CuGaSe2 Layers Research Articles

Dependence of Crystal‐Field Energy on Strain/Stress Sensed by Temperature Variation of Chalcopyrite Semiconductor (Optical) Band‐Gap for Efficient Band‐Gap Tuning in the CIS/CIGS Photovoltaic

Chalcopyrite selenide single crystals and epitaxial layers (CuIn1−xGaxSe2, x = 0.00, 0.08, 0.19, 1.00) were characterized by temperature‐dependent photoreflectance (PR), photoluminescence (PL), photoluminescence–excitation (PLE), and variable excitation‐energy photoluminescence (VEPL) spectroscopy. The transition energies Ea, Eb, and Ec of both CuInSe2 (CIS) and CuGaSe2 (CGS) layers sensed by PR were higher than the energies of single crystals. CuInSe2 and CuGaSe2 grown on GaAs(001) underlie compressive and tensile stresses, respectively, which lead to band‐gap broadening in CIS and band‐gap narrowing in CGS. The increase of the Ea, Eb, Ec energies of tensely stressed CuGaSe2 layers to energies higher than those of the bulk originates from the stress dependence of the non‐cubic crystal field. Band‐gap scanning of the CuGaSe2 layer with continuous‐wave Ti:sapphire‐laser confirmed the absence of correlation between band‐gap readjustment and intrinsic defects. The energy of the band‐edge exciton EFE, in the PL‐spectra, was lower than the Ea transition energy, in the PR‐spectra, which is assigned to partial quenching of ΔCF with the increase of external tensile stress by gallium‐segregation at the chalcopyrite/GaAs‐interface. The stress dependence of ΔCF is negligible in CuInSe2 and linear, with a rate of 9 meV/100 MPa, in CuGaSe2. It is revealed that the energy band‐gap of photovoltaic chalcopyrite absorbers can be tuned by simultaneous built‐in and external lattice‐tuning.

Over 11% Efficient CuGaSe2 Solar Cells Without Using KCN Treatment

CuGaSe2 (CGS) has a bandgap energy of 1.68 eV and is theoretically very suitable to be in tandem with silicon or Cu(In,Ga)Se2 solar cells. However, due to the high quantity of surface defects, high‐performance CGS usually relies on a KCN surface treatment, which is a high‐toxic process and restricts its further development. Herein, by reducing the exposure time to air of the CGS layer as far as possible and developing a new absorber growth procedure to reduce the surface Cu x Se generation, grain size is successfully increased and interface recombination is reduced without the use of KCN. Combined with a simple annealing process, the defect concentration is successfully decreased and the depletion width is broadened dramatically and the power conversion efficiency is promoted to 11.05% without any traditionally used KCN treatment. This work provides a nontoxic and low‐pollution way to fabricate high‐efficient CGS solar cells.

Improved CuGaSe2 absorber properties through a modified co-evaporation process

The present study deals with CuGaSe2-thin-films for solar cell applications. With the aim of achieving chemically and structurally homogeneous CuGaSe2 layers, it is proposed a modified co-evaporation process, which implies two so-called relaxation sequences during films growth. The resulting layers are characterized by scanning electron microscopy and Raman spectroscopy. By comparing their characteristics with those of CuGaSe2 grown without relaxation sequence, it is demonstrated that the modified process yields improved distribution of elements throughout the whole layer. The performance of the resulting solar cells is improved exclusively through increased quantum efficiency in the large wavelengths; further output voltage increase would require alternative junction partners.

Formation of Ga double grading in submicron Cu(In,Ga)Se2 solar cells by pre-depositing a CuGaSe2 layer

By introducing a pre-deposited CuGaSe2 layer, a steep Ga back grading has been formed in a submicron CIGS layer with a high-temperature process for the first time.

