A room-temperature bonding technique for integrating wide bandgap materials such as gallium nitride (GaN) with thermally conducting materials such as diamond could boost the cooling effect on GaN devices and facilitate better performance through higher power levels, longer device lifetime, improved reliability, and reduced manufacturing costs. The technique could have applications for wireless transmitters, radars, satellite equipment, and other high-power and high-frequency electronic devices.

The technique, called surface-activated bonding, uses an ion source in a high-vacuum environment to first clean the surfaces of the GaN and diamond, which activates the surfaces by creating dangling bonds. Introducing small amounts of silicon into the ion beams facilitates forming strong atomic bonds at room temperature, allowing the direct bonding of the GaN and single-crystal diamond to fabricate high-electron-mobility transistors (HEMTs).

The resulting interface layer from GaN to single-crystal diamond is just four nanometers thick, allowing heat dissipation up to two times more efficient than in the state-of-the-art GaN-on-diamond HEMTs by eliminating the low-quality diamond left over from nanocrystalline diamond growth. Diamond is currently integrated with GaN using crystalline growth techniques that produce a thicker interface layer and low-quality nanocrystalline diamond near the interface. Additionally, the new process can be done at room temperature using surface-activated bonding techniques, reducing the thermal stress applied to the devices.

“This technique allows us to place high thermal conductivity materials much closer to the active device regions in gallium nitride,” said Samuel Graham, the Eugene C. Gwaltney Jr. School Chair and professor in Georgia Tech’s George W. Woodruff School of Mechanical Engineering. “The performance allows us to maximize the performance for gallium nitride on diamond systems. This will allow engineers to custom design future semiconductors for better multifunctional operation.”

The research, conducted in collaboration with scientists from Meisei University and Waseda University in Japan, was reported February 19 in the journal ACS Applied Materials and Interfaces. The work was supported by a multidisciplinary university research initiative (MURI) project from the U.S. Office of Naval Research (ONR).

For high-power electronic applications using materials such as GaN in miniaturized devices, heat dissipation can be a limiting factor in power densities imposed on the devices. By adding a layer of diamond, which conducts heat five times better than copper, engineers have tried to spread and dissipate the thermal energy. 

However, when diamond films are grown on GaN, they must be seeded with nanocrystalline particles around 30 nanometers in diameter, and this layer of nanocrystalline diamond has low thermal conductivity – which adds resistance to the flow of heat into the bulk diamond film. In addition, the growth takes place at high temperatures, which can create stress-producing cracks in the resulting transistors.

“In the currently used growth technique, you don’t really reach the high thermal conductivity properties of the microcrystalline diamond layer until you are a few microns away from the interface,” Graham said. “The materials near the interface just don’t have good thermal properties. This bonding technique allows us to start with ultra-high thermal conductivity diamond right at the interface.” 

By creating a thinner interface, the surface-activated bonding technique moves the thermal dissipation closer to the GaN heat source.

“Our bonding technique brings high thermal conductivity single crystal diamond closer to the hotspots in the GaN devices, which has the potential to reshape the way these devices are cooled,” said Zhe Cheng, a recent Georgia Tech Ph.D. graduate who is the paper’s first author. “And because the bonding takes place near room temperature, we can avoid thermal stresses that can damage the devices.”

That reduction in thermal stress can be significant, going from as much as 900 megapascals (MPa) to less than 100 MPa with the room temperature technique. “This low stress bonding allows for thick layers of diamond to be integrated with the GaN and provides a method for diamond integration with other semiconductor materials,” Graham said.

Beyond the GaN and diamond, the technique can be used with other semiconductors, such as gallium oxide, and other thermal conductors, such as silicon carbide. Graham said the technique has broad applications to bond electronic materials where thin interfacial layers are advantageous.

“This new pathway gives us the ability to mix and match materials,” he said. “This can provide us with great electrical properties, but the clear advantage is a vastly superior thermal interface. We believe this will prove to be the best technology available so far for integrating wide bandgap materials with thermally conducting substrates.”

In future work, the researchers plan to study other ion sources and evaluate other materials that could be integrated using the technique. 

