You are browsing the archives of heat sink.

ORNL’s James Klett holds an LED streetlamp. The lamp will use heat sinks of graphite foam
(samples in his left hand) to extend the life of the LEDs and cut operating costs.

Around 1997, Oak Ridge National Lab’s James Klett and Timothy Burchell discovered how to make graphite foam, a material that had at least one amazing property: It transfers heat like crazy.

If this property of the foam seems a little counter-intuitive, that’s because foam materials are often associated with with heat insulation properties. But in this case, the foam acts as a super heat radiator. A story in an ORNL newsletter said the stuff worked so well that if you put an ice cube on a hockey puck-sized chunk of the graphite foam, and put the foam on you hand, “the cube melts from your body heat as if it were on a hot griddle.”

At the time, Klett, a researcher in the lab’s Metals and Ceramics Division, noted that, “Graphite foam is as thermally conductive as aluminum at one-fifth the weight. It has a very high surface-area-to-weight ratio and a high heat transfer coefficient. This interests engineers and designers because products that use energy wage an ongoing battle with heat,” he says.

He said the key to the foam’s conductivity is its unusual graphite crystal structure that is full of air pockets, making it only 25% dense and lightweight. A network of graphite “ligaments” in the foam wicks heat away from its source.

Klett shows that ice held against the graphite foam will melt quickly because the heat from the hand holding the foam is transferred rapidly through the foam. As a result, this hand feels the cold fast.

When they made their discovery, Klett and Burchell were building on a legacy of carbon innovations that go back to at least the 1960s when Johhn Googin developed the first method to produce carbon foams was used as high-temperature furnace insulation. Klett and Burchell also developed a commercial carbon-carbon disk brakes system.

Over the past decade, Klett, Burchell and ORNL have licensed the special foam for numerous applications – especially with mechanical and electronic heat-transfer applications – and the material garnered an R&D 100 award.

Now, the foam’s ability to act as an efficient heat sink is being put to new uses in the world of energy-efficient lighting. On Friday, ORNL announced that it has licensed the foam to LED North America for use in commercial LED lighting systems such as in the large arrays now being manufactured for street lamps and parking garages.

The lab says passive cooling materials, such as the foam, are needed to increase LED efficiency and lifetime. ORNL reports that each 10° decrease in temperature can double the life of the lighting components. “While this technology will reduce temperatures and increase the life of the LED lighting systems, what it will really do is save municipalities millions of dollars every year in replacement fixture costs as well as maintenance,” Klett said.

Besides being lightweight, Klett says the foam is easy to machine and use in manufacturing. These advantages give it a growing edge compared to traditional heat transfer materials, such as copper or aluminum.

LED North America president Andrew Wilhelm predicts that the foam will double the life of the LED units. He also says the first lamps using the foam will be installed later this year in an ORNL parking lot.

Share

Diamond crystal with a carbide film (white). Credit: © Fraunhofer IFAM

Researchers at one of Germany’s Fraunhofer Institute think they have come up with a better material to dissipate heat in electronics, using a material that combines traditional materials like copper with diamond power.

A novel material is needed because of the new challenges arising as more and more previously separated electronic units are being combined on single chips. More components mean more heat in a shrinking volume, yet components and connecting elements need to be kept in the safe zone – 90°C to 130°C.

Currently, copper or aluminum plates are attached to the chip’s components to act as heat sinks. But, the expansion and contraction of the metal plate (which can expand about three or four times as much as the silicon and other ceramic material in the rest of the chip) often leads to cracks and breaks at solder points.

Fraunhofer staffers at the Institute for Manufacturing Engineering and Applied Materials Research IFAM in Dresden have been looking for an alternative to the copper or aluminum plates. Their ideal material would be something that conducts heat better and has a small expansion coefficient, and these researchers think they have found it.

“We did this by adding diamond powder to the copper. Diamond conducts heat roughly five times better than copper,” says Thomas Schubert, project manager at IFAM. “The resulting material expands no more than ceramics when heated, but has a conductivity one-and-a-half times superior to copper. This is a unique combination of properties.”

The researchers faced one major hurdle: How to bond the diamond and copper? “One ingredient we can use to achieve this [bond] is chrome. Even small amounts form a carbide film on the diamond surface, and this film easily bonds to copper,” Schubert explains.

Fraunhofer is partnering on this project with Siemens and Plansee, among others, as part of the EU project “ExtreMat.” ExtreMat is an interesting effort to accelerate novel material development. Its website describes it as a project that “assembles this critical mass of experts from different materials-related industries, research centers, universities and science institutes in Europe in a multi-sectorial and cross-cutting approach. The ExtreMat project consortium currently comprises 37 institutions in 12 European countries. It forms the unique pool of expertise, competencies and facilities which is needed to fulfill the ambitious objectives of this Integrated Project.”

The Fraunhofer effort is part of a special ExtreMat subproject “to develop novel heat sink composites and to test their performance under extreme loading conditions. By extreme, they mean up to 1000°C. The route that’s been laid out is to use a copper matrix reinforced with ceramic and intermetallic (nano) particles or fibers. Besides something with a thermal conductivity similar to copper, they are looking for materials that also have a heat flux removal capability of up to 20 MW/m2.

Share