Steven W. Johnston

Dilute nitride GaInNAs and GaInNAsSb solar cells by molecular beam epitaxy

Dilute nitride films with a roughly 1 eV band gap can be lattice-matched to gallium arsenide and germanium, and therefore could become a critical component in next-generation multijunction solar cells. To date most dilute nitride solar cells have been plagued with poor efficiency, due in large part to short diffusion lengths. This study focuses on two techniques aimed at improving the quality of dilute nitride films grown by molecular beam epitaxy: the utilization of biased deflection plates installed in front of the nitrogen plasma source, and the introduction of antimony during growth. Results from GaInNAs cells grown with and without deflection plates, and GaInNAsSb solar cells are reported. The use of biased deflection plates during GaInNAs growth improved every aspect of solar cell performance. For the GaInNAs devices grown with deflection plates, the dark current density, open-circuit voltage, and fill factor were the best of the devices studied. The GaInNAsSb cells had the highest quantum efficienc...

Evidence of the Meyer–Neldel rule in InGaAsN alloys and the problem of determining trap capture cross sections

Deep-level transient spectroscopy measurements have been performed on the quaternary semiconductor InGaAsN. A series of as-grown and annealed metalorganic chemical-vapor-deposited and molecular-beam-epitaxy samples with varying composition were studied. We observed a deep hole trap with activation energy ranging between 0.5 and 0.8 eV in all samples. The data clearly obey the Meyer–Neldel rule (MNR) with an isokinetic temperature of 350 K. We show that great care must be used in extracting capture cross sections (σ) from materials that obey the MNR. In fact, we argue that it is probably not possible to determine σ from the detrapping rate alone. One must measure both trapping and detrapping rates.

Measuring temperature-dependent activation energy in thermally activated processes: A 2D Arrhenius plot method

Thermally activated processes are characterized by two key quantities, activation energy (E(a)) and pre-exponential factor (nu(0)), which may be temperature dependent. The accurate measurement of E(a), nu(0), and their temperature dependence is critical for understanding the thermal activation mechanisms of non-Arrhenius processes. However, the classic 1D Arrhenius plot-based methods cannot unambiguously measure E(a), nu(0), and their temperature dependence due to the mathematical impossibility of resolving two unknown 1D arrays from one 1D experimental data array. Here, we propose a 2D Arrhenius plot method to solve this fundamental problem. Our approach measures E(a) at any temperature from matching the first and second moments of the data calculated with respect to temperature and rate in the 2D temperature-rate plane, and therefore is able to unambiguously solve E(a), nu(0), and their temperature dependence. The case study of deep level emission in a Cu(In,Ga)Se(2) solar cell using the 2D Arrhenius plot method reveals clear temperature dependent behavior of E(a) and nu(0), which has not been observable by its 1D predecessors.

Surface Passivation of CdTe Single Crystals

Low open-circuit voltages (850-870 mV), due to excessive bulk and surface recombination, currently limit CdTe photovoltaic efficiencies. Here, we study surface recombination in single crystals with single-photon excitation time-resolved photoluminescence (1PE-TRPL) to measure minority carrier lifetimes. Typically, minority carrier lifetimes of untreated undoped CdTe material as measured by 1PE-TRPL are ~100 ps or less, even though their bulk lifetimes as measured by two-photon excitation TRPL can reach 100 ns. Such short 1PE-TRPL lifetimes indicate very high surface recombination velocities exceeding 100 000 cm/s. Here, we examine treatments that can reduce surface recombination and discuss different ways of evaluating their efficacy.

