We developed a method of growing thin Si film at $600^{\circ}C$ by hot wire CVD using a very thin large-grained poly-Si seed layer for thin-film Si solar cells. The seed layer was prepared by crystallizing an amorphous Si film by vapor-induced crystallization using $AlCl_3$ vapor. The average grain size of the p-type epitaxial Si layer was about $20{\mu}m$ and crystallographic defects in the epitaxial layer were mainly low-angle grain boundaries and coincident-site lattice boundaries, which are special boundaries with less electrical activity. Moreover, with a decreasing in-situ boron doping time, the mis-orientation angle between grain boundaries and in-grain defects in epitaxial Si decreased. Due to fewer defects, the epitaxial Si film was high quality evidenced from Raman and TEM analysis. The highest mobility of $360cm^2/V{\cdot}s$ was achieved by decreasing the in-situ boron doping time. The performance of our preliminary thin-film solar cells with a single-side HIT structure and $CoSi_2$ back contact was poor. However, the result showed that the epitaxial Si film has considerable potential for improved performance with a reduced boron doping concentration.

ZnSe/$Zn_3P_2$ heterojunctions with a substrate configuration were fabricated using a series of cost-effective processes. Thin films of ZnTe and $Zn_3P_2$ were successively grown by close-spaced sublimation onto Mo-coated glass substrates. ZnSe layers thinner than 100nm were formed by annealing the $Zn_3P_2$ films in selenium vapor. Surface selenization generated a high density of micro-cracks which, along with voids, provided shunt paths and severely deteriorated the diode characteristics. Annealing the $Zn_3P_2$ film at $300^{\circ}C$ in a $ZnCl_2$ atmosphere before surface selenization produced a dense microstructure and prevented micro-crack generation. The mechanism of micro-crack generation by the selenization was described and the suppression effect of $ZnCl_2$ treatment on the micro-crack generation was explained. ZnSe/$Zn_3P_2$ heterojunctions with low leakage current ($J_0$ < $1{\mu}A/cm^2$) were obtained using an optimized surface selenization process with $ZnCl_2$ treatment. However, the series resistance was very high due to the presence of an electrical barrier between the ZnTe and $Zn_3P_2$ layers.

In this paper, indium reduced materials for transparent conductive electrodes (TCE) were fabricated and their physical properties were evaluated. Two of materials, indium-zinc-tin oxide (IZTO) and aluminum (Al) were selected as TCE materials. In case of IZTO nanoparticles, composition ratios of In, Zn and Sn is 8:1:1 were synthesized. Size of the synthesized IZTO nanoparticles were less than 10 nm, and specific surface areas were about $90m^2/g$ indicating particle sizes are very fine. Also, the IZTO nanoparticles were well crystallized with (222) preferred orientation despite it was synthesized at the lowered temperature of $300^{\circ}C$. Composition ