Table of contents

The full performance of GaN devices for high power applications is not exploited due to their self-heating. Possible solutions are the integration of materials with high heat conductivity i.e., single crystalline diamond and graphene layers. We report the growth of single crystalline (0001)-oriented GaN thin films on (100), (110) and (111) diamond single crystals studied by transmission electron microscopy (TEM) in cross-sections. As for graphene, we show a high quality GaN layer that was deposited on patterned graphene layers and 6H-SiC. The atomic structures of the interfaces in the heterostructure are studied using aberration-corrected scanning TEM combined with energy dispersive x-ray and electron energy-loss spectroscopy.

We present the results of a conventional and high-resolution electron microscopy investigation of thick (up to 15 μm) semipolar GaN layers grown on Si(001) offcut substrates with 3C-SiC and AlN buffer layers. GaN and AlN layers have been grown by chloride vapor phase epitaxy. The silicon carbide buffer layers were produced by a new method of solid-phase synthesis on 4 and 7° offcut Si(001) substrates. It is shown that the use of solid-phase synthesis for the formation of SiC layer allows one to grow a semipolar GaN layer. The asymmetrical defect structure of the semipolar layer is revealed.

Strained Si_1−yC_y epilayers have been grown on Si (001) by reduced pressure chemical vapor deposition, using low-cost precursors disilane and trimethylsilane. Substitutional C incorporation has been achieved in strained epilayers up to y = 1.5%, while higher C content of at least 2.4% is observed in relaxed layers. These results are comparable to the highest concentrations achieved using more highly reactive, but expensive, precursors. These relatively high C content epilayers were found to form defects throughout growth attributed to the clustering of C adatoms, which result in localized accelerated amorphous growth and, consequently, hillocks forming on the epilayer surface. The formation, size and distribution of these surface defects has been analyzed through the use of various microscopic techniques. The size and density of these structural defects increases with both C content and epilayer thickness. In our layers of fixed growth time, substitutional C compositions above 1.5% causes hillocks to fuse on the surface; subsequently amorphous growth occurs, which forms an amorphous layer over the crystalline Si_1−yC_y epilayer and hence prevents further epitaxy or reliable device fabrication. The results of this investigation suggest that substitutional C composition of below 1.5% could be achieved without the need for expensive and volatile precursors or complex growth processes, assuming sufficiently thin layers are grown.

Cross section transmission electron microscopy has been used to analyse dislocation filter layers (DFLs) in five similar structures of GaAs on Si that had different amounts of strain in the DFLs or different annealing regimes. By counting threading dislocation (TD) numbers through the structure we are able to measure relative changes, even though the absolute density is not known. The DFLs remove more than 90% of TDs in all samples. We find that the TD density in material without DFLs decays as the inverse of the square root of the layer thickness, and that DFLs at the top of the structure are considerably more efficient than those at the bottom. This indicates that the interaction radius, the distance that TDs must approach to meet and annihilate, is dependent upon the TD density.

Fe₃Si/Al/Fe₃Si/GaAs(001) structures were deposited by molecular-beam epitaxy and characterized by transmission and scanning electron microscopy, and x-ray diffraction. The first Fe₃Si film on GaAs(001) grew epitaxially as a (001) oriented single crystal. The subsequent Al film grew almost {111} oriented in a fibrous texture although the underlying Fe₃Si is exactly (001) oriented. The growth in this orientation is triggered by a thin transition region which is formed at the Fe₃Si/Al interface. In the end, after the growth of the second Fe₃Si layer on top of the Al, the final properties of the whole stack depended on the substrate temperature T_s during deposition of the last film. The upper Fe₃Si films are mainly {110} oriented although they are poly-crystalline. At lower T_s, around room temperature, all the films retain their original structural properties.

