We developed a pyrometer that operates near the high-temperature bandgap of GaN, thus solving the transparency problem once a {approx} 1 {micro}m thick GaN epilayer has been established. The system collects radiation in the near-UV (380-415 nm) and has an effective detection wavelength of {approx}405 nm. By simultaneously measuring reflectance we also correct for emissivity changes when films of differing optical properties (e.g. AlGaN) are deposited on the GaN template. We recently modified the pyrometer hardware and software to enable measurements in a multiwafer Veeco D-125 OMVPE system. A method of synchronizing and indexing the detection system with the wafer platen was developed; so signals only from the desired wafer(s) could be measured, while rejecting thermal emission signals from the platen. Despite losses in optical throughput and duty cycle we are able to maintain adequate performance from 700 to 1100 C.
Experimental studies have been performed on the velocity-field characteristics of AlGaN/GaN heterostructures. A pulsed voltage input in combination with a four-point measurement was used in a 50 {Omega} environment to determinethe drift velocity of electrons in the two-dimensional electron gas as a function of the applied electric field. These measurements show an apparent saturation velocity near 3.1 x 10{sub 7} cm/s, at a field of 140 kV/cm. A comparison of these studies shows that the experimental velocities are close to previously published simulations based upon Monte Carlo techniques.
The goal of this one year LDRD was to improve the overall efficiency of InGaN LEDs by improving the extraction of light from the semiconductor chip. InGaN LEDs are currently the most promising technology for producing high efficiency blue and green semiconductor light emitters. Improving the efficiency of InGaN LEDs will enable a more rapid adoption of semiconductor based lighting. In this LDRD, we proposed to develop photonic structures to improve light extraction from nitride-based light emitting diodes (LEDs). While many advanced device geometries were considered for this work, we focused on the use of a photonic crystal for improved light extraction. Although resonant cavity LEDs and other advanced structures certainly have the potential to improve light extraction, the photonic crystal approach showed the most promise in the early stages of this short program. The photonic crystal (PX)-LED developed here incorporates a two dimensional photonic crystal, or photonic lattice, into a nitride-based LED. The dimensions of the photonic crystal are selected such that there are very few or no optical modes in the plane of the LED ('lateral' modes). This will reduce or eliminate any radiation in the lateral direction so that the majority of the LED radiation will be in vertical modes that escape the semiconductor, which will improve the light-extraction efficiency. PX-LEDs were fabricated using a range of hole diameters and lattice constants and compared to control LEDs without a photonic crystal. The far field patterns from the PX-LEDs were dramatically modified by the presence of the photonic crystal. An increase in LED brightness of 1.75X was observed for light measured into a 40 degree emission cone with a total increase in power of 1.5X for an unencapsulated LED.
The AlGaInN material system is used for virtually all advanced solid state lighting and short wavelength optoelectronic devices. Although metal-organic chemical vapor deposition (MOCVD) has proven to be the workhorse deposition technique, several outstanding scientific and technical challenges remain, which hinder progress and keep RD&A costs high. The three most significant MOCVD challenges are: (1) Accurate temperature measurement; (2) Reliable and reproducible p-doping (Mg); and (3) Low dislocation density GaN material. To address challenge (1) we designed and tested (on reactor mockup) a multiwafer, dual wavelength, emissivity-correcting pyrometer (ECP) for AlGaInN MOCVD. This system simultaneously measures the reflectance (at 405 and 550 nm) and emissivity-corrected temperature for each individual wafer, with the platen signal entirely rejected. To address challenge (2) we measured the MgCp{sub 2} + NH{sub 3} adduct condensation phase diagram from 65-115 C, at typical MOCVD concentrations. Results indicate that it requires temperatures of 80-100 C in order to prevent MgCp{sub 2} + NH{sub 3} adduct condensation. Modification and testing of our research reactor will not be complete until FY2005. A new commercial Veeco reactor was installed in early FY2004, and after qualification growth experiments were conducted to improve the GaN quality using a delayed recovery technique, which addresses challenge (3). Using a delayed recovery technique, the dislocation densities determined from x-ray diffraction were reduced from 2 x 10{sup 9} cm{sup -2} to 4 x 10{sup 8} cm{sup -2}. We have also developed a model to simulate reflectance waveforms for GaN growth on sapphire.
We develop a reciprocal-space model that describes the (hkl) dependence of the broadened Bragg peakwidths produced by x-ray diffraction from a dislocated epilayer. We compare the model to experiments and find that it accurately describes the peakwidths of 16 different Bragg reflections in the [010] zone of both GaN and AlN heterolayers. Using lattice-distortion parameters determined by fitting the model to selected reflections, we estimate threading-dislocation densities for seven different GaN and AlGaN samples and find improved agreement with transmission electron microscopy measurements.
Ohmic contacts on p-type GaN utilizing Pd/Ir/Au metallization were fabricated and characterized. Metallized samples that were rapid thermally annealed at 400 C for 1 min exhibited linear current-voltage characteristics. Specific ohmic contact resistivities as low as 2 x 10{sup -5} {Omega} cm{sup 2} were achieved. Auger electron spectroscopy and x-ray photoelectron spectroscopy depth profiles of annealed Pd/Ir/Au contact revealed the formation of Pd- and Ir-related alloys at the metal-semiconductor junction with the creation of Ga vacancies below the contact. The excellent contact resistance obtained is attributed to the formation of these Ga vacancies which resulted in the reduction of the depletion region width at the junction.
