Hole Capture Cross Section Research Articles

Regulation of CN-related optical transitions and non-radiative capture cross-section by biaxial strain in AlN

Carbon-related 4.7 eV absorption band and small in-plane strains in AlN may have some significant effects on its application in optoelectronic devices. Based on the accurate hybrid density functional calculation, we investigate the transition energy levels, photo-transition processes, and hole capture cross-sections of CN defect. We propose that the transition from −1 to 0 charge states of CN defect may be responsible for the 4.7 eV absorption band in AlN. In addition, the CN defect-related absorption and emission peaks are linearly dependent on the biaxial strain in the range of −3% to +3%, and the hole non-radiative capture rate by the CN center at the −3% biaxial strain is only 3.65% of that at the +3% biaxial strain. This work provides an effective approach for regulating the charge carrier capture ability of the defect center and improving device performance.

Comprehensive investigation of material properties and operational parameters for enhancing performance and stability of FASnI3-based perovskite solar cells

Recent advancements in the efficiency of lead-based halide perovskite solar cells (PSCs), exceeding 25%, have raised concerns about their toxicity and suitability for mass commercialization. As a result, tin-based PSCs have emerged as attractive alternatives. Among diverse types of tin-based PSCs, organic–inorganic metal halide materials, particularly FASnI3 stands out for high efficiency, remarkable stability, low-cost, and straightforward solution-based fabrication process. In this work, we modelled the performance of FASnI3 PSCs with four different hole transporting materials (Spiro-OMeTAD, Cu2O, CuI, and CuSCN) using SCAPS-1D program. Compared to the initial structure of Ag/Spiro-OMeTAD/FASnI3/TiO2/FTO, analysis on current–voltage and quantum efficiency characteristics identified Cu2O as an ideal hole transport material. Optimizing device output involved exploring the thickness of the FASnI3 layer, defect density states, light reflection/transmission at the back and front metal contacts, effects of metal work function, and operational temperature. Maximum performance and high stability have been achieved, where an open-circuit voltage of 1.16 V, and a high short-circuit current density of 31.70 mA/cm2 were obtained. Further study on charge carriers capture cross-section demonstrated a PCE of 32.47% and FF of 88.53% at a selected capture cross-section of electrons and holes of 1022 cm2. This work aims to guide researchers for building and manufacturing perovskite solar cells that are more stable with moderate thickness, more effective, and economically feasible.

Reassessing iron–gallium recombination activity in silicon

In this work, we present a comprehensive re-evaluation of the iron–gallium (FeGa) recombination parameters in silicon using injection-dependent lifetime spectroscopy (IDLS). Ga-doped silicon wafers (of varying resistivities) with precise concentrations of intentional iron contamination in the silicon wafer bulk, through ion implantation and distribution, were used. The presence of interstitial Fei and FeGa, and their lifetime-limiting effects in these silicon wafers, were confirmed through measuring the effective minority carrier lifetime changes during the conditions that are known to cause FeGa dissociation and association. The presence of Fe was also confirmed by deep-level transient spectroscopy. To ensure accurate IDLS analysis of the FeGa defect in silicon, a lifetime linearization scheme was employed to effectively filter out interference by other defects. Error analysis was employed to find the combination of defect parameters that best fit the experimental data and to ascertain the range of uncertainty associated with the IDLS best-fit results. The optimal fitting of the experimental IDLS by Shockley–Read–Hall statistics produced an electron capture cross section σn=2.3×10−14cm2, hole capture cross section σp=1.1×10−14cm2, and a trap energy level Et=EV+0.2−0.01+0.02eV for the FeGa defect in silicon. The extracted defect parameters are also verified by experimentally measuring the crossover point of Fei and FeGa lifetime curves.