Type II emission optoelectronic devices using GaAsSb strain reduction layers (SRL) over InAs quantum dots (QDs) have aroused great interest. Recent studies have demonstrated an extraordinary increase in photoluminescence (PL) intensity maintaining type II emission after a rapid thermal anneal (RTA), but with an undesirable blueshift. In this work, we have characterized the effect of RTA on InAs/GaAs QDs embedded in a SRL of GaAsSb by transmission electron microscopy (TEM) and finite element simulations. We find that annealing alters both the distribution of Sb in the SRL as well as the exchange of cations (In and Ga) between the QDs and the SRL. First, annealing causes modifications in the capping layer, reducing its thickness but maintaining the maximum Sb content and improving its homogeneity. In addition, the formation of Sb-rich clusters with loop dislocations is noticed, which seems not to be an impediment for an increased PL intensity. Second, RTA produces flatter QDs with larger base diameter and induces a more homogeneous QD height distribution. The Sb is accumulated over the QDs and the RTA enlarges the Sb-rich region, but the Sb contents are very similar. This fact leaves the type II alignment without major changes. Atomic-scale strain analysis of the nanostructures reveal a strong intermixing of In/Ga between the QDs and the capping layer, which is the main responsible mechanism of the PL blueshift. The improvement of the crystalline quality of the capping layer together with higher homogeneity QD sizes could be the origin of the enhancement of the PL emission.

Transmission and scanning electron microscopy were used to examine the growth of gallium nitride (GaN) on polycrystalline diamond substrates grown by metalorganic vapour phase epitaxy with a low-temperature aluminium nitride (AlN) nucleation layer. Growth on unmasked substrates was in the (0001) orientation with threading dislocation densities ≈7 × 10⁹ cm⁻². An epitaxial layer overgrowth technique was used to reduce the dislocation densities further, by depositing silicon nitride stripes on the surface and etching the unmasked regions down to the diamond substrate. A re-growth was then performed on the exposed side walls of the original GaN growth, reducing the threading dislocation density in the overgrown regions by two orders of magnitude. The resulting microstructures and the mechanisms of dislocation reduction are discussed.

The effects of the template on the optical and structural properties of Al_0.75Ga_0.25N/Al_0.8Ga_0.2N multiple quantum well (MQWs) laser active regions have been investigated. The laser structures for optical pumping were grown on planar c-plane AlN/sapphire as well as on thick epitaxially laterally overgrown (ELO) AlN layers on patterned AlN/sapphire. Two ELO AlN/sapphire templates were investigated, one with a miscut of the sapphire surface to the m-direction with an angle of 0.25°, the other with a miscut angle of 0.25° to the sapphire a-direction. The MQWs are studied by atomic force microscopy, plan-view cathodoluminescence (CL) at room temperature and 83 K as well as transmission electron microscopy using high-angle annular dark-field imaging and energy-dispersive x-ray spectroscopy. The results are compared to optical pumping measurements. It was found that the surface morphology of the templates determines the lateral wavelength distribution in the MQWs observed by spectral CL mappings. The lateral wavelength spread is largest for the laser structures grown on ELO AlN with miscut to sapphire a-direction caused by the local variation of the MQW thicknesses and the Ga incorporation at macrosteps on the ELO-AlN. A CL peak wavelength spread of up to 7 nm has been found. The MQWs grown on planar AlN/sapphire templates show a homogeneous wavelength distribution. However, due to the high threading dislocation density and the resulting strong nonradiative recombination, laser operation could not be achieved. The laser structures grown on ELO AlN/sapphire show optically pumped lasing with a record short wavelength of 237 nm.

In this work the growth of complex n-type, high Ge content superlattice structures by reduced pressure chemical vapor deposition is presented. The structures feature 50 repeats of a 14 layer period, which includes a main quantum well that is between 13 and 21 nm wide. The total epitaxy thickness is approximately 8 μm. Diffusion and segregation in the structures was minimized by using a low growth temperature. Materials characterization shows the structures to be of good crystalline quality, with the thickness of all layers close to the design, abrupt interfaces, and uniformity throughout the structures. High angle annular dark field scanning transmission electron microscopy is shown to be an ideal technique for measuring layer thickness and interface quality in these structures.

