The semiconductor minority carrier lifetime contains information about several important material properties, including Shockley–Read–Hall defect levels/concentrations and radiative/Auger recombination rates, and the complex relationships between these parameters produce a non-trivial temperature-dependence of the measured lifetime. It is tempting to fit temperature-dependent lifetime data to extract the properties of the Shockley–Read–Hall recombination centers; however, without a priori knowledge of the distribution of the Shockley–Read–Hall states across the bandgap, this fit problem is under-constrained in most circumstances. Shockley–Read–Hall lifetime data are not well-suited for the extraction of Shockley–Read–Hall defect levels but can be used effectively to extract minority carrier recombination lifetimes. The minority carrier recombination lifetime is observed at temperatures below 100 K in a Si-doped n-type InGaAs/InAsSb superlattice, and deviation from its expected temperature-dependence indicates that the capture cross section of the defect associated with Si-doping has an activation energy of 1.5 meV or a characteristic temperature of 17 K. This lower temperature regime is also preferrable for the analysis of the physics of defect introduction with displacement-damage-generating particle irradiation.
The effect of majority carrier concentration and minority carrier lifetime on the performance of mid-wave infrared ( λ cutoff = 5.5 μ m ) nBn detectors with variably doped InGaAs/InAsSb type-II superlattice absorbers is investigated. The detector layer structures are grown by molecular beam epitaxy such that their absorbing layers are either undoped, uniformly doped with a target density of 4 × 1015 cm−3, or doped with a graded profile, and variable-area mesa detector arrays are fabricated. Each material's temperature-dependent minority carrier lifetime is determined by time-resolved photoluminescence, and majority carrier concentration is extracted from capacitance–voltage measurements. Detector performance is evaluated with dark current and photocurrent measurements, from which quantum efficiency and shot-noise-limited noise-equivalent irradiance are calculated. The two doped detectors have lower dark current densities compared to their undoped counterpart due to the reduction in diffusion current as well as suppression of depletion current. Although both intentionally doped devices exhibit lower minority carrier lifetimes relative to the undoped device, the device with graded doping maintains a comparable quantum efficiency to the undoped device. Ultimately, the graded doping structure exhibits the highest sensitivity with a shot noise-limited noise-equivalent irradiance of 6.3 × 1010 photons/cm2 s in low-background light conditions, within a factor of 4× of an infrared detector pixel with Rule 07 dark current density and unity quantum efficiency. A detailed analysis of the dark current, quantum efficiency, and minority carrier lifetime provides insight into the material and device design factors that must be considered to realize a device with optimal sensitivity.
Capacitance–voltage measurements are a powerful technique to determine doping profiles of semiconductor pn junctions and Schottky barrier diodes. The measurements were recently extended to III-V-based mid-wavelength nBn infrared detectors, and absorber doping densities have been extracted using the widely accepted Schottky approximation, where the potential drop across the device is assumed to be across the depleting absorber layer. However, this approach is limited to when the absorber region of the nBn is under high reverse bias and thus is only able to extract the absorber region doping profile. Here, we introduce a semi-analytical model that is capable of extracting barrier dopant polarity, doping concentration, and thickness, as well as contact and absorber layer doping concentrations, all from a capacitance–voltage measurement. Rather than solely considering the potential drop across the depleting layers, it considers the potential drop across the accumulating layer as well. This negative charge accumulation occurs for the contact and absorber layers in the case of reverse and forward biases, respectively. This allows for a single model to be applied to a capacitance–voltage curve at forward and reverse biases and it can provide regions of bias where the absorber transitions from depletion to accumulation. We compare the agreement of the semianalytical model with modeling results from commercially available finite element method software and experimental capacitance–voltage data. Finally, we show that the method is consistent with the Schottky approximation of extracting absorber doping densities at high reverse bias and discuss the model's limitations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.