The realization of commercial inertial fusion energy (IFE) depends critically on the availability of diode-pumped solid-state lasers (DPSSLs) that simultaneously deliver high pulse energy, high peak power, and high wall-plug efficiency at an economically viable cost per Watt. A central bottleneck has been the pump source itself: conventional single-junction (SJ) diode laser bars suffer from efficiency roll-off at the high pulse energies required for fusion-class DPSSLs, driven primarily by intrapulse heating. As a result, today ' s pump architectures often operate well below the efficiency optimum of SJ bars, fundamentally limiting system scalability and cost reduction Here we demonstrate that dual-junction (DJ) diode laser bars fundamentally change this paradigm. By vertically integrating two active junctions within a single bar geometry, DJ devices nearly double the differential quantum efficiency while operating at substantially reduced current density. Experimental results at wavelengths of 880 nm (Nd-3.-doped gain media) and 940 nm (Yb3+:YAG) show that DJ bars shift the efficiency maximum toward dramatically higher pulse energies and bar powers - precisely where fusion-relevant DPSSL pump systems must operate. At pulsewidths and operating points relevant to IFE, DJ bars maintain efficiencies exceeding 60% at kilowattclass bar powers, while SJ bars have already entered a regime of pronounced efficiency degradation. This shift of the efficiency maximum toward the real operating point represents an important system-level breakthrough. It enables significantly higher usable power per bar without sacrificing efficiency, directly resulting in a reduced number of bars, interconnects, cooling interfaces, testing, and assembly steps required in large-scale diode pump arrays. The resulting reduction in system complexity and component count provides a clear pathway toward lower $/W at the pump-module level - one of the most critical enablers for economically viable fusion laser drivers While challenges remain in managing intrapulse heating and junction-to-junction spectral alignment, our results show that these effects are tractable through epitaxial and thermal design optimization. Overall, dual-junction diode laser bars move the performance sweet spot of diode pumping to where fusion lasers actually operate. By aligning high efficiency with high absolute pump energy in a single device platform, DJ technology emerges as a key enabling innovation for next-generation DPSSLs and marks a pivotal step toward practical inertial fusion energy.

The slope efficiency and drive voltage of broad area AlInGaAs laser diodes near 865 nm is observed to decrease significantly under quasi-CW pulsed operation at currents well above threshold, in a manner that cannot be explained by thermal effects or carrier leakage over heterojunction barriers. Simulations show that the slope efficiency reduction is explicable by increased free carrier absorption in the waveguide region. Empirical formulas are presented to represent these effects in a closed analytic form suitable for use in simulators for diode-pumped laser systems.

The development, test and deployment of inertial fusion energy facilities depend critically on the economic supply of diode laser pumps. Efforts from innovation in automation to yield and volume scaling in semiconductor production and the development of ultra-low-cost packaging technologies are essential to lower purchase cost in Euros/dollars per Watt. It is equally important to substantially increase reliable output powers and conversion efficiency of the semiconductor devices, via continuous improvement in III-V technology. We therefore summarize current developments in industry and research in the performance scaling of diode pumps for fusion application. First, we show how efficient 1-cm diode bars with 1000 W output power are migrating from the laboratory towards mass production, enabled by design improvement and techniques such as high-throughput passivation. Next, we summarize recent research, that shows a path to scaling 1-cm bars toward multi-kilowatt powers, whilst efforts in defect elimination promise increased fabrication yield and longer lifetimes.

A method is proposed for determination of the pulse-to-pulse variation in junction temperature and emission wavelength of a semiconductor laser diode during a train of pulses. This approach, based on impulse response functions, enables predictions for pulse trains with arbitrary pulse-to-pulse variations in output power, pulse width, and pulse delay using a limited set of experimental characterization data. The use of this approach is illustrated by application to a particular device structure.