The formation of three-dimensional truncated pyramids after the deposition of AlN/GaN superlattices onto (0001) AlN/sapphire templates has been analysed by atomic force microscopy as well as transmission electron microscopy. V-pits in AlN layers and the formation of nano-mounds around the v-pit edges are suggested to be responsible for the pyramid formation. Keeping the individual AlN layer thickness at 2.5 nm in the 80xAlN/GaN superlattice, the transformation to the three-dimensional pyramids is observed when the individual GaN layer thickness exceeds 1.5 nm. A subsequent overgrowth of the pyramidal structures by AlGaN results in inhomogeneous Ga distribution in the layers and laterally inhomogeneous strain states. Nevertheless, compared to the growth on planar layers, the overgrowth of the truncated pyramids leads to a slight reduction in dislocation density from 1 · 10¹⁰ cm⁻² (for GaN thickness of 1 nm in SL) to 7 · 10⁹ cm⁻² (for GaN thickness of 2 nm in SL). The non-planar growth front and thus the compositional inhomogeneity in AlGaN vanish gradually with increasing AlGaN thickness. As a result, homogeneous 4 μm thick Al_0.5Ga_0.5N buffer layers suitable for the fabrication of UV-B LED structures can be obtained.

We demonstrate a method to determine the indium concentration, x, of In_xGa_1-xN thin films by combining plasmon excitation studies in electron energy-loss spectroscopy (EELS) with a novel way of quantification of the intensity of x-ray lines in energy-dispersive x-ray spectroscopy (EDXS). The plasmon peak in EELS of InGaN is relatively broad. We fitted a Lorentz function to the main plasmon peak to suppress noise and the influence from the neighboring Ga 3d transition in the spectrum, which improves the precision in the evaluation of the plasmon peak position. As the indium concentration of InGaN is difficult to control during high temperature growth due to partial In desorption, the nominal indium concentrations provided by the growers were not considered reliable. The indium concentration obtained from EDXS quantification using Oxford Instrument ISIS 300 x-ray standard quantification software often did not agree with the nominal indium concentration, and quantification using K and L lines was inconsistent. We therefore developed a self-consistent iterative procedure to determine the In content from thickness-dependent k-factors, as described in recent work submitted to Journal of Microscopy. When the plasmon peak position is plotted versus the indium concentration from EDXS we obtain a linear relationship over the whole compositional range, and the standard error from linear least-squares fitting shows that the indium concentration can be determined from the plasmon peak position to within Δx = ± 0.037 standard deviation.

GaAs/AlGaAs core–shell nanowires (NWs) were grown on Si(111) by Ga-assisted molecular beam epitaxy via the vapor–liquid–solid mechanism. High-resolution and scanning transmission electron microscopy observations showed that NWs were predominantly zinc-blende single crystals of hexagonal shape, grown along the [111] direction. GaAs core NWs emerged from the Si surface and subsequently, the NW growth front advanced by a continuous sequence of (111) rotational twins, while the AlGaAs shell lattice was perfectly aligned with the core lattice. Occasionally, single or multiple stacking faults induced wurtzite structure NW pockets. The AlGaAs shell occupied at least half of the NW's projected diameter, while the average Al content of the shell, estimated by energy dispersive x-ray analysis, was x = 0.35. Furthermore, molecular dynamics simulations of hexagonal cross-section NW slices, under a new parametrization of the Tersoff interatomic potential for AlAs, showed increased atom relaxation at the hexagon vertices of the shell. This, in conjunction with the compressively strained Al_0.35Ga_0.65As shell close to the GaAs core, can trigger a kinetic surface mechanism that could drive Al adatoms to accumulate at the relaxed sites of the shell, namely along the diagonals of the shell's hexagon. Moreover, the absence of long-range stresses in the GaAs/Al_0.35Ga_0.65As core–shell system may account for a highly stable heterostructure. The latter was consolidated by temperature-dependent photoluminescence spectroscopy.

