The mid-infrared (mid-IR) wavelength range (~2-30 μm) is of significant importance for a range of applications in fields ranging from environmental monitoring, industrial process control, bio-medical imaging, security and defense, and pharmaceutical and food quality control. The mid-IR’s technological importance is the result of its position as the spectral home of distinct vibrational and rotational absorption resonance signatures of a wide range of molecular species, as well as the thermal emission of biological and mechanical objects for a wide range of temperatures. To enable many of the proposed mid-IR applications, compact, high efficiency, and low-cost mid-IR emitters are required. The recent development of the quantum cascade laser (QCL) has provided an increasingly efficient and high-power, compact, and now commercially-available light source for mid-IR optical systems. The QCL is a particularly attractive mid-IR source, as it allows significant wavelength flexibility, resulting from the laser designer’s ability to control the electronic and optical properties of the semiconductor heterostructure via bandstructure engineering. These lasers emit across almost all of the mid-IR, and are capable of continuous wave, room-temperature, and single-mode operation, with high output powers, for most, if not all wavelengths of importance for mid-IR sensing and countermeasure applications.

It is difficult to reach short wavelengths (2-4 μm) using traditional QCLs made from the InGaAs/AlInAs lattice-matched to InP material system, which requires strain-balanced InGaAs/AlInAs materials, with increased design complexity resulting from constraints on well and barrier thicknesses imposed by the strain balance. All QCLs suffer from extremely short non-radiative recombination lifetimes resulting primarily from phonon scattering, but also ionized impurity scattering and interface roughness effects. These short non-radiative lifetimes (in the ~ps range) lead to large threshold current densities which limits QCL wall-plug efficiency, and makes QC-based devices non-ideal sub-threshold emitters. In effect, QCLs, while currently the device of choice for applications requiring a compact source of coherent mid-IR light, make surprisingly poor light emitting diodes (LEDs). Due to the optical selection rules governing the QCL intersubband transitions, these devices can only emit TM polarized light, preventing surface emission without additional outcoupling structures.

Certain protocols for quantum cryptography rely on a robust source of single photons.[1] Semiconductor quantum dots (QDs) offer several advantages over other approaches for single photon generation. QDs can be incorporated into scalable cavity/waveguide structures, and quantum key distribution with QD sources has already been demonstrated.[2]

MBE of low-density InGaAs or GaSb QDs is well established, but still requires extremely close control over growth. QD density and size are dramatically affected by small changes in MBE parameters (growth rate, temperature and thickness). This sensitivity is a challenge for QD uniformity.

In contrast, we present a material system with a remarkably straightforward growth process, which delivers single photon emission. We see spectrally sharp emission lines from bulk InAlAs grown on InP(111)A by MBE (a). Using a combination of cross-sectional STM (b), and k.p simulations, we identify excitons trapped in indium-rich nanoclusters as the origin of these spectral features. In-rich regions form spontaneously during growth via nanoscale InAlAs phase-segregation. Individual nanoclusters exhibit a median emission linewidth of 137 μeV, and median fine structure splitting of 28 μeV.

Photon statistics of nanocluster emission confirm pure, single photon emission, with second-order correlation values as low as g⁽²⁾(0)=0.05 +0.17/-0.05 under CW excitation (c), and g⁽²⁾(0)=0.24±0.02 under pulsed excitation. Our results suggest that, once optimized, In-rich nanoclusters could be a promising candidate for on-demand, single photon generation.

In the past decade, quantum ring nanostructures (QRNs) have gained widespread attention as potential candidates for the study of fundamental physics such as quantum interference [1,2]. One of the promising ways to grow QRNs is to employ the droplet epitaxy (DE) technique, which facilitates the fabrication of QRNs on both lattice matched and mismatched systems [3,4]. Most of the fabricated QRNs involve GaAs/AlGaAs and InAs/GaAs systems.

In this work, we discuss the formation of InAs QRNs on GaSb (100) substrate using DE. Since adatoms surface migration is one of the key criteria for the realization of high quality and uniform QRNs, several growth parameters such as substrate temperature (T_s), amount of indium (In) supply, and surface reconstructions are optimized. By controlling these parameters, we can manipulate the migration of In adatoms. The In droplets are crystallized to InAs QRNs using different arsenic (As) flux (beam equivalent pressure), which in turn modifies the shape and size of these QRNs.