Investigation of a minority carrier trap in a NiO/β-Ga2O3 p–n heterojunction via deep-level transient spectroscopy

The properties of a minority carrier (hole) trap in β-Ga2O3 have been explicitly investigated using a NiO/β-Ga2O3 p–n heterojunction. Via deep-level transient spectroscopy, the activation energy for emission (E emi) and the hole capture cross section (σp ) were derived to be 0.10 eV and 2.48 × 10−15 cm2, respectively. Temperature-enhanced capture and emission kinetics were revealed by the decrease in the capture time constant (τc ) and emission time constant (τe ). Moreover, it was determined that the emission process of the minority carrier trap is independent of the electric field. Taking carrier recombination into account, a corrected trap concentration (N Ta) of 2.73 × 1015 cm−3 was extracted, together with an electron capture cross section (σn ) of 1.42 × 10−18 cm2. This study provides a foundation for the comprehension of trap properties in β-Ga2O3, which is crucial for overcoming self-trapped hole effects when obtaining p-type β-Ga2O3 materials and performance enhancement of β-Ga2O3-based power devices.

Total ionizing dose radiation hardening technology based on double-charge multiple-step ion implantation

This paper describes the total ionizing dose (TID) radiation hardening of buried oxide in separation by implantation of oxygen silicon on insulator (SOI) substrates. In this study, 0.5-μm partially depleted SOI complementary metal–oxide–semiconductor (CMOS) transistors and circuits were prepared using double-charge multiple-step ion implantation and annealing, and the changes in the drive current, leakage current, and threshold voltage of the CMOS transistor under different TID radiation conditions were compared. Under a TID of 500 k Rad(Si), radiation-hardened H-type N-channel metal–oxide–semiconductor transistors had a threshold voltage drift of <100 mV in the worst case, and H-type P-channel metal–oxide–semiconductor transistors had a threshold voltage drift of <150 mV. Under a TID of 1 Mrad(Si), CMOS transistors showed no significant increase in leakage current caused by TID radiation. Under a TID of 500 k Rad(Si), the standby current of the radiation-hardened 32-bit DSP was <1.5 mA. The high-density integrated circuits prepared using this technology performed well under harsh ionizing radiation environments. Finally, this technology was compared with similar foreign technologies and TID-unhardened wafers with hardened wafers. This technology reached the international advanced level among equivalent technologies regarding hole capture cross section and maximum threshold voltage drift in the model. PACS85.30.De 85.30.-e.

Origin of Ga vacancy-related YL center in n-type GaN: A first-principles study

The widely existed yellow luminescence (YL) in n-type GaN is attributed to Ga vacancy-related point defects. However, the nature of Ga vacancy-related YL centers, especially the identification of defect types, is still unclear. Based on first-principles calculations, we systematically investigated the defect formation energies, “0/+“-related optical transition processes and the hole non-radiative trapping of VGa-ON, VGa-VN, VGa-3H and VGa-ON-2H defects in n-type GaN. By static coupling method, the calculated capture cross-sections of VGa-ON and VGa-VN at 300 K are 3.52 × 10−14 cm2 and 6.22 × 10−14 cm2, respectively, which are in good agreement with the hole capture cross-sections of YL center measured by deep-level transient spectroscopy (DLTS) experimentally. While the calculated capture cross-section of 2.46 × 10−13 cm2 for VGa-ON-2H is three times of the maximum experimental data, and the value 1.02 × 10−12 cm2 for VGa-3H is 1–2 orders of magnitude larger than the experimental results. The calculated hole capture cross-sections of these defects centers are quite different, but the transitions between 0 and + charge states of the above defects are all consistent with the experimental YL emission process. This work provides the correlation between the VGa-related defects and the YL center in n-type GaN, bringing essential information to a long-standing controversy.

Photo-conductivity as a transmission phenomenon: Application to the study of β−Ga2O3 thin film

The electrical conduction in semiconductors can be regarded as the flow of particles through an active medium. Starting from the conduction approach as a particle transmission problem, simple and closed expressions for transport coefficients are obtained in a bipolar model of a homogeneous semiconductor at non-steady-state conditions. These coefficients simplify the calculation of the conductivity as a time function and the evaluation of transport phenomena resulting from multiple processes. Noteworthily, they are independent of the dimensions of the sample, and thereby, suitable for the macroscopic case. The formulation described here can be used for the characterization of photoconductive samples from transport measurements and for the verification of band structure models. As an example, we apply this method to analyze the photo-response of a β-Ga2O3 thin film, obtaining estimates for electron (200 cm2/V⋅s) and hole (20 cm2/V⋅s) mobilities, recombination (2.53 × 10−24 cm2) and capture (3.91 × 10−20 cm2) cross-sections, concentrations of defects (deep defects: 3.068 × 1013 1/cm3, shallow defects: 7.556 × 1012 1/cm3) and the density of states of the bands.