Results are presented of a study of {113}-defect formation in vertical Si nanowire n-type tunnel field effect transistors with nanowire diameters ranging from 40 to 500 nm. The nanowires are etched into an epitaxial moderately As doped n-type layer grown on a heavily As doped ${{\rm{n}}}^{+}$ Si substrate. p⁺ contacts on the nanowire are created by epitaxial growth of a heavily B doped layer. Using focused ion beam cutting, samples for irradiation are prepared with different thicknesses so that the nanowires are fully or partially embedded in the sample thickness. {113}-defects are created in situ by 2 MeV e-irradiation in an ultra-high voltage electron microscope between room temperature and 375 °C. The observations are discussed in the frame of intrinsic point defect properties, taking into account the role of dopants and capping layers. The important impact of the specimen thickness is elucidated.

Transmission electron microscopy is used to examine the structure and composition of In_xGa_1−xN nanorods grown by plasma-assisted molecular beam epitaxy. The results confirm a core–shell structure with an In-rich core and In-poor shell resulting from axial and lateral growth sectors respectively. Atomic resolution mapping by energy-dispersive x-ray microanalysis and high angle annular dark field imaging show that both the core and the shell are decomposed into Ga-rich and In-rich platelets parallel to their respective growth surfaces. It is argued that platelet formation occurs at the surfaces, through the lateral expansion of surface steps. Studies of nanorods with graded composition show that decomposition ceases for x ≥ 0.8 and the ratio of growth rates, shell:core, decreases with increasing In concentration.

Focused ion beams (FIBs) are widely applied during manufacturing and for failure analysis, as a preparation tool for cross sectional scanning electron microscopy or for the extraction of lamellae for (scanning) transmission electron microscopy investigation of nanoelectronic devices. The impact of the ion beam milling on surface contamination is investigated by time-of-flight secondary ion mass spectroscopy, while the electrical surface damage is analyzed by a micro four-point probe. It is shown that the redeposition of milled Ga and Cu reaches levels below sensitivity (5 × 10¹⁰ at cm⁻²) at less than 10 mm from FIB structures while the lateral range of electrical surface damage is an order of magnitude smaller. The major source of the redeposition is the resputtering of sputtered material from the sample that was previously deposited on the SEM column. The 2D distribution of the redeposition is asymmetric and is simulated well based on a simplified model of the column and sample configuration. The electrical surface damage mainly relates to the beam tails. Pt deposits for surface protection require much lower Ga⁺ ion doses, and therefore have less impact on the wafer surface contamination. However, the range of electrical surface damage is larger for Pt deposits due to increased beam scattering in the low vacuum during the Pt deposition. With these contamination and damage levels and ranges, 'wafer return', i.e. continuing the wafer processing after the FIB, can be considered feasible for back-end of line processes with the loss of only the analyzed die or, potentially, also its neighbor. For front-end of line processes the acceptable contamination levels are more stringent and the feasibility of wafer return will be more process specific.

Copper containing transmission electron microscopy (TEM) specimens frequently show corrosion after focused ion beam (FIB) preparation. This paper reveals that the corrosion product is a Cu–S phase growing over the specimen surface. The layer is identified by energy-dispersive x-ray spectroscopy, and lattice spacing indexing of power spectra patterns. The corrosion process is further studied by TEM on cone-shaped specimens, which are intentionally stored after FIB preparation with S flakes for short time. Furthermore, a protective method against corrosion is developed by varying the time in the FIB vacuum and the duration of a subsequent plasma cleaning.

High-pressure phases of silicon such as Si-XII/Si-III exhibit attractive optical, electrical and chemical properties, but until now, it has been technologically impossible to produce a significant quantity of Si-XII or Si-III. In this study, to explore the possibility of generating high-pressure silicon phases efficiently, comparative nanoindentation experiments were conducted. Effects of the loading rate, unloading rate and maximum indentation load were investigated, and key factors affecting the high-pressure phase formation were identified. A new nanoindentation protocol is proposed that introduces an intermediate holding stage into the unloading process. The resulting end phases under the indent were detected by a laser micro-Raman spectrometer and compared with those formed in conventional nanoindentation. The results indicate that high-pressure phases Si-XII and Si-III were successfully formed during the intermediate holding stage even with a very high loading/unloading rate. This finding demonstrates the possibility of rapid production of high-pressure phases of silicon through fast mechanical loading and unloading.

