Xinmin Cao

Simulation of a-Si/a-SiGe thin film tandem junction solar cells

Amorphous silicon (a-Si) based thin film tandem junction solar cells are simulated based on a uniform field collection model. From the photovoltaic parameters of a single junction a-Si top cell and a few amorphous silicon–germanium (a-SiGe) bottom cells, the optimized a-Si/a-SiGe tandem cell can be predicted. The simulation results are in good agreement with the experiment. The highest efficiency a-Si/a-SiGe tandem cells are obtained with a combination of a-SiGe characteristics and a relatively large mismatch in the short circuit current between the top and bottom cells. A key reason for this behaviour is that the tandem cell may exhibit a larger fill factor than either one of the component cells under a certain current mismatch.

Fabrication and Characterization of Triple-junction Amorphous Silicon Based Solar Cell with Nanocrystalline Silicon Bottom Cell

Recent research activities and results on the fabrication and characterization of high-efficiency triple-junction hydrogenated amorphous silicon (a-Si:H) based solar cells with hydrogenated nanocrystalline silicon (nc-Si:H) bottom cells at the University of Toledo (UT) are briefly summarized and reported in this paper. Using VHF PECVD technique, new deposition regimes have been developed in UT multi-chamber load-locked PECVD deposition system for the preparation of high quality a-Si:H, a-SiGe:H and nc-Si:H i-layers at deposition rates in the range of 2-15 A/s. Incorporating various improvements in device fabrication and characterization, 7.8% initial and 7.4% stable active-area (0.25 cm2) cell efficiencies have been achieved for VHF nc-Si n-i-p single-junction solar cells. Initial efficiency of 11.0% for a-Si/nc-Si tandem-junction was obtained. 12.4% initial and 11.0% stable cell efficiencies for a-Si/a-SiGe/nc-Si triple-junction solar cells have also been achieved. We also report 7.2% initial efficiency for single-junction nc-Si:H cells having nc-Si:H i-layer deposited at high rate using RF PECVD at a high pressure of 8 Torr

Analysis and optimization of thin film photovoltaic materials and device fabrication by real time spectroscopic ellipsometry

Methods of spectroscopic ellipsometry (SE) have been applied to investigate the growth and properties of the material components used in the three major thin film photovoltaics technologies: (1) hydrogenated silicon (Si:H); (2) cadmium telluride (CdTe); and (3) copper indium-gallium diselenide (CuIn1-xGaxSe2 or CIGS). In Si:H technology, real time SE (RTSE) has been applied to establish deposition phase diagrams that describe very high frequency (vhf) plasmaenhanced chemical vapor deposition (PECVD) processes for hydrogenated silicon (Si:H) and silicon-germanium alloy (Si1-xGex:H) thin films. This study has reaffirmed that the highest efficiencies for a-Si:H and a-Si1-xGex:H component solar cells of multijunction devices are obtained when the i-layers are prepared under maximal H2 dilution conditions. In CdTe technology, the magnetron sputter deposition of polycrystalline CdTe, CdS, and CdTe1-xSx thin films as well as the formation of CdS/CdTe and CdTe/CdS heterojunctions has been studied. The nucleation and growth behaviors of CdTe and CdS show strong variations with deposition temperature, and this influences the ultimate grain size. The dielectric functions ε of the CdTe1-xSx alloys have been deduced in order to set up a database for real time investigation of inter-diffusion at the CdS/CdTe and CdTe/CdS interfaces. In CIGS technology, strong variations in ε of the Mo back contact during sputter deposition have been observed, and these results have been understood applying a Drude relaxation time that varies with the Mo film thickness. Ex-situ SE measurements of a novel In2S3 window layer have shown critical point structures at 2.77±0.08 eV, 4.92±0.005 eV, and 5.64±0.005 eV, as well as an absorption tail with an onset near 1.9 eV. Simulations of solar cell performance comparing In2S3 and the conventional CdS have revealed similar quantum efficiencies, suggesting the possibility of a Cd-free window layer in CIGS technology.

Spectroscopic aspects of front transparent conductive films for a-Si thin film solar cells

This work demonstrates a method to optimize the indium tin oxide (ITO) thin films as front transparent electrode to maximize the efficiency of substrate type amorphous silicon (a-Si) based thin film solar cells. It shows that the total light intensity absorbed by the a-Si layer can be predicted by combining a multilayer optical simulation with the nonuniform solar spectrum and the spectroscopic response of the absorption coefficient of the a-Si film. Consequently, an optimized ITO film can be identified. The photovoltaic performances of experimentally obtained a-Si single junction solar cells confirm the simulation results, indicating an ITO film about 56 nm thick leads to the highest efficiency. Furthermore, it is shown that the ITO films should be deposited at relatively low temperature around 132 °C to avoid damage to the a-Si top p-layer and p-i-n junction. It is found that introducing a small fraction, ∼0.61% flow ratio, of O2 in the sputtering Ar gas reduces the sheet resistivity of the ITO film and...

Amorphous silicon germanium solar cells deposited on stainless steel at elevated pressure

This work reports our efforts on improving the deposition rate for amorphous silicon germanium (a-SiGe) absorber layer by increasing the PECVD process pressure. We demonstrate that at an elevated pressure of 1∼4 Torr, the deposition rate reaches 3.5∼4 Å/sec, which is about 4 times higher than previous low pressure processes (0.3∼0.6 Torr). Deposited at such a high rate, the single junction a- SiGe solar cells exhibit an efficiency comparable to that achieved at low pressure (12.5% initial, 10.4% stabilized). Furthermore, tandem junction cells using the high rate deposited a-SiGe as bottom cell show an initial efficiency as high as 13.27% and a stabilized efficiency of 11.50% after light soaking.

