Gallium nitride nanowires grown
Demonstrated reproducible growth of isolated GaN nanowires 30 to 250 nm in diameter and 5 to 25 micrometers in length using gas-source molecular beam epitaxy (MBE), without the use of metal catalysts. Low growth rates at substrate temperatures near 820 °C were combined with high nitrogen flux to form the wires, which grew out of an irregular matrix layer containing deep faceted holes. X-ray and variable-temperature photoluminescence measurements, in collaboration with the Optical Materials Metrology Project, and transmission electron microscopy studies, in collaboration with the NIST Materials Science and Engineering Laboratory, showed that the GaN nanowires have a high degree of crystalline perfection.
Contact: Dr.
Kris Bertness |
 |
Nanowire ultraviolet LED demonstrated
Developed and demonstrated, in collaboration with the Optical Materials Metrology Project, operation of an ultraviolet light-emitting diode based on gallium nitride nanowires. The LED demonstration required growth of isolated, low-defect-density GaN nanowires on silicon, structural characterization as feedback to growth parameters, removal of the wires from the growth substrate and dispersal on an n-type GaN substrate, optimization of electrical contacts, and spectroscopy of the LED output to verify emission from the nanowires. The LED emission center wavelength near 385 nm was believed to be the shortest yet demonstrated for this type of device at the time.
Contact: Dr.
Kris Bertness
Single nanowire photoluminescence measured
Grew GaN nanowires to enable variable-temperature, polarization-resolved photoluminescence (PL) measurements of individual wires. The nanowires enabled the first unambiguous room-temperature measurement of photoluminescence anisotropy from unstrained wurtzite gallium nitride that arises from the symmetry, band structure, and selection rules for radiative electronic transitions in the material. They also were used to demonstrate the utility of PL measurements in understanding the post-processing induced strain in otherwise strain-free nanowires. (See Optical Materials Metrology Project accomplishments.)
Contact: Dr.
Kris Bertness
Nanowire laser
Demonstrated optically pumped laser action in an as-grown GaN nanowire, in collaboration with the Optical Materials Metrology Project. Using a q-switched pump source at 266 nm, the spectral width of initial superluminescence narrowed and shifted to longer wavelengths with increased pump intensity, and the output vs. pump intensity was superlinear. In addition, the lasing spectrum showed distinct longitudinal modes corresponding to the approximately 25 micrometer long, MBE-grown wires.
Contact: Dr.
Kris Bertness
Single nanowire field-effect transistors
Fabricated and characterized high-quality metal-semiconductor field effect transistors (MESFETs) based on individual gallium nitride nanowires (NWs), in collaboration with the Optical Materials Metrology Project. The Schottky gates exhibited excellent two-terminal Schottky diode rectification behavior and ideality factors significantly lower than those previously reported for individual GaN NWs. In addition, the Schottky-gated MESFETs showed low threshold gate voltages, high transconductances, near-ideal subthreshold swings, and large on/off drain-source current ratios. The MESFET design shows promise for significant improvement in the performance of GaN NW field-effect devices.
Contact: Dr.
Kris Bertness
Nanowire Oscillators with Record Q
Grew nanowires that have a mechanical quality factor at least 10 times higher than reported values for other nanoscale devices, and comparable to that of commercial quartz crystals. Collaborators at the University of Colorado in Boulder, demonstrated high Q factors in wires that are 30 to 500 nm in diameter and 5 to 20 micrometers long, having resonance frequencies between 400 kHz and 2.8 MHz. To measure the resonance properties, the as-grown nanowires on pieces of silicon substrate were attached to the top of a piezoelectric stack in a scanning electron microscope, and an ac signal was applied. The output signal from the SEM secondary electron detector as a function of applied frequency was used to determine the resonance frequencies and Qs of the nanowires. Q near room temperature for various nanowires ranged from 2,700 to above 60,000, with a typical wire exhibiting a Q of 38,000, at least 10 times higher than previously reported values for a-axis oriented GaN nanowires, carbon nanotubes, and single-crystal silicon microstructures of similar surface-to-volume ratio. The Q values observed are comparable to those of commercial quartz crystal resonators used in feedback oscillators. Positive feedback to the piezoelectric stack caused oscillations of the nanowires with closed loop Q exceeding 106. Because GaN nanowires are piezoelectric and are grown on silicon, it makes them compatible with existing microelectronics processing methods and suggests their increasing importance in nanoelectromechanical (NEMS) systems.
Contact: Dr.