The impact of a single event on the performance of CMOS current mirrors (CMs) is studied experimentally in this paper. Both basic and cascode CMs based on bulk Si and PDSOI substrates are employed to demonstrate the permanent effects of damage generated by heavy-ion strikes. The results show that the mismatch of the CMs (bulk Si/PDSOI basic/cascode CMs) changes after heavy-ion irradiation, which means that the accuracy of the output current may need re-evaluation when CMs are operated in a harsh environment. For output impedance, a drastic reduction of 40% is observed for small (W/L = 0.5 μm/0.25 μm) PDSOI basic CMs. This may limit the application of CMs when high output impedance is required to provide a large gain or common mode rejection ratio. Different types of performance degradation after heavy-ion irradiation are classified, and the characteristics are also statistically compared between different types of CM. The mechanisms of these changes are then discussed and traced back to the damage induced by the random heavy-ion strikes. These results demonstrate the permanent effects of damage generated by heavy-ion strikes in CMOS CMs, and provide insights into the impact of heavy-ion irradiation on analog circuits.

The absorption of infrared radiation by a polar semiconductor nanosphere is investigated, using a new extension of the Mie theory. The theory takes into account the spatial dispersion or nonlocality and determines the generalized Mie coefficients using an additional boundary condition of the continuity of the normal component of the displacement field at the surface of the system. Examples of calculated infrared absorption spectra are presented, by employing the typical values of doped InAs and InSb nanospheres. In the presence of the spatial nonlocal effects, the blue shift of the main absorption peaks, due to the coupled surface plasmon-optical phonon modes, are found to occur, while two groups of subsidiary peaks, due to coupled bulk plasmon-optical phonon excitations, appear in the optical spectra of the polar semiconductor nanospheres.

In this report, we present results of an experimental investigation of a near mid-gap trap energy level in InAs_{10 ML}/GaSb_{10 ML} type-II superlattices. Using thermal analysis of dark current, Fourier transform photoluminescence and low-frequency noise spectroscopy, we have examined several wafers and diodes with similar period design and the same macroscopic construction. All characterization techniques gave nearly the same value of about 140 meV independent of substrate type. Additionally, photoluminescence spectra show that the transition related to the trap centre is temperature independent. The presented methodology for thermal analysis of dark current characteristics should be useful to easily estimate the position of deep energy levels in superlattice photodiodes.

Structural and optical characterization of multi-stack InAs/InGaAs submonolayer quantum dots (SML-QDs) grown under the Stranski–Krastanov growth mode was performed via a polarization-dependent study. Various angular in-plane (plane perpendicular to growth direction) and out-of-plane (plane parallel to the 45°-polished facet) polarization-dependent spectral photocurrents were measured to investigate the three-dimensional quantum confinement in multi-stack 0.3 monolayer InAs SML-QDs (the shape of SML-QDs embedded in an SML-QD based photodetector). Scanning transmission electron microscopy images revealed the interdiffusion of indium atoms between SML-QDs and the InGaAs quantum well due to an insufficient amount of indium, which agrees well with the tendency of s- to z-polarized response ratios.

We have investigated the feasibility of selectively growing SiGe:B layers at 450 °C, 20 Torr in a 300 mm industrial reduced pressure chemical vapor deposition tool. A reduced H₂ carrier gas mass-flow has been used in order to have acceptable growth rates at such a temperature, which is very low indeed. We have first of all studied on blanket Si wafers the in situ boron doping of SiGe with Si₂H₆, GeH₄ and B₂H₆. A growth rate increase by a factor close to 7 together with a Ge concentration decrease from 53% down to 32% occurred as the diborane mass-flow increased. Very high B⁺ ion concentrations were obtained in layers that were single crystalline and smooth. Their concentration increased almost linearly with the B₂H₆ mass-flow, from 1.8 up to 8.3 × 10²⁰ cm⁻³. The associated resistivity dropped from 0.43 down to 0.26 mΩ cm. We have then tested whether or not selectivity versus SiO₂ could be achieved by adding various amounts of HCl to Si₂H₆ + GeH₄ +B₂H₆. Single crystalline growth rates of intrinsic SiGe(:B) on Si were very similar to poly-crystalline growth rates on SiO₂-covered substrates irrespective of the HCl flow. Straightforward selectivity was thus not feasible with a co-flow approach. As a consequence, a 450 °C deposition/etch (DE) process was evaluated. Growth occurred at 20 Torr with the above-mentioned chemistry, while the selective etch of poly-SiGe:B versus c-SiGe:B was conducted at 740 Torr with a medium HCl mass-flow (F(HCl)/F(H2) = 0.2) and a high H₂ flow. A 2.2 etch selectivity was achieved while retaining single crystalline if slightly rough SiGe:B layers.

