No description available.

We are living in an era of digital electronics. The number of robots have already exceeded the human population of the entire earth. An article in the Guardian newspaper dated 30th March 2018 suggests that 10 million UK workers will be jobless within 15 years as they will be replaced by robots. These astonishing facts shed light on the importance of knowledge and how important it is to use it wisely for our benefit without ultimately destroying us. Knowledge in all forms is accessible without going to a library or buying a newspaper. Furthermore to access information, we often use sleek devices such as smart phones, using highly developed multimedia platforms which consume large amounts of power. In 2016, IBM found that humans create 2.5 quintillion bytes of data daily. Since high computing usage is related to large power consumption, the basic building block of electronics i.e. the transistor is required to be more power efficient. This is now possible through spintronics, where the spin of an electron is exploited instead of the charge. A new class of exotic materials called topological insulators are predicted to exhibit efficient spintronic applications. These materials can conduct spin polarised current on their surface while remaining completely insulting from the inside. Moreover, doping topological insulators with magnetic impurities unlocks new avenues for spin memory devices in the form of a single spin polarized dissipationless conduction channel. In topological insulators, there is always a contribution from the inside (bulk) in addition to surface conduction, thereby yielding charge transport rather than spin transport. On this basis, the aim of my PhD was to explore techniques to grow, characterize, fabricate and measure devices on topological Dirac materials, with the hope to experimentally distinguish the bulk from the surface states and also exploit their exotic properties arising from opening of the bulk band gap by intentional magnetic doping. Samples consisted of thin films of Bi2Se3, Sb2Te3, Cr doped Sb2Te3, bilayers of Dy doped Bi2Te3/Cr doped Sb2Te3 and Cd3As2 nanowires. It was found that a seed layer of an undoped topological insulator was a crucial first step to ensure high quality growth by molecular beam epitaxy, followed by the desired stoichiometry. By physically doping Sb2Te3 with Cr, a successful control of the magnetic and electrical properties such as coercivity, anomalous Hall resistance RA xy, Curie temperature Tc, carrier density and mobility were achieved. A substitutional Cr doping ranging from 7.5% to 38% was attained revealing a Tc reaching up to 186 K. Gated electrical measurements displayed a change in RA xy and carrier density by ~ 50% on applicating of just -3 V gate bias in a sample with 29% doping. A comparison between electrical transport, Magneto-optical Kerr effect and terahertz time domain spectroscopy measurements revealed that the mechanism of magnetization was RKKY mediated. Furthermore, the bilayer structure displays a clear exchange bias coupling arising from the proximity of the antiferromagnetic Dy doped Bi2Te3 layer with the ferromagnetic Cr doped Sb2Te3 layer. Electrical transport measurements on Bi2Se3 Hall bars fabricated using Ar+ milling and wet chemical etching were compared. The results showed a more bulk type response in the chemical etched sample even though Ar+ milling was responsible for creating more disorder in the system leading to a higher carrier density and lower mobility. A thickness dependent study on Sb2Te3 thin films revealed a single conducting channel associated with a coupled surface and bulk state for a 12 nm sample, compared to, two conducting channels associated with the top and bottom surfaces for the 25 nm sample. Electrical transport on Dirac semimetal Cd3As2 nanowires reveal an ultra-high mobility of 56884 cm2V-1s-1 at 1.8 K from analysis of Shubnikov-de Haas oscillations. By studying various Dirac materials, new avenues for practical device applications can be explored.

GaInAsSb is a promising material for mid-infrared devices such as lasers and detectors because it is a direct band gap material with large radiative coefficient and a cut-off wavelength that can be varied across the mid-infrared (from 1.7 to 4.9 μm) while remaining lattice matched to GaSb. On the other hand, the potential of the alloy is hampered by predicted ranges of concentration where the constituents of the alloy become immiscible when the crystal is grown near thermodynamic equilibrium at typical growth temperatures. There have been efforts to extend the wavelength of GaInAsSb alloys through such techniques as digital alloy growth and non-equilibrium growth, but most of the compositional range has for a long time been inaccessible due to immiscibility challenges. Theoretical studies also supported the existence of thermodynamic immiscibility gaps for non-equilibrium growth conditions. Lower growth temperatures lead to shorther adatom diffusion length. While a shorter adatom diffusion length suppresses phase separation, too short an adatom length is associated with increased defect formation and eventually loss of crystallinity. On the other hand, hotter growth temperatures move epitaxial growth closer to thermodynamic equilib- rium conditions, and will eventually cause phase separation to occur. In this study thick 2 μm; bulk GaInAsSb layers lattice-matched to GaSb substrates were grown across the entire (lattice-matched) compositional range at low growth temperatures (450°C), including the immiscibility region, when grown under non-equilibrium conditions with MBE. High quality epitaxial layers were grown for all compositions, as evidenced by smooth morphology (atomic force microscopy), high structural quality (X-ray diffraction), low alloy fluctuactions (electron dispersive spectroscopy in cross sectioned samples), and bright room temperature photoluminescence. Because initial theoretical efforts have suggessted that lattice strain can influence layer stability, we have studied effects of strain on alloy stability. Unstable and metastable alloys were grown hot enough for the onset of phase separation, then progressively strained and characterized. We show that strain is effective in suppressing phase separation. Finally, we performed time-resolved carrier lifetime measurements for InAsSb alloy with low concentrations of Ga to investigate the role of Ga in influencing nonradiative carrier recombination. There have been studies on non-Ga containing antimonide structures (InAsSb, InAs/InAsSb) that show long carrier lifetimes, which suggest that Ga plays a role in reducing carrier lifetime, because Ga-containing structures such as InAs/GaSb superlattices have much shorter carrier lifetimes. Ga may reduce carrier lifetime through native defects that increase background carrier concentration, or that create mid-gap electronic states. Here, a series of GaInAsSb alloys were grown with low to zero Ga concentration. No difference in carrier lifetime was observed between Ga and Ga-free structures, and minority carrier lifetimes > 600 ns were observed. Additional work remains to be done to obtain background carrier densities in the samples with Hall measurements.