Using a nondestructive, in situ near-infrared defect imaging technique, we obtained time-resolved images of dark line defect (DLD) growth in the cavities of several high-power 790-nm diode lasers and calculated individual DLD growth rates. A clear longitudinal spatial dependence in the DLD growth rate was observed, in which the DLD growth rate was faster near the anti-reflective (AR) facet. Longitudinal spatial hole burning simulation results indicate that the total photon density increases toward the AR facet, while the carrier density decreases toward the AR facet. These results suggest that photon absorption plays a key role in enhancing DLD growth beyond the contribution of the traditional carrier capture mechanism. Asymmetric DLD growth that was preferentially directed toward the high-reflectivity facet was also observed, providing further support for the role of photon absorption in the growth process. To account for these findings, we propose a photon-accelerated mechanism for the enhancement of DLD growth in which the slower carrier capture process in the traditional recombination-enhanced dislocation glide mechanism is replaced by a faster photon absorption process.

An in situ and nondestructive technique is developed to image the formation and evolution of dark line defects in the cavity of a high-power diode laser. The technique uses broadband near infrared emission that originates in the laser's core layers and enables defects to be imaged with high spatial resolution through the substrate. In particular, it enables defect imaging through the substrate of shorter wavelength lasers, even when the substrate is opaque near the lasing wavelength. The evolution of dark line defects during aging is studied in several devices, with correlations established between the observed characteristics of defect growth and changes in device parameters such as optical power, operating wavelength, threshold current, and slope efficiency. Gradual degradation is found to be associated with dark line defects that slowly propagate from dark spots that are present in the device interior in its fresh (unaged) condition, rather than propagating from spots that are formed during aging, suggesting a strategy to screen fresh devices for expected reliability. This defect growth phenomenon is found to be particularly evident in the dark spots near the output facet.

Using a high sensitivity infrared camera, we image the optical cavity of an operating high-power diode laser through a window etched in the substrate and observe weak IR emission from the waveguide core region. The IR intensity maps show dark spots in the cavity that subsequently grow into line defects (all oriented in the same direction) as the laser ages. This technique holds promise as a nondestructive, in situ approach to study the formation and evolution of defects in an operating device. We also use CCD-based thermoreflectance to generate high-resolution facet temperature profiles of the same lasers during aging, with the results suggesting that the slow degradation of optical power that occurs prior to laser failure relates more to cavity defect formation than facet defect (hotspot) formation.

Power handling capabilities of broad-area high-power diode lasers are limited by the heat extraction capabilities of the device packaging. Traditional methods of heat extraction rely on conductive heat extraction from the diode chip and an emitting facet in contact with either quiescent or naturally convecting air. This leads to a thermal profile in the lasing direction of the cavity and a hot emitting facet. A hot facet accelerates material degradation, reducing the mean time to failure and limiting the safe operating power. Direct contact between the facet and a liquid coolant could enable higher levels of heat extraction compared to traditional cooling pathways. An innovative approach to cooling high-power, broad-area diode lasers via total immersion in liquid coolant is proposed and tested. In this study, we demonstrate that single emitters can operate with the emitting facet in direct contact with static coolant, with no negative change to device power or efficiency. Thermal analysis and models show that immersed diodes operate with improved thermal pathways, yielding lower total thermal resistance with the greatest improvement to thermal resistance at the facet-fluid interface.

Thin films of ZnO were grown by metal organic chemical vapor deposition (MOCVD) in a vertical injection rotating disk reactor (RDR) system on sapphire substrates. Kinetics of ZnO growth by MOCVD were studied and an optimal growth window for a RDR tool was determined. Experimental growth conditions were chosen based on calculations of Reynolds Number (Re) and mixed convection parameter in order to select a growth window with stable gas flow and uniform heat transfer. Growth parameters were systemically varied within this window to determine the optimal growth conditions for this MOCVD tool and to study how these parameters affect film growth and quality. Properties of ZnNiO films grown by MOCVD were also studied to determine the effects of Ni incorporation on structural, optical, and magnetic properties.