It is shown that in high-power, large optical cavity laser diodes at high injection currents, the optical losses due to nonuniform carrier accumulation in the optical confinement layer can ensure the laser operation in the fundamental transverse mode. An experimental demonstration of switching from second order mode to fundamental mode in large optical cavity lasers with current and/or temperature increase is reported and explained, with the calculated values for the switching current and temperature in good agreement with the measurements. The results experimentally prove the nonuniform nature of carrier accumulation in the confinement layer and may aid laser design for optimizing the output.

We have investigated the effects of post-metallization-annealing (PMA) in oxygen atmosphere on recessed-gate GaN metal-oxide-semiconductor heterostructure field effect transistors (MOSHFETs). The flat band voltage of MOS is a function of bulk and interface charges in the oxide, which strongly depends on a post-annealing process as well as deposition conditions. A positive threshold voltage shift enabling normally-off operation has been achieved by an O₂ PMA process where the GaN MOSHFET employed an ICPCVD SiO₂ gate oxide with a Ni/Au metal gate. According to the analysis using energy dispersive x-ray spectroscopy in transmission electron microscopy and x-ray photoelectron spectroscopy, it is suggested that the improved SiO₂/GaN interface quality with an enhanced metallic-like Ga level was responsible for the positive shift in threshold voltage.

This paper presents a novel mathematical model of the bipolar resistive switching (BRS) of the metal-insulator-semiconductor-metal (MISM) in a Pt/Ta₂O₅/TaO_x/Pt memristor. The proposed model is based on quantum mechanics and describes the BRS behaviour based on electron band theory and the physical characteristics of the metal-insulator-semiconductor (MIS) system. It also includes the physical characteristics of the insulator layer. The novelty of the proposed model lies in incorporating the tunnelling probability factor (TPF) between the semiconductor and the metal layers and therefore demonstrating its effect on the conduction mechanism. In addition, the effect of continuous variation of the interface traps densities and the ideality factor during BRS is modelled using the semiconductor properties and the characteristics of the MIS system. Thus, the model emphasizes the dependency of the memristor current on the physical characteristics of the insulator layer. Moreover, the electric field equation for the active region is derived for the MISM structure which is used, together with the Mott and Gurney rigid point-ion model and the Joule heating effect, to model the oxygen ion migration mechanism. Finally, the model also demonstrates the self-limiting growth of the doped region. Extensive simulation is carried out on the proposed model and the results are correlated against the experimental data which show that the proposed model is in good agreement with the physical characteristics of the MISM memristor.

We have investigated dual-gate AlGaN/GaN metal-insulator-semiconductor high-electron mobility transistors (MIS-HEMTs) using Si₃N₄ as the gate dielectric by comparison with single-gate MIS-HEMTs. It is shown that the presence of the second gate induces a slight reduction in the maximum output current, transconductance and breakdown voltage, but with the advantages of 5 dB enhanced power gain and higher f_T/f_max. Combined with a physics-based device simulation, the breakdown characteristics of the dual-gate device are revealed to be dependent on the second gate. These results demonstrate that the incorporation of dual-gate configuration into the MIS gate is a potential alternative for GaN-based high-power and high-frequency applications.