abstract: III-V multijunction solar cells have demonstrated record efficiencies with the best device currently at 46 % under concentration. Dilute nitride materials such as GaInNAsSb have been identified as a prime choice for the development of high efficiency, monolithic and lattice-matched multijunction solar cells as they can be lattice-matched to both GaAs and Ge substrates. These types of cells have demonstrated efficiencies of 44% for terrestrial concentrators, and with their upright configuration, they are a direct drop-in product for today’s space and concentrator solar panels. The work presented in this dissertation has focused on the development of relatively novel dilute nitride antimonide (GaNAsSb) materials and solar cells using plasma-assisted molecular beam epitaxy, along with the modeling and characterization of single- and multijunction solar cells. Nitrogen-free ternary compounds such as GaInAs and GaAsSb were investigated first in order to understand their structural and optical properties prior to introducing nitrogen. The formation of extended defects and the resulting strain relaxation in these lattice-mismatched materials is investigated through extensive structural characterization. Temperature- and power-dependent photoluminescence revealed an inhomogeneous distribution of Sb in GaAsSb films, leading to carrier localization effects at low temperatures. Tuning of the growth parameters was shown to suppress these Sb-induced localized states. The introduction of nitrogen was then considered and the growth process was optimized to obtain high quality GaNAsSb films lattice-matched to GaAs. Near 1-eV single-junction GaNAsSb solar cells were produced. The best devices used a p-n heterojunction configuration and demonstrated a current density of 20.8 mA/cm2, a fill factor of 64 % and an open-circuit voltage of 0.39 V, corresponding to a bandgap-voltage offset of 0.57 V, comparable with the state-of-the-art for this type of solar cells. Post-growth annealing was found to be essential to improve Voc but was also found to degrade the material quality of the top layers. Alternatives are discussed to improve this process. Unintentional high background doping was identified as the main factor limiting the device performance. The use of Bi-surfactant mediated growth is proposed for the first time for this material system to reduce this background doping and preliminary results are presented. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2017

