United States

Abstract Body: In recent years, III-V semiconductor nanowires (NWs) have significantly attracted researchers due to one-dimensional architecture, quantum confinement effects, and a higher tolerance for stress-strain mismatch that allow greater freedom in engineering combinations of material systems in a variety of various nanowire architectures to meet the demands of next-generation optoelectronic devices. Dopant incorporation in a well-controlled manner is essential to realize advanced devices in NW configuration successfully. Unfortunately, the knowledge obtained on dopant incorporation and carrier concentration from the thin film studies cannot be directly translated to NWs due to the dopant’s influence on the growth kinetics and different growth mechanisms along with the axial and radial directions. The commonly used measurement techniques in thin films for the determination of carrier concentration namely, Hall effect, field-effect transistor, and capacitance-voltage require highly sophisticated lithography steps. In the last decade, several characterization methods have evolved for the assessment of dopants in NWs namely, off-axis electron holography for which the sample preparation is complex, and it requires additional information on the NW thickness and homogeneity; secondary ion mass spectrometry and atom probe tomography requires a standard of known dopant concentration and is destructive. Therefore, a combination of conductive-atomic force microscopy (C-AFM), scanning Kelvin probe microscopy (SKPM), X-ray photoelectron spectroscopy (XPS), and ultraviolet photoelectron spectroscopy (UPS), provide a set of excellent characterization techniques for doping assessment, as they do not involve any complex contact fabrications. In this work, we evaluate the carrier concentration and incorporation of Beryllium (Be) dopants in GaAs NWs and Tellurium (Te) dopants in GaAsSb NWs grown using Ga-assisted molecular beam epitaxy (MBE) with variation in Be cell temperatures (TBe) from 750 oC to 950 oC and Te cell temperatures (TGaTe) from 500 oC to 570 oC, respectively. The extensive research interest in GaAs and GaAsSb NWs stems from its narrow bandgap region covering the important optical telecommunication wavelengths of 1.3 µm and 1.55 µm. The current-voltage measurements using C-AFM of the vertical single NW showed a significant enhancement in current for GaAs NWs doped with TBe at 950 oC and GaAsSb NWs doped with TGaTe at 550 oC. The carrier concentration of Be-doped GaAs and Te-doped GaAsSb NWs are found using COMSOL Multiphysics fitting. From XPS and UPS, the atomic percent, work function, and carrier concentration of Be-doped GaAs NWs and Te-doped GaAsSb NWs are determined. A shift in Fermi level towards the valence band with increasing TBe for GaAs NWs, provided evidence towards higher Be incorporation with TBe - 950 oC doped NWs having a work function of 5.6 eV. Whereas the Fermi level is shifted towards the conduction band in Te–doped GaAsSb NWs. The values of electron density from UPS concur very well with the values of electron density determined from C-AFM. The change in surface potential observed in doped NWs in SKPM characterization also provided strong evidence of Be and Te incorporation. Hence, these surface analytical tools without any contact fabrication are found to be powerful characterization techniques for direct measurement of the dopant levels in NWs, which are critical for bandgap engineering design of optoelectronic devices. Acknowledgement This material is based upon research supported by the Air Force Office of Scientific Research (AFOSR) under grant number W911NF1910002.

EMC 2021

Doping Assessment of Ga-Assisted MBE Grown Be-Doped GaAs and Te-Doped GaAsSb Nanowires for Infrared Photodetector Application

ups.

skpm

c-afm

nanowires

The Electronic Materials Conference (EMC) is the premier annual forum on the preparation, characterization and use of electronic materials. This year’s event will feature a plenary session, topical sessions, a poster session and more.

EMC is coordinated with the Device Research Conference (DRC), which will be held during the same week, from June 20-23. This recognizes the strong interaction between device and electronic materials research and provides fruitful exchange of information between attendees of both Conferences.

