials will tend to expand/contract during melting or solidification. Most metals will contract during solidification. Typically, this is less than a 1% volume change. In general a change in volume is not desirable no matter if it is expansion or contraction. The volume change can result in some mobility of the liquid. With the repeated phase change of a metal, you can see dewetting due to oxidation. This could result in voids. If the TIM is under significant pressure, the liquid can also squeeze out resulting in a possible electrical short of adjascent circuitry.
Q: How does hybrid TIM made with metal particles and viscous organic oils compare with metal-metal hybrid TIM in terms of performance?
A: The thermal conductivity of a liquid metal is going to be much higher than any organic material that would be used to embed particles. The overall performance of a solid liquid all metal TIM will transfer the most overall heat.
Q: Do these matters require burn in?
A: When using an all liquid or a solid/liquid hybrid, no burn in is required. When using a compressible metal TIM, the performance will improve over time as the material requires time to achieve the plastic deformation required to fill all air gaps. This can be accomplished with a 12-hour pre-load of pressure onto the TIM.
Q: You talked about pump-out due to expansion/contraction, but how important is it when using a Hybrid liquid metal to be able to control the gap between the 2 faces and match the volume of the TIM to it so you don’t have squish-out during initial assembly under the pressure of the heatsink?
A: Process optimization is critical with using a hybrid metal TIM solution. If you apply too much liquid, it is certainly possible to get squeeze out of the metal. The thicker the solid metal preform, the more liquid it can absorb. There is not a single optimized condition as every application may require slightly different ratios of the two materials.