Two dimensional numerical simulations of carrier dynamics during time-resolved photoluminescence decays in two-photon microscopy measurements in semiconductors

We use two-dimensional numerical simulations to analyze high spatial resolution time-resolved spectroscopy data. This analysis is applied to two-photon excitation time-resolved photoluminescence (2PE-TRPL) but is broadly applicable to all microscopic time-resolved techniques. By solving time-dependent drift-diffusion equations, we gain insight into carrier dynamics and transport characteristics. Accurate understanding of measurement results establishes the limits and potential of the measurement and enhances its value as a characterization method. Diffusion of carriers outside of the collection volume can have a significant impact on the measured decay but can also provide an estimate of carrier mobility as well as lifetime. In addition to material parameters, the experimental conditions, such as spot size and injection level, can impact the measurement results. Although small spot size provides better resolution, it also increases the impact of diffusion on the decay; if the spot size is much smaller than the diffusion length, it impacts the entire decay. By reproducing experimental 2PE-TRPL decays, the simulations determine the bulk carrier lifetime from the data. The analysis is applied to single-crystal and heteroepitaxial CdTe, material important for solar cells, but it is also applicable to other semiconductors where carrier diffusion from the excitation volume could affect experimental measurements.

Applications of imaging techniques for solar cell characterization

Minority-carrier lifetime, diffusion length, resistance, and shunting are all useful parameters for monitoring material quality, processing, and cell performance. These data may be quickly obtained, and even potentially used during manufacturing, when collected from recently developed imaging techniques. Point-by-point measurements provide quantitative data that is valuable as a research tool, even though data acquisition time may be lengthy. Imaging data can often be collected in seconds with better resolution, and while it may initially appear only qualitative, correlations and calibrations are possible and can transform the image to a set of values. We present several examples of photoluminescence imaging on multi-crystalline Si wafers compared to microwave reflection lifetime mapping. We also present electroluminescence imaging and dark lock-in thermography on several cells of different efficiency and compare to diffusion length, lifetime, and sheet resistance.

Relationship of Open-Circuit Voltage to CdTe Hole Concentration and Lifetime

We investigate the correlation of bulk CdTe and CdZnTe material properties with experimental open-circuit voltage (Voc) through fabrication and characterization of diverse single-crystal solar cells with different dopants. Several distinct crystal types reach Voc > 900 mV. Correlations are in general agreement with Voc limits modeled from bulk minority-carrier lifetime and hole concentration.

Modeling minority-carrier lifetime techniques that use transient excess-carrier decay

Details of the operation of a photoconductive decay technique called resonant-coupled photoconductive decay are revealed using modeling and circuit simulation. The technique is shown to have a good linear response over its measurement range. Experimentally measured sensitivity and linear response compare well to microwave reflection at low injection levels. We also measure the excess-carrier decay rate by an infrared free-carrier transient absorption technique and show comparable high injection level lifetimes.

Evidence of the Meyer-Neldel Rule in InGaAsN Alloys: Consequences for Photovoltaic Materials

We present data showing the potential adverse effects on photovoltaic device performance of all traps in InGaAsN. Deep-level transient spectroscopy measurements were performed on InGaAsN samples grown by both metal-organic chemical vapor deposition and RF plasma-assisted molecular-beam epitaxy. For each growth technique, we studied samples with varying nitrogen composition ranging from 0% to 2.2%. A deep hole trap with activation energy ranging between 0.5 and 0.8 eV is observed in all samples. These data clearly obey the Meyer-Neldel rule, which states that all traps have the same emission rate at the isokinetic temperature. A fit of our trap data gives an isokinetic temperature of 350 K, which means that both deep and shallow traps emit slowly at the operating temperature of solar cells-thus, the traps can be recombination centers.

CdTe single-crystal heterojunction photovoltaic cells

Most recent gains in CdTe photovoltaic (PV) device efficiency have been in short-circuit current density (Jsc) and fill factor (FF), rather than open-circuit voltage (Voc). Because Jsc is nearing its theoretical limit, further improvements in device efficiency will require increasing Voc beyond 860 mV and increasing FF. Voc and FF may be improved by increasing both the carrier concentration and minority-carrier lifetime of the CdTe. However, Voc may be limited for other reasons, including Fermi-level pinning and surface recombination. In this study, we used doped CdTe single crystals to test whether higher carrier concentration and lifetime can overcome traditional Voc barriers. In our work to date, we have fabricated heterojunction CdTe PV cells with Voc up to 929 mV, FF of ~60%, and efficiencies of 10%.