![](https://assets.underline.io/uploads/markdown_image/1/image/eb26c24a18d5d1030616afa4f993cf89.png)
	
<a href="https://www.mrs.org/docs/default-source/meetings-events/mrs-conference-services/2021/emc-2021/2021-emc-program-at-a-glance-6-16-21.pdf?sfvrsn=e39f950c_4">Latest Program-at-a-Glance</a>

<a href="https://drive.google.com/file/d/1eiS1ao4p83VNy8xgHs3Rls7gTQjx7Rc4/view?usp=sharing">Full Program with Abstracts</a>

technical paper

Abstract Body: The first reports of applying the Boltzmann Sigmoidal model on pitch-dependency in patterned nanowires (NWs) of dilute nitride GaAsSbN on p-Si (111) substrates by self-catalyzed plasma-assisted molecular beam epitaxy are presented in this study. Morphological and structural characterizations of the patterned NWs are accomplished via field emission scanning electron microscope (FESEM) and high-resolution transmission electron microscope (TEM), respectively. Top, middle, and bottom NW segments captured via selected area electron diffraction (SAED) characterization confirmed the single-crystal zincblende structure of the NWs. The absence of satellite spots in the SAED pattern indicates the NWs are free of planar defects. Sizeable bandgap tuning of ~75 meV, as ascertained from 4K photoluminescence (PL), is demonstrated over a pitch length variation of 200 nm to 1200 nm. Axial and radial growth rates reveal a logistic sigmoidal growth trend with respect to pitch, which differs from those commonly observed in other patterned non-nitride III-V NWs. The Boltzmann sigmoidal model offers additional insight into the shift of the PL spectral arising from differences in Sb and N incorporation from pitch induced variation in secondary fluxes. The extracted inflection points, p0, of 636 nm and 806 nm, for axial growth rate correlate very well with the homogeneous broad peaks revealed by PL analysis at these pitch lengths. Furthermore, the inflection point indicates the pitch length where the pitch dependent effects on the secondary fluxes play a prominent role in axial and radial growth of patterned NWs. Additionally, the values of the sigmoidal slope, α, correlate well with the range of pitches where Sb and N incorporation are observed with broad PL spectra. Results indicate that the extracted parameters obtained via the sigmoidal model of inflection point, and slope can be used as a rubric for optimal pitch length for patterned array design of homogeneous nitride NWs and can be extended to other highly mismatched alloy compounds. This work is financially supported by National Science Foundation (Award No. 1649517) and through the Title III HGBI Ph.D. Fellowship.

Analysis of Patterned GaAsSbN Nanowires via Boltzmann Sigmoidal Model

Abstract Body: Two-dimensional materials based on transition metal carbides have been intensively studied due to their unique properties including metallic conductivity, hydrophilicity, and structural diversity and have shown great potential in several applications. While MXenes based on magnetic transition elements show interesting magnetic properties, not much is known about the magnetic properties of titanium-based MXenes. Here, we measured magnetic properties of Ti3C2Tx MXenes synthesized by different chemical etching conditions. Our measurements indicate that there is a paramagnetic–antiferromagnetic (PM-AFM) phase transition, and the transition temperature depends on the synthesis procedure of MXenes. Our observation indicates that the magnetic properties of these MXenes can be tuned by the extent of chemical etching which can be beneficial for the design of MXene-based spintronic devices.

Tuning the Magnetic Properties of Two-Dimensional MXenes by Chemical Etching

Abstract Body: Metamaterials and metasurfaces are promising optical materials for many applications in optics and optoelectronics as the optical characteristics can be engineered via sub-wavelength features. Hyperbolic metamaterials and metasurfaces—a particular class of these designer materials—have been widely studied due their uncommon optical characteristics that enable effects such as negative refraction, light beam steering, subdiffraction imaging, spontaneous and thermal emission engineering, and broadband optical absorption. These materials have been demonstrated across much of the optical spectrum using myriad approaches. Recently, negative refraction of optical surface modes was accomplished using a hyperbolic metasurface comprising coupled nanoridge arrays. For applications in super-resolution imaging, hyperbolic metasurfaces are of particular interest because a larger fraction of the electromagnetic field will interact with the object being imaged, rather than residing inside of the bulk of the material. One approach to imaging using metasurfaces is to leverage structured illumination imaging (SIM), a paradigm where an object is illuminated via a sub-diffraction pattern that can be externally controlled, such as by varying the angle of the incident light. In SIM, a super-resolution image is computed from several diffraction-limited images. Exciting and localizing modes with large effective modal indices provide a path towards increasing the resolution beyond conventional SIM approaches, such as localized surface plasmon SIM. However, coupling to these high-index modes is challenging. In this talk, we will demonstrate hyperbolic metasurfaces appropriate for SIM. These metasurfaces utilize finite-length coupled nanoridge arrays. The coupling between nanoridges is used to engineer the effective modal index of the supported optical modes. These optical modes are then confined by introducing small gaps (90 nm) along the nanoridges, resulting in standing waves that can serve as an illumination pattern for SIM. Large arrays of coupled-nanoridge metasurfaces are fabricated using silver on Si via electron beam lithography and standard metal lift-off. The arrays are characterized using angle-, polarization-, and wavelength-dependent reflection spectroscopy. Optical modes corresponding to multipolar plasmon resonances (n = 2 and n = 3) are identified, including a mode that is identified as a localized hyperbolic mode. The energy and dispersion of all the observed modes agree well with numerical models of the effective modal index. Numerical simulations of the reflection of free-space light also agree well with the measurements. A numerical demonstration of the imaging capabilities for these modes at 458 nm of 9 nm quantum dots, including realistic material losses, predicts a 3.3X improvement in the resolution over diffraction-limited images.

