Advancing electronics technology with a pioneering study of gold alloys

A silent revolution is occurring in electronics: Many 20th-century solid state electronic devices are being replaced by tiny microelectromechanical systems, or MEMS. They outperform their solid state counterparts in some applications, use less power and are often cheaper to make.

MEMS’ commercial breakthrough came in the automotive industry where they were used in pressure sensors and actuators in seatbelts, airbags and stability systems.

They are now found almost everywhere in the electronic world, including in inkjet printers, Wii game controllers, iPhones, digital cameras and disposable blood-pressure sensors.

As with most electronics devices, the vast majority of MEMS devices are silicon-based – mainly because of the availability of the cheap, high-quality material. Silicon is also an almost perfectly elastic material, as it exhibits no time-dependent mechanical behavior. And it resists fatigue. One switch can go through trillions of cycles without breaking.

But silicon-based MEMS have their limitations, says Rick Vinci, professor of materials science and engineering.

“There are many potential MEMS applications, for example in satellite communications, that require a much higher current-carrying capacity than silicon can provide,” says Vinci. “This is something that can only be achieved with metals.”

When the dimensions of a metal component are reduced to those of a typical MEMS device, however, its mechanical properties start to change with time. The component becomes viscoelastic, making it challenging to design stable MEMS devices using metals.

For the past 10 years, Vinci’s group has studied the time-dependent behavior of thin metal films with a novel gas pressure bulge test apparatus designed at Agere by Walter Brown, a member of the National Academy of Engineering who is now adjunct professor of materials science and engineering.

Brown’s device uses gas pressure to inflate a thin metal membrane – a little like blowing up a soap bubble. The response of the membrane to the applied pressure helps determine its mechanical properties.

“The unique thing about this apparatus in comparison to other bulge test devices is that it uses the capacitance [the ability to hold an electric charge] between the bulged film and a fixed electrode, rather than optical measurements, to determine membrane displacement,” says Brown.

“It’s a fast and accurate way of measuring very small changes in displacement.”

For a 1-micron-thick film of pure gold, Vinci’s team has found a 27-percent reduction in elastic modulus (deformity) over three days at room temperature under constant strain conditions.

“The restoring force available to reopen a MEMS switch based on such a membrane would become significantly reduced over time and ultimately lead to failure,” says Vinci.

By measuring the properties as a function of temperature, Vinci and his team were able to determine the activation energy for this viscoelastic behavior. It turned out to be extremely small, on the order of 0.1eV.

“This suggested that a partial or step-wise movement known as a ’double-kink’ mechanism, a type of defect, is responsible for the time-dependent behavior,” says Vinci.

“You can test this theory by adding a small amount of an impurity such as vanadium oxide into the gold film, as impurities normally inhibit movement.”

Sure enough, when Vinci’s team did this, the vanadium oxide nanoparticles significantly reduced the amount of stress relaxation in the film. This helped to confirm the “double-kink” hypothesis and also to point the way to controlling stress relaxation, which will be key if metals are to be used in MEMS applications. Vinci remains optimistic.

“We believe these gold alloys could be the basis of the next generation of MEMS devices,” he says.

Vinci’s work on gold films is funded in part by Raytheon Company, BAE Systems, DARPA and the Common-wealth of Pennsylvania.