Dans des conditions de croissance spécifiques, des sections nanofilaires d’AlGaN peuvent croître en épitaxie sur des bases nanofilaires de GaN. De telles croissances, effectuées par épitaxie par jets moléculaires dans le cadre dans le cas présent, permettent la caractérisation ultérieure de petits volumes d’AlGaN exempt de défauts étendus communément observés dans les couches planaires. Cette absence de défauts rend ces fils prometteurs pour les dispositifs optoélectroniques émettant dans l’ultraviolet. Cependant, la réalisation de tels composants nécessite de mieux comprendre les propriétés fondamentales des fils.La question des inhomogénéités d’alliage à l’échelle nanométrique reste notamment à éclaircir. Afin d’y voir plus clair, ces dernières ont été dans un premier temps étudiées dans cette thèse. Pour nos expériences, des nanofils d’AlGaN non-intentionnellement dopés (NID) ont été crûs dans des conditions variées afin d’ajuster potentiellement les fluctuations de composition de l’alliage et ainsi sonder éventuellement des centres de localisation de porteurs de taille et composition différentes. Il a premièrement été observé au moyen de méthodes de caractérisation structurale que la longueur des sections plus riches Al qui nucléent préférentiellement au sommet des fils de GaN peut être ajustée en variant les paramètres cinétiques de croissance, mettant en lumière un mécanisme de croissance gouverné par la cinétique. Des études optiques ont ensuite démontré que les fluctuations de composition induisent de la localisation et présentent un comportement de type boîte quantique. Ce dernier a été observé quel que soit les conditions de croissance explorées dans ce travail. Il est ensuite démontré que les régions plus riches Ga spontanément formés durant la synthèse de l’AlGaN partagent des propriétés µ-optiques similaires sur une plage de longueur d’onde d’émission donnée, pour toutes les conditions de croissance utilisées dans cette étude. De telles régions, émettrices de photons uniques, sont présentes à très petite échelle, puisque elles ont été également mises en évidence dans des nanodisques quantiques d’AlGaN très fins.En outre, le dopage des nanofils d’AlGaN, surtout de type p, est loin d’être totalement compris. En particulier, En particulier, le problème de l’incorporation ainsi que de l’activation optique et électrique dans les fils demeure nébuleux. Cette question a été étudiée pour des jonctions pn nanofilaires d’AlGaN dopées avec des atomes Mg et Si. Premièrement, des signatures propres à l’incorporation des dopants dans les nanofils ont été mises en exergue au travers de techniques de caractérisation structurale, avant que des jonctions pn AlGaN soient mises en évidence électriquement. De plus, des analyses optiques ont mis en lumière des dopants de type n et p optiquement actifs. Néanmoins, les dopants Mg ne sont que partiellement actifs électriquement en raison de la passivation par l’hydrogène mise en évidence par l’observation de complexes Mg-H. Pour résoudre ce problème, des recuits post-croissance ont été effectués. En parallèle, des jonctions pn nanofilaires d’AlN ont été préliminairement examinées et présentent des caractéristiques morphologiques intéressantes. En effet, des creux profonds ont été observés dans les fils et associés au dopage Mg effectué à basse température de croissance. La morphologie des fils peut être ajustée en jouant sur les paramètres cinétiques de croissance et sur l’effet surfactant des atomes Mg. En augmentant la température, les creux disparaissent tandis que la forme du sommet des fils, usuellement hexagonale, change pour devenir « étoilée », mettant en exergue des conditions de croissance très éloignées de l’équilibre thermodynamique. L’activation électrique des dopants n’a pas été observée jusqu’à présent dans ces jonctions pn d’AlN. / Using specific growth conditions, AlGaN nanowire (NW) sections can be grown in epitaxy on top of GaN NW templates. Such NW growth, performed by plasma-assisted molecular beam epitaxy in the present case, allows the subsequent characterization of very small volume of material free of extended defects commonly observed in planar structures. This absence of defects makes these NWs very promising for optoelectronic devices operating in the ultraviolet. However, achieving such devices requires a better understanding of the NW fundamental properties.The issue of alloy inhomogeneity at nanoscale has notably remained obscure so far. In order to make it clearer, the latter has been first investigated in the present work, especially through optical characterization. For our experiments, non-intentionally doped (NID) AlGaN NWs have been grown in various conditions in order to potentially tune the compositional fluctuations within the AlGaN alloy and therefore possibly probe for carrier localization centers of different size and Al composition. It has been firstly observed through structural characterization that the length of Al-rich sections preferentially nucleating on top of GaN NWs can be tuned by varying the growth kinetical parameters, emphasizing a growth mechanism governed by kinetics. Optical studies have then evidenced that compositional fluctuations induce carrier localization and exhibit a quantum dot-like behavior. The latter has been observed whatever the growth conditions explored in this work. Our results are consistent with the spontaneous formation during growth of tiny Ga-richer regions shown to share similar micro-optical features over a given emission wavelength range for all investigated growth conditions. Such regions exhibiting the single-photon emission character are present at very small scale, as signs of their existence have been also evidenced in thin NID AlGaN quantum disks.In addition, doping in Al(Ga)N NW, especially p-type, is far from being fully comprehended. In particular, the issue of dopant incorporation as well as optical and electrical activation in such NWs remains unclear. The latter has been examined in Al(Ga)N NW pn junctions doped with Mg and Si atoms. First, signatures specific to dopant incorporation in NWs have been highlighted through structural characterization, before evidencing AlGaN pn junctions electrically. Moreover, optical analysis have exhibited optically active both dopant types. Nonetheless, Mg dopants are but partially active electrically due to passivation by hydrogen emphasized by the observation of Mg-H complexes. To cope with the latter issue, post-growth annealing experiments have been attempted. Concomitantly, AlN NW pn junctions have been also preliminarily investigated and present interesting morphological features. Indeed, deep hollows have been observed in NWs and associated with Mg doping carried out at low growth temperature. The NW morphology can be tuned by varying growth kinetical parameters and by using the surfactant effect of Mg atoms. When increasing growth temperature, these hollows disappear, while the NW top shape has been observed to switch from hexagonal to star-like, emphasizing growth conditions very far from thermodynamical equilibrium. Electrical activation of dopants has not been evidenced so far in AlN NW pn junctions.

A large variety of material systems for light emission and detection were studied: from very small band gap semiconductors for infra-red (IR) detectors to wide band gap semiconductors for ultra violet (UV) emission as well as CdSe/ZnSe QDs for single photon emitters and rare earth doped oxides for laser fabrication. The growth and characterization aspects were tackled. This work will focus on the relations between the growth procedures and the optical properties. The information that can be gained from optical studies as well as the limitations of those ones will be explained in each case. Following that, a number of projects will be presented. The main one will be based on how to circumvent the problems linked with p-type doping of wide bandgap semiconductors. This project, based on field effect hole injection in wide band-gap semiconductors addresses the major challenge of fabricating efficient deep UV emitters.