“We have the ability to choose processing conditions as well as the substrate and semiconductor material to engineer heterogenous substrates for wide bandgap devices,” Graham said. “That allows us to choose the materials and integrate them to maximize electrical, thermal, and mechanical properties.”

In addition to the researchers already mentioned, the paper included co-corresponding author Fengwen Mu from Meisei University and Waseda University in Japan, Luke Yates from Georgia Tech, and Tadatomo Suga from Meisei University.

This research was supported by the U.S. Office of Naval Research (ONR) through MURI Grant No. N00014-18-1-2429. Any findings, conclusions, and recommendations are those of the authors and not necessarily of the Office of Naval Research.

CITATION: Zhe Cheng, Fengwen Mu, Luke Yates, Tadatomo Suga and Samuel Graham, “Interfacial Thermal Conductance across Room-Temperature-Bonded GaN/Diamond Interfaces for GaN-on-Diamond Devices” (ACS Appl. Mater. Interfaces, 2020, 12, 8376?8384). https://doi.org/10.1021/acsami.9b16959

Research News
Georgia Institute of Technology
177 North Avenue
Atlanta, Georgia  30332-0181  USA

Media Relations Contact: John Toon (404-894-6986) (jtoon@gatech.edu).

Meet Oliver Brand, executive director of the Institute for Electronics and Nanotechnology (IEN) at Georgia Tech.

IEN is one of Georgia Tech's 10 interdisciplinary research institutes (IRIs) within the Georgia Tech Research enterprise.

What is your field of expertise and why did you choose it?

My research is in the area of Micro Electro Mechanical Systems or MEMS and, in particular, the development of micro-scale physical, chemical and biological sensors, which are fabricated using processes similar to the ones used to make integrated circuits. I was first introduced to this area at the beginning of my Ph.D. in the early 1990s and was initially fascinated by images of these micrometer-sized devices. We quickly learned that these beautiful, tiny structures and devices can have many useful applications. 
 
What makes the way in which your IRI enables campus research unique? 
My field of research MEMS and micro/nanotechnology in general rely on expensive equipment to fabricate these tiny devices and systems. To enable research in this area, IEN maintains extensive core facilities for fabrication and characterization at the micro- and nanoscale. These facilities are not only used by Georgia Tech researchers, but by dozens of companies and other academic institutions. Many of these companies are start-ups that couldn’t do the work they do without access to these facilities. In addition, nanotechnology is a highly interdisciplinary field and the IRIs help bring together the interdisciplinary teams to tackle big problems. 
 
What couldn’t have happened without your IRI? 
IEN supports an incredibly talented team of research faculty and staff that have the goal of enabling nanotechnology research, development and commercialization done by Georgia Tech faculty, students, and external partners. Many nanotechnology research accomplishments at Georgia Tech and beyond could have simply not happened without the dedication of this team. 
 
What impact is your research having on the world? 
Let me pick a timely example, the development of COVID-19 tests. Many of these tests have nanotechnology components in order to detect the virus that has nanometer dimensions by itself. Over the past year, we have been heavily involved in an NIH-funded program called Rapid Acceleration of Diagnostics or RADx, where we assist project teams developing novel tests with engineering expertise but also access to the core facilities I mentioned earlier. Some of the tests we have supported now already are commercially available. 

Learn more about IEN.

The Nano Group at the Georgia Tech Research Institute develops novel applications for carbon nanotubes and graphene based on their highly distinct electrical and physical properties. This group has developed thriving partnerships both in industry and government that led to research across a variety of ranging from functional fabrics and 2D materials to photovoltaics and supercapacitors.

This installment of the Faces of Research Q&A series is with Jud Ready, who leads the Nano group, is deputy director of innovation initiatives at Georgia Tech's Institute for Materials (IMat), and is an adjunct professor in the School of Materials Science & Engineering.

What is your field of expertise and why did you choose it?
I am a materials scientist. I applied for admission to Georgia Tech in 1989 for aerospace engineering. I was declined admission to that major, but was instead admitted to Tech as a "UEC" (undecided engineering candidate). A then-requirement of that "major" was to take three one-hour courses in three different majors during your freshman year to help you decide by your sophomore year what you wanted to do with your life. The three I chose were:  civil engineering (CE), materials science engineering (MSE), and electrical engineering (EE). After AE, MSE most resonated with my interests of becoming an astronaut.  Microgravity materials science was my interest at the time, but my interests evolved towards materials used in electronics, and ultimately my doctorate was on copper metallurgy (corrosion of printed wiring boards).

