Ultrathin materials show rusty promise for next-gen electronics

Story

September 14, 2017

Move over, silicon: Two semiconductors share an odd “rusty” trait with silicon when it comes to insulating circuitry, but go way beyond to outperform it in other ways that show great promise for next-generation (next-gen) electronics.

Electrical engineers at Stanford University are working with two semiconductors – hafnium diselenide (HfSe2) and zirconium diselenide (ZrSe2) – and discovered that these ultrathin materials share or exceed some of silicon’s rather unusual, but highly desirable, properties. While their work is still in the experimental stage, the engineers report that these materials may very well open the door to the thinner, more energy-efficient chips that will be demanded by devices of the future.

The semiconductor industry has relied on silicon for electronics since the 1960s – it’s had an extremely good run – but next-gen electronics will require computer chips to become merely a few atoms thick. After more than 50 years of adhering to Moore’s Law, which essentially states that the number of transistors per square inch on circuits will double each year, chipmakers are on a quest to find alternative materials that behave like silicon, yet are capable of reaching way beyond the wall silicon has hit in terms of scaling.

One of silicon’s downright weirdest properties that helped establish it as the foundation for electronics is that it acts as a good natural insulator, thanks to the silicon dioxide, aka “silicon rust,” it produces.

Chipmakers typically expose silicon to oxygen during manufacturing to isolate the circuitry. Other semiconductors don’t share silicon’s “rusty” property and must be layered with other insulators, which poses many engineering challenges and can affect reliability.

But when the Stanford engineers put the hafnium and zirconium diselenides to the test, they made an intriguing discovery: These materials also form this elusive, high-quality insulating rust layer when exposed to oxygen. (Figure 1.)

As they began shrinking the diselenides to atomic thickness, the engineers realized that these materials share yet another of silicon’s quirky properties: They can switch transistors on within a specific range of band gap. If it’s too low, circuits leak and become unreliable; if it’s too high, a chip will require too much energy to operate. Surprisingly, both ultrathin materials fall into the same ideal band gap as silicon.

It’s actually a big deal because it means that these diselenides can be crafted into tiny circuits a mere three atoms thick, roughly two-thirds of a nanometer. This tiny size simply isn’t possible with silicon: Eric Pop, an associate professor of electrical engineering at Stanford, points out that engineers haven’t been able to make silicon transistors thinner than about five nanometers before the material properties begin to change in undesirable ways.

By combining thinner circuits and desirable high-k dielectric insulation (a material’s ability to concentrate an electric field), it essentially means that it may be possible to make these ultrathin diselenide semiconductors into transistors 10 times smaller than anything made with silicon today.

Don’t expect silicon to disappear overnight, but “for consumers, this could mean a much longer battery life and much more complex functionality if these semiconductors can be integrated with silicon,” Pop says.

What’s next? There’s much more work ahead to refine the electrical contacts between transistors on the group’s ultrathin diselenide circuits, according to Michal Mleczko, a postdoctoral scholar working with Pop. These sorts of connections “have always proven a challenge for any new semiconductor, and the difficulty becomes greater as we shrink circuits to the atomic scale,” he adds.

Another big hurdle standing in the way of ever-tinier electronics is how to find a way to better control the oxidized insulators to ensure that they remain both thin and stable. Once these two key pieces are resolved, the engineers can start to integrate the new materials with others and then scale up to working wafers, complex circuits, and eventually complete systems.

Despite the need for further research and work, “a path to thinner, smaller circuits – and more energy-efficient electronics – is within reach,” Pop says.

This work received support from the Air Force Office of Scientific Research, the National Science Foundation, Stanford Initiative for Novel Materials and Processes, the Department of Energy’s Office of Basic Energy Services’ Division of Materials Sciences, and a Natural Sciences and Engineering Research Council of Canada fellowship.

	(Click graphic to zoom)

Figure 1 | In this greatly enlarged cross-section of an experimental chip, the bands of black and white reveal alternating layers of hafnium diselenide – an ultrathin semiconducting material – and the hafnium dioxide insulator. Credit: Michal Mleczko.