Copper and aluminum have been used to make cooling systems for computers and other applications for years. But sometimes those coolers' performance is insufficient, due to the characteristics of those elements. Researchers from Chalmers University of Technology in Sweden may have a solution. In a recent research paper, they proposed using graphene and carbon fiber to make heatpipes with significantly increased performance.
A heatpipe is a fairly simple device. A tube made of copper or aluminum is partially filled with water or ammonia-based compound. Next, the air is removed, and then the tube is sealed. One of the pipe's end is applied to a hot spot, another is placed somewhere where the temperature is lower. Once the temperature of the working liquid gets too high on the hot end, it vaporizes into a gas and quickly transfers itself into a colder part — the so-called condenser section — of the tube, thus, transporting heat. Once the gas gets cold, it condenses back into liquid and returns to the hotter area. The cycle can continue virtually endlessly.
In PCs, the condenser section is usually equipped with a heatsink, so performance and efficiency of such heatpipes greatly depends on the performance and efficiency of the radiator. But there are also applications that cannot use heatsinks, and their performance depends on thermal conductivity of the pipes themselves. That makes thermal conductivity of the heatpipes' materials crucial.
Thermal conductivity of copper is about 400W/mK at atmospheric pressure and 20 degrees Celsius (68 degrees Fahrenheit) but varies with temperature. By contrast, thermal conductivity of graphene is in the range of 3,000 - 5,000W/mK at room temperature.
(Image credit: Chalmers University of Technology/Phys.org)
Researchers from Chalmers University of Technology, along with scientists from China and Italy, point to graphene films as a way to increase the thermal conductivity of heat pipes and improve performance of cooling solutions meant to be installed into constrained spaces. The tubes would be aided by carbon fiber wicker-enhanced inner surfaces.
Evidently, heatpipes made of graphene will be far more efficient than heat pipes made of copper. The scholars tried 150mm-long pipes with a 6mm outer diameter and found that their thermal transfer coefficient was approximately 3.5 times better than that of a similar copper-based heat pipe.
But that efficiency comes at a price. According to estimates, graphene costs about $100 per gram. By contrast, copper is $0.0079 per gram, as of December 2020. Graphene is very light, and not a lot would be needed for a heatpipe, (depending on dimensions, of course), but graphene tubes would still be prohibitively expensive for many applications.
Stay On the Cutting Edge: Get the Tom's Hardware Newsletter
Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.
In fact, graphene heatpipes hardly make much sense for already-efficient desktop PCs that can be equipped with large copper radiators and multiple fans. But for space and weight constrained environments, like avionics and automotive and space electronics, such pipes may be what the doctor ordered.
Interestingly, the researchers said graphene heat pipes could also be used with traditional heatsinks with fans.
"The condenser section, the cold part of the graphene enhanced heatpipe, can be substituted by a heatsink or a fan to make the cooling even more efficient when applied in a real case," said Ya Liu, Ph.D. Student at the Electronics Materials and Systems Laboratory at Chalmers.
Anton Shilov is a contributing writer at Tom’s Hardware. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.
I think the limiting factor is the fin-to-air interface rather than the ability of the heat pipes to transfer but maybe I'm wrong.
I would like to see someone try a cooler with a textured surface or micro perforations on the aluminum fins that increases effective surface area and air contact.
I've often wondered why the heatpipes aren't soldered to the fins 360 degrees around but as the article notes air coolers are already quiet effective and cost efficient.
What we need is a new standardized form factor for the CPU Place that is FIXED . once this is done , it will be very easy to make heatpipes coolers span to the rear of the Case and uses 120/140 fan to mount a very huge radiator(fins) that continue from the rear of the case to the middle of the motherboard allowing for a very huge cooler to be mounted on the case itself and not on the board.
I think the limiting factor is the fin-to-air interface rather than the ability of the heat pipes to transfer but maybe I'm wrong.
Much of the heat has been extracted before the last few cooling fins, so better heat pipes will help noticeably.
However, to support your statement, there is a molecular layer of (mostly) nitrogen adsorbed onto the surface of the fins which has a significant effect on the fin-to-air thermal conductivity. It seems like I read a research brief on some attempt to directly address that issue (short of using higher fan RPM).
Taken together, an improved heat sink could be half the size (perhaps leaving more room for an integral Peltier/TEC, if you don't mind making a lot more heat, though the current TEC designs are not terribly bulky).
I wonder if the CPU manufacturers could use Graphene to make a small heatpipe cap to place over the bare CPU chips (there seems to be a ton of empty space inside), replacing the solder and lid. This would probably function much better than even a delidded CPU would.
However, to support your statement, there is a molecular layer of (mostly) nitrogen adsorbed onto the surface of the fins which has a significant effect on the fin-to-air thermal conductivity.
The air you breathe is already 75-80% nitrogen, so there effectively is a "film of nitrogen" on everything exposed to typical air.
The air you breathe is already 75-80% nitrogen, so there effectively is a "film of nitrogen" on everything exposed to typical air.
Yes, a "film of nitrogen" is adsorbed on to most or all surfaces exposed to atmosphere. The question is how to minimize the thickness and/or strength of the adsorbed layer to improve heat transfer to air that is freely moving. I wish I could find the article I was thinking of, but here is an image that might help to illustrate the issue: Multi-Layer Adsorption