To fabricate the InAs/GaSb QRNs, a 200 nm thick GaSb buffer layer is grown on GaSb (100) substrate with V/III flux ratio of 5~6 at T_s = 500° C. We modify the surface stoichiometry and achieve surface reconstructions of 2x5, 1x3 and 1x1 before In droplet formation via closing of the Sb valve at different T_s. In order to form In droplets, the 2x5 surface with residual Sb is maintained by closing of Sb at T_s= 325 °C. After a while, the surface reconstruction is converted into 1x3. To passivate the residual Sb on the surface, sub-mono layer (ML) of Ga is introduced at T_s=200 °C. After supply of 1.5 ML of Ga on the surface, the RHEED pattern changes to 2×1 indicating group-III termination stoichiometry. 1.5 ~ 2 ML of In is deposited on the surface with a growth rate of 0.142 ML/s at various T_s (50 ~ 350 °C). The T_s dependent-droplet formation can allow droplet densities of 10⁷ ~ 10¹¹/cm². During In deposition at various T_s, the RHEED pattern shows either halo or 4x1 reconstructions depending upon the T_s.

The As flux of 1x10^-5 Torr is supplied at T_s =200 °C to crystalize the In droplets (2 ML) into InAs QRNs. Further, the QRNs crystal quality is improved by continuously providing As flux till T_s= 316 °C. Fig. 1 shows the surface morphology of InAs/GaSb QRNs observed by atomic force microscopy.

Ordered arrays of high quality semiconductor single quantum dots that can produce on-demand single photons with high spectral uniformity and can be integrated on-chip with passive light manipulation structures is a critical need for realization of on-chip nanophotonic quantum information systems. Here, we demonstrate triggered single photon emission at liquid nitrogen temperature from spatially ordered array of spectrally uniform site-controlled InGaAs single quantum dots synthesized on spatially ordered nanomesa tops using designed surface-curvature stress gradient directed preferential atom migration to mesa tops, dubbed substrate-encoded size-reducing epitaxy (SESRE) [1-3]. Owing to control on the growth kinetics at the atomistic level and correspondingly growth induced control on the shape and size, these InGaAs mesa-top single quantum dots (MTSQDs) arranged in a 5x8 array spread over 1000μm², show spectral uniformity (standard deviation) of ~8.3nm, an order of magnitude better compared to the self-assembled quantum dots over similar areas. Furthermore, these MTSQDs show uniform single photon emission behavior with g⁽²⁾(0)=0.24±0.07 at 77K, a first for the InGaAs/GaAs system and indicative of strong quantum confinement achieved in such MTSQDs. Further optical studies of MTSQD and MTSQD-light passive elements integrated structure are underway.

+ Author for correspondence: madhukar@usc.edu

[1] A. Madhukar, Thin Solid Films 231, 8 (1993).

[2] K.C. Rajkumar, A. Madhukar, P. Chen, A. Konkar, L. Chen, K. Rammohan, D.H. Rich, J. Vac. Sci. Technol. B 12, 1071 (1994).

[3] J. Zhang, Z. Lingley, S. Lu, A. Madhukar, J. Vac. Sci. Technol. B 32, 02C106 (2014).

III-V quantum dot lasers grown on silicon are proving to be a promising light source for silicon photonics [1]. Previous demonstrations have relied on intentionally offcut silicon substrates to suppress antiphase domains from III-V on silicon heteroepitaxy, while exact on-axis silicon substrates are needed for compatibility with CMOS process flows. We report the first demonstration of an electrically pumped quantum dot laser grown on exact silicon substrates without offcut. The epitaxial laser stack was grown on a GaP/Si (001) template provided by NAsP III-V GmbH, consisting of a 775 μm thick (001) on-axis p-doped Si substrate, with 200 nm thick n-doped Si homo-epitaxial buffer and a subsequent 45 nm thick n-doped GaP nucleation layer. An InAs quantum dot laser embedded in a GaAs/AlGaAs GRINSCH waveguide was then grown in MBE. The active region consisted of seven stacks of InAs quantum dot layers (2.75 MLs deposited at 0.11 ML/s, V/III ratio of 35) embedded in 8 nm In_0.15Ga_0.85As quantum wells, which were separated by partially p-doped GaAs barriers. The same active structure was also grown on a GaAs substrate for comparison, both of which were concurrently processed into deeply etched ridge waveguide lasers with varying stripe widths.

