Chandra Goradia

Theory of the InP shallow homojunction solar cell

Abstract Based on a parametric variation study using a fairly comprehensive model, we have arrived at a near-optimum design of the shallow homojunction InP solar cell. Our results predict that using long lifetime material, an optimized front contact design, a two-layer antireflection coating and the other geometrical and material parameters of our near-optimum design, a beginning-of-life total area efficiency of 22% or slightly higher is realistically achievable under AM 0, 25 °C. Using a lifetime damage coefficient obtained by curve-fitting our model to measured illuminated current-voltage curves at beginning-of-life and at several 1 MeV electron fluences, on two InP cells, we predict a degradation of roughly 10% in efficiency after irradiation with a fluence of 1015 1 MeV electrons cm−2 for the optimally designed cell. The paper also attempts to explain the reasons for the superior radiation tolerance of the InP cell compared with that of silicon and GaAs solar cells.

Theory of high efficiency (Cd,Zn)S/CuInSe2 thin film solar cells

Abstract A theoretical model is presented which allows the fitting of calculated results to both dark and illuminated current versus voltage characteristics at several temperatures above 300 K for two relatively high efficiency Boeing (Cd,Zn)S/CuInSe2 thin film solar cells. The model is based on the assumptions that (1) the device is an n-(Cd,Zn)S/p-CuInSe2 heterojunction, (2) interface recombination is the dominant loss current mechanism both in the dark and under illumination and (3) the interface recombination centers are negatively charged and their density increases under illumination. The model naturally explains the large A factors observed (about 1.4 – 2.2), the temperature dependence of the current-voltage characteristics, the non-translation between the illuminated and dark current-voltage characteristics and the observed low open-circuit voltages. In addition, results of parametric variation studies are presented which show that a maximum open-circuit voltage of about 550 mV is possible in these devices if the effective dopings in the selenide and sulfide are greater than 5 × 10 15 cm −3 and greater than 5 × 10 16 cm −3 respectively, while the negatively charged interface recombination level density is below 6 × 10 10 cm −2 .

Key factors limiting the open circuit voltage of n/sup +/pp/sup +/ indium phosphide solar cells

The problem of the presently obtained best open-circuit voltage (V/sub oc/) values of n/sup +/pp/sup +/ InP solar cells being considerably smaller than those predicted by basic theory is theoretically investigated. The values of some key parameters in the cells are obtained with a computer model by closely matching theoretical and measured curves for illuminated I-V, log I/sub sc/-V/sub oc/, and spectral response characteristics of a high-efficiency (17.9%) InP solar cell made by the Spire Corporation. An optimally designed InP solar cell with nearly the best efficiency that it is capable of providing in the n/sup +/pp/sup +/ shallow homojunction structure is described and modeled using these parameters. The performance parameters of the optimally designed solar cell are calculated, and it is shown that such a cell is capable of having an efficiency of 22.61% under 1 AM0 at 27 degrees C (300 K). Using a baseline solar cell design which is a slightly modified and improved version of the Spire 6 cell, the key factors which limit the open-circuit voltage V/sub oc/ of such a solar cell are discussed.<<ETX>>

Effective first layer antireflective coating on InP solar cells grown by chemical oxidation

Commonly used first layer antireflection (AR) coatings for InP solar cells, such as ZnS, Sb/sub 2/O/sub 3/, SiO/sub 2/ and SiO, deposited either by electron-beam or by resistive evaporation, destroy the stoichiometry of the emitter surface. Consequently, the surface recombination velocity (SRV) at the emitter surface is significantly increased, leading to a reduction in the values of solar cell performance parameters. This can be prevented by growing, after contacting, a thin native oxide layer on the emitter surface. Best results are obtained using a phosphorus-rich chemical oxide grown by chemical oxidation using a newly developed etchant (PNP) based on HNO/sub 3/, o-H/sub 3/PO/sub 4/ and H/sub 2/O/sub 2/. The chemical oxide grown on p/sup +/-InP emitters, using the PNP etchant, passivates the surface and can be used as a first layer AR coating.<<ETX>>

High voltage thermally diffused p/sup +/n solar cells

The possibility of fabricating thermally diffused p/sup +/n InP solar-cells with high open-circuit voltage without sacrificing the short circuit current is discussed. The p/sup +/n InP junctions were formed by Cd and Zn diffusion through a 3-5 nm thick anodic or chemical phosphorus-rich oxide cap layer grown on n:InP:S (with N/sub D/-N/sub A/=3.5*10/sup 16/ and 4.5*10/sup 17/ cm/sup -3/) Czochralski LEC-grown substrates. After thinning the emitter from its initial thickness of 1 to 2.5 mu m down to 0.06-0.15 mu m, the maximum efficiency was found when the emitter was 0.2 to 0.3 mu m thick. Typical AM0, 25 degrees C values of 854-860 mV were achieved for V/sub oc/, J/sub sc/ values were from 25.9 to 29.1 mA/cm/sup 2/ using only the P-rich passivating layer left after the thinning process as an antireflection coating.<<ETX>>