Silicon substrates offer a variety of advantages over conventional LED substrates, including both their low cost and their compatibility with the existing semiconductor manufacturing infrastructure. Researchers have sought to avail themselves of these benefits by developing GaN LEDs on silicon substrates. Integration of the two materials, however, presents several technical challenges that have proven difficult to address. The main challenges are the large lattice mismatch between GaN and silicon and the large difference in thermal expansion coefficients. Solutions have begun to emerge in recent years, with Toshiba being the first to commercialize GaN-on-Si technology on 200mm substrates in 2013, and several companies now shipping GaN-on-Si LEDs. Toshiba has developed a novel buffer layer structure and MOCVD growth technology that controls strain, wafer bow, and cracking while also suppressing formation of dislocations. This technology allows for the growth of GaN epitaxial layers with minimal cracking and dislocation densities in the low 108 cm-2. Control of strain and wafer bow is critical for the epitaxial process, and this requires very fine control over both absolute wafer temperature and temperature uniformity. Toshiba’s technology controls temperature uniformity very well, and Toshiba has demonstrated wavelength uniformity with σ=2nm across a 200mm wafer. Thin film LEDs fabricated from these materials have efficiencies on par with the best efficiencies of LEDs grown on conventional substrates. Blue LED wall-plug efficiencies of ~83% are demonstrated.

In this paper effects of NH3 doping on ZnO thin films grown by metal organic chemical vapor deposition (MOCVD) on c-plane sapphire substrates using diethyl zinc (DEZn) and O2 precursors and N2 as the carrier gas have been studied. NH3 flow rates were varied from 0.1% to 4% in the growth runs. All the runs were done at 500°C at 10 Torr pressure. The XRD measurements show a single ZnO (002) peak. Raman data for the samples confirms presence of ZnO:N modes at 275cm−1, 510cm−1 and 575 cm−1 and 645cm−1. The PL results for Zn rich films show weak broad peaks centered at 480nm and 650nm with no ZnO band edge emission, while oxygen rich films show weak ZnO band edge emission and a strong broad orange peak centered at 650nm. Hall effect measurements indicate that all of the as-grown films are highly resistive. Some are weakly p-type with carrier concentration of 4.24 × 1014 cm−3 and mobility of 16.55 cm2/Vs. Annealing in N2 ambient for 60 minutes at 800°C enhances the PL band edge emission and converts all the films to highly conducting n-type, with carrier concentration on the order of 8 × 1018 cm−3, mobility on the order of 12 cm2/Vs and resistivity of 0.063 Ω-cm.

A series of ZnO thin films with different thicknesses grown on sapphire substrates by metalorganic chemical vapor deposition (MOCVD) have been studied by different characterization techniques. The optical properties are investigated by photoluminescence (PL), optical transmission (OT) and 1st order derivatives, various angle scanning ellipsometry (VASE). Rutherford Backscattering (RBS) shows the atomic Zn:O ratios with a few percentage aviation from 1:1, and thicknesses in range of 10~230 nm, roughness layer with 10~30nm, which are corresponding to results from atomic force microscopy (AFM), and scanning electron microscopy (SEM). The optical and structure characterization measurements have confirmed the good quality of these epitaxial ZnO materials.

InGaN alloys are widely researched in diverse optoelectronic applications. This material has also been demonstrated as a photovoltaic material. This paper presents the study to achieve optimum electrically active p-type InGaN epi-layers. Mg doped InGaN films with 20% In composition are grown on GaN templates/sapphire substrates by MOCVD. It is found that the hole concentration of p-type InGaN depends strongly on the Mg flow rate and V/III molar ratio and hole concentration greater than 2×1019 cm-3 has been achieved at room temperature. The optimum activation temperature of Mg-doped InGaN layer has been found to be 550–600°C, which is lower than that of Mg-doped GaN. A solar cell was realized successfully using the InGaN epi-layers presented here.

GaN thin films have been grown on Si(111) substrates using an atomic layer deposition (ALD)-grown Al2O3 interlayer. This thin Al2O3 layer reduces strain in the subsequent GaN layer, leading to lower defect densities and improved material quality compared to GaN thin films grown by the same process on bare Si. XRD ω-scans showed a full width at half maximum (FWHM) of 549arcsec for GaN grown on bare Si and a FWHM as low as 378arcsec for GaN grown on Si using the ALD-grown Al2O3 interlayer. Raman spectroscopy was used to study the strain in these films in more detail, with the shift of the E2(high) mode showing a clear dependence of strain on Al2O3 interlayer thickness. This dependence of strain on Al2O3 thickness was also observed via the redshift of the near bandedge emission in room temperature photoluminescence (RT-PL) spectroscopy. The reduction in strain results in a significant reduction in both crack density and screw dislocation density compared to similar films grown on bare Si. Screw dislocation density of the films grown on Al2O3/Si substrates approaches that of typical GaN layers on sapphire. This work shows great promise for the use of oxide interlayers for growth of GaN-based LEDs on Si.