Over 100% magnetoresistance ratio at room temperature in magnetic tunnel junctions with CuGaSe2 spacer layer

We investigated the structure and magneto-transport properties of magnetic tunnel junctions (MTJs) consisting of half-metallic Co2FeGa0.5Ge0.5 ferromagnetic electrodes and a semiconductive CuGaSe2 interlayer. The magnetoresistance (MR) ratio reaches more than 100% at room temperature and nearly 250% at 30 K with the resistance area product (RA) smaller than 1 Ω μm2. Microstructural analysis clarified that Co and Fe atoms diffuse into the CuGaSe2 layer, which can be a possible reason for the scattering of RA values but poorly affect the MR ratio itself. Our findings stress the potential of semiconductor barriers in MTJs for spintronics application that requires low RA.

Effect of sodium diffusion on the properties of CIGS solar absorbers prepared using elemental Se in a two-step process

The influence of Na diffusion from various glass substrates during a high-temperature slenization process on the microstructure and morphology of two-step formed CIGS absorber layers is investigated. In order to minimise the CIGS absorber formation time, elemental Se vapour is used to prepare the CIGS absorber. The grain sizes of the CIGS films are found to increase with increasing sodium in the glass substrates (extra clear glass, soda-lime glass, borosilicate glass). TiN and SiN thin films are used as diffusion barrier layers inserted between the glass substrate and the Mo rear conatct to tune the Na diffusion from the soda-lime glass. The interdiffusion between the In-rich CuInSe2 surface layer and the Ga-rich CuGaSe2 layer is promoted by the barrier layer, leading to larger CIGS grains. Efforts are also taken to understand the differences in Na diffusion (from the glass substrates) and their effects on the MoSe2 intermediate layer formation during the high-temperature CIGS absorber formation processes. We find that excess amounts of Na and Se are essential for the MoSe2 growth. The excessive Na in the form of Na2Sex at the CIGS/Mo interface works as a Se source and catalyses the MoSe2 formation. The Se flow in the two-step CIGS formation process must be sufficiently high to obtain high-efficiency CIGS solar cells.

Probing diffusion of In and Ga in CuInSe2/CuGaSe2 bilayer thin films by x-ray diffraction

The bilayer thin films of CuInSe2 (CIS) and CuGaSe2 (CGS) are fabricated by a molecular beam deposition (MBD) technique by sequential depositions of CGS followed by CIS and vise versa on Mo-coated soda-lime glass (SLG) and Mo/Al2O3-coated (Na blocking) SLG substrates. The thicknesses of CIS/CGS layers are adjusted to have the overall x=[Ga]/([In]+[Ga]) at approximately 0.4 similar to what is generally used in CIGS thin film solar cells. The growth temperature and the Cu-ratio y=[Cu]/([In]+[Ga]) of the CIS and CGS layers are systematically varied for Cu-rich (y>1) and Cu-poor (y<1) conditions. X-ray diffraction (XRD) technique is used to analyze the shift of diffraction peaks of the preferred orientations of the bilayers compared to those of the single-layer CIS, CGS and CIGS thin films. The XRD results show that higher Ga diffusion is observed in the Cu-rich CIS/CGS bilayer rather than others. The results suggest that the diffusion of Ga is enhanced by the excess Cu–Se phase that is dependent upon the substrate temperature. The Na from the SLG substrate is one of the important factors affecting Ga diffusion. The results show alloying CIGS patterns rather than separated CIS and CGS patterns as observed in the bilayers with Na.

Study of single crystal CuInSe2 thin films and CuGaSe2/CuInSe2 single quantum well grown by molecular beam epitaxy

High quality CuGaSe2 and CuInSe2 single crystalline layers are grown on GaAs (001) by employing the deposition sequence of migration enhanced epitaxy using a solid source molecular beam epitaxy system. When CuGaSe2 is grown on CuInSe2 at moderate temperatures, severe interdiffusion takes place at the heterojunction of CuGaSe2/CuInSe2. This problem has been solved by optimizing the growth temperature and deposition rates of the constituent elements. Thus, we have successfully grown CuGaSe2/CuInSe2 single quantum well with sharp interfaces on GaAs (001) for the first time. Intense photoluminescence from the single quantum well with 10nm well width is demonstrated.