High quality thermally diffused p/sup +/-n InP structures

Cd diffusion and Zn diffusion into n-InP:S (N/sub D/ =3.5*10/sup 16/ and 4.5*10/sup 17/ cm/sup -3/) were performed by a closed ampoule technique at diffusion temperatures from 500 to 600 degrees C by using either high-purity Cd and Zn or Cd/sub 3/P/sub 2/ and Zn/sub 3/P/sub 2/. The Czochralski LEC grown substrates with etch pit densities (EPDs) from 3*10/sup 4/ to 7*10/sup 4/ cm/sup -2/ were used. Diffusions were performed through either bare surfaces or using SiO/sub 2/ (50-100 AA thick) and phosphorus-rich anodic and chemical oxides (25-50 AA thick) as cap layers. Specular surfaces have been obtained after Cd diffusion from Cd/sub 3/P/sub 2/ through P-rich oxide cap layers with a very low surface dislocation density which goes through a minimum of 400-800 cm/sup -2/ at the diffusion temperature of 560 degrees C. AM0 250 degrees C V/sub oc/ values as high as 860 mV from solar cells made on these structures are reported.<<ETX>>

Predicted performance of near-optimally designed indium phosphide space solar cells at high intensities and temperatures

The authors calculated the expected performance dependence of near-optimally designed shallow homojunction n/sup +/pp/sup +/ InP solar cells on incident intensities up to 200 AMO and temperatures up to 100 degrees C (373 K). Both circular and rectangular cells were considered, the former for use in a Cassegrainian concentrator array at 100 AM0, 80-100 degrees C and the latter for use in a Slats concentrator array at 20 AM0, 80-100 degrees C. With efficiencies near 22% at 80 degrees C, both the circular and rectangular InP shallow homojunction solar cells compare very favorably to GaAs cells of the same design and may be preferable to the GaAs cells for space applications because of the superior radiation tolerance of the InP cells.<<ETX>>

Applicability of the Meyer–Neldel rule to solar cells

A comparison of data taken on high quality silicon, GaAs, and GaInAs solar cells with those taken on a variety of homojunction, heterojunction, and metal‐insulator‐semiconductor devices indicates that while the Meyer–Neldel rule may be applicable to certain types of solar cells it is not applicable to well‐behaved, diffusion‐controlled homojunction devices. It cannot be used, therefore, as a universal rule to predict maximum achievable solar cell voltages.

Experiment-based projected high efficiency thermally diffused p/sup +/n (Cd,S) InP solar cells for space applications

By drastically reducing the defect densities of p/sup +/n (Cd,S) InP diffused structures the authors have succeeded in fabricating p/sup +/n InP solar cells with measured AM0, 25/spl deg/C V/sub /spl prop// values exceeding 880 mV, without anti-reflection (AR) coating. Experiment-based projected maximum achievable AM0, 25/spl deg/C efficiency of these cells is 21.3%. Preliminary investigation of the performance parameters of p/sup +/n (Cd,S) InP structures and solar cells after irradiation with 10/sup 13/ cm/sup -2/ 3MeV protons indicate higher radiation tolerance of this configuration as compared to n/sup +/p InP configuration due to its better annealing properties.<<ETX>>

Defect density reduction of n/sup +/p and p/sup +/n InP structures fabricated by closed ampoule thermal diffusion

A significant reduction of defect densities of n/sup +/p and p/sup +/n InP structures fabricated by closed-ampoule thermal diffusion was obtained after optimizing the diffusion process. For n/sup +/p structures, the lowest etch pit density (EPD) of 6*10/sup 5/ cm/sup -2/ was achieved after S diffusion into InP:Cd (N/sub A/=1.2*10/sup 16/ cm/sup -3/) substrates using a thin In(PO/sub 3/)/sub 3/-rich anodic oxide diffusion cap layer at a diffusion temperature of 660 degrees C, while the lowest EPD after S diffusion into InP:Zn (N/sub A/ approximately 2*10/sup 16/ cm/sup -3/) under similar diffusion conditions was 8*10/sup 6/ cm/sup -2/. For p/sup +/n structures, surface EPD values as low as 4*10/sup 2/ cm/sup -2/ were achieved in the case of Cd diffusion into InP:S (N/sub D/=3.5*10/sup 16/ cm/sup -3/) substrates at a diffusion temperature of 560 degrees C using a thin In(PO/sub 3/)/sub 3/-rich chemical oxide diffusion cap layer, while the lowest EPD in the case of Zn diffusion was 3*10/sup 5/ cm/sup -2/. The differences are explained by the large number of In/sub 2/S/sub 3/, InS, and Zn/sub 3/P/sub 2/ surface and deep precipitates detected in the case of n/sup +/-p(S, Zn) and p/sup +/-n (Zn, S)InP structures.<<ETX>>