Device-quality GaN thin films have been grown on Si(111) substrates using an Al2O3 transition layer, and initial devices show performance similar to comparable devices on sapphire. X-ray diffraction rocking curve scans show a linewidth of 378 arcsec for the GaN (0002) reflection. Comparison of these layers to GaN layers grown on bare Si substrates shows a significant reduction in strain with the use of the Al2O3 transition layer. Raman spectroscopy measurements verify this reduction in strain, as shown by the shift of the GaN E2(high) with variations in Al2O3 layer thickness. GaN-based devices were also grown and fabricated using this process. Devices on Si showed an IQE of ~32%, which is comparable to the ~37% observed for similar devices on sapphire. The devices on Si also showed better efficiency at high current densities compared to the devices on sapphire, despite the longer peak emission wavelength on Si, which may be due to a difference in thermal conductivity between Si and sapphire. A growth process has been developed for high-quality GaN on Si, and initial device results show that Si is a viable substrate technology for MOCVD growth of GaN-based devices.

InGaN alloys are widely researched in diverse optoelectronic applications. This material has also been demonstrated as a photovoltaic material. This paper presents the study to achieve optimum electrically active p-type InGaN epi-layers. Mg doped InGaN films with 20% In composition are grown on GaN templates/sapphire substrates by MOCVD. It is found that the hole concentration of p-type InGaN depends strongly on the Mg flow rate and V/III molar ratio and hole concentration greater than 2×10 19 cm -3 has been achieved at room temperature. The optimum activation temperature of Mg-doped InGaN layer has been found to be 550–600°C, which is lower than that of Mg-doped GaN. A solar cell was realized successfully using the InGaN epi-layers presented here.

p -type, n-type and semi-insulating ZnO:N thin films were successfully grown by metal organic chemical vapor deposition on c-plane sapphire using diethyl zinc and O2 precursors, N2 carrier gas, and NH3 as dopant. NH3 flow rates were varied from 0.2% to 4% in the growth runs. The resulting films were characterized for their structural, optical, and electrical properties by scanning electron microscopy, x-ray diffraction (XRD), Raman spectroscopy, photoluminescence (PL), and Hall effect measurements. XRD show a single ZnO (002) peak; Raman data show the presence of ZnO:N modes at 275, 510, 575, and 645cm−1; and PL results show broad peaks at 480 and 600nm corresponding to deep N incorporation for all the samples. Hall effect show n-type films with carrier concentrations of 6.57×1018cm−3, p-type with carrier concentrations of 4.24×1014cm−3, and semi-insulating with resistivity on the order of 1.5×105Ωcm.

In this report we present recent results for MOCVD growth of high indium content InGaN films on ZnO substrates. Growth was attempted on both bulk ZnO as well as ZnO epilayers grown on sapphire by MOCVD. ZnO is an attractive alternative substrate for III-Nitrides because of its superior lattice match: specifically ZnO is perfectly matched with In0.18Ga0.82N and low cost of substrates. Stable InGaN films with >18% indium were achieved on the bulk substrates and were characterized by HRXRD, PL, and optical transmission. Varying the growth parameters - primarily growth temperature and In/(In + Ga) flow ratio - was found to affect the optical and structural properties of the films. By growing on a better matched substrate the high indium composition InGaN epitaxial films experience less strain and can therefore be grown thicker without creating relaxation-induced extended crystal defects. This is important, as high indium content InGaN films cannot be grown on GaN thick enough for full above-bandgap absorption without introducing detrimental extended crystal defects. This limitation is thought to be a limiting factor in the achievable ISC in InGaN solar cells.