The University of New Mexico

An NSF Research Experience for

Undergraduates program in





  1. Fabrication of nanostructured substrates for cubic GaN growth (Faculty Mentor: Steven R. J. Brueck)

    Device-grade films of cubic GaN offer potential device improvements associated with: high mobilities, high p-doping concentrations and elimination of the piezoelectric field that can impact recombination lifetimes in polar wurzite GaN, and potentially impact the high-power droop issues.

    The starting point for this project is previous work at UNM (Fig. 1) that showed phase segregation for GaN grown in <111> V-grooves on a <001> Si substrate at nanoscale regime [Lee 2004a], [Lee 2004b], [Lee 2005]. Si substrates provide an inexpensive, large-area, readily available substrate that would impact the economics of GaN light-emitting diodes (LEDs) and provide the possibility of their integration with Si microelectronics. GaN preferentially nucleates on the <111> faces of the V-groove, and the symmetry forces generation of a <001> cubic phase region in the center of the V-groove. We would like to: 1) understand the limits of this process, e.g. at what point does the higher ground-state energy of the cubic GaN overwhelm the symmetry that drives the phase separation, and 2) develop techniques to isolate and enlarge the cubic GaN regions. The goal is to provide sufficiently large regions of cubic GaN for detailed optical and electrical characterization and ultimately to serve as a platform for further device studies.

    During this project, the undergraduate student will fabricate a series of nanopatterned substrates with varying groove spacings and dimensions for these investigations. The interferometric lithography technology for these studies is available in the mentor's laboratory, as are a cadre of graduate students and research scientists who can assist in training the student. GaN growth will be performed at RPI on Prof. Wetzel's MOCVD system. The student will also be trained in characterization techniques including photoluminescence (PL) and scanning electron microscopy (SEM).

    This will clearly benefit both the project, by enabling a systematic investigation of the limits and characteristics of this unique phase separation process, and the student by involving him/her in an essential way in an exciting ongoing research project that will have a strong impact on our understanding of nanoscale crystal growth, and, if successful, will also impact important device technologies including LEDs and blue/uv diode lasers.

    Fig. 1. Phase-separated GaN grown in a flat-bottom Si V-groove. The GaN in the center (B) is cubic, while the two nucleation nanocrystals (A) on the <111> Si sidewalls have wurzite structure.

  2. Narrow-bandgap nano-pillar growth and optical characterization (Faculty Mentor: Ganesh Balakrishnan)

    This project will require undergraduate students to investigate growth of nanopillars of (In)GaSb on GaAs, using selective-area molecular beam epitaxy (MBE). Epitaxy takes place in apertures that have been masked by SiO2 on GaAs (111)B substrates. Using this method, nanopillars have been grown with high aspect ratios up to 180:1 (9 µm long, 50 nm diameter, with the lateral dimensions determined by the diameter of the oxide apertures, as shown in Fig. 2. The nanopillars are hexagonal in cross section with the {110} facets exposed. In contrast to the vapor-liquid-solid (VLS) growth mode, the growth of the nano-pillars is not due to the presence of any metal-based catalysts but is rather due to the diffusion of the group III and V ad-atoms on the mask and subsequently on the nano-pillar itself. Thus the nanopillar is an equilibrium crystal shape (ECS) and is highly faceted and reproducible [Wong 2007], [Tran 2007], [Liang 2007]. Furthermore, the highly faceted nature of the nano-pillar combined with the growth of heterostructures allows for the realization of unique quantum-dot-like quantum-confined structures that may be tailored to precise dimensions. Thus, the undergraduate students will have the opportunity to learn and understand MBE as well as characterization concepts in nano-photonics through this project.

    The MBE laboratory at CHTM has been at the forefront of narrow-bandgap optoelectronics and nanotechnology research [Tatebayashi 2008], [Gradkowski 2009a], [Gradkowski 2009b], [Tatebayashi 2009], [Huang 2009]. This is perhaps the only laboratory of its kind with an emphasis on teaching MBE to undergraduate students [Balakrishnan 2005a], [Balakrishnan 2005b], [Balakrishnan 2005c], [Mehta 2006], [Balakrishnan 2007]. Currently, 3 undergraduates work in the MBE laboratory.

