Manufacturing Bits: Sept. 6

DARPA ALD; infrared cameras; total recall.


The University of Colorado at Boulder has developed an atomic layer deposition (ALD) technology that can be performed at room temperatures.

The technology, dubbed electron-enhanced ALD (EE-ALD), has been developed as part of the Local Control of Materials Synthesis (LoCo) program at the U.S. Defense Advanced Research Projects Agency (DARPA).

The LoCo program is developing technologies, such as deposition, that can be done at or near room temperature. The Department of Defense (DoD) is interested in developing technologies using ALD, which deposits materials one layer at a time at high temperatures. But the temperatures employed during the process exceed the limits of many DoD-relevant substrates.

At times, traditional ALD deposits materials at temperatures around 800 degrees Celsius. With EE-ALD, researchers deposited a gallium nitride (GaN) film on a silicon substrate at 27 degrees Celsius or 80 degrees Fahrenheit.

EE-ALD could devise a new class of thin-film microelectronics. Researchers from the University of Colorado have built a custom EE-ALD tool for use in processing 150mm wafers. During the process, researchers were able to control the electron energy during the ALD cycles. This, in turn, enabled them to tune the process to handle material deposition or removal. The removal process is sometimes called atomic layer etch (ALE.)

GaN film deposited on a silicon substrate at 27 degrees Celsius (Source: University of Colorado Boulder)

GaN film deposited on a silicon substrate at 27 degrees Celsius (Source: University of Colorado Boulder)

The group is also exploring other methods to etch specific materials, such as aluminum nitride and hafnium oxide. “Looking forward, the EE-ALD approach could serve not just as a tool for integrating incompatible materials but also more generally to build and etch device architectures at atomic scales, an increasingly important capability as circuit geometries shrink,” said Tyler McQuade, DARPA program manager, on the agency’s Web site.

Infrared cameras
In a separate move, DARPA has awarded HRL Laboratories funding to develop wafer-scale semiconductors for infrared detectors and sensors. This, in turn, would enable tiny infrared cameras.

HRL, a research venture between Boeing and General Motors, will also develop methodologies to integrate infrared materials with silicon read-out chips.

The project is part of DARPA’s Wafer Scale Infrared Detectors (WIRED) program. Digital cameras in smartphones can be manufactured quickly and inexpensively, but making tiny infrared cameras and sensors is expensive. The WIRED program, according to DARPA, will address the gaps in short wave infrared (SWIR) and mid-wave infrared (MWIR) imaging. This will be done by developing a low-cost detector technology using wafer-scale fabrication techniques.

“Time-consuming, conventional die-level serial processes will be replaced with streamlined wafer-level processes,” said Rajesh Rajavel, a senior scientist at HRL, on the venture’s site.

Total recall
Earlier this year, DARPA also awarded HRL funding for a two-year project in the RAM Replay program.

This project hopes to develop a next-generation electrode-embedded head cap. The cap will enable learning during sleep, which, in turn, will increase a person’s ability to integrate and recall information.

Researchers will devise a high-definition transcranial current stimulation (HD-tCS) head cap. The technology will supposedly tag specific memories and skills during sleep. The memory tagging and reactivation technique is called STAMP or Spatio-Temporal Amplitude-Modulated Pattern.

Sleep learning? (Source: HRL Laboratories)

Sleep learning? (Source: HRL Laboratories)

The STAMP method will be evaluated as volunteers perform operationally relevant tasks developed by the University of Southern California. Prior to human trials, the project’s research protocol will be approved by two institutional review boards at U.C. Berkeley and University of New Mexico. With partners Neuroelectrics and Soterix Medical, the team will also work to transition the memory-enhancing technology into a commercial product.

“We’ll develop a first-of-its-kind cognitive model of memory replays during sleep/wake stages to predict task performance that can be personalized to control the stimulation intervention for each user,” said Praveen Pilly, project leader and research staff member in HRL’s Center for Neural and Emergent Systems. “Students, employees, and athletes could potentially benefit, as could those with traumatic brain injury who have impaired memory.”