Controlled Nanowire Growth Promises Better Materials

A team at Lawrence Berkeley National Laboratory (LBNL; Berkeley, CA) says it has gained better control over the properties of gallium nitride (GaN) nanowires by learning how to decide in which direction the crystals will grow.

Nanowires are crystals only a few nanometers in diameter but up to several microns in length. The nanowires hold promise for a variety of optoelectronic devices, including blue-emitting LEDs, nanoscale UV lasers, and sensors for miniscule amounts of fluids.

"Basically, we came up with a fairly simple trick to grow this nanowire in the preferred direction," says Piedong Yang, assistant professor of chemistry at the University of California, Berkeley and a faculty scientist at LBNL. "It's really achieved by carefully choosing the right substrate."

Yang and his team of researchers worked with two substrates, lithium aluminum oxide (LiAlO₂) and magnesium oxide (MgO). The shape of the crystal—what is known as its symmetry—and the lattice constant of the substrate surface control the direction in which the GaN crystal grows. The LiAlO₂ has twofold symmetry, which means it will look the same if rotated 180°. One plane of the GaN crystal has the same twofold symmetry. When GaN is metal-organic chemical vapor deposited on the substrate, the resulting formation naturally aligns with the plane that matches it.

The MgO substrate, on the other hand, has threefold symmetry, meaning it looks the same if rotated only 120°. That matches the symmetry of a different plane of the GaN, so GaN grows in a different, but predictable, direction on MgO. Lastly, a cross section of the GaN crystal grown on LiAlO₂ is triangular, whereas a cross section of the crystal grown on MgO is hexagonal.

Though the GaN was the same in both crystals, the light emission from the two wires differed by 100 meV, with one emitting around 370 nm and the other at 390 nm.

"This is sort of the first proof of concept that you do have the capacity to control the growth direction of gallium nitride," Yang says.

Being able to control the direction of growth means that researchers can tune various physical properties of the material, such as bandgap, index of refraction, thermal and electrical conductivity, and piezoelectric polarization. The standard way of controlling these physical properties is by varying the size of the nanostructure, but his method provides another means of control, Yang says.

Transmission electron microscope images show GaN forming a triangular cross section (top) when grown on LiAlO₂ and a hexagonal cross section (bottom) when grown on MgO.

"Basically we add another knob to tune these physical properties," he says.

Charles Lieber, a professor of chemistry at Harvard University (Cambridge, MA) who works with nanowires, calls the Berkeley work "a nice piece of science but not something that I would call a breakthrough, since the basic ideas—of substrate controlled growth orientation and structures (triangular and hexagonal)—have been demonstrated in previous publications for several nanowire systems. To me the much bigger issue is whether one can achieve selective n- and p-type doping needed to make active photonic devices," he says.

Zhong Lin Wang, a professor of materials science and engineering at Georgia Institute of Technology (Atlanta, GA), says the method of using single-crystal substrates to grow nanowires was probably first demonstrated for zinc oxide a few years ago. But the fact that Yang achieved it with GaN is important. "The result is not surprising from my point of view, but good progress for GaN," Wang says.

Yang says he hopes to experiment with other substrates to achieve different growth directions. In the meantime, his team is busy trying to learn all the details of the materials it has already produced. The team is also trying to build LEDs and field-effect transistors out of its nanowires to see how well they work.