    The students recruited under the REU program will be assimilated into this well-established research program. The summer interns will have a rigorous one week training program on the MBEs and characterization labs. They will then spend the rest of the 9 weeks on the growth and characterization of the nano-pillars. These students will be aided in their research by highly accomplished and dedicated members of the MBE lab (the faculty mentor, two research professors, one senior research scientist, one post-doc). The local students will undergo a more rigorous training for a month, followed by 7 months of research during the academic year. The students will have full access to the "undergraduate-MBE reactor" a GEN-II antimonide tool and will be able to conduct research and maintain the tool during the program. These students will be provided a graduate student/post-doc mentor to monitor and help with the research. The students will also be exposed to a state-of-the-art GEN 10 robotic MBE that will allow them to work with industry standard robotics and control systems for modern MBEs.

    Fig. 2. Nanopillars formed using selective area growth.

  3. High-temperature cadmium-free nanophosphors for daylight quality white LEDs (Faculty Mentor: Marek Osinski)

    Commercially available white-LEDs are produced by combining a blue InGaNGaN LED chip with a yellow phosphor, typically, yttrium aluminum garnet (YAG) doped with cerium. They emit an intense bluish white light dominated by blue emission from InGaN chip. This spectrum may have significant adverse health effects, as it has been recently discovered [Brainard 2001], [Thapan 2001] that human body clocks are highly sensitive to blue light (460 - 480 nm). Excessive blue light may disrupt circadian rhythms, which in turns has been linked with a host of pathological conditions in night shift workers, including breast and colorectal cancer, cardiovascular disease, depression and anxiety disorders, obesity, and type 2 diabetes.

    We suggest the problem of excessive blue light emission can be overcome by exploiting the size-dependent broad color tunability of colloidal quantum dots (QDs). While CdSe/ZnS QDs exhibit high luminescence efficiency, high chemical stability, and a size-controlled color variation due to quantum confinement effects, they contain highly toxic cadmium, which will limit their applicability to solid-state lighting. Instead, we investigate novel Cd-free nanophosphors based on InP:Cu and ZnSe:Mn cores. As shown in Fig. 4, these QDs can emit light at high temperatures, which is important for high-power white-light LEDs that generate significant heating. shows the imitated effect of simple blending of different-sized QDs synthesized in Prof. Osinski's lab.

    The REU students involved in this project will learn how to synthesize colloidal QDs, how to control their size, and how to measure their optical properties (photoluminescence, absorption, quantum efficiency, lifetime,). The initial work will focus on spanning a large part of the visible spectrum, from blue-green to red. Subsequently, the students will be involved in dispersion of these nanophosphors in silicone, and in demonstration of daylight quality white light emission from blue InGaN/GaN LEDs combined with nanophosphors.

    Fig. 3. Red emission from InP:Cu / ZnSe QDs at 126 ºC.

  4. Nanophotonics in optically coupled semiconductor quantum dot - gold nanoparticle assemblies (Faculty Mentor: Ravi K. Jain)

    This project focuses on the science and technology of optically coupled semiconductor QD - gold nanoparticle (AuNP) assemblies. Optical excitation of surface plasmons within metallic nanoparticles (NPs) of dimensions between 5 and 100 nm can result in strong local optical field enhancements, which can greatly enhance the fluorescence emission from neighboring light-emitting molecules or nanoparticles. Interactions between NP-localized surface plasmons and QDs are still relatively poorly understood. Experiments-to-date have yielded confusing and contradictory results, partly due to difficulties in reliable sample preparation, and partly due to unresolved questions on critical issues that affect such field-induced fluorescence enhancements [Gueroui 2004], [Song 2005], [Wang 2009a], [Wang 2009b]. As such, detailed parametric studies of linear and nonlinear fluorescence enhancements in QDs are strongly desirable. Moreover, there is a strong need to use monodisperse particles, both for the metallic NPs and the QDs in future experiments, along with a strong need to develop reproducible experimental techniques to control the interparticle spacing (of the order of 10 nm) with very high precision (< 1 nm). Such fundamental understanding is expected to strongly impact the design of related plasmonic devices, such as plasmon-enhanced LEDs and solar cells.