Verification of phototransistor model for Cu(In,Ga)Se2 solar cells

Previous studies of Cu(In,Ga)Se2 thin film solar cells showed that the long-term stability critically depends on the bias across the junction. As a result of a dark anneal the current–voltage (IV)-characteristics in the dark showed a blocking behavior with increasing anneal time. In the final stage the device exhibits an open circuit voltage (Voc) which is independent from the illumination intensity, a crossover of the dark and illuminated IV-characteristics and Voc saturation for decreasing temperatures. These characteristics also occur in the initial state prior to the endurance test, however, at low temperature (<200K) measurements. We suggested a phototransistor model to explain the observed characteristics. The prerequisite of this model is the existence of a Schottky barrier at the back contact. In this contribution more insights into this phototransistor model and its experimental verification will be given and discussed. Finally we suggest how to avoid the effects of the back barrier with the help of a CuGaSe2 layer at the back of the absorber and a Ga gradient through the absorber. These measures will be verified with simulations and compared to measurements on co-evaporated devices.

Structural analysis of Cu(In,Ga)Se2 films fabricated by using sputtering and post-selenization

We made Cu(In,Ga)Se2 (CIGS) films by using sputtering and post-selenization. First, we deposited stacked metallic films by using the sputtering method and then carried out post-selenization for the precursor stacked metals with different amounts of Se powder, 0.160 g, 0.321 g, 0.642 g, and 0.964 g. We found that with a small amount of Se, separated CuInSe2 and CuGaSe2 layers were formed, which was confirmed by X-ray diffraction (XRD), Raman spectroscopy, and energy-dispersive X-ray analysis. For larger Se amounts, the CIGS phase was observed in the results of XRD and Raman spectroscopy. These results indicate that the amount or the partial pressure of Se plays an important role in the reaction kinetics for stacked precursor metals to form the CIGS phase.

Characteristics of CuGaSe2 layers grown on GaAs substrates

CuGaSe2 films are grown on GaAs (001) substrates by migration-enhanced epitaxy. Electron probe X-ray micro analysis measurement shows that the CuGaSe2 films are stoichiometric regardless of the Cu/Ga supply ratio, which indicates that migration-enhanced epitaxy is useful for making stoichiometric CuGaSe2 films. Hall effect measurement indicates the p-type conductivity with the hole mobility as high as 2700cm2/Vs at low temperatures. The photoluminescence spectrum at 4K is dominated by the band-edge emission which includes both A-band free and bound exciton emissions. The emission due to donor-acceptor pairs and other deep levels are observed very weakly. Moreover, at temperatures above 180K, a new emission band appears in the higher energy region of the band-edge emissions. The peak is tentatively attributed to the emission including the B-band free exciton. These results indicate that the optical properties of the CuGaSe2 films grown by using migration-enhanced epitaxy are fairly good.

Growth of CuGaSe2 Layers on Closely Lattice-Matched GaAs Substrates by Migration-Enhanced Epitaxy

CuGaSe2 single-crystal films are grown on the As-stabilized (2×4) surface of (001) GaAs by migration-enhanced epitaxy (MEE), where Cu+Ga and Se are alternately deposited. The growth process is monitored by refraction high-energy electron diffraction (RHEED) in the [110] azimuth. Under the Cu-enriched growth condition, a deformed 4-fold pattern is observed in both Cu+Ga and Se deposition periods. The deformed 4-fold pattern is found to be related to the segregation of Cu2Se on the CuGaSe2 surface as confirmed by the results of X-ray diffraction (XRD) measurement. By reducing the beam equivalent pressure of Cu (Cu-BEP), clear 4-fold patterns appear in both Cu+Ga and Se deposition periods instead of deformed 4-fold patterns. Further reduction of Cu-BEP results in clear 4- and 2-fold patterns for Cu+Ga and Se deposition periods. Under these growth conditions, Cu2Se-segregation-free CGS growth is achieved. Thus, the CuGaSe2 single-crystal layers without Cu2Se-segregation are successfully grown on GaAs(001) substrates by optimizing the Cu-BEP.