Hole capture cross section of the Al acceptor level in 4H-SiC

We performed current deep level transient spectroscopy (I-DLTS) on Al-doped p-type 4H-SiC epilayers. The epilayers exhibited a peak at 100–120 K in the I-DLTS spectra. The correspondence of the deep-level concentrations with the net acceptor concentration implies that the peak originates from the Al acceptor level. The observed activation energy of the peak was low compared to the Al acceptor level reported by Hall measurements owing to the Poole-Frenkel effect. We estimated capture cross section σ for holes of the Al acceptor level at the peak temperature based on the I-DLTS peak height dependence on the injection pulse widths and analyzed the temperature dependence of σ for holes.

Analytical model of the modulated photoluminescence in semiconductor materials

Modulated photoluminescence (MPL) is an optoelectronic characterization technique of semiconductor materials. Going to high frequencies enables one to characterize fast phenomena, and so materials with a short lifetime such as chalcogenides or III–V absorbers. Some typical signatures have already been experimentally observed. However, physical mechanisms and quantitative analyses are not well understood yet. Here, using both an analytical approach and a full numerical modeling, we study how the energy position of a defect level, its electron and hole capture cross sections, its density, influence the frequency dependence of the MPL phase. We show that quantitative information can be extracted. We also study the effect of additional surface recombination, and of non homogeneities created by carrier generation profiles or asymmetric top surface and bottom surface recombination velocities, where diffusion of the carriers plays a role and can be limiting at high frequency. Finally we apply our model to an experimental result to extract defect parameters of the sample. Our analysis highlights the usefulness of MPL and the importance of having a proper modeling of the experiment.

The Hydrogenation Impact on Electronic Properties of p-Diamond/n-Si Heterojunctions.

The undoped polycrystalline diamond films (PDFs) have been deposited on n-type silicon (Si) by Hot Filament Chemical Vapor Deposition (HF CVD) technique. The reaction gases are a mixture of methane and hydrogen. The obtained PDFs were characterized by scanning electron microscopy (SEM) and Raman spectroscopy which, in addition to the diamond phase, also confirms the presence of sp hybridized carbon bonds. As-grown CVD diamond layers are hydrogen terminated and show p-type conductivity. The effect of the level of hydrogenation on the electrical properties of p-diamond/n-Si heterojunctions has been investigated by temperature dependent current–voltage (J-V/T) characteristics. The obtained results suggest that the energy distribution of interface states at the grain boundary (GB) subjected to hydrogenation becomes shallower, and the hole capture cross-section can be reduced. Hydrogenation can lead to a significant reduction of the GB potential barrier. These results can be interesting from the point of view of hydrogen passivation of GBs in microelectronics.

Evaluation of dominant loss mechanisms of PERC cells for optimization of rear passivating stacks

In this paper a comprehensive theoretical analysis has been carried out for Passivated Emitter Rear Contact (PERC) solar cells with the help of numerical modeling using Sentaurus TCAD simulation software. For the rear passivation we have explored various dielectric layers, such as, Aluminium Oxide (Al2O3), Silicon Oxide (SiO2), Hafnium Oxide (HfO2) and Aluminium Nitride (AlN) along with the Silicon Nitride (SiNx) capping layer. Optical generation near the rear side the passivating stacks was analyzed. Recombination current densities were also evaluated for assorted defects with various concentration, hole and electron capture cross sections and position of quasi Fermi level for surface passivation. Thicknesses of the passivating and capping layers were optimized with respect to optical and electrical losses. Field effect passivation (FEP) property of the rear passivating stacks was also investigated. For simulated textured Al-BSF solar cell on silicon wafer of bulk lifetime 250µs, an efficiency of 19.5% was obtained. This solar cell has been used here for reference. We have achieved efficiency of 22.6% for PERC solar cells of rear contact area of 8% on similar type of wafer. It was noticed from few articles that industrial PERC solar cells were fabricated using rear metal contact less than 10% which is well validating our study. It was also perceived that employing Passivated Emitter Rear Totally Diffused (PERT) structure with optimized design parameters further improved the efficiency to ∼ 24%.

