Graphene and Beyond: The Wonder Materials That Could Replace Silicon in Future Tech

Researchers at the University of Manchester create exotic materials only 1 to 2 nanometers thick, in a vacuum chamber.

Formulate silicon just right, shape it into a transistor, and it can be both a conductor and an insulator, depending on the charge you run through it—a fundamental property without which the entire digital revolution, and the internet, and everything from TikTok to Covid vaccines would be impossible.

But silicon is showing its age. The reliable biennial doubling in the computational power of microchips, known as Moore’s Law, has been slowing, and could soon come to an end. It’s pretty much impossible, using current methods, to get the elements etched into silicon, like transistors, below about 3 nanometers in their smallest dimension. (To put that in perspective, a 3-nanometer film can be as few as 15 atoms thick.) So the tech industry is in search of other wonder materials to take good old silicon’s place—or at least combine with it to vastly increase its capabilities.

A prototype device made using silicon and graphene that can detect coronavirus particles, displayed by UT Austin’s Deji Akinwande and a colleague.

PHOTO: UNIVERSITY OF TEXAS AT AUSTIN DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING

Researchers on the bleeding edge of physics, chemistry and engineering are experimenting with exotic-sounding substances to be used in microchips. They include graphene, black phosphorus, transition metal dichalcogenides, and boron nitride nanosheets. Collectively, they’re known as 2-D materials, since they are flat sheets only an atom or two thick. Largely unknown just 20 years ago, they are now regularly fabricated in labs, using methods as mundane as a blender and as tricky as high-temperature vapor deposition.

Some of the results of this research can already be found in devices on sale today, but the bulk are expected to turn up over the next decade, bringing new capabilities to our gadgets. These will include novel features such as infrared night-vision mode in smartphones, and profound ones such as microchips that are 10 times faster and more power-efficient. This could enable new forms of human-computer interaction, such as augmented-reality systems that fit into everyday eyeglasses.

Sounds like science fiction? In fact, some of it is. Many potential applications of these 2-D materials won’t be realized anytime soon, for any number of reasons, including how difficult they are to work with and integrate into existing electronics, or to fabricate at a billions-of-units-a-year scale.

In their quest, researchers must sort through all of these potential materials. A small army is examining them in simulation and in the lab, at dozens of universities, and at IBM, Samsung, TSMC, GlobalFoundries and just about every other big chip design or manufacturing firm in the world. They’re looking for the right combination of desirable traits and manufacturability—the ability to be produced reliably in large quantities.

A nanoscale device created in the University of Manchester’s ultrahigh vacuum chamber is about the size of a grain of pollen.

PHOTO: THE UNIVERSITY OF MANCHESTER

The granddaddy of all 2-D materials is graphene. If we could shrink ourselves down, Magic School Bus-style, and hover just above its surface, graphene would look like a flat plane of hexagons made from carbon atoms. Graphene is like the graphite in a pencil, only arranged in a flat crystal.

Its existence was theorized in the 1940s, but it was not properly synthesized and characterized until 2004, by a pair of researchers. (It won them a Nobel Prize.)

Graphene is strong, and has a talent for conducting heat, so it has already found applications such as keeping smartphones and their batteries cool, and extending the life of athletic gear. As its properties are more akin to other conductors like gold or copper, it’s not likely to replace silicon. But it does have a number of unique characteristics that make it useful when combined with traditional silicon microchips.

San Diego-based startup Cardea Bio already markets such a combined system. Its new sensor, a “biology-gated transistor,” attaches biologically active molecules, for example certain antibodies, to a sheet of graphene, which is in turn attached to silicon. Graphene isn’t only a great conductor, it’s exquisitely sensitive to anything touching it that might interfere with its conductivity.

“The beautiful thing about biology is that it’s technology—there’s organized complexity to it,” says Cardea Bio Chief Executive Michael Heltzen. Graphene allows the analog world of biology to be translated into the digital world that human engineers and human-built systems can manipulate and gather data from, he adds.

Startup Cardea Bio’s “biology-gated transistor" can interact with particles such as antibodies or microbes.