Lanthanide monopnictides (Ln-V), often referred to as rare-earth monopnictides, have recently received considerable attention due to the epitaxial relationship that can exist between the Ln-V and III-V and their effect on the electrical properties of the composite system. Most Ln-V materials are semimetallic; however, recent work suggests that TbAs is an indirect gap semiconductor.[1] To explore the fundamental properties of TbAs, we first present the growth, characterization, and comparison of TbAs:In_0.53Ga_0.47As and TbAs:GaAs. Comparison of fluence dependent optical pump terahertz probe measurements show both systems are capable of saturation, suggesting a band gap in TbAs, but complete saturation occurs at differing pump fluences. Spectrophotometry shows a strong matrix dependent band gap. Temperature dependent Hall effect provides information on carrier concentrations and Fermi level placement. The comparison of these two samples allows us to propose a band diagram where the TbAs nanoparticle forms a type II, staggered, heterojunction in InGaAs, and a type I, straddled, heterojunction in GaAs.[2]

To further explore the fundamental properties of TbAs, we next present the first molecular beam epitaxial growth and characterization of complete TbAs films. For this study, two TbAs films with differing thicknesses (nominally, 70nm and 350nm) were grown by solid-source MBE. High-resolution X-ray diffraction scans show a broad TbAs peak separated from the substrate peak, indicating the growth of a TbAs film. Reciprocal space mapping shows that both the thick and thin TbAs films have relaxed and have some mosaic nature. Electron diffraction patterns obtained by transmission electron microscopy (TEM) match the expected pattern for TbAs. Spectrophotometery shows a thickness dependent optical band gap, 435 meV and 517 meV for the thick film and thin film respectively. Temperature dependent Hall effect measurements suggest that the TbAs films are degenerately doped and show that the carrier concentration of the thin film is approximately four time larger than that of the thick film. This suggests that measured band gap has been increased by a Burstein-Moss shift. Studying the growth and characterization of TbAs films and nanoparticles embedded within In_0.53Ga_0.47As and GaAs provides a more complete and fundamental understanding of TbAs.

[1] L.R. Vanderhoef, A.K. Azad, C.C. Bomberger, D.R. Chowdhury, D.B. Chase, A.J. Taylor, J.M.O. Zide, and M.F. Doty, Phys. Rev. B 89, 045418 (2014).

[2] C.C. Bomberger, L.R. Vanderhoef, A. Rahman, D. Shah, D.B. Chase, A.J. Taylor, A.K. Azad, M.F. Doty and J.M.O. Zide, Applied Physics Letters 107, 102103 (2015).

The recent success in obtaining epitaxial rare earth nitride (REN) thin films has been central in improving the understanding of their fundamental properties and in particular demonstrating, for most of them, their intrinsic ferromagnetic semiconducting nature with a wide variety and complementary magnetic properties across the series. The presence of large spin and orbital moments, contrasting with the quenched orbital moment in traditional metal-based ferromagnets, gives additional and very unfamiliar magnetic properties.[1] Among the RENs, SmN stand out for its unusual physical properties. It is the only known semiconductor with a nearly vanishing net magnetic moment directed opposite to the fully aligned spin moment.[2] Our recent observations also suggests that heavily donor-doped SmN by nitrogen vacancies becomes superconducting with ferromagnetism remaining intact at low temperature.[3] An interesting aspect is that we and others have demonstrated that the RENs are epitaxy-compatible materials with group III-nitrides (GaN, AlN, InN), the technologically important nonmagnetic semiconductor family used for the fabrication of white and blue light emitting diodes (LEDs) and high power electronics.[1] This opens the possibility to introduce spintronics in LEDs and transistors.

Developing spintronics structures based on these two nitride families will rely on (i) the understanding and the ability to control, at the atomic scale, the interface structure and chemical stability and (ii) growing high quality epitaxial SmN thin layers, typically of the order of tens nanometer thick. Here we present results on the growth of rocksalt SmN on hexagonal (wurtzite) GaN and AlN (0001) surfaces. Combined in situ X-ray photoelectron spectroscopy and reflection high-energy electron diffraction studies show that Ga segregates at the surface during the first stages of the epitaxial growth of SmN on GaN surfaces. The Ga surface segregation can be simply suppressed by growing a few monolayers of AlN before starting the SmN growth, resulting in a significant improvement of the crystallinity of SmN thin films assessed by X-ray diffraction. The effect of the growth temperature on the structural properties of thick SmN layers on AlN (0001) is also presented. We demonstrate that the growth temperature controls the orientation of the SmN layers. A clear transition in the growth orientation of SmN from (111) to (002) is observed when the temperature is increased above 700°C, resulting in purely (002) oriented SmN layer at 800°C. Interestingly in-situ scanning tunnelling microscopy images revealed a concomitant change of the surface morphology.