    Preliminary studies will be performed by the REU students (over the summer or over 32 weeks during an AY) by fabricating - and studying the optical properties of - "layered" QD-AuNP assemblies similar to that shown schematically below in Fig. 5. During the fabrication and linear and nonlinear optical characterization, undergraduate students will not only gain hands-on experience on several important optical characterization techniques (such as lock-in detection and photon counting instruments) and advanced laser systems (such as argon, mode-locked Ti-sapphire), but will also obtain fundamental insight and invaluable basic scientific knowledge related to quantum-confined semiconductors and localized surface plasmons.

    Fig. 4. Cross-sectional view of the structure of a QD-AuNP "layered" multi-particle assembly.

  5. Interaction between colloidal nanoparticles and high-Q optical whispering-gallery modes in liquid phase spherical resonators (Faculty Mentor: Mani Hossein-Zadeh)

    This project combines the relatively young fields of optofluidics and nanophotonics in the context of high-Q optical whispering gallery (WG) resonators and NPs. Optical interactions with colloidal NPs (especially colloidal QDs) have been subject of research for a long time. One of the subdivisions of this field focuses on interaction of high-Q resonant optical fields with the NPs [Fan 2001], [Poitrasa 2003], [Gaponik 2006]. High-Q WG modes in micro-droplet resonators are known for their small mode volume, large intensity and long photon lifetime. Recent results have shown the possibility of stable and efficient coupling to these resonators using fiber tapers embedded in an aqueous medium [Hossein-Zadeh 2006]. So far, most of these studies have been based on the interactions with the evanescent fields leaking out of a solid optical resonator [Moreels 2006], [Min 2006], [Zhu 2009], [Charlebois 2010]. In this REU project, we intend to use microdroplets made of colloidal solution as the resonator, which will enable a direct and strong interaction between the NPs and the resonant optical field.

    The first phase of the project will focus on characterizing the quality factor of the colloidal droplet resonator as a function of NP size, density, and type. The activities in this phase will include: identifying a colloidal solution of NPs and the proper background liquid, building the experimental setup, and finally characterization. The students will learn a great deal about NPs, optical microresonators, measurement tools and techniques, nonlinear optical interactions, and overall basics of nanophotonics and optofluidics. The first phase of this project is simple enough (both theoretically and experimentally) that with a basic knowledge of electromagnetics, optics, and quantum mechanics, the students can bring themselves up to speed in a relatively short period of time. More than background knowledge and experience, this project demands personal characteristics such as motivation, creativity, and patience.

    Fig. 5. Schematic diagram of the typical experimental setup used for optical coupling to micro-droplet resonators.

  6. Infared detectors with quantum dots and superlattices (Faculty Mentor: Sanjay Krishna)

    In this project, the REU students will be involved with the characterization of two emerging infrared detector technologies, namely (i) Photoconductors based on intersubband transitions in self-assembled quantum-dot infrared photodetectors (QDIPs) and (ii) Photovoltaic detectors based on interband transitions in Type II InAs/GaSb strain layer superlattices (SLSs).

    InAs/GaAs QDIPs are extrinsic, intraband, photoconductive photodetectors. InAs QDs embedded in a strain-relieving InGaAs quantum well are known as dot-in-a-well (DWELL) heterostructures. QDIPs using a DWELL heterostructure not only permit greater control over wavelength tunability, but they have also demonstrated excellent device performance [Raghavan 2004]. Prof. Krishna's group has also been involved with the integration of DWELL detectors with plasmonic structures [Rosenberg 2009], [Chang 2010].

    SLS-based detectors have been a subject of active research in the recent past and exploit a type-II band structure that exists in the InAs/GaSb system. Since the electron-hole overlap is controlled by the thicknesses of the constituent layers, it enables one to fabricate a small bandgap material using "mid-bandgap" semiconductors. This is predicted to lead to a decrease in the dark current due to reduced tunneling and lower Auger recombination rates. By optimizing the oscillator strength in this material system, a large quantum efficiency and responsivity can be obtained [Rogalski 2006].

    Prof. Krishna's research group has successfully demonstrated the first mid-wave infrared camera ( cut-off = 4.2µm) with a 320 256 focal plane array (FPA) based on type-II InAs/GaSb SLS with an nBn design [Kim 2008]. A thermal image taken with this camera at a detector temperature of 77 K and integration time of 16.3 ms is shown in Fig. 8 below. The bright areas of the image represent warmer regions whereas the dark areas exhibit colder regions. In the figure, the facial imprint of the fingerprints after touching a cold can are clearly visible, demonstrating the good imaging quality of camera.