What makes Georgia Tech research institutes unique?
The "embedded" connection to the academic side of Georgia Tech gives us an unfair advantage over our competitors by having so many innovative academic resources (people, students, labs, etc.) right at our fingertips.  

What impact is your research having on the world?
We are in the final stages of integrating a spacecraft that will map the water-ice resources on the moon: gtri.gatech.edu/newsroom/lunar-flashlight. We expect it to launch Dec. 27, 2022.

What do you like to do in your spare time when you are not working on your research or teaching?
I am a former Boy Scout and current scout leader. I enjoy camping, volunteerism, and community service. I am also a youth-league sports coach for my children and am now — after three kids — in my 13th season overall! Basketball is my favorite sport; I love to dribble the ball up and down the court. I am also an enthusiastic supporter of Georgia Tech athletics, and am an instrument-rated airplane and helicopter pilot.

Genetic screens and drug screens play an essential role in the understanding of gene functions and the development of therapeutics. Genetic screens can identify genes contributing to a defect or disease state, while drug screens search for treatments that can restore normal function.

Traditionally, large-scale screens have been performed on single cells because of their small size and cost-efficiency, but with the limitation of missing the complexity of whole animals.

Screening animals allows for tackling phenotypes more relevant to human diseases, in particular dynamic phenotypes such as brain activity, muscle activity, or behavior. Such screens present huge benefits for the study of neuropsychological disorders and behavioral disorders. However, the challenge is that handling animals is far more challenging than handling single cells  – especially when the need is to scale up to the thousands.

A team of researchers led by Hang Lu, a professor in Georgia Tech’s School of Chemical and Biomolecular Engineering, published a paper in the journal Small on a technological platform that would solve this bottleneck using droplet microfluidics.

Encapsulating microscopic animals in tiny droplets transposes the problem into a different perspective. Instead of manipulating animals directly, one now aims at handling a “passive” object containing an animal, Lu said. Working with droplets also has several advantages: no cross contamination, small volumes, and a gentle way of trapping and transporting the animal.

“Using droplet microfluidics allows for processing animals faster, user bias-free, and with nanoliter amounts of reagents. This platform is ideal to study the animal reactions to dynamic alterations of its chemical environment,” Lu said.

Studying C. elegans

As a proof of application, the team worked with C. elegans, a freely living nematode that varies in size from a couple hundred microns at the larval stage to one millimeter once adult. C. elegans is a widely used model organism in genetics that was first introduced in the 1960s. Several Nobel Prizes have been rewarded to researchers working on this model.

Besides having its genome fully sequenced and connectome fully determined, this mainly-hermaphroditic nematode has a large progeny, short reproduction cycle, and is inexpensive to maintain. Altogether, these attributes have established C. elegans as a key organism for large-scale screens. Traditionally, the screens have been performed manually and required months, if not years, of graduate student time and been often limited to somewhat simple animal responses.

The droplet platform may speed up that process as well as broaden the scope of assays performed on the animals. “We have developed a droplet toolbox that allows us to manipulate the animals and their environment and perform elaborate automated protocols” explained Guillaume Aubry, a research scientist in the Lu lab.

The team members first created an animal encapsulation system where each animal is packaged in an aqueous droplet. Then they created various droplet manipulators to alter the animal’s environment. In particular, a liquid exchanger allows researchers to swap the animal from one droplet to another. Finally, they established a method for programming protocols and performing tasks automatically.

Increased Throughput, Applications

The researchers demonstrated a throughput of more than a hundred animals per hour for cases of observing the animal behavior in response to a chemical stimulant. They also showed that this platform can be used to monitor brain activity simultaneously, which opens countless applications in neuroscience to decipher neuronal circuitry, dynamics, and correlate brain activity with behavior.

The team explained that the platform is adaptable to other small animals in the micro- to millimetric range. Because the design principle relies on droplets, the system is scalable to accommodate the animal size. As an example, the team scaled down the system to work with the first larval stage of C. elegans, which is considerably smaller, and demonstrated monitoring the larvae’s neuronal activity to a sudden change of chemical environment.