We investigate the differences between the growth of horizontal and vertical semimetallic ErSb nanowires in a semiconducting GaSb matrix through the use of in-situ Scanning Tunneling Microscopy (STM), as well as ex-situ measurements. These structures may be used in several applications, including buried contacts, and terahertz devices, where orientation of the nanowires play a large role.[1] For this reason it is important to understand the mechanism behind formation of horizontal nanowires so that they may be incorporated into device structures, and so that the varying growth modes in the ErSb/GaSb system may be adapted to other rare earth-III/V material systems.

We identify a change in the growth mode of horizontal nanowires when compared to the previously reported growth of vertical nanowires and nanoparticles.[2] Horizontal nanowire growth is accompanied by the formation of large macrosteps forming on the surface which play an important role in the resulting distribution of nanowires.

We also observe a second previously unreported growth mode for horizontal nanowire formation dependent on substrate temperature. The higher temperature growth mode corresponds to the previously reported nanostructure formation, whereas the lower temperature growth mode results in thinner nanowires that do not grow with an embedded growth mode and do not appear to generate macrosteps on the surface during growth.

⁺Author for correspondence: cpalmstrom@ece.ucsb.edu

[1] M. Hanson, S. Bank, J. Zide, J. Zimmerman, A. Gossard, J. Crystal Growth 301-302 4(2007).

[2] J. Kawasaki, B. Schultz, H. Lu, A. Gossard, C Palmstrøm, Nanoletters 13 2895(2013).

performance benefits, design flexibility, and novel device structures, such as metal-base

transistors [1]. We have reported on the use of RF-plasma MBE to grow 4 – 100 nm-thick

metallic hexagonal β-Nb₂N thin films epitaxially on hexagonal SiC substrates [2,3].

Epitaxial growth of smooth, low-resistivity, metallic, single-phase β-Nb₂N films was

demonstrated in a 775 – 850 °C temperature window that is compatible with III-N MBE

growth. This suggests that β-Nb₂N has the potential to create high quality semiconductor /

metal / semiconductor heterostructures that have not been realizable in the past. In this

presentation, we will show that high-quality AlN / Nb₂N and GaN / AlN / Nb₂N films can

be obtained using standard RF-plasma MBE growth optimization techniques. In-situ

RHEED patterns indicate that the III-N films grown on β-Nb₂N are naturally N-polar using

our standard MBE growth conditions, but the polarity of the overgrown III-N films can be

converted to metal-polar through the use of a low-temperature buffer layer. X-ray

diffraction and electrical measurements demonstrate the high quality of AlN [4,5] and GaN

films grown on epitaxial β-Nb₂N. Finally, we note that the chemical differences between b-

Nb₂N and the comparatively inert III-N materials and SiC enable novel device processing,

such as epitaxial lift-off, creating novel heterogeneous integration opportunities [6].

+ Author for correspondence: scott.katzer@nrl.navy.mil

[1] S. M. Sze and H. K. Gummel, Solid-State Electron. 9, 751 (1966).

[2] D. S. Katzer, N. Nepal, D. J. Meyer, B. P. Downey, V. D. Wheeler, D. F. Storm, M. T. Hardy, Appl. Phys.

Express 8, 085501 (2015).

[3] D. S. Katzer, N. Nepal, D. J. Meyer, B. P. Downey, V. D. Wheeler, D. F. Storm, M. T. Hardy, MRS

Advances, (2016), doi:10.1557/adv.2016.27

[4] N. Nepal, D. S. Katzer, D. J. Meyer, B. P. Downey, V. D. Wheeler, D. F. Storm, M. T. Hardy, Appl. Phys.

Express 9, 021003 (2016).

[5] B. P. Downey, D. S. Katzer, N. Nepal, D. J. Meyer, D. F. Storm, V. D. Wheeler and M.T. Hardy, accepted

for publication in Electronics Lett. (2016).

[6] D. J. Meyer, B. P. Downey, T. J. Anderson, D. S. Katzer, N. Nepal, V. D. Wheeler, D. F. Storm, and M. T.

Hardy, accepted for presentation at CS-MANTECH, Miami, FL, May 16-19, 2016.

