The gallium gradient in Cu(In,Ga)Se2 (CIGS) layers, which forms during the two industrially relevant deposition routes, the sequential and co‐evaporation processes, plays a key role in the device performance of CIGS thin‐film modules. In this contribution, we present a comprehensive study on the formation, nature, and consequences of gallium gradients in CIGS solar cells. The formation of gallium gradients is analyzed in real time during a rapid selenization process by in situ X‐ray measurements. In addition, the gallium grading of a CIGS layer grown with an in‐line co‐evaporation process is analyzed by means of depth profiling with mass spectrometry. This gallium gradient of a real solar cell served as input data for device simulations. Depth‐dependent occurrence of lateral inhomogeneities on the µm scale in CIGS deposited by the co‐evaporation process was investigated by highly spatially resolved luminescence measurements on etched CIGS samples, which revealed a dependence of the optical bandgap, the quasi‐Fermi level splitting, transition levels, and the vertical gallium gradient. Transmission electron microscopy analyses of CIGS cross‐sections point to a difference in gallium content in the near surface region of neighboring grains. Migration barriers for a copper‐vacancy‐mediated indium and gallium diffusion in CuInSe2 and CuGaSe2 were calculated using density functional theory. The migration barrier for the InCu antisite in CuGaSe2 is significantly lower compared with the GaCu antisite in CuInSe2, which is in accordance with the experimentally observed Ga gradients in CIGS layers grown by co‐evaporation and selenization processes. Copyright © 2014 John Wiley & Sons, Ltd.

In the third part of this series, we study the influence of trap states in the band gap of semiconductors on the time-resolved luminescence decay (TRL) after a pulsed excitation. The results based on simulations with Synopsys TCAD® and analytical approximations are given for p-doped Cu(In,Ga)Se2 as a working example. We show that a single trap can be mostly described by two parameters which are assigned to minority carrier capture and emission. We analyze their influence on the luminescence decay and study the difference between a single trap and an energetic Gaussian trap distribution. It is found that trap states artificially increase the TRL decay and obscure the recombination dynamics. Thus, there is a demand for experimental methods which can reveal the recombination of minority carriers in a TRL experiment without trapping effect. In this regard, a variation of the device temperature, the excitation frequency, the injection level, as well as a bias illumination may be promising approaches. We study these methods, discuss advantages and disadvantages, and show experimental TRL for prove of concept. At the end, we validate our approach of simulating only band-to-band radiative recombination although photoluminescence spectra often exhibit free-to-bound radiative recombination of charge carriers.

Thin-film solar cells based on polycrystalline absorbers have reached very high conversion efficiencies of up to 23-25%. In order to elucidate the limiting factors that need to be overcome for even higher efficiency levels, it is essential to investigate microscopic origins of loss mechanisms in these devices. In the present work, a high efficiency (21% without anti-reflection coating) copper indium gallium diselenide (CIGSe) solar cell is characterized by means of a correlative microscopy approach and corroborated by means of photoluminescence spectroscopy. The values obtained by the experimental characterization are used as input parameters for two-dimensional device simulations, for which a real microstructure was used. It can be shown that electrostatic potential and lifetime fluctuations exhibit no substantial impact on the device performance. In contrast, nonradiative recombination at random grain boundaries can be identified as a significant loss mechanism for CIGSe solar cells, even for devices at a very high performance level.

In the second part of this series, we studied TRL decay on semiconductor layers and thin film homostructures after a pulsed excitation by simulation with Synopsys TCAD® and by mathematical approximation. Again, our working example is Cu(In,Ga)Se2. We investigate the influence of the excitation pulse length, axial diffusion, bulk-defects, and defects at the contacts, as well as space charge on the TRL-decay separately by quasi one-dimensional simulations of semiconductor layers and semiconductor homostructures. Material parameters like defect density, carrier mobility, and surface recombination velocity are varied in a wide range, such that the calculations are applicable to other semiconductors. We further study the influence of multi-pulse excitation. We show how material parameters such as carrier lifetime and carrier mobility can be extracted from the TRL transients and how the samples can be characterized by excitation dependent measurements in the open circuit case. We can explain some effects found in luminescence experiments, like an increased decay in semiconductor junctions due to the electric field in the space charge region. However, we also discuss the effect of charge storage which may lead to decreased decay. It is revealed that under high injection conditions single layers within a semiconductor stack can be characterized in terms of carrier lifetime.

Time-resolved luminescence (TRL) is a non-destructive, non-invasive, and contactless characterization method. We studied TRL decay on semiconductor layers and thin film homostructures after a steady state illumination by simulation with Synopsys TCAD® and by analytical approximate solution of the appropriate equations. First, we show that the luminescence decay time in general equals the minority carrier lifetime only for a homogeneous and time-independent carrier lifetime. Then, we investigate the influence of photon recycling, injection level, charge carrier diffusion, defects in the bulk and at the surfaces, as well as space charge on the TRL decay separately by quasi one-dimensional simulations of semiconductor layers and semiconductor homostructures. We further study the influence of sample non-homogeneity as may be found in polycrystalline semiconductors. We show how carrier lifetime can be extracted from the TRL transients and how the samples can be characterized by excitation dependent measurements in the open circuit case. We can explain some effects found in luminescence experiments, like a decrease of the decay time with an increasing excitation, a maximum in the decay time due to saturated bulk-defects and curved luminescence transients due to high injections or sample inhomogeneities. Furthermore, we are focussing on the question, how single layers within a semiconductor stack can be characterized.

No abstract

The main objective of time-resolved photoluminescence (TRPL) is to characterize minority carrier recombination in semiconductors. However, trap states in the band gap can lead to artificially long decay times thus distorting the measured minority carrier lifetime. In this work, we propose to measure TRPL under elevated temperature and excitation in order to reduce minority carrier trapping. Taking three Cu(In,Ga)Se2 layers as examples, we show that the decay time decreases with increasing temperature—in accordance with simulations. Under increasing excitation, the decay time can become smaller due to trap saturation but also can become larger due to asymmetric hole and electron lifetimes. By comparison of simulation and experiment, we can find the energy, the density, and the electron capture cross-section of the trap which in the present example of Cu(In,Ga)Se2 films gives values of ∼200 meV, ∼1015 cm−3, and ∼10−13 cm2, respectively.

We investigate light-induced degradation of Cu(In,Ga)Se2 (CIGSe) layers by means of time-resolved photoluminescence (TRPL) measurements. Illumination in the range of minutes with 1 sun white light equivalent leads to a strong reduction of the carrier lifetime as determined by TRPL. Ambient storage in the dark, however, does not cause degradation. X-ray photoelectron spectroscopy of the CIGSe surface reveals a light induced enhancement of Na 1s and O 1s core-level emission. The position of the O 1s peak at 531.6 eV is related to a Na-O-CIGSe bonding complex. The light-induced degradation of the CIGSe layer finally translates into inferior open circuit voltages due to the dominance of interface recombination in completed solar cell devices. This study has implications for laboratory research and may need to be regarded in CIGSe module production.