Below are the questions asked during the event, along with their respective answers.
Q: How do you decide which materials to mix with the liquid metals to make a hybrid solution?
A: There are several factors that go into determining what materials should be mixed with the liquid metals. First is the thermal conductivity of that material. Ideally, you want something higher in thermal conductivity to improve the overall performance of the system. You may also select materials based on solubility, corrosion resistance, and tendency to oxidize.
Q: Can these liquids or solid-liquid hybrids be mounted in a vertical position?
A: As long as you are working with a relatively thin bond line, these materials can be oriented vertically. The adhesive forces of the metals to the other surfaces prevent it from flowing out of the interface. This is similar to water between glass microscope slides. At some point, if you massively over apply the liquid metal to the interface, it will be possible to have the liquid flow.
Q: Does the physical structure of these hybrids change over time during thermal cycling or power cycling?
A: Depending on the composition of these hybrid TIMs, it is possible for their structure to change over time. For example, if there are solid particles that are soluble in Ga, they will dissolve into the liquid over time. Likewise, it is possible for the liquid metal to dissolve metal from surfaces it is in contact with. This all depends on the metals being used and the exposure temperature.
Q: Is the liquid metal TIM compatible with immersion cooling technique?
A: To my knowledge, this has not been studied but it seems like it would be risky. As the immersion liquid comes into contact with the liquid metal, I can see the possibility of displacement that potentially pushes some of the liquid metal out from under the interface.
Q: Can Ga based liquid metals be applied to anodized aluminum heat sinks?
A: There are published papers that indicate that anodization can prevent the gallium dissolution. This is not something that we have specifically studied but at least seems like a promising approach.
Q: Is there a way to keep this pumping action contained? Like some kind of seal around the sides?
A: It is definitely possible to create a seal using an epoxy or silicone ring. This is an added process step but does add a layer of protection.
Q: The CPU case is normally Aluminium – so the liquid metal suits are FPGAs only and similar chips?
A: Gallium based alloys are definitely not a drop-in replacement when coming in direct contact with aluminum. There are engineered solutions around this such as adding a metalized layer of something like Ni to the aluminum where it can contact the TIM. Alternatively, it may be possible to move away from aluminum entirely although that can be a significant design change.
Q: Can you give some application example which is already using this material for TIM. Automotive application?
A: There isn’t a lot of high volume use of liquid metal-based TIMs today. There are a number of niche applications. However, there has been a renewed interest over the past couple of years due to the required heat dissipation of higher power semiconductor devices.
Q: Tim can you talk to the bond layer thickness versus a traditional grease, and where LM is advantageous?
A: LM has a much higher bulk thermal conductivity than any thermal grease. This means that as you increase the bond line thickness, LM will degrade in performance much less than a grease would.
Q: Hi Tim, any suggestion for sealing material?
A: It is definitely possible to create a seal using an epoxy or silicone ring. This is an added process step but does add a layer of protection.
Q: Tim, but what is the total thickness with the solid-liquid hybrid versus traditional grease?
A: For a thermal grease, the key for success is having as thin a bond line as possible. This is application dependent but greases are often 25 microns or less. An SLH approach can be much thicker because of the higher thermal conductivity. They allow for much more ability to deal with a lack of planarity between objects. The SLH is often between 25-150 microns in thickness.
Q: What is the solid material thickness in the structure?
A: This is application-specific, but in many cases, the solid structure is between 25-150 microns.
Q: When will the SLH constructs be available for purchase and use? What Technology Readiness Level (TRL) is this at currently?
A: The SLH technologies consist of commercially available materials that have been used in other industries for many years. From a supply perspective, they are readily available. They are being tested in many use conditions today to better understand how they perform in an array of applications.
Q: What will be the minimum thickness achieved with SLH?
A: The minimum thickness is probably around 25 microns but that is highly dependent on what is being used as the solid material.
Q: Thank you for the great presentation. I would like to know if any problems occur when the temperature goes below the range shown?
A: Solidification of the liquid metals is a significant hurdle to overcome. As these alloys solidify, they expand by about 3%. Repeated solidification and re-melting creates additional oxidation that can hinder performance. Sealing around the liquid can stop airflow and oxidation.
Q: What typical applications do you envisage for this technology. Do you have specific examples you can discuss?
A: The most pressing need today is for TIM1 applications where the thermal interface material is in direct contact with the backside of the die. This is where the most heat dissipation is required and where metals can have the biggest impact.
Q: Are liquid metals being widely used in high heat applications?
A: I know liquid metals are being used as a TIM in applications where the junction temperature is reaching at least 115C. Not sure if this qualifies as high heat per the question.
Q: I may have missed it but what is the range of electrical conductivity of your SLH’s?
A: SLH TIMs are generally in the range of 30-70 W/mK. This is highly dependent on the solid material used in conjunction with the liquid.
Q: Are there any limitations regarding the applicable area dimensions?
A: There are no real limitations. As the area grows, planarity typically becomes a bigger issue. Managing the bond line between non-flat surfaces can become difficult with liquid or hybrid materials.
Q: Can liquid metal be used as an anti-corrosion coating for copper tubes in evaporators with alu fins?
A: Unfortunately, this is outside my area of expertise.
Q: Are there SLH’s available in small quantities so we can test compared to our existing thermal grease?
A: Yes
Q: Have you tested Gallium based LM’s with magnesium?
A: This has not been tested at this point.
Q: What Ni plating thickness on Aluminum is required to prevent the dissolving issue?
A: In this case, you need to be sure that the Ga doesn’t dissolve the Ni so it can come in contact with the Al. If you are running at 100C or less, the solubility of Ni is almost zero so you can get away with a very thin layer. If for some reason, this is seeing very high temperatures for long periods, you will need a greater Ni thickness.
Q: How about liquid metal compatibility with AlN ceramic?
A: I would expect this to be ok but it is not something that I am aware has been tested before.
Q: Can you talk about the oxidation processes of gallium? Is a gasket the only way to limit it?
A: Like many metals gallium will oxidize over time. Generally, this oxide is self-limiting at about 5 angstroms. However, if you disrupt the surface, the oxides can continue to grow. Sealing is one way to limit oxidation because you are limiting airflow. You can also play with the alloy to add elements that are more resistant to oxidation.
Q: How do you limit the travel of the liquid metal?
A: Volume of liquid is the best way to control the movement. If you are able to apply a thin layer, the adhesive forces will cause the liquid to stick to the surfaces and not easily move. If a higher volume of liquid is needed, you will have to consider some type of sealing process.
Q: Are the SLH constructs commercially available? What TRL (technology readiness level) would you rate them for military/space applications?
A: The SLH technologies consist of commercially available materials that have been used in other industries for many years. From a supply perspective, they are readily available. They are being tested in many use conditions today to better understand how they perform in an array of applications.
Q: What about the compatibility with copper and Gallium? Is there any test done by you?
A: There are a number of studies showing the solubility rates of copper into gallium at a range of temperatures. We have done some studies and have some limited data.
Q: What’s the chemical impact of liquid metal as a TIM onto Aluminum Alloy?
A: This is a difficult question to answer without specifics on the alloy. Certain alloys are probably ok but this would need to be validated on a case-by-case basis.