The composition and structure of TiAl-based metallizations have been investigated depending on the Ti and Mo barriers. The lowest contact resistivity of 4 × 10⁻⁶ Ω.cm² for a Ti barrier and 7 × 10⁻⁶ Ω.cm² for a Mo barrier is obtained at a Ti/Al ratio of 0.43 after annealing at 800 °C. The scanning transmission electron microscope (STEM) and energy dispersive spectroscopy (EDS) analyses reveal that Mo is not an effective barrier for the Au in-diffusion and Al out of diffusion during annealing. The intensive diffusion processes lead to the formation of the semimetal TiN compound at the interface and intermetallic phases of Au, Al, and Ti, the structure and composition of which depend on the barrier metal.

III-V semiconductor nanowires have attracted intensive research interest because of their promising optical and electronic properties that can be manipulated by tailoring nanowire composition and morphology. Therefore, it is crucial to measure and control the diameter distribution of the grown nanowires. In this study, we analyze the diameter distribution of Au-catalyzed InAs nanowires. Au colloidal nanoparticles dispersed on InAs (111) B substrates and nanoparticles obtained by the thermal annealing of Au films were used as catalysts for InAs nanowire growth. The annealing time and temperature, the thickness of the Au film and the colloid sizes were systematically varied not only to understand their influence on nanowire diameter distribution, but also to find the optimal parameters for realizing samples with uniform and controlled diameter distribution. Morphological characterization was performed by scanning electron microscopy measurements and the image analysis was carried out using in-house-developed automated image analysis software to accurately determine the diameter distribution of the nanowires. A description of the image analysis software is also presented. The thermal annealing of films turned out to be the most suitable method for uniformity and density control, while the colloidal nanoparticles yielded narrow and more reproducible diameter distributions.

A type-II GaAs/GaAlAs short-period superlattice (SPSL) used as an electro-optic medium for the spectral range 820–850 nm is studied in a vertical microcavity geometry. SPSL is sandwiched between two GaAlAs distributed Bragg reflectors. Optical power reflectance (OR) spectra are measured as a function of applied reverse bias at different tilt angles and temperatures. All spectra reveal a blue shift of the reflectivity dip upon applied voltage which evidences a negative electrorefraction of the electro-optic medium. The shift enhances up to ∼0.6 nm once the exciton resonance is brought close to the wavelength of the reflectivity dip. As opposed to those modulators based on quantum–confined Stark effect, no increased absorption is observed at an applied bias, because the integrated intensity of the reflectivity dip in the OR spectra is virtually constant. This indicates a low absorption loss with applied bias and consequently a high potential for the increased dynamic range of the related modulator.

Memristive devices with different underlying physical mechanisms are investigated and compared with respect to their utilization in passive cross-bar arrays for computing and memory applications. Niobium oxide-based metal–insulator–metal structures in various configurations exhibit abrupt filamentary resistive switching, filamentary resistive switching together with a threshold switching effect and analog switching characteristics. It is found that the initial electroforming step, which is mandatory for filamentary cells, causes problems if no individual selector device ensuring internal current compliance is applied. In contrast, cells based on analog switching are forming free and could be operated without difficulty. Thus they might be of value for utilization as passive circuit elements.

Within the context of reducing production costs, thin (<90 μm) silicon foils intended for photovoltaic applications have been fabricated from standard (100)Si wafers using a low-temperature (<150 °C) stress-induced lift-off process. A multi-frequency electron spin resonance (ESR) study was performed in order to evaluate, at atomic scale, the quality of the material in terms of defects, including identification and quantification. Generally, a complex ESR spectrum is observed, disentangled as the superposition of three separate signals. This includes, most prominently (∼91% of total density) the D-line (Si₃ ≡ Si· dangling bonds in a disordered Si environment), a set (∼6%) of highly anisotropic signals ascribed to dislocations (K1-like), and a triplet, identified as the Si-SL5 N-donor defect. Defect density depth profiling from the lift-off side shows all signals disappear in tandem after etching off a ∼33 μm thick Si layer, indicating a highly correlated−equal in relative terms−distribution of the three types of defects over the affected top part of the Si foil. The defect density is found to be highly non-uniform laterally, with the density peaking near the crack initiation point, from which defect generation spreads. It is thus found that the SLIM-Cut method for fabrication of thin Si foils results in the introduction of defects that would unacceptably impair the functionality of photovoltaic cells built on these substrates. Fortunately, this may be cured by etching off a thin top Si layer, resulting in a most useful thin Si foil of standard high quality.