Field-Effect Passivation of Undiffused Black Silicon Surfaces

Black silicon (b-Si) surfaces typically have a high density of extreme nanofeatures and a significantly large surface area. This makes high-quality surface passivation even more critical for devices such as solar cells with b-Si surfaces. It has been hypothesized that conformal dielectrics with a high fixed charge density ( <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>f</b></sub> ) are preferred as the nanoscale features of b-Si result in a significant enhancement of field-effect passivation. This article uses 1-D, 2-D, and 3-D numerical simulations to study surface passivation of b-Si, where we particularly focus on the charge carrier control by | <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>f</b></sub> | up to 1 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">13</sup> cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> under accumulation conditions. We will show that there is a significant space charge region compression in b-Si nanofeatures, which affects the charge carrier population control for moderate | <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>f</b></sub> | up to ≈1 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">12</sup> cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> . The average surface minority charge carrier density can be reduced by 70% in some cases, resulting in an equivalent reduction in area-normalized surface recombination losses if the effective surface recombination velocity ( <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> ) is limited by minority carriers. This provides a possible solution for the empirical <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> ∝ 1/ <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>f</b></sub> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sup> reported previously. We will also show that the situation is more complicated for surface passivation films where the ratio between the electron and hole capture cross section (σ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>n</b></sub> / σ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>p</b></sub> ) is higher than 10 for p-type surfaces. For commonly used surface passivation films with a | <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>f</b></sub> | larger than ≈1 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">12</sup> cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> , there is little space charge compression for b-Si. Consequently, <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> simply scales with the surface area, i.e., there is no enhanced reduction of surface recombination by field-effect passivation on b-Si.

First-principles identification of deep energy levels of sulfur impurities in silicon and their carrier capture cross sections

Studies of the deep energy levels and nonradiative carrier capture induced by sulfur doping in silicon were initiated 60 years ago; however, the defect configurations, their deep energy levels, and the carrier capture cross sections are still not well understood. In this study, we focus on SSi substitution, and perform a first-principles study of its defect configurations and the deep energy levels using hybrid exchange-correlation functional. We discover a new distortive configuration for SSi + besides the previously obtained structure with higher symmetry. For both SSi + configurations, the deep transition levels (0/+) and (+/2+) are determined as 0.35 eV and 0.68 eV below the conduction band minimum, respectively. As a benchmark calculation, the hole-capture cross sections for neutral and +1 charged states are obtained based on the distortive structure. The cross section for SSi + agrees with the experiment, demonstrating the multi-phonon process for SSi + capturing a hole, whereas the cross section of SSi 0 is significantly lower than the experimental data because hole capture by SSi 0 is an Auger-type process. Our calculations provide a benchmark for the evaluation of the cross section of carrier capture in semiconductors using multi-phonon nonradiative recombination theory.

Single-Photon Sources Based on Novel Color Centers in Silicon Carbide P–I–N Diodes: Combining Theory and Experiment

HighlightsTheory of electrically driven single-photon sources based on color centers in silicon carbide p–i–n diodes.New method of determining the electron and hole capture cross sections by an optically active point defect (color center) from the experimental measurements of the single-photon electroluminescence rate and second-order coherence.The developed method is based on the measurements at the single defect level. Therefore, in contrast to other approaches, one point defect is sufficient to measure its electron and hole capture cross sections.Point defects in the crystal lattice of SiC, known as color centers, have recently emerged as one of the most promising single-photon emitters for non-classical light sources. However, the search for the best color center that satisfies all the requirements of practical applications has only just begun. Many color centers in SiC have been recently discovered but not yet identified. Therefore, it is extremely challenging to understand their optoelectronic properties and evaluate their potential for use in practical single-photon sources. Here, we present a theoretical approach that explains the experiments on single-photon electroluminescence (SPEL) of novel color centers in SiC p–i–n diodes and gives the possibility to engineer highly efficient single-photon emitting diodes based on them. Moreover, we develop a novel method of determining the electron and hole capture cross sections by the color center from experimental measurements of the SPEL rate and second-order coherence. Unlike other methods, the developed approach uses the experimental results at the single defect level that can be easily obtained as soon as a single-color center is identified in the i-type region of the SiC p–i–n diode.Graphic