Hybridization of Multi-Plasmon Modes on Coupled Nanoridge Array Metasurfaces for Super-Resolution Imaging

Abstract Body: Cadmium oxide (CdO) is one of the most promising material systems for low-loss infrared (IR) plasmonics to date. Through reactive co-sputtering using high power impulse magnetron sputtering (HiPIMS) and radio frequency (RF) sputtering, doped CdO demonstrates high mobility films with tunable carrier concentrations across a range corresponding to epsilon-near-zero (ENZ) modal frequencies spanning the mid-IR and reaching into the near-IR. Specifically, indium-doped CdO (In:CdO) yields carrier densities ranging between 2x1019 cm-3 and 5x1020 cm-3 while maintaining mobilities between 300-400 cm2/Vs across this range, thus allowing for the widest accessibility of the IR spectrum (1650-5325 cm-1) of a single plasmonic material grown by sputtering. Fully accessing the spectrum of plasmonic applications, including surface plasmon oscillations, ENZ modes, and strong coupling phenomena requires that one maintains these appealing transport properties over a film thickness range spanning a few tens to a few thousands of nanometers. Extensive CdO fabrication experiments using high-power impulse magnetron sputtering reveal a strong dependence of carrier density and mobility on film thickness when thickness drops below 100 nm, a particularly significant range for ENZ modes. We attribute this behavior to well-reported surface charge accumulation and depletion layers caused by Fermi level pinning. In this presentation, we demonstrate the reversal between surface charge accumulation and depletion in In:CdO. Further, we show how these surface effects can influence optical properties, and how they can be controlled by understanding and manipulating CdO defect chemistry. Lastly, we will show preliminary results of plasmonic tunability via ferroelectric switching.

Mid-Wave to Near-IR Optoelectronic Properties and Epsilon-Near-Zero Behavior in In-Doped CdO

Abstract Body: Ultrawide-bandgap (UWBG) Al-rich (x ≥ 0.4) AlGaN metal oxide heterostructure field-effect-transistors (MOSHFETs) are ideal device candidates for solar-blind detection due to their internal transistor gain and inherent solar blind detection (λ ≤ 300nm) ability. We previously reported the deep-trap energies and physical locations responsible for slower transient time in Al0.65Ga0.35N/Al0.45Ga0.55N MOSHFETs and suggested using back-barrier (BB) to reduce the trapping effects by screening the channel from the growth interface. We, now, report the optical response and current transient behaviors of AlxGa1-xN MOSHFETs with a fixed Al0.65Ga0.35N back-barrier and linearly graded AlxGa1-xN back-barrier. The transfer curves (Id-Vg) shift towards more negative (~0.63V for fixed BB and ~1.05V for graded BB) under light (λ ~ Eg,channel) in both cases indicates the presence of photoelectrons. The MOSHFETs recovered with light turned off, although a small threshold shift remained (Vth,shift), by 0.15 and 0.45V (traps ~ 8×1011 and 6.5×1011 cm-2) in 3 minutes for fixed and graded BB MOSHFETs respectively. Peak responsivities of 3.1×105 and 2.3×105A/W in the saturation mode for fixed and graded BB MOSHFETs were quenched by one order with strong 365nm UV light used to fill the traps. Consequently, the off-transient time of over 10 minutes reduced to ≤ 12s under 365nm light, while the recovery transient became more symmetric, indicating the presence of deep-traps that were filled in the back-barrier MOSHFETs. These results show that the BB does not have a significant impact on the speed of these devices due to the presence of these deep traps, We will present gate length (LG) dependent transient measurements with gated-transmission line model (GTLM) MOSHFET devices without BB to isolate transients arising from the access region and from under the channel. Our initial analysis suggests that the transients are similar for both, strongly indicating that the growth interface where the BB was placed is still responsible for slowing down the device. This result is surprising given that the density of those interface traps is ~5×1011 cm-2, much smaller compared to the polarization charge introduced by the BB ~1013cm-2, suggesting that clustering of point defects near dislocations may play a role as in GaN pn diodes.