Kris Bertness
|
 |
 |
Ink-jet printer placement of nanowires
Demonstrated, with an industrial collaborator, that thermal ink jet print heads can be used to place GaN nanowires on patterned substrates. The nanowires were dispersed in ethylene glycol-water and alcohol-water solutions. To avoid clogging, the ink jet heads were chosen with drop weights from 72 to 165 ng (volumes about 72 - 165 pL). The method was used in combination with dielectrophoresis to place single nanowires across narrow gaps in metal patterns. The placement accuracy is much higher than that with micropipette placement (for which the smallest possible drop volume is only about 200 nL). Ink jet technology holds promise for low-cost, rapid, massively parallel placement and processing of nanowires for optoelectronic, electronic, and sensor applications.
Contact: Dr.
Kris Bertness
Uniformity of MBE-grown quantum dots
Assessed, the uniformity of self-assembled InGaAs quantum dots grown by MBE under a variety of conditions. Lateral variations in the dot density and height were determined from atomic force micrographs. The growth rate was found to have the largest influence on lateral uniformity; the most uniform dot distributions were found for dots deposited at low rates, 0.15 monolayers/s. Dots deposited at a rate of 1.11 monolayers/s had large variations in both height and density, but these variations decreased as the dot density increased. The uniformity of dots deposited quickly also improved for the top layer of dots in stacked layers, even though these layers had decreased dot densities. There were negligible differences in the lateral height and density uniformities of dots as functions of continuous versus pulsed growth, wafer diameter, and In mole fraction. Achieving regular in-plane spatial distributions and sufficiently uniform size distributions is of great importance but a major challenge for the manufacture of quantum dot devices for applications including optical sources, detectors, and solar cells.
Contact: Dr.
Alexana Roshko
Temperature dependence of MBE- and MOCVD-grown quantum dot density characterized
Determined the temperature dependence of the dot density for pulsed and continuously grown MBE samples as well as for MOCVD grown samples. Assuming an Arrhenius dependence of the dot density on temperature, both the activation energies (-1.61 and -1.64 eV) and the pre-exponentials (10-8 and 9x10-9 dots/um2, for the pulsed and continuous-growth MBE samples, respectively) are nominally identical. For the MOCVD sample the dependence is quite different. The activation energy is lower (-3.63 eV) and the pre-exponential is smaller (10-22 dots/um2). The difference in fit for the MBE and MOCVD dots is probably due to the fact that in MOCVD the adsorption and decomposition of the organometallics, as well as the diffusion and nucleation are controlled by the substrate temperature, while in MBE the flux is controlled independently and only the diffusion and nucleation depend on the substrate temperature.
Contact: Dr.
Alexana Roshko
Uniformity of MBE- and MOCVD-grown quantum dots compared
Determined that MBE growth of InGaAs/GaAs quantum dots at low, continuous rates leads to uniformity in dot density equal to or better than that from MOCVD growth by collaborators. A nine by nine array of AFM images, taken over the central regions of 5.0 and 7.5 cm diameter wafers, was analyzed to determine the distribution of dot heights and densities. For MBE samples grown in a pulsed growth mode with a slow rotation rate the standard deviation in the dot density was typically 15 to 25 % of the average density. However, for samples with a continuous low growth rate and fast rotation rate, the standard deviation in the dot density was typically 3 to 4 % of the average. This is better than the 10 to 15 % standard deviation in dot density found for most MOCVD samples. MBE-grown samples, both pulsed and continuously grown, have much narrower dot size distributions than their MOCVD-grown counterparts as shown in the histograms and AFM images in Figure 2.
Contact: Dr.
Alexana Roshko |
 |
AlGaAs composition standards offered to the industry
NIST's AlGaAs composition Standard Reference Material (SRM) number 2841. This SRM consists of an epitaxial layer of AlxGa1‑xAs with certified mole fraction x of Al grown on a GaAs substrate. It is intended for use as a reference standard for analytical methods that measure the composition of thin films, such as electron microprobe analysis (EMPA), photoluminescence (PL), Auger electron spectroscopy (AES) and x-ray photoelectron spectroscopy (XPS). The specified Al mole fraction for each of these AlGaAs SRMs, 2842 is near x = 0.2, with an expanded uncertainty of 0.0014 (k = 2 coverage). Also given in the SRM certificate is the uncertified reference photoluminescence peak energy value. Standard Reference Material (SRM) 2842 for nominal mole fraction x=0.3 is also offered by NIST.
Contact: Dr.
Kris Bertness |
 |
Source gas purity instrumentation calibrated
Conducted tests of commercial cavity ring-down instrumentation in collaboration with the manufacturer and a semiconductor source gas manufacturer; the tests enabled calibration of the instrument for moisture in phosphine with record sensitivity at the sub-10-ppb level. When tested with arsine gas, this instrument was unable to provide useful moisture measurements because of spectral interference from arsine absorption lines. The manufacturer is pursuing the use of a different spectral region.
Contact: Dr.
Kris Bertness
|