Relating Gain Degradation to Defects Production in Neutron-Irradiated 4H-SiC Transistors

Wide-gap semiconductor SiC is generally considered as a superior candidate for high-radiation applications, due to its higher displacement energy for both Si and C lattice atoms. In this work, we find that the current gain ( $\beta $ ) of 4H-SiC bipolar junction transistors (BJTs) can severely deteriorate after neutron irradiation. Deep-level transient spectroscopy (DLTS) reveals that two major carrier-killer centers, Z1/2 and EH6/7, were produced in neutron-irradiated devices. Surprisingly, we find that the shallow Z1/2 center can play a dominant role over the deep EH6/7 defects in carrier recombination under high-injection conditions, which contributes to the unusual smaller hole capture cross section than that of the electron one, as indicated by the first-principles calculations. Finally, technology computer aided design (TCAD) simulation confirms that the enhanced carrier recombination from the Z1/2 centers is responsible for the degraded performance of SiC BJTs after irradiation. Our findings not only provide a deep insight into the underlying physics for displacement damage in SiC BJTs, but also are of interest for technological applications field related to high-energy particle irradiation or implantation.

Sub-Pico-Second Hole Generation Lifetime in Thin Film IGZO Sputtered and Annealed on P-Silicon Substrate

Indium-gallium-zinc-oxide (IGZO) sputtered and annealed in the absence of oxygen shows large oxygen vacancy concentration leading to n-type doping of the thin film. However, because of low intrinsic carrier concentration minority carrier (hole) concentration remains weak. Thus, measurement of thermal hole generation lifetime in unipolar IGZO is challenging. In this work, utilizing the charge transient of an appropriately biased depleted HfO2/IGZO thin film MOS capacitor fabricated on a p-silicon substrate, we extracted hole thermal generation lifetime of IGZO. The measured lifetime was found between 0.62–0.064 ps whence IGZO film received 400 °C to 700 °C anneal treatments followed by its direct sputtering on p-silicon substrate. Further, a morphology analysis of the IGZO was presented. Difference in the lifetime which occurred due to amorphous to the nano-crystalline transition of the IGZO thin film with increased anneal temperature was furthermore validated through XRD, SEM and AFM measurements. Hole capture cross-sections as we obtained from surface generation velocities at the IGZO/HfO2 interface and energy distributions of interface state density were found in agreement, which indirectly demonstrates the accuracy of the measured hole generation lifetime employing the transient capacitance extrapolation, since under the steady state condition hole generation through the interface states of IGZO/HfO2 tends to balance its bulk (IGZO) counterpart

First-principles study on non-radiative carrier captures of point defects associated with proton generation in silica

The non-radiative carrier captures of point defects associated with proton generation in both crystalline (α-quartz) and amorphous silica (a-SiO2) have been investigated by using first-principles calculations based on the density functional theory. Major point defects related to proton generation are considered, including the dimer and puckered configuration of oxygen vacancy, doubly hydrogenated oxygen vacancy, and hydroxyl E′ center. The carrier capture cross sections of the defects are calculated in terms of a one-dimensional static coupling method. The neutral dimer and doubly hydrogenated oxygen vacancies show a large hole capture cross section, and the positively charged puckered oxygen vacancy and the hydroxyl E′ center show a large electron capture cross section. These results further support that dissociation of molecular hydrogen at a positively charged dimer oxygen vacancy and that of a positively charged doubly hydrogenated oxygen vacancy are the main reactions in silica to generate protons that play a key role in ionization damage of silicon devices.