This work presents an analysis of the capabilities of GaN Schottky diodes for frequency multipliers and mixers at millimeter wavelengths. By using a Monte Carlo (MC) model of the diode coupled to a harmonic balance technique, the electrical and noise performances of these circuits are investigated. Despite the lower electron mobility of GaN compared to GaAs, multipliers based on GaN Schottky diodes can be competitive in the first stages of multiplier chains, due to the excellent power handling capabilities of this material. The performance of these circuits can be improved by taking advantage of the lateral Schottky diode structures based on AlGaN/GaN HEMT technology.

Time-resolved photoluminescence (PL) measurements of Al_0.82In_0.18N/GaN heterostructures revealed a large enhancement of low temperature PL from the GaN layer and a strong temperature dependence of this effect. Analysis of different phenomena that might affect the photoexcited carrier dynamics suggests that the enhanced GaN PL should be attributed to photoexcited hole transfer from the AlInN layer. The hole transport most probably takes place via dense sub-band edge valence band states related to nanoscale In-rich clusters. Scanning near-field optical microscopy data support this assignment.

Buffer leakage is an important parasitic loss mechanism in AlGaN/GaN high electron mobility transistors (HEMTs) and hence various methods are employed to grow semi-insulating buffer layers. Quantification of carrier concentration in such buffers using conventional capacitance based profiling techniques is challenging due to their fully depleted nature even at zero bias voltages. We provide a simple and effective model to extract carrier concentrations in fully depleted GaN films using capacitance–voltage (C–V) measurements. Extensive mercury probe C–V profiling has been performed on GaN films of differing thicknesses and doping levels in order to validate this model. Carrier concentrations as extracted from both the conventional C–V technique for partially depleted films having the same doping concentration, and Hall measurements show excellent agreement with those predicted by the proposed model thus establishing the utility of this technique. This model can be readily extended to estimate background carrier concentrations from the depletion region capacitances of HEMT structures and fully depleted films of any class of semiconductor materials.

Density functional theory calculations have been applied to study the structural and electronic properties of layered -GaSe, γ-InSe, β-GaS and GaTe compounds. The optimized lattice parameters have been obtained using vdW-DF2-C09 exchange-correlation functional, which is able to describe dispersion forces and produces interlayer distances in close agreement with experiments. Based on the calculated electronic band structures, the energy position of the charge neutrality level (CNL) in the III–VI semiconductors has been estimated for the first time. The room-temperature values of CNL are found to be 0.80 eV, 1.02 eV, 0.72 eV and 0.77 eV for -GaSe, β-GaS, GaTe and γ-InSe, respectively. The persistent p-type conductivity of the intentionally undoped -GaSe, β-GaS and GaTe and n-type conductivity of γ-InSe crystals are discussed and explained using the concept of CNL. We also estimated the barrier heights for a number of metal/semiconductor and semiconductor/semiconductor interfaces assuming partial Fermi level pinning at the CNL. A reasonable agreement between our calculations and the available experimental data has been obtained.

The electrical characteristics of p-channel pentacene thin-film transistors (TFTs) were analyzed at different operating temperatures ranging from 253 to 353 K. An improvement in the drain current and field-effect mobility of the pentacene TFTs is observed with increasing temperature. From the Arrhenius plots of field-effect mobility extracted at various temperatures, a lower activation energy of 99.34 meV was obtained when the device is operating in the saturation region. Such observation is ascribed to the thermally activated hole transport through the pentacene grain boundaries. On the other hand, it was found that the Au/pentacene contact significantly affects the TFTs electrical characteristics in the linear region, which resulted in a higher activation energy. The activation energy based on the linear field-effect mobility, which increased from 344.61 to 444.70 meV with decreasing temperature, implies the charge-injection-limited electrical behavior of pentacene TFTs at low temperatures. The thermally induced electrical characteristic variations in pentacene TFTs can thus be studied through the temperature dependence of the charge injection and transport processes.