An analysis of temperature-dependent photocurrent-spectra in photoconductive CuGaSe2 layers

The photocurrent (PC) variation in photoconductive CuGaSe2 (CGS) layers had been investigated as a function of temperature. Three peaks A, B, and C of the PC spectra were associated with the band-to-band transitions. Thus, the parameters of the crystal-field splitting (Δcr) and spin-orbit splitting (Δso) were directly acquired through the PC measurement. The Δcr and Δso were 0.0903 eV and 0.2130 eV at 10 K, respectively. From the relations of peak position and temperature, the temperature dependence of the band-gap energy is well described by Eg(T)=Eg(0)−(8.63×10−4)T2/(336+T). Also, the Eg(0) is estimated to be 1.7952, 1.8855, and 2.0985 eV at the valence-band states of Γ7(A), Γ6(B), and Γ7(C), respectively. However, the behavior of the PC was different from that generally observed in other semiconductors, the PC intensities decreased with decreasing temperature. From the relation of log Jph versus 1/T, where Jph is the PC density, two dominant levels were observed, one at high temperature and the other at low temperature. By comparing the results of the PC and photoluminescence, we confirmed that the activation energy of 82.7 meV at high temperatures is related to the dissociation energy of donor level due to Se vacancies. Consequently, we suggest that in photoconductive CGS layers, the trapping center due to native defects and impurities limits the PC signal with decreasing temperature.

Comparison between photoluminescence spectroscopy and photoreflectance spectroscopy in CuGaSe2 epilayer

Photoluminescence (PL) spectroscopy and photoreflectance (PR) spectroscopy are very useful techniques for studying the properties of materials. In this paper, the same material of Cu-rich metal-organic vapour phase epitaxy (MOVPE) grown CuGaSe2 layer is investigated in a temperature range from 20 to 300K to compare these two techniques. Both PL and PR spectra appear red shifted, less intense and broadened. The temperature dependence of interband transitions is studied by using the Manoogian-Leclerc equation. The values of the band gap energy at T = 0K and the effective phonon temperature are estimated. The temperature dependences of intensities and broadenings of PL and PR spectral lines are also analysed. Based on the results of the comparison, the features and applications of the PL and PR can be shown in detail.

A Neutral Barrier at CGS Grain Boundaries - Compositional and Structural Dependencies

AbstractUnlike other photoactive layers (e.g. Si) polycrystalline chalcopyrite layers achieve higher efficiency than monocrystalline ones. Although grain boundaries may play an important role for this phenomenon, there is currently no commonly accepted model for the electronic structure of grain boundaries. First experimental results on CuGaSe2 absorbers indicated a small neutral barrier for majority carriers (20-40meV) at sigma 3 grain boundaries, which is the predominant grain boundary in polycrystalline chalcopyrite absorbers [1]. These results are in discrepancy with theory, which predicts a 10 times higher barrier; we suspect that the copper excess might reduce the barrier height.Here we present a study using Hall-effect and Kelvin Probe Force Microscopy (KPFM) measurements to investigate the composition dependence of the barrier height of epitaxially grown CuGaSe2 layers containing a sigma 3 grain boundary as a function of the Cu/Ga ratio. First results show that the barrier height is independent of the copper content. In addition, we present initial results on sigma 9 grain boundaries which are observed to occur in high efficiency absorbers.

Diffusion of Cu, In, and Ga in In2Se3/CuGaSe2/SnO2 thin film photovoltaic structures

This article is a study of copper, indium, and gallium interdiffusions in In2Se3/CuGaSe2/SnO2/glass thin film heterostructures annealed at different temperatures. The use of CuGaSe2 material in place of Cu(In,Ga)Se2 is only required by the indium diffusion studies. The CuGaSe2 layers were grown by close-spaced vapor transport for two types of sources having different grain sizes. The In2Se3 films were deposited by thermal evaporation. The heterostructures were annealed in vacuum at different temperatures and analyzed by secondary ion mass spectroscopy (SIMS). The copper, indium, and gallium SIMS concentration profiles show that the copper diffuses up to the In2Se3 film surface and that the indium can diffuse far away from the In2Se3/CuGaSe2 interface towards SnO2. The copper, indium, and gallium diffusions were studied and the interdiffusion parameters were computed. The simultaneous interdiffusions of copper and indium induces the formation of a p–n junction responsible for the photovoltaic effect of the Zn/In2Se3/CuGaSe2/SnO2/glass photocells, the SnO2 side being lighted. This hypothesis is supported by results carried out from electron beam induced current measurements, showing a notable shift of the junction from the In2Se3/CuGaSe2 interface through the CuGaSe2 layer in terms of the annealing temperature, resulting in an increasing of the photovoltages up to 650 mV.