Forward-to-reverse bias step-recovery measurements were performed on In.07Ga.93N/GaN and Al.36Ga.64N/Al.46Ga.54N quantum-well (QW) light-emitting diodes grown on sapphire. With the QW sampling the minority-carrier hole density at a single position, distinctive two-phase optical decay curves were observed. Using diffusion equation solutions to self-consistently model both the electrical and optical responses, hole transport parameters tp = 758 {+-} 44 ns, Lp = 588 {+-} 45 nm, and up = 0.18 {+-} 0.02 cm2/Vs were obtained for GaN. The mobility was thermally activated with an activation energy of 52 meV, suggesting trap-modulated transport. Optical measurements of sub-bandgap peaks exhibited slow responses approaching the bulk lifetime. For Al.46Ga.54N, a longer lifetime of tp = 3.0 us was observed, and the diffusion length was shorter, Lp = 280 nm. Mobility was an order of magnitude smaller than in GaN, up = 10-2 cm2/Vs, and was insensitive to temperature, suggesting hole transport through a network of defects.
High-resolution, variable temperature PL experiments were performed in the spectral region associated with recombination processes involving the ground and excited states of the neutral donor bound excitons. High-resolution infrared measurements in combination with high-sensitive SIMS unambiguously identified Si and O shallow donors and yield their ground state binding energies. These binding energies are in excellent agreement with values obtained by the analysis of the two-electron-satellite PL spectra considering the participation of ground and excited state donor bound excitons. This work clarifies conflicting aspects existing in donor identification and the binding energies of the impurities and excitons.
Solid-state light sources emitting at wavelengths less than 300 nm would enable technological advances in many areas such as fluorescence-based biological agent detection, non-line-of-sight communications, water purification, and industrial processing including ink drying and epoxy curing. In this paper, we present our recent progress in the development of LEDs with emission between 237 and 297 nm. We will discuss growth and design issues of deep-UV LEDs, including transport in Si-doped AlGaN layers. The LEDs are designed for bottom emission so that improved heat sinking and light extraction can be achieved by flip chipping. To date, we have demonstrated 2.25 mW of output power at 295 nm from 1 mm x 1 mm LEDs operated at 500 mA. Shorter wavelength LEDs emitting at 276 nm have achieved an output power of 1.3 mW at 400 mA. The heterostructure designs that we have employed have suppressed deep level emission to intensities that are up to 330 x lower than the primary quantum well emission.
Using in situ wafer-curvature measurements of thin-film stress, we determine the critical thickness for strain relaxation in Al{sub x}Ga{sub 1-x}N/GaN heterostructures with 0.14 {le} x {le} 1. The surface morphology of selected films is examined by atomic force microscopy. Comparison of these measurements with critical-thickness models for brittle fracture and dislocation glide suggests that the onset of strain relaxation occurs by surface fracture for all compositions. Misfit-dislocations follow initial fracture, with slip-system selection occurring under the influence of composition-dependent changes in surface morphology.
Optically detected magnetic resonance (ODMR) has been performed on Si-doped GaN homoepitaxial layers grown by organometallic chemical vapor deposition on free-standing GaN templates. In addition to intense excitonic bandedge emission with narrow linewidths (<0.4 meV), these films exhibit strong shallow donor-shallow acceptor recombination at 3.27 eV. Most notably, ODMR on this photoluminescence band reveals a highly anisotropic resonance with g{sub {parallel}} = 2.193 {+-} 0.001 and g{sub {perpendicular}} {approx}0 as expected for effective-mass shallow acceptors in wurtzitic GaN from k {center_dot} p theory. This previously elusive result is attributed to the much reduced dislocation density and impurity levels compared to those typically found in the widely investigated Mg-doped GaN heteroepitaxial layers. The possible chemical origin of the shallow acceptors in these homoepitaxial films will be discussed.