A Comparative Study of Slow Current Transients in Al0.4Ga0.6N Channel MOSHFET’s with Back Barriers

Abstract Body: Owing to their noncentrosymmetric crystal structure, III-nitride semiconductors exhibit internal polarization fields which can be exploited to dramatically expand the design space of polar heterostructures. Due to their strength (1-10 MV/cm), these built-in electric fields can be harnessed for the realization of novel device functionalities via band-structure engineering and polarization-induced doping. This fundamentally different doping technique completely dispenses with the need of impurity donors or acceptors to generate free carriers, enabling the realization of highly mobile 2D electron and hole gases (2DEGs and 2DHGs) in nitride heterostructures. While polarization-induced doping has been previously engineered in nitride structures grown atop sapphire templates, SiC wafers, and single-crystal GaN substrates; further improvements in device performance are expected when 2DEGs and 2DHGs are grown on high-quality single-crystal AlN substrates. This is possible thanks to the high thermal conductivity and high breakdown electric field of bulk AlN crystals. These outstanding material properties allow not only improved thermal management for high-power devices but also takes advantage of the widest bandgap within the family of nitride semiconductors. This last benefit, however, also prevents efficient generation of free carriers via traditional impurity doping techniques. In this scenario, we leverage the intense built-in spontaneous and piezoelectric polarization fields to generate both types free carriers via polarization engineering. In this presentation, we report, for the first time, the epitaxial growth and transport properties of polarization-induced 2DEGs and 2DHGs in GaN/AlN heterostructures, homoepitaxially grown on single-crystal AlN substrates. The simultaneous demonstration of complementary 2DEG and 2DHG in undoped heterostructures and on a single substrate platform, enables an unmatched level of integration for high-power and high-frequency nitride-based electronics, raising hopes for all-nitride CMOS technology.

Polarization-Induced 2D Electron and Hole Gases Homoepitaxially Grown on Single-Crystal AlN Substrates

