Industrial Physics Forum 2013: Frontiers in nanomanufacturing

By Devin Powell

In the wake of President Obama’s call for 15 new manufacturing hubs across the country, the third session of this year’s Industrial Physics Forum focused on Frontiers in Nanomanufacturing. Four speakers addressed the promises and pitfalls of nanoscale approaches to innovation.

Nanomanufacturing matchmaking

To kick off the session, Alexander Liddle of the National Institute of Standards and Technology’s campus in Gaithersburg, Maryland, made the economic case for investing in new manufacturing technologies. Materials shaped at the nanoscale often have new or enhanced properties. Advances in nanomanufacturing could thus give the United States an edge and help to curb its growing trade imbalance.

“Nanomanufacturing is key to revitalizing the US manufacturing industry,” said Liddle.

The challenge, he said, will be to match emerging technologies with appropriate products. Finding the right balance between cost and value is important, as is identifying the throughput and precision with which new materials can be produced.

Optical lithography, which uses light to pattern a material’s surface, seems to have already found a killer application: integrated circuits. A lithography machine has a price tag of about $50 million. But it consistently churns out 175 semiconductor wafers per hour an industry-friendly cost of about $10 per wafer layer.

Nanoparticles that self assemble to form colloidal structures , on the other hand, may be chasing the wrong application. Liddle said those hoping to create colloidal crystals with useful optical properties will be hampered by the many performance-compromising defects that tend to accumulate during the process. Window films that change color depending on the temperature may be a more suitable product. Tolerant of imperfections, such films would take advantage of the material’s low cost, high throughput, and unusual thermal properties.

New measurement techniques will be crucial for migrating appropriate technologies from the lab to the factory, said Liddle. He’s working on high-speed techniques using microwaves that reveal, in real time, the internal structure of a sheet of polymer embedded with carbon nanotubes.

Transitioning transistors

One industry that could benefit from a new nanotech breakthrough is integrated circuitry. Computer chips stopped getting faster around 2003, said Thomas Theis of the Semiconductor Research Corporation . Though transistors keep getting smaller, the clock speed of chips has been more or less static.

Energy limitations have held computers back. Today’s silicon transistors must be run at about 1 V to flip a bit from 0 to 1. That voltage creates heat, limiting the densities of modern chips.

“This situation has already robbed the industry of a little bit of vitality,” said Theis. “But the market is still there for faster computers, if we can deliver them.”

In search of new materials with better properties, SRC launched a network of universities called STARnet , with $194 million in funding over five years. A smaller project, the Nanoelectronics Research Initiative , is spending $25 million over 5 years on efforts in academia to create new transistors with steeper on/off switching voltages that would consume less power.

Last year, the field of prospective NRI candidates was winnowed to eight promising approaches. One idea would achieve a small switching voltage by exploiting a quantum phenomenon called tunneling. Another exploits ferroelectric materials with polarities that can be rapidly switched by electric fields.

“So far the clear winner in the search for the next switch is not there,” said Theis. To continue the search, STARnet recently expanded its network. NRI is expected to do the same later this year.

Transistors at the atomic scale

Reflecting on the trend toward ever-smaller transistors, Michelle Simmons of the University of New South Wales said that the tiny switches will, in the coming years, shrink much faster than many expect them to.

“Commercial devices will hit the atomic scale within the next five years,” she said.

Today’s commercial transistors are already starting to feel the influence of atoms. Intel is working on its smallest transistor to date, which measures only 14 nanometers across. At that scale the positions of individual atoms can change the properties of a device.

Simmons and her team have developed a way to precisely place individual atoms in silicon. Though her ultimate aim is to create a silicon-based quantum computer, her technique could help to miniaturize the components of traditional computers.

To build an atomically exact device, Simmons first coats a wafer of silicon with hydrogen. Using a sharp atomic probe microscope, she then knocks off a few hydrogen atoms. Phosphene gas adheres to the bare patch and, when heated, replaces a silicon atom on the surface with a phosphorous atom.

“This is the first time anyone has deterministically placed a phosphorous atom in a device,” said Simmons.

Encapsulated beneath additional layers of silicon, one or more phosphorous atoms can function as a transistor or a gate . Chains of phosphorous form nanowires that, unlike copper wires, maintain a low electrical resistivity at the small scales.

Nanosolar for terawatts

Already a mature industry with a defined manufacturing process, photovoltaics has spawned a supply chain that stretches across the world. In 2011 the growth of photovoltaic manufacturing worldwide equaled that of gas turbines. Thanks to increasing supply that has outstripped demand, the price of a solar panel has dropped precipitously .

In the future, said Caltech’s Harry Atwater, the cost of a solar cell will be dominated not by the materials in the cell itself, but by the module’s other components. “Efficiency becomes paramount,” said Atwater, who is exploring new nanoscale approaches to harvesting the sun’s energy.

Trapping light within a cell could help to counter a major source of inefficiency: the conversation of photons to heat. Hemispherical bumps on the surface of a silicon cell could thus improve its efficiency, Atwater has shown. So could parabolic structures shaped like cups that collect light and concentrate it. Silicon wires embedded in polymers have been used to make stretchy photovoltaic films with an efficiency of 28.8%.

The biggest improvements in efficiency may come not from a single material but from combining different materials. Efficiencies of 40% have been achieved by layering materials specialized for different wavelengths on top of each other and allowing light to filter through the stack. Ultimately, Atwater hopes to split sunlight into its component colors and channel each directly to its own custom-tailored panel.

Devin Powell is a freelance science writer based in Washington, DC. His stories have appeared in Science News, Wired, US News & World Report, and other outlets.