This SAND report is the final report on Sandia's Grand Challenge LDRD Project 27328, 'A Revolution in Lighting -- Building the Science and Technology Base for Ultra-Efficient Solid-state Lighting.' This project, which for brevity we refer to as the SSL GCLDRD, is considered one of Sandia's most successful GCLDRDs. As a result, this report reviews not only technical highlights, but also the genesis of the idea for Solid-state Lighting (SSL), the initiation of the SSL GCLDRD, and the goals, scope, success metrics, and evolution of the SSL GCLDRD over the course of its life. One way in which the SSL GCLDRD was different from other GCLDRDs was that it coincided with a larger effort by the SSL community - primarily industrial companies investing in SSL, but also universities, trade organizations, and other Department of Energy (DOE) national laboratories - to support a national initiative in SSL R&D. Sandia was a major player in publicizing the tremendous energy savings potential of SSL, and in helping to develop, unify and support community consensus for such an initiative. Hence, our activities in this area, discussed in Chapter 6, were substantial: white papers; SSL technology workshops and roadmaps; support for the Optoelectronics Industry Development Association (OIDA), DOE and Senator Bingaman's office; extensive public relations and media activities; and a worldwide SSL community website. Many science and technology advances and breakthroughs were also enabled under this GCLDRD, resulting in: 55 publications; 124 presentations; 10 book chapters and reports; 5 U.S. patent applications including 1 already issued; and 14 patent disclosures not yet applied for. Twenty-six invited talks were given, at prestigious venues such as the American Physical Society Meeting, the Materials Research Society Meeting, the AVS International Symposium, and the Electrochemical Society Meeting. This report contains a summary of these science and technology advances and breakthroughs, with Chapters 1-5 devoted to the five technical task areas: 1 Fundamental Materials Physics; 2 111-Nitride Growth Chemistry and Substrate Physics; 3 111-Nitride MOCVD Reactor Design and In-Situ Monitoring; 4 Advanced Light-Emitting Devices; and 5 Phosphors and Encapsulants. Chapter 7 (Appendix A) contains a listing of publications, presentations, and patents. Finally, the SSL GCLDRD resulted in numerous actual and pending follow-on programs for Sandia, including multiple grants from DOE and the Defense Advanced Research Projects Agency (DARPA), and Cooperative Research and Development Agreements (CRADAs) with SSL companies. Many of these follow-on programs arose out of contacts developed through our External Advisory Committee (EAC). In h s and other ways, the EAC played a very important role. Chapter 8 (Appendix B) contains the full (unedited) text of the EAC reviews that were held periodically during the course of the project.
AlGaN/GaN test structures were fabricated with an etched constriction. A nitrogen plasma treatment was used to remove the disordered layer, including natural oxides on the AlGaN surface, before the growth of the silicon nitride passivation film on several of the test structures. A pulsed voltage input, with a 200 ns pulse width, and a four-point measurement were used in a 50 {Omega} environment to determine the room temperature velocity-field characteristic of the structures. The samples performed similarly over low fields, giving a low-field mobility of 545 cm{sup 2} V{sup -1} s{sup -1}. The surface treated sample performed slightly better at higher fields than the untreated sample. The highest velocity measured was 1.25 x 10{sup 7} cm s{sup -1} at a field of 26 kV cm{sup -1}.
Wickenden, A.E.; Koleske, Daniel K.; Henry, R.L.; Twigg, M.E.; Fatemi, M.
The relationship of GaN resistivity to film microstructure and impurity compensation are investigated using transmission electron microscopy, secondary ion mass spectroscopy, X-ray diffraction, and resistance measurements. Unintentionally doped GaN films grown by MOCVD at varying pressures exhibit increased grain size, reduced carbon and oxygen impurity incorporation, reduction in the density of threading dislocations (TDs) with an edge component, and reduced resistivity with increasing growth pressure. Variation in resistivity over eight orders of magnitude is observed as a result of varying the MOCVD growth pressure in a controlled experiment. Our results suggest that disclocations play an important role in the resistivity of GaN. Evidence is presented of impurities segregating at TDs having an edge component, and acting as compensating centers. The control of such compensation as a function of MOCVD growth conditions is outlined. Published by Elsevier B.V.
To observe the effects of polarization fields and screening, we have performed contacted electroreflectance (CER) measurements on In{sub 0.07}Ga{sub 0.93}N/GaN single quantum well light emitting diodes for different reverse bias voltages. Room-temperature CER spectra exhibited three features which are at lower energy than the GaN band gap and are associated with the quantum well. The position of the lowest-energy experimental peak, attributed to the ground-state quantum well transition, exhibited a limited Stark shift except at large reverse bias when a redshift in the peak energy was observed. Realistic band models of the quantum well samples were constructed using self-consistent Schroedinger-Poisson solutions, taking polarization and screening effects in the quantum well fully into account. The model predicts an initial blueshift in transition energy as reverse bias voltage is increased, due to the cancellation of the polarization electric field by the depletion region field and the associated shift due to the quantum-confined Stark effect. A redshift is predicted to occur as the applied field is further increased past the flatband voltage. While the data and the model are in reasonable agreement for voltages past the flatband voltage, they disagree for smaller values of reverse bias, when charge is stored in the quantum well, and no blueshift is observed experimentally. To eliminate the blueshift and screen the electric field, we speculate that electrons in the quantum well are trapped in localized states.
We demonstrate that when vertical threading dislocations in (0001) GaN are imaged in plan-view by transmission electron microscopy, a surface-relaxation contrast operates in addition to that due to the strain fields of dislocations passing through the specimen. We show that all three dislocation types (edge, screw, and mixed) can be detected in the same image using g = (11{bar 2}0) and 18{sup o} specimen tilt from [0001], allowing total densities to be assessed properly. The type of an individual dislocation can also be readily identified.
Epitaxial Si-doped GaN layers were studied using a low-energy positron beam, where the grain size varies from 0.2 to 2-5 μm. The concentration of the negatively charged Ga vacancies was found to be independent of the grain size, which showed that Ga vacancies exist in the grain interior. Positrons were also observed to get trapped also at other negatively charged centers.