Abstract Body: The use of p-type superlattices (SL) is valuable for III-nitride semiconductor-based LEDs and pFETs (p-type Field effect transistors) because the polarization effects create a periodic oscillation of the energy bands, enhancing the ionization of the deep acceptors (Ex: Mg) [1-3]. Recently, the existence of acceptor traps 0.8eV above the valence band of GaN, located at the positive polarization interfaces in III-nitride systems, was shown [4]. This acceptor trap level was the main source of holes in systems which had insufficient or no Mg doping in them. The present study experimentally demonstrates the effect of these acceptor traps on the charge and mobility of holes in N-polar p-type GaN/(AlN/AlGaN) superlattices. N-polar modulation doped GaN/(AlN/Al0.24Ga0.75N) 7-period superlattices were grown using metalorganic chemical vapor deposition (MOCVD). Each superlattice period was 20nm thick composed of 5nm GaN:Mg /5nm unintentionally doped (u.i.d) GaN/ x nm u.i.d. AlN/ (5-x) nm u.i.d AlGaN/ 5nm AlGaN:Mg, where the AlN layer thickness, x, was 0 nm ≤ x ≤ 1.6 nm. In addition to the AlN thickness series, the Mg doping was varied for each AlN thickness with 3.75×1018 cm-3 ≤ [Mg]≤ 2.5×1019 cm-3. The Mg doping in the samples was calibrated using Secondary Ion Mass Spectroscopy, and the SL quality and composition was measured using X-Ray Diffraction. Mg activation was carried out at 800○C and Pd/Au contacts were deposited to prepare the samples for room temperature Hall measurements. The experimental results were compared to those calculated using STR-FETIS. When the band-bending of both GaN and AlN was < 0.8eV, the sheet charge concentration of holes (ps) measured using Hall matched with the calculated values. For each of the samples where band-bending of either GaN or AlN was ≥ 0.8eV(without accounting for the impact of trap ionization on band-bending), the measured ps was greater than the amount of Mg dopants put into the sample. This showed that the excess holes came from the acceptor traps at the positive polarization interfaces, here - AlGaN/GaN and AlN/AlGaN interfaces. The acceptor trap ionized when either band-bending was ≥ 0.8eV and the Fermi level was pinned at this trap level leading to the very high measured ps. A low sheet resistance (~3.4kΩ/■) was observed for N-polar modulation doped GaN/AlN/AlGaN SLs, with ps~1.3×1014 cm-2 and µp~14 cm2/Vs. For the samples in this study, the two-dimensional hole gas (2DHG) exists at the GaN/AlN interface or a GaN/AlGaN interface, if no AlN present. It was observed for each of the AlN thickness series (with various Mg doping levels) the µp decreased in samples where the acceptor traps ionized. It was further observed within those set of samples that µp decreased to a great extent (>50% decrease) for those samples where AlN band-bending was ≥ 0.8eV – independent of the GaN band-bending in the same sample. The 2DHG is closer to the AlN/AlGaN positive polarization interface (PPI-1) (0 to 1.6nm) compared to the AlGaN/GaN positive polarization interface (PPI-2)(10nm). Therefore, the more dominant source of scattering (and hence the decreased mobility) are the ionized acceptor traps which exist at the AlN/AlGaN positive polarization interface, the closer interface. While the ionized traps at AlGaN/GaN interface also cause scattering, their effect is not as pronounced because they are far away from the 2DHG. This explains the higher mobility observed in in samples where only the GaN band-bending was ≥ 0.8eV, and the AlN band-bending was < 0.8eV leading to the ionization of traps at only the AlGaN/GaN positive polarization interface (PPI-2). Further details of simulations, experiments and methodology will be presented at the conference. [1] P. Kozodoy, et al., APL(1999) [2] A. Raj, et. al., IEEE-EDL (2020) [3] C.J. Zollner, et. al., OME (2020) [4] A. Krishna, et. al., APL (2020)

Effect of Acceptor Traps at Positive Polarization Interfaces on the Charge and Mobility of Holes in N-Polar P-Type GaN/(AlN/AlGaN) Superlattices

Abstract Body: β -Ga2O3 has attracted considerable technological and scientific interest due to its ultra-wide ~ 4.8 eV band gap and its extremely high projected ~ 8 MV/cm breakdown field, both of which can advance high-power and high-speed electronics. However, to reach the full potential of these applications, an understanding and control of electrically - active native point defects and impurities in β -Ga2O3 is needed. Besides their contribution to shallow dopants, deep levels of such extrinsic states can compensate free carriers and have a major impact on their recombination and transport, and possibly their ultimate breakdown voltage strength. Luminescence studies [1-3] have identified commonly observed emissions at 2.5 eV, 2.8 eV, 3.0 eV, 3.5 eV and 3.8 eV, some of whose nature can be related to native defects. The deep level defect nature of extrinsic impurities such as Ir in β -Ga2O3 is of interest since high quality single crystal growth techniques such as edge-defined film-fed growth (EFG) make use of iridium dies and crucibles and because large EFG crystals can be grown with quality that is suitable for technological applications. Hence, the impact of Ir in β -Ga2O3 is of both scientific and practical concern. We used depth-resolved cathodoluminescence spectroscopy (DRCLS) and surface photovoltage spectroscopy (SPS) to explore the nature of optical transitions at Ir- Tamura Corp. (010) UID EFG interfaces. Low-kV e-beam excitation through 25 nm-thick, e-beam-deposited Ir-Ga2O3 diodes provided DRCL spectra with near-nm depth resolution. Near - surface and bulk DRCLS of adjacent bare Ga2O3 reveals a prominent ~3.2 eV emission similar to a photoluminescence/cathodoluminescence feature reported previously [4,5], while our SPS identifies a similar transition located 3.15 eV± 0.1 eV below the conduction band. These results are consistent with various theoretical and experimental studies that attribute the ~3.2 eV emission to recombination involving self-trapped holes. Near-nm scale DRCLS at the Ir-Ga2O3 interface shows the emergence of 2.1 eV and 2.6 eV features, which is consistent with transitions associated with the complementary IrGa(II) (0/+) and IrGa(II) (+/0) levels predicted theoretically [6]. The ~3.2 eV emission peak energy maximum also shifts downward towards the Ir-Ga2O3 interface, which correlates with the emergence of a sharp ~2.9 eV emission that can possibly be identified with a IrGa(I) charge state transition. These measurements provide a distinction as well as possible interplay between deep level emission due to Ir impurities and possible self-trapped polarons that can introduce deep level recombination that affect free carrier transport, density, and dielectric breakdown. These results also suggest that Ir is readily incorporated into the Ga2O3 lattice and can produce a major luminescence feature that may produce significant free carrier recombination and compensation. Research efforts are currently underway to perform rapid-thermal annealing on the Ir-deposited sample to determine the activation energy and diffusion coefficient for Ir in Ga2O3. DNR, MSH, and LJB gratefully acknowledge AFOSR Grant No. FA9550-18-1-0066 (Ali Sayir). [1] H. Gao et al., Appl. Phys. Lett. 112, 242102 (2018). [2] H. Gao et al., J. Phys. D: Appl. Phys. 53 465102 (2020). [3] Y. Wang et al., Sci. Rep. 8, 18075 (2018). [4] S. Yamaoka et al., Physical Review B 95, 094304 (2017). [5] T. Onuma et al., Appl. Phys. Lett. 103, 041910 (2013). [6] J.R. Ritter et al., Appl. Phys. Lett. 113, 052101 (2018).