Temperature dependent carrier lifetime, diffusion coefficient, and diffusion length in Ge0.95Sn0.05 epilayer

The development of new technology, which would be able to shift photosensitivity of Si devices to the mid-infrared range, preserving the benefits of cheap silicon readout circuits, is of high priority for short-wave infrared photo-detection in defense, medical, night vision, and material production applications. Group IV GeSn-based materials have recently shown promising optoelectronic characteristics, allowing extension of the detection range to the mid-infrared region. However, the electronic properties of the material are not well understood and need further research. In this work, we provide temperature dependent studies of carrier lifetime, diffusion coefficient, and diffusion length in Ge0.95Sn0.05 epilayer on silicon by applying contactless light induced transient grating technique. The observed temperature dependence of lifetime was explained by the recombination of carriers on vacancy-related defects. The electron and hole capture cross sections were calculated. The temperature dependence of the diffusion coefficient indicated hole mobility limited by phonon and defect scattering. Weakly temperature dependent diffusion length of ∼0.5 μm verified material suitability for efficient submicrometer-thick optoelectronic devices.

Photoluminescence efficiency of Al-rich AlGaN heterostructures in a wide range of photoexcitation densities over temperatures up to 550 K

Time-resolved photoluminescence and light-induced transient grating techniques were applied for temperature and excitation dependent internal quantum efficiency and carrier diffusion investigation in silicon-doped epilayer and AlxGa1-xN MQWs (x > 0.6). Decrease of photoluminescence efficiency at high excitations correlated with increase of diffusion coefficient as well as nonradiative recombination rate, indicating excitation and temperature dependent localized exciton density reduction. Excitation-dependent lifetime decrease was less pronounced than the corresponding drop of the time-integrated photoluminescence efficiency dominated by thermal dissociation of excitons. At lowest excitations excitons were found captured to vacancy complexes. With increase of excitation exciton ionization appeared, leading to enhanced nonradiative recombination of holes on aluminum vacancies and simultaneous droop of PL efficiency due to saturated bimolecular free carrier plasma recombination. At initial recombination stages and high excitations diffusive recombination on dislocations was also revealed. Modelling provided free exciton binding energies of 104, 140 meV in the MQWs and in the layer; exciton localization energies in 12-35 meV range; localized exciton lifetimes in 2-4 ns range; free exciton and carrier radiative recombination rate coefficients of rex = (0.6 0.2)109(T/300)-1 1/s and Brad = (7 1)10-10(T/300)-3/2 cm3/s, respectively. Vacancy complex capture cross section for excitons was determined to be = (2 1)10-16(300/T)2 cm2. Electron and hole capture cross sections to aluminum vacancy were modeled (1.5 1)10-13 cm2 and (7 1)10-13 cm2, respectively. Dislocation Coulomb radius for free carrier recombination was found to be 12-15 nm.

Carrier recombination processes in Fe-doped GaN studied by optical pump–probe techniques

We applied time-resolved free carrier absorption and light-induced transient grating techniques for monitoring carrier dynamics in lightly Fe-doped GaN bulk crystals. Comparison of measured decay times at single- and two-photon carrier photoexcitation with those reported for higher Fe doping revealed two different branches of lifetime dependence on iron density. A deeper insight was reached by using different wavelengths for probing carrier dynamics, which enabled discrimination of electron and hole capture processes by Fe ions. The capture cross section for holes, σh = (4 ± 1) × 10−15 cm2 for the Fe2+* state (5E → 4T2 transition) was determined to be ten times larger than that for electrons, σe = (4 ± 1) × 10−16 cm2 for the Fe3+* state (4T1 → 3T1 transition). Thermal activation of the [Fe2+(5T2),hVB] complex, where hVB is a free hole, required an energy of 315 meV. Light-induced transient grating decay revealed decreasing diffusivity of electrons with Fe doping due to initial carrier capture to Fe3+ states. Simultaneous measurements of carrier diffusion coefficient and lifetime at low excitation conditions (3 × 1017 cm−3) revealed a decrease of bipolar diffusion length from 2 μm in the undoped GaN down to 0.08 μm in the Fe-doped samples at 300 K.