Nanowires: lighting the future

Group III nitrides—such as aluminum gallium indium nitride (AlGaInN)—are technologically important semiconductors that absorb and emit light efficiently over a very broad and useful energy range from the UV to visible to IR wavelengths. They are the basis for commercial products such as visible LEDs and blue laser diodes (e.g., Blu-Ray). Recently, 1D nanostructures based on III-nitride semiconductors, including nanowires and nanorods, have attracted attention as potential nanoscale building blocks for enhanced performance or functionality optoelectronics, sensing, photovoltaics, and electronics. In particular, the unique ability of nanowires to relieve strain laterally can lead to defect-free material, improved alloy compositions that allow green-red wavelengths to be achieved, and growth on arbitrary, cheap substrates.

These unique attributes could potentially enhance the performance and lower the cost of LEDs based on nanowires, with implications for solid-state lighting and display applications. However, before their promise can be fully realized, several challenges must be addressed in the areas of controlled nanowire synthesis, understanding and controlling nanowire structural, electrical, thermal, and optical properties, and nanowire device integration. Our group at Sandia National Laboratories is currently pursuing nanowire research with the goal of addressing these many challenges.

III-nitride nanowires can be fabricated by a variety of techniques, including bottom-up approaches and top-down lithographic techniques. Bottom-up methods are generally used for synthesizing nanowires, and they often involve a nanoscale metal catalyst particle to direct the 1D growth. The advantages of using this approach include nanowires free of detrimental crystal defects known as dislocations and the ability to grow on inexpensive, lattice-mismatched substrates such as glass and metal foil, which we have demonstrated in our lab. Figure 1(a) shows vertically aligned GaN nanowires grown using nickel catalyst nanoparticles by metal-organic chemical vapor deposition (MOCVD).¹ Structural analysis shows that these nanowires are single-crystalline and free of dislocations.

Figure 1. (a) Vertically aligned bottom-up gallium nitride (GaN) nanowire growth using a nickel catalyst. (b) Ordered GaN nanorod array fabricated using a two-step dry plus wet-etch process.

Figure 2. (a) Monochromatic 566nm cathodoluminescent (CL) image revealing surface-defect luminescence. (b) Composite CL image of a GaN/InGaN core-shell nanowire showing multicolor emission resulting from a nonuniform indium (In) distribution.

Figure 3. (a) Schematic for vertically integrated, electrically injected nanorod-based GaN/InGaN LED shown in (b), in which a p-doped GaN layer has been coalesced into a planar layer over the nanorods. MQW: Multiquantum well. p, n: Doping. Al₂O₃: Sapphire.

Bottom-up nanowire growth methods, including catalyst-free methods such as selective-area growth, do have the disadvantage of requiring highly specific growth conditions to increase the on-axis growth rate while minimizing lateral growth. This can lead to nonoptimal material quality and less flexibility in material design, such as doping and heterostructures. Thus, we recently developed a new top-down approach for fabricating ordered arrays of high-quality GaN-based nanorods with controllable height, pitch, and diameter. This top-down method allows construction of nanorods from high quality, arbitrarily doped films grown by MOCVD using standard, optimized conditions. Figure 1(b) shows a periodic GaN nanorod array made from a two-step dry plus selective wet-etch process we recently developed. Since both our bottom-up and top-down methods employ industry-standard MOCVD growth, we expect that commercial adoption should be straightforward.

Before nanowires can become useful for devices, it is first necessary to understand and ultimately control and optimize their properties. To this end, we are employing a variety of optical, electrical, and structural nanocharacterization techniques, which are made challenging by the small dimensions of the nanowires. For example, spatially resolved cathodoluminescence (CL) experiments are being used to map the frequencies and intensities of light emission from these nanowires with nanoscale resolution, allowing us to observe and pinpoint the location of defects to the nanowire surface region: see Figure 2(a).² Using a combination of CL, scanning transmission electron microscopy, and energy dispersive x-ray spectroscopy, we have also mapped the distribution of indium in InGaN alloy shell layers grown around GaN nanowires: see Figure 2(b).³ Combined with finite element models, we find that relaxed strain in nanowires allows for high indium incorporation in the InGaN layers, which may lead to highly efficient III-nitride-based LEDs in the green to red wavelengths that are difficult to achieve with current thin-film based architectures.

Single-nanowire devices have been fabricated and studied by making electrical contacts using traditional lithography or with nanoscale probe tips manipulated inside of scanning or transmission electron microscopes. This has allowed us to examine in real time the microstructural changes and breakdown mechanism of individual nanowire devices under high electrical power.⁴ For LEDs, integrated devices consisting of hundreds of thousands to even millions of individual nanowire LEDs are needed and have also been fabricated via a vertical integration scheme: see Figure 3(a). Understanding the present limitations and improving the performance of these integrated nanowire-based LED structures is currently a priority area of research for us.

In summary, III-nitride-based nanowires possess unique and attractive attributes for realizing improved performance LEDs and other devices. Further research and development, however, is needed to realize their full potential. We expect that continued advances in the areas of nanowire fabrication, characterization, and devices should lead to surprising new insights into these interesting structures.

The bottom-up nanowire growth and surface-defect CL studies were supported by the US Department of Energy, Office of Basic Energy Sciences (BES), Division of Materials Sciences and Engineering. The GaN/InGaN core-shell studies and top-down nanorod fabrication were supported by Sandia's Solid State Lighting Science Energy Frontier Research Center, funded by DOE BES. We also acknowledge support from the National Nuclear Security Administration's (NNSA) Laboratory Directed Research and Development program. Sandia National Laboratories is a multiprogram laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the US DOE's NNSA under contract DE-AC04-94AL85000.