“Ultimately this platform will help to bridge the gap between environment, genotype, and phenotype,” Aubry said. “The platform opens up exciting possibilities in particular for the study of neurodevelopmental disorders. With this scalable tool, one can envision monitoring the development of single individuals and assessing their behavioral and neuronal activities over time. Another exciting avenue is to take further advantage of the nanoliter volumes to pursue drug screens, for example on the effect of neuronal activity.”

Funding

The authors acknowledge National Institute of Health (NIH R21NS117066, R01NS096581, and R01AG056436) for funding. This work was performed in part at the Georgia Tech Institute for Electronics and Nanotechnology, a member of the National Nanotechnology Coordinated Infrastructure (NNCI), which is supported by the National Science Foundation (Grant ECCS-1542174).

Citation

G. Aubry, M. Milisavljevic, and Hang Lu “Automated and Dynamic Control of Chemical Content in Droplets for Scalable Screens of Small Animals” Small 18, 2200319 (2022) DOI: 10.1002/smll.202200319 https://onlinelibrary.wiley.com/doi/full/10.1002/smll.202200319 (Supplemental Movie 5 and Supplemental Movie 7 of the Supporting Information of the paper)

Alex Weidenbach is a graduate research assistant and Ph.D. student at Georgia Tech working with W. Alan Doolittle. In the following Q&A, Weidenbach briefly discusses his work in the IEN cleanroom and gives advice to current and future users.

How long have you been using the IEN Cleanroom?  

I was hired as an intern at the Institute for Electronics and Nanotechnology (IEN) in 2014 during my undergraduate studies at Georgia Tech. In this role, I worked with both the processing and equipment teams and gained a wide range of skills in just about every aspect of cleanroom work. Those skills ultimately led to a job at Axion Biosystems, a local BioMEMS company, upon graduation. While at Axion, I continued to work in the inorganic cleanroom on proprietary research on microelectrode arrays (MEAs). I also developed some new fabrication processes to increase Axion’s manufacturing throughput. I returned to Georgia Tech to pursue my Ph.D., and I am currently in my fifth year of Ph.D. studies under Professor W. Alan Doolittle. All in all, I have been working in the inorganic cleanroom for the past seven years, and I have done everything from tool maintenance to consulting to academic research.

What tools do you use when you are in the cleanroom and what are you doing? 

I primarily use the Denton Discovery 2 sputterer to co-deposit lithium-containing films to make memristive devices for neuromorphic computing applications. To make these films into devices, I frequently use a plethora of tools in the IEN cleanroom which includes the SCS G3P8 Spinner, Karl Suss TSA MA-6 Mask aligner, Heidelberg MLA 150, Vision RIE, Plasma Therm ICP, CtrLayer AET RTP, and the SSI RTP. I regularly need photolithography, dry etching, and rapid thermal annealing to finish my devices.

What is/has been your favorite project you have worked on in the IEN cleanroom?

My own research on memristive devices has been the most rewarding work I’ve done in the IEN cleanroom, though I’m not sure there has been a project that I did not enjoy. I find fabrication to be mentally satisfying and personally fulfilling, so I find enjoyment in the cleanroom work itself. I especially enjoy interacting with other users, recommending tools, and trying to help improve their process flows.

What advice do you have for people thinking about using a tool in the IEN cleanroom?

My advice to future cleanroom users is to make sure you get trained on multiple tools that can perform the same process. Having backup tools ready in case your favorite tool goes down or you run into issues is an absolute must, and it will save you countless hours in the future. Plan ahead and get trained on as many tools as possible. Also, take care to understand how the tool works and what exactly the tool is doing rather than just learning how to operate it. By knowing what is going on inside the chamber of the tool you are using you can more easily debug your process when you inevitably run into problems or challenges with your devices.

What is your favorite thing about the IEN Cleanroom? 

My favorite things about the IEN cleanroom are the number of tool options and the amount of space available to quickly prototype new devices and explore fabrication processes. There are not many general-use cleanrooms set up to do what IEN does at the scope in which it operates. Having so many tools available really makes exploring new fabrication techniques and replicating research easier.