Impacts of an intrinsic a-Si buffer layer between the p-type nc-Si layer and the intrinsic a-SiGe layer in single junction solar cells

This paper reports numerical modeling and experimental investigation for the impacts of an intrinsic a-Si buffer layer between the p-type nc-Si layer and the intrinsic a-SiGe layer with a narrow bandgap of 1.40–1.55eV on the performances of a-SiGe single junction solar cells. The effects of bandgap and thickness of the buffer layer were simulated by using Analysis of Microelectronic and Photonic Structures (AMPS) computer model developed at Penn State University. The results obtained by the simulation show that the intrinsic a-Si buffer layer can lead to an increase in the open circuit voltage (Voc), but cause a decrease of the fill factor (FF) and the conversion efficiency (Eff), depending on how large the band gap and thickness of the buffer layer are. Our experimental results are consistent with the simulations results; i.e., a thick and wide-bandgap buffer layer between i and p layers can cause a serious deterioration in FF. An optimal a-SiGe single junction solar cell without the a-Si buffer layer has achieved an efficiency of 9.41% with Voc=0.576V, Jsc=23.34 mA/cm2, and FF=70.0%.

Numerical simulation and experimental investigation of a-Si/a-SiGe tandem junction solar cells

This paper reports numerical modeling and experimental investigation of a-Si/a-SiGe tandem junction solar cells, based on a field-aided collection model. The performances of the composed a-Si/a-SiGe tandem cells could be predicted from the numerical modeling and experimental photovoltaic parameters of the component cells. An optimized choice for the tandem cells has been proposed by the aid of numerical simulation and verified by experimental results for the cases of a fixed a-Si top cell or a fixed a-SiGe bottom cell, respectively. It is found that 1) the highest efficiency (∼13%) of the tandem cells can be achieved with a certain mismatch in Jsc between the top and bottom cells; 2) a small amount of Ge can be incorporated into a-Si based top cells to obtain an optimized current match with the bottom cell; 3) the Jsc of the simulated tandem solar cells is always a little greater, rather than exactly equal to the limiting Jsc of the component cells.

High rate deposition of a-Si and a-SiGe solar cells near depletion condition

Amorphous silicon (a-Si) and amorphous silicon germanium (a-SiGe) absorber layers are deposited at high rates of 7∼8 Å/sec using RF plasma enhanced chemical vapor deposition. The single junction a-Si top and a-SiGe bottom cells deposited at such a high rate exhibit initial efficiencies of 10.06% and 9.96%, respectively, while the process is not yet fully optimized. A tandem junction cell made using the high rate deposited a-Si and a-SiGe shows an initial efficiency as high as 11.04%. A combination of proper RF power density, gas pressure, and H2 dilution enables the intrinsic layers being deposited near a depletion condition and is responsible for the promising performances.

Effects of hydrogen dilution grading in active layer on performance of nanocrystalline single junction bottom component and corresponding a-Si:H based triple junction solar cells

Using hydrogen dilution grading in nanocrystalline silicon (nc-Si:H) intrinsic layer, considerably high spectral response has been achieved in the longer wavelength (650nm-1000nm) of solar spectrum from single junction solar cells, to be used as bottom junction in monolithic triple junction a-Si:H based thin film solar cells. The intrinsic layers have been deposited with a high deposition rate of 8 Aring/s, using VHP PECVD deposition technique from Si2H6 diluted with H2. In the present work, using R (R=[H2]/[Si 2H6]) from an initial value 50 to different final values viz. 40, 37.5 and 35, single junction nc-Si:H cells have been fabricated and the same single junction components have been applied as bottom junction to fabricate a-Si:H/a-SiGe:H/nc-Si:H triple junction structure where the properties of top and middle junctions were kept fixed. The short circuit current of the triple junction cells is shown to be limited by the bottom junction and the fill factor has been improved in triple junction cells compared to the same obtained from the corresponding single junction cells used as nc-Si:H bottom junction. Comparison of the fill factors under red (>630nm) illumination and that under blue (<510nm) illumination for these single junction nc-Si:H cells indicated structural quality of p/i interface and bulk of the i-layer of nc-Si:H solar cells. A conversion efficiency of 11.2% has been achieved from triple junction solar cell. The light induced degradation studies are currently in process for triple junction cells under a 100 mW/cm2 white light illumination. Using spectroscopic ellipsometry, evaluation of crystalline volume fraction at different level of growth is undergoing, which will enable fine-tuning of grading in hydrogen dilution

Impacts of n-type interface on the performances of a-Si based solar cells

This paper reports experimental results and numerical simulations for the impact of n-type interface on the performance of n-i-p type a-Si based solar cells deposited on stainless steel (SS) foil substrate. Here the n-type interface includes the interfaces of n-a-Si layer with i-a-Si layer and SS substrate. The obtained results show that 1) the extra interfacial layer between SS and the n-a-Si layer will cause a rollover behavior of the light I-V curve, and H2 plasma pretreatment on the SS surface may eliminate the extra layer and significantly improve the cell performance; 2) the contamination of n-dopants for the sequential growth of i-layer will seriously deteriorate the cell performances and cause a crossover of the dark and light I–V curves when the n-i-p a-Si solar cells were deposited in a single chamber mode; and 3) the i-a-Si buffer layer between i-a-SiGe and doped a-Si layers could improve the cell performance, but if the thickness of the buffer layer was beyond a critical value, it may cause a bending of the light I–V curves.