PHOTO: CARDEA

Cardea Bio’s system is currently being sold in an instrument used by researchers, but the chips could someday scan liquids for specific organic molecules—that is, just about anything an organism large or small puts into the environment—says Mr. Heltzen. The company announced that, as part of a project at the Georgia Tech Research Institute, funded by the Defense Advanced Research Projects Agency, it is creating a sensor that could detect coronavirus particles in a stream of air. Such a device inside a building could detect SARS-CoV-2, soon after anyone exhales particles of the virus. If it works, it could eventually lead to systems that could be reprogrammed to detect other pathogens.

Other graphene-silicon team-ups on the horizon include ultrathin, ultrasensitive cameras, says Deji Akinwande, a professor at the University of Texas at Austin who researches 2-D materials. That’s because graphene can yield optical sensors a hundred times more sensitive to light than ones made with silicon. In addition, because it can “see” in a wider range of the electromagnetic spectrum, graphene-based materials could make possible tiny, inexpensive, high-resolution infrared cameras of the sort that could fit into smartphones. Already in the prototype stage, this technology could give our smartphone cameras the ability to see the heat generated by objects.

In addition, 2-D materials’ handling of light could lead to even more meaningful upgrades to our devices by the middle of this decade, Dr. Akinwande and his colleagues predict. Light would be a far faster and more efficient means of communication within and between microchips and other components inside a computer, hastening the replacement of electrons with photons inside microchips and communications networks.

Another area where 2-D materials could have a big impact, says Dr. Akinwande, would be in stacking microchips one atop another, like a high-rise. Silicon stacking is already common in flash memory and in mobile-device chips where space is at a premium, such as inside the Apple Watch.

Dr. Akinwande examines graphene-based transistor arrays used to take electronic measurements of biological processes.

PHOTO: UNIVERSITY OF TEXAS AT AUSTIN DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING

Because 2-D materials are just an atom or two thick, they can be either grown atop silicon microchips, or they can be grown separately, and then carefully placed. This has two advantages over just stacking layers of silicon, says Deep Jariwala, an engineering professor at the University of Pennsylvania who specializes in nanotechnology. The first is that many can be stacked without adding appreciable height to a chip. The second is that some 2-D materials, especially graphene, dissipate heat so well, engineers could use them to create chip high-rises that run even faster than conventional microchips, without burning themselves out.

At the University of Manchester, researchers created an ultra-clean facility for stacking 2-D materials atop one another. Because these materials are easily damaged by air, all such manipulations must be done in a vacuum chamber.

Translating this kind of cutting-edge manufacturing into something that can happen in the world’s giant microchip factories—the so-called “fabs” run by companies like TSMC, Samsung and GlobalFoundries—is the real trick to bringing 2-D materials to the real world, says Peter Barrett, a venture capitalist at Playground Global who invests in companies that work on next-generation microchips and their materials.

“The story of silicon’s success is its manufacturability,” says Mr. Barrett.

Perhaps, years or decades from now, once some of these novel 2-D materials are understood well enough, and billions have been spent on rolling them out at the scale of the global semiconductor industry, one or more of them might come to replace silicon in some of its primary applications inside our computers, says Dr. Jariwala.

One such 2-D material that has shown promise because, unlike graphene, it is a good semiconductor, is molybdenum disulfide, he adds. It has already been used to create flexible electronics and a simple microprocessor. And it is hardly alone in its suitability as a potential silicon replacement: It’s part of a large family of hundreds, if not thousands, of promising materials. One challenge, as with so many of these substances, is that making and handling them can be difficult.

In the meantime, says Mr. Barrett, novel applications of existing silicon chips, like quantum computing and brain-mimicking “neuromorphic” computing, will drive engineers to push silicon to its absolute physical limits, and in the process pave the way for what comes next.

From shrinking electronics and cracking codes to cloud computing and artificial intelligence, the demands on our hardware are growing faster than current tech can keep up. Enough demand, coupled with enough progress in the lab, he adds, could eventually justify the huge investment required to bring 2-D materials to center stage.

Write to Christopher Mims at christopher.mims@wsj.com