    Since the summer REU program is limited in time and scope, the emphasis for the undergraduate student will be on the radiometric characterization of the FPA rather than the growth and fabrication, which are much more complicated and require months of training. In summer of 2008, an NNIN REU student from Johns Hopkins University (Ms. Jennifer Hou) was advised by Prof. Krishna on the characterization of infrared FPAs based on QDs (see Section D.6.C). At the end of a very successful REU effort, Ms. Hou was the second author on a paper submitted for the proceedings of an international conference [Shenoi 2008b]. We will capitalize on our past efforts to build a successful REU program.

    Nature Communications 19 April 2011

    Fig. 6. Thermal image taken with 320 256 InAs/GaSb SLS nBn camera developed in collaboration with QmagiQ LLC. A two-point non-uniformity correction was undertaken.

  7. Optical characterization of thermal properties on the nanoscale (Faculty Mentor: Kevin J. Malloy)

    Novel devices for addressing the ongoing energy crisis are a necessity. One such device that leads to many potential applications is a thin film heat switch. These devices would have the ability to switch between highly thermally conductive and thermal insulating states, either within or normal to the film plane. With such a device, heat flow could be controlled in a similar way that electronic switches control the flow of electrons and optical devices control the flow of electrons. New forms of heat engines that enable energy scavenging of waste heat and refrigeration or heating of various areas could be envisioned.

    While large-scale mechanical heat switches are known, mechanical thin film versions have yet to be manufactured. Another approach is to develop materials with controllable thermal conductivities. Several of these materials are known, typically utilizing changes in the electronic contribution to thermal conductivity.

    We are investigating controllable thermal conductivity in nanomaterial systems. The archetypical system involves various calamatic liquid crystals. As shown schematically in Fig. 9, depending on the orientation of the mesogen director, the thermal conductivity can vary by a factor of 4. Our current research incorporates additional components, such as carbon nanotubes, into the liquid crystal as a means of improving the contrast in thermal conductivity.

    Several means are used to measure the thermal properties of these materials. The electrical "3-omega" technique is routinely used in our laboratory. An appropriate REU research project would be to establish an optical method of characterizing thermal properties, such as the time-domain based transient thermal reflectance technique or the frequency-domain equivalent thermal reflectivity phase shift technique. All of the required equipment exists in our laboratory, lacking only an enthusiastic undergraduate to set it up.

    Fig. 7. Edge view of a liquid crystal heat switch showing the interdigitated electrodes (rectangles) and the liquid crystal directors (indicated by the orientation of the ellipses). The electric field direction (curved arrows) and the liquid crystal directors can be switched between being predominately parallel to the film of liquid crystal (top panel) to mainly perpendicular to the film (bottom panel).

  8. Plasmonically-enhanced nanopillar single photon avalanche diodes (Faculty Mentor: Majeed Hayat)

    This proposed project investigates a new platform for single photon avalanche diodes (SPADs) based on a three-dimensional (3D) plasmonic grating self-aligned to a patterned III-V nanopillar array for both enhanced absorption and enhanced avalanche multiplication within the nanopillar. Traditional planar SPADs are fundamentally limited by the lateral junction area being equivalent with the optical collection area. The requisite large collection thickness of absorbing material to maximize photon detection efficiency (PDE) results in deleterious effect for dark current, dead time and timing jitter. All of these factors collectively limit the photoncount rate of a SPAD. This proposal offers a transformative device platform to solve a technologically important problem of high efficiency with small foot-print and simplified fabrication in single photon detectors necessary for quantum communications. The proposed ideas are substantiated by strong initial work and proof of concept. Our team includes worldrenowned experts in nanostructure epitaxy and device fabrication (Huffaker, Co-PI, UCLA), SPAD modeling and design (Hayat, Co-PI, UNM), carrier-dynamics modeling (Wu, visiting scholar, NTU) and SPAD characterization (Bienfang, unfunded, NIST).

    Along with device innovation, lead by UCLA team, this collaborative effort will elucidate new physical phenomena in semiconductor nanostructures, plasmonic antennae and offer the first theoretical investigation of 3D impact ionization in nanopillars. This latter component will be carried out by Hayat's team at UNM. Our cumulative result is to transform state-of-art SPAD performance in detection efficiency, dark count rate and jitter.

    Fig. 8.