Deep Levels and Self-Trapped Excitons at Iridium/Edge-Fed Grown β-Ga2O3 Interfaces

Abstract Body: β-gallium oxide (Ga2O3) is a promising material for applications in power electronics and as a transparent conducting oxide due to its wide bandgap of 4.8 eV and high breakdown electric field of 8 MV/cm. Fundamental properties such as doping, mobility, optical absorption, luminescence, and charge trapping are determined by the device’s ensemble of defects. Thermodynamic simulations model the limiting case where the system has reached equilibrium; however, all real systems are limited by both kinetics and thermodynamics during device processing and operation. Here, we present a coupled reaction-diffusion solver used to study time-dependent, one- and zero-dimensional systems, going beyond thermodynamic predictions for defects. By using a detailed reaction mechanism, we can model carrier-defect interactions, such as trapping and emission, resulting in an in-depth description of time-dependent device behavior and accurate device characterization. The solver uses the finite volume method to approximate spatial derivatives and implicit time stepping for time derivatives. This allows us to model surface defect formation and subsequent diffusion into the bulk, as well as bulk defect formation. Using an implicit time stepping scheme enables us to study an extraordinary range of time scales, from femtoseconds to years, unlocking possibilities to simulate device lifetime reliability in addition to device processing.

Time-Dependant Simulations of Defect Kinetics and Thermodynamics in Gallium Oxide

Abstract Body: Rare-earth chromites (RCrO3) have Néel temperature (TN, above which these become paraelectric) in the range of 100-280 K, which is much higher than the magnetic transitions observed in rare-earth manganites (RMnO3). These family of compounds have been studied for the magnetoelectric-multiferroic (ME-MF) for their potential applications in various devices including magnetic field sensors, energy harvesting devices, and magnetic refrigeration (MR), to name some. In case of ME-MF based devices, it is desirable to have the transitions close to room temperature. In this talk, the structure, dielectric, and magnetic properties of some RMnO3 and RCrO3 based materials will be presented with a focus on their band gap and mutliferroic behavior and the structure-property relations.

Downloads

Next from EMC 2021

Analysis of Patterned GaAsSbN Nanowires via Boltzmann Sigmoidal Model

Similar lecture

A Te Doped GaAsSb Ensemble Nanowire Photodetector for Near-Infrared Application

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EMC 2021

Analysis of Patterned GaAsSbN Nanowires via Boltzmann Sigmoidal Model

Similar lecture

A Te Doped GaAsSb Ensemble Nanowire Photodetector for Near-Infrared Application

Downloads