To build its microphones, the company has partnered with facilities that fabricate other types of chips. As with other chips, Akustica’s are made of layers of semiconductor materials, insulating materials, and metals. To add a microphone, the engineers modified the design of a few layers to make a membrane out of the same metal used in the circuitry. After this, Gabriel says, the chip is post-processed, which includes steps to etch away some of the silicon on the chip to reveal the buried metal membrane. This final processing, he adds, can be done inexpensively and quickly.
Because it’s relatively inexpensive to mass-produce semiconductors, the Akustica technique “could become the lowest-cost method to building a microphone,” says Jonathan Bernstein, a researcher at Cambridge, MA-based Draper Laboratory, which works with small microphone systems. However, he adds, there is a tradeoff when using a semiconductor process to make a device like a microphone. A potentially higher-quality acoustic system could be made, he says, if the design weren’t limited to the same materials used in the circuit.
And while integrating a microphone into a circuit could be advantageous for some applications, it may not be necessary for others, says Levent Degertekin, professor of mechanical engineering at the Georgia Institute of Technology in Atlanta. “Hearing-aid manufacturers don’t want those capabilities built in,” he says, because they have their own digital circuitry that works well. Instead, he says, the best application right now for single-chip microphones is the laptop market.
Akustica, which spun out of a Carnegie Mellon University research project in 2001, began shipping its first microphone earlier this year. The device is currently used in some Fujitsu laptops, and last month the company announced a second product combining two microphones on one wire to capture more sound. It is expected to appear in products by year-end.