Ryuichi Shimokawa

Japanese Indoor Calibration Method for the Reference Solar Cell and Comparison with the Outdoor Calibration

Details of the established Japanese indoor calibration method for solar cells were introduced and various kinds of reference cells were calibrated using the indoor and outdoor methods for comparisons. It was found that both calibrated values were very consistent and nearly equal for crystalline Si cells, while the total irradiance detected with an absolute spectro-radiometer in the indoor calibration was about 2% higher than that detected with a pyranometer EKO MS-801 confirming to the current World Radiometric Reference. Also, the calibration errors due to a deviation from the cosine response for a low incident angle were found to be about 2% for an inert gas-sealed crystalline Si cell and about 1% for unpackaged and module-packaged crystalline Si cells.

Effect of atmospheric parameters on solar cell performance under global irradiance

Abstract The effect of incident sun angle and atmospheric conditions on solar-cell performance under global irradiance is investigated by theoretical analysis for reference-cell constructions of four kinds: module-packaged crystalline Si, module-packaged amorphous Si, unpackaged crystalline Si and inert gas-sealed crystalline Si. The dependence of solar-cell response on the incident beam angle (deviation from the cosine response) is quantitatively confirmed by calculations and measurements. The effect of deviation from the cosine response on cell sensitivity under global horizontal irradiance is investigated. The effect of incident sun angle and atmospheric conditions on solar-cell performance under global irradiance is then quantified.

Epitaxial Vapor Growth of Gallium Antimonide

Epitaxial vapor growth of gallium antimonide in the closed tube process is studied using polycrystal gallium antimonide and hydrogen chloride as the source crystal and the transport agent respectively. The transport rate is maximum when the hydrogen chloride concentration initially charged in the reactor is about 10-7 mole per unit volume. The surface morphology of the grown layer is strongly dependent on the pretreatment of substrate surface: The grown layer is mirror-smooth when the substrate surface is vapor-etched in situ for about 30 minutes preceding to the deposition. The electrical properties of the grown layers are studied by using the Pauws method. The hole concentration in the grown layer obtained in the routine experiment is comparable to or smaller than the lowest value ever reported (3~4×1016 holes/cm3), and the hole mobility ranges from 700 to 800 cm2/Vsec without any indication of compensation.

Epitaxial vapor growth of gallium antimonide

Abstract The epitaxial vapor growth of gallium antimonide in the conventional closed tube process was studied by using polycrystalline gallium antimonide as a source material. The transport agents used were hydrogen chloride, antimony penta-chloride and iodine. The growth rates were found to range from 0.5 to 10 μm/ hr when the initially charged amount of the transport agent was from 10 −8 to 10 −6 mole/cm 3 . The electrical properties of the grown layers were studied by the Van der Pauw method 1 ). The minimum of the hole concentration in the grown layer was 1.5−2 X 10 16 /cm 3 and the mobility was larger than 700 cm 2 /V sec. The p-n junction diodes were also prepared by using the same vapor growth technique. They had good rectification characteristics. The surface morphology of the grown layer was strongly dependent on the pretreatment of substrate surface: The surface of the grown layer was very smooth when the substrate surface was vapor-etched for about 30 min prior to the deposition.

Photon Collection Enhancement by White Rear Cover Reflection and the Design of Reference Cells for Module Performance Measurement

The effect of rear cover sheet reflection in solar modules on cell performance was investigated by shading and LBIC techniques. It was found that the photon collection enhancement was caused by the light reflection on the white rear cover sheet (white Tedlar) and 12–13% of the incident photons on the white rear cover sheet were guided to the cell by total internal reflection. Also, it was confirmed from the spectral response measurement that a single cell package without surrounding cells and white rear cover sheet could be used as a reference cell for the evaluation of solar module performance.

Possible Errors due to Deviation from the Cosine Response in the Reference Cell Calibration under Global Irradiance

The dependence of solar cell response on the incident beam angle was quantitatively examined using a monochromatic light beam with a wavelength range of 350 to 1150 nm. The results obtained for module-packaged, unpackaged and inert gas-sealed crystalline Si reference cells were compared with theoretical calculations. Next, the effect of incident sun angle on solar cell performance was theoretically investigated under global horizontal irradiance. The possible errors due to deviation from the cosine response were calculated in the reference cell calibration under global irradiance.

Applicability of the reference cell method to polycrystalline silicon cells with the spectral response shift under illumination

Abstract In current practice, the reference cell method is used to evaluate the photovoltaic performance of single-crystal cells. The applicability of this method to polycrystalline silicon cells with non-linear response to illumination intensity was studied. First, spectral response of EFG (edge-defined film fed growth) and CAST (casting, SISLO) silicon cells were measured as a function of the bias light intensity (1 mW cm−2 – 100 mW cm−2) and the bias light spectral distribution; based on these data, the non-linearity of the short-circuit current with respect to the illumination intensity was estimated. Next, the measurement error in the reference cell method due to spectral mismatch between reference cell and test cell was calculated for various spectral irradiances.

Global radiation model and angular distribution of the diffuse irradiance

Abstract The horizontal global radiation computer codes for the evaluation of solar cell performance under various atmospheric conditions were established. The numerical calculation method of the radiation transfer equation, including the doubling and adding method, was employed to avoid enormous computing capacities. Extensive horizontal global radiation data sets including the spectral energy distributions and angular distribution were generated. The spectral irradiances are presented graphically, and compared with experimental results and with other previously published theoretical results. We originally generated angular radiation data sets of the horizontal diffuse irradiance under various atmospheric conditions and found that it had a peak at the zenith angle of about 50° even if the turbidity and solar elevation change widely.

Applicability of the reference cell method to the performance measurement of amorphous silicon solar cells with a spectral response shift

Abstract The applicability of the reference cell method to newly developed amorphous silicon cells, which have a spectral response shift due to bias light or bias voltage, was investigated. Spectral response models, which can effectively represent the relative spectral responses of amorphous silicon solar cells with various cell configurations, qualities and bias effects, were developed on the basis of measured data. The variation in cell sensitivities under various atmospheric conditions was examined, and the error in short-circuit current measurements due to spectral mismatch was analysed.

Bias Light Wavelength Dependence of Minority Carrier Diffusion Length of EFG Silicon Ribbon Solar Cells

Local spectral photoresponse of EFG ribbon solar cells was measured as a function of bias light wavelength and intensity. The bias light wavelength was found to have small influence on the spectral response in local positions with longer (89 µm) and shorter (7 µm) diffusion length, but large influence on that in positions with medium diffusion length (20 µm). The effective minority carrier diffusion length estimated from the spectral response measurement was found to increase semilogarithmically with the intensity at various bias light wavelengths. The effect of bias light wavelength on diffusion length is explained by the penetration depth of bias light into solar cells.