  9. Flexible, thin-film solar cells (Faculty Mentor: Luke F. Lester)

    Commercial thin-film flexible photovoltaics are paving the way to low-cost electricity that is conformal to a variety of surfaces. Current conversion efficiencies under standard environmental conditions are in the 3-15% range for flexible, thin film devices [Lester 2010]. For comparison, solar cells built from heavy, stiff, and fragile inorganic materials can have efficiency as high as 39% [Emcore 2010]. A review of current commercially-available flexible thin-film solar technology shows that the best performing products made with amorphous Si [PowerFilm 2010] and thin-film copper/indium/gallium selenide (CIGS) [SoloFilm 2010] demonstrate ~0.13 W/g. For small autonomous objects that have constrained size and weight limitations, it is highly desirable to substantially increase this specific power by at least an order of magnitude by utilizing high-efficiency devices on a light, flexible materials. The objective of this flexible, thin-film solar cell project is to realize a specific power > 2 W/g and to integrate cells onto novel, lightweight membranes suitable for wings on flying objects or "skin" on crawlers (Fig. 7). To date, undergraduates working on this project have won the Best Paper Award in the 2010 IEEE Southwest Section Student Paper Contest. Future undergraduate participants will continue to focus on maintaining high efficiency in rigid, but very small area InAs quantum dot solar cells that can be integrated onto flexible surfaces such as aluminum/stainless steel sheets, KAPTON, or Nylon-66 nanopaper. GaAs cells embedded with InAs quantum dots will be studied for their proven efficiency in small-area devices. Where necessary, the students will learn how to deposit metal interconnects onto non-conductive membranes and to attach arrays of individual solar cell chips onto the surfaces of the different flexible materials. Efficiency testing of the novel flexible solar technologies will be conducted under standard illumination conditions before and after stress tests, designed to assess the durability of the integrated system.

    Fig. 9. A solar powered "bat", one of the possible autonomous robots enabled by highly efficient flexible solar technology.

  10. Electrical and optical modeling/simulation of GaN-based nanowire coaxial LEDs (Faculty Mentor: Daniel Feezell)

    GaN-based coaxial (core-shell) LEDs offer a wide range of advantages over conventional planar LEDs. For example, the active region of coaxial LEDs is formed on nonpolar {11 ?00} m-plane GaN sidewalls (Fig. 1), which eliminates the quantum confined Stark effect (QCSE) and improves the radiative recombination efficiency of the LEDs. The recent evolution [Hersee, 2006] of a catalyst free, scalable, and repeatable MOCVD growth process for GaN nanowires has enhanced the commercial viability of these LEDs. Indeed, companies such as Nanocrystal Corporation and glo AB are currently working on the commercialization of nanowire GaN-based LEDs. However, many challenges still exist, such as determining the optimal device design for uniform current injection and efficient light extraction.
    The student participating in this REU project will model and simulate the electrical and optical characteristics of the nanowire coaxial LED structure shown in Fig. 1. The student will be integrated into the research group and advised by senior-level researchers (including the faculty mentor, a post-doc, and an MOCVD systems engineer), and take an active role in improving the performance of these devices. A basic knowledge of optics and semiconductor device physics is sufficient to participate in this project.
    Analytical and numerical models will be developed for the current injection analysis and various devices geometries will be examined to optimize and reduce current crowding effects. Additionally, the student will examine these effects on an array of coaxial LED structures, as shown in Fig. 2. The goals of this part of the project include the determination of the path of least electrical resistance for the electrons and holes, both in the space charge region (InGaN/GaN MQW) and the neutral (n-GaN and p-GaN) regions. This information will be combined with the light extraction analysis (below) to determine the optimal electrode geometry, device dimensions, and epitaxial layer structure.
    The light extraction efficiency of an LED is defined as the fraction of photons generated in the active region that escape the device into free space. This quantity is primarily determined by the device geometry and optical loss processes. While light extraction techniques are well developed for planar LED structures, nanowire coaxial LEDs present unique challenges. The REU student will use commercial optical ray-tracing software (Light Tools) to model the nanowire LED structure and simulate the light extraction efficiency for various device configurations. The effects of device geometry, nano surface texturing, and semi-transparent electrodes will be explored. The goal is to improve the light extraction efficiency in a manner that is compatible with the designs proposed to optimize the current injection uniformity.

    Fig. 10. GaN nanowires.