Lanthanide monopnictide (Ln-V), also known as rare-earth monopnictide, nanoparticles epitaxially embedded within III-V semiconductors drastically change the electrical properties of the composite material. These changes make them interesting for applications including photoconductive (PC) switches for terahertz sources and detectors. There has been recent interest in narrowing the band gap of the PC switch’s host material, with much of the work focusing on altering the group III atoms. This primarily modifies the conduction band and alters the nanocomposites’ electrical properties. For example, ErAs:GaAs pins the Fermi level within the band gap and results in a high dark resistance, while ErAs:In_0.53Ga_0.47As pins the Fermi-level near the conduction band edge and causes a reduced dark resistance.[1,2] Alternatively, the band gap can be reduced by changing the group V atoms. Previous work shows that incorporating small amounts of Bi into GaAs causes a narrowing of the band gap due to valence band anticrossing (VBAC), while the conduction band remains relatively unchanged.[3] Thus, it is expected that ErAs:GaBiAs will have similar electronic properties to ErAs:GaAs with a narrower band gap.

In this presentation, we explore the first growth and characterization of ErAs:GaBiAs. We present the effects of growing slightly group III rich, including the presence of Ga droplets and how Er and Bi affect droplet formation and the incorporation of one another. We also explore the growth and characterization of films grown under near-stoichiometric conditions. Morphology and composition are studied via XRD, RSM, RBS, and TEM. Spectrophotometery is used to determine the band gap reduction due to bismuth incorporation. Hall effect is used to determine carrier concentration, mobility, and dark resistance. High dark resistance with relatively high mobility, similar to that seen in ErAs:GaAs, is observed. Finally, optical-pump optical-probe measurements were used to determine the carrier lifetimes. As with ErAs:GaAs, ErAs nanoparticles incorporated in GaBiAs result in reduced carrier lifetimes. These results show that the incorporation of Bi into GaAs allows for band gap tuning while maintaining the desired band alignment with ErAs nanoparticles, suggesting a route to terahertz sources and detectors driven by inexpensive fiber lasers.

[1] C. Kadow, A.W. Jackson, A.C. Gossard, S. Matsuura, and G.A. Blake, Applied Physics Letters 76, 3510 (2000).

[2] D.C. Driscoll, M. Hanson C. Kadow, and A. C. Gossard, Applied Physics Letters 78, 1703 (2001).

[3] K. Alberi, O.D. Dubon, W. Walukiewicz, K.M. Yu, K. Bertulis, and A. Krotkus, Applied Physics Letters 91, 051909 (2007).

Chalcopyrite ternary heterovalent semiconductor compounds can undergo a transition between an ordered chalcopyrite structure and a disordered zinc-blende-like phase. Unlike in adamantine alloys, the disorder results in a band-gap reduction relative to the band gap of the ordered lattice. ZnSnN₂ represents an interesting member of the chalcopyrite family of materials, due to its earth abundant constituant elements and a band gap close to the range needed for solar cells. Density functional theory (DFT) calculations predict the ordered phase to have an orthorhombic lattice and a direct band gap of 2.0 eV [1, 2]. Using special quasirandom structures (SQS) to model the disordered Zn and Sn cation sub-lattice, DFT simulations predict that the band gap for the disordered phase will be close to 1.0 eV and will have a hexagonal lattice [1]. This almost 1.0 eV reduction presents an opportunity for band gap engineering by controlling the disorder of the cation sublattice.

A series of films has been grown by plasma-assisted MBE in order to investigate the possibility of controlled cation disorder as well as its effects on physical and electronic properties of the material. By varying the growth conditions, specifically either the metal to the nitrogen flux or the substrate, we have confirmed the existence of the both phases of the crystal via synchrotron x-ray diffraction and in-situ RHEED. All of the films though have a high free carrier concentration ( > 10¹⁹ cm^-3). Taking into account the Burstein-Moss shift and calculating the effective masses of the carriers from parabolic fits to the density results, the optically measured absorption edge variation is consistent with an order-induced band gap change between 1.0 and 2.0 eV [3].

[1] N. Feldberg, J. D. Aldous, W. M. Linhart, L. J. Phillips, P. A. Stampe, R. J. Kennedy, D. O. Scanlon, G. Vardar, R. L. Field, T. Y. Jen, R. S. Goldman, T.D. Veal, S.M. Durbin, Appl. Phys. Lett., 103 (2013) 042109.

[2] Punya, A., Lambrecht, W. R., and van Schilfgaarde, M. (2011). Physical Review B, 84, 165204.

[3] Veal, T. D., Feldberg, N., Quackenbush, N. F., Linhart, W. M., Scanlon, D. O., Piper, L. F., and Durbin, S. M. (2015) Advanced Energy Materials, 5.