Composition dependent doping and transport properties of CuGaSe2

ABSTRACTChalcopyrites are doped by intrinsic defects, therefore their doping behavior depends on their composition. The doping and transport properties of epitaxial CuGaSe2 layers prepared under varying Cu excess have been investigated by temperature dependent Hall effect and conductivity measurements. Two acceptors, 134 meV and 80 meV deep, and a high degree of compensation, increasing with decreasing Cu excess, are found. The temperature dependence of the mobility indicates scattering with phonons, demonstrating high quality material. Defect scattering dominates at lower temperatures for CuGaSe2 grown under moderate Cu excess. CuGaSe2 grown under little or no Cu excess shows transport in a defect band at lower temperatures.

Optical properties of CuGaSe2 and CuAlSe2 layers epitaxially grown on Cu(In0.04Ga0.96)Se2 substrates

Photoluminescence (PL) study has been performed on epitaxial layers of CuGaSe2 and CuAlSe2 grown by metalorganic chemical vapor epitaxy on CuGa0.96In0.04Se2 substrates prepared by the traveling heater method. PL properties of epilayers are compared with each other for those grown on GaAs (100), CuGa0.96In0.04Se2 (100), CuGa0.96In0.04Se2 (112), and randomly oriented CuGa0.96In0.04Se2 substrates. PL results are discussed in terms of the lattice mismatches and stress in the epilayers.

Epitaxial Growth of ZnSe Films on CuGaSe2 Films

ZnSe epitaxial layers are grown on CuGaSe2 layers on GaAs(l00) substrates by MBE. The ZnSe/CuGaSe2/GaAs(100) structures are characterized by X-ray diffraction (XRD) and photoluminescence (PL) measurements and transmission electron microscopy (TEM). ZnSe layers are grown coherently on CuGaSe2 layers that were grown coherently on GaAs substrate. The properties of ZnSe layer are sensitive to the degree of lattice relaxation of underlying CuGaSe2 epitaxial layer. Cu-acceptor bound exciton emission are clearly seen in PL spectra from ZnSe, though Cu-related deep level emissions dominate the spectra.

Heteroepitaxy and characterization of CuGaSe2 layers grown by low-pressure metalorganic chemical-vapor deposition

CuGaSe2 chalcopyrite compounds were grown heteroepitaxially on both GaAs and GaP substrates by means of the low-pressure metalorganic chemical-vapor deposition method. Optical and structural properties were characterized comprehensively by photoreflectance (PR), photoluminescence (PL), x-ray diffraction, transmission electron microscopy, transmission electron diffraction, and electron-probe microanalysis. The CuGaSe2 epilayers had c(001) surface on GaAs(001) substrates and a(100) surface on GaP(001) substrates, respectively, the results being similar to the case of CuAlSe2. Energies of A, B, and C excitons associated with uppermost valence bands were determined from analysis of PR spectra, and the energies of good-quality epilayers are close to those of the bulk crystal. The slight increase of the crystal-field splitting in the valence bands were discussed in terms of the lattice strain in the epilayer caused by the lattice mismatch. Low-temperature PL spectra exhibited an intense peak at 1.71 eV, the energy being in good agreement with the A-exciton energy. A weak peak due to a free-to-acceptor transition was also observed at 1.66 eV. A broad PL peak at 1.76 eV was observed together with the intense band-edge PL at 1.67 eV, and the peak was assigned to relate to the B-exciton transition.