Below are the questions asked during the event, along with their respective answers.
Q: How can reflow technology help in the reflow of solder TIMs for TIM1 applications?
A: When forming a solder joint as a thermal interface material, the two critical aspects that need to be monitored are voiding and wetting. High levels of voids in the solder joint will result in poorer heat transfer. Similarly, if wetting is poor, there will be areas of high interfacial resistance which also impacts heat transfer. The reflow profile and atmosphere used should be optimized by assessing these two attributes as response variables in a DOE. Additionally, the use of vacuum reflow technology can be considered to help reduce the amount of voiding.
Q: Are there any significant handling and use concerns when dealing with liquid metals?
A: Liquid metals used for thermal applications typically consist of gallium, indium, and/or tin. All three of these metals are quite benign and do not possess any significant environmental or human health risks. The one important precaution to consider is that gallium is classified as a corrosive material. It is incompatible with certain metals, most notably aluminum. Precaution should be taken to ensure gallium is not in contact with any critical aluminum surface.
Q: How big of a risk is it to use metals since they are electrically conductive?
A: Metals are electrically conductive. They should not be used in applications where the thermal interface material also needs to provide electrical isolation. In addition, you will want to avoid contact electrical circuitry as there is risk of shorting.
Q: What percentage of void would be bad for heat dissipation performance? Is there any rule of thumb number?
A: Voiding will impact thermal performance. More voiding will have more impact. The amount of voiding that is acceptable will be highly dependent on the heat dissipation requirements of any IC or module. In most thermal applications of high-performance devices, people are looking to keep voiding less than 10%. Some applications have even more stringent requirements.
Q: Is there any dissipation performance difference in the open void and closed void of the structure?
A: In addition to the amount of voiding, the location of the void can impact thermal performance. Specifically, if the void is located at either of the interfaces themselves, it is likely to be more impactful than if the void was in the bulk of the solder. The reason for this is that solder has very good heat conduction in all directions. Therefore, it would be relatively easy for heat to travel around a void in the bulk. If the void is at the interface, it will effectively reduce the area in which the heat can pass from one object to another and reduce the amount of heat dissipation that would be possible.
Q: What is the gap size and variance that the heat-spring technology can accommodate?
A: The Heat-Spring can accommodate an overall gap of around 350 microns. In terms of lack of planarity, it can accommodate up to about 150 microns in thickness variation.
Q: Could you give us some numbers in terms of power generated (watts) by these quantum chips?
A: These high-performance packages are often over 200 watts. Overall wattage is an important part of understanding heat dissipation but you also need to look at chip size. A lower wattage chip that is small could require a higher performance TIM than a higher wattage one that is much larger.
Q: Any recommendation on TIM for automotive application…
A: There are many automotive thermal applications with wildly varying requirements. Metal TIMs can be a viable option in some of these applications where high heat dissipation is needed. For example, a power module for an electric vehicle could use a metal TIM when attaching it to the heat-sink.
Q: Slide 5, item 5: what is meant by “reduced Rth versus air over large gaps” as a primary desired TIM function?
A: For thermal interface materials, it is easy to say that I want the highest thermal performance. However, in reality, there are often trade-offs. One example is in the area of gap filling. Many of these applications do not require super-high thermal performance (but it does need to be better than nothing (or air). In these cases, you are looking for something with ease of use to fill that large gap and are often willing to sacrifice on thermal performance to get something that can be easily implemented in a production process.
Q: These solutions seem rather expensive and the majority of TIMs being used are TIM2. What is your opinion on TIM2 and what do you see as their future?
A: Cost is certainly an important factor in looking at TIMs. Nobody is going to use a more expensive material than they actually need. The examples that I shared are growing areas where the use of metal TIMs make a lot of sense. There will always be plenty of applications where a very low cost grease is “good enough.” In TIM2 applications, the key is to look at whether you value the attributes that a metal TIM can bring. These attributes could include thermal performance, pump-out resistance (long life), and/or compatibility with cooling systems like liquid immersion or cryogenics.
Q: Are patterned metallic TIMs suitable for TIM 2 applications? How does the cost compare to traditional TIMs?
A: Patterned metallic TIMs are being used in TIM2 applications today. They are going to be substantially more expensive than most dispensed grease type materials. When comparing to other pre-cut pad materials, they can be similar to some of those high thermal conductivity PCM options. The key is whether the attributes of the metal TIM make it worth the cost. In a growing number of applications, that answer is yes.
Q: Any partner product using patterned metal TIMs at this moment?
A: Patterned metallic TIMs are currently used in a number of production applications.
Q: Can you speak a little more about the “mismatch” you said occurs during the fitment between the dye and the heat spreader? This was a problem with liquid metals but with compressible heat springs, this is negated right?
A: The mismatch is a mismatch in CTE. Regardless of the TIM material used, that mismatch will be there. Some TIMs are more resilient at absorbing the pumping action that can occur when a device is powered on and off. Generally solid or high viscosity fluids will be more resistant to this “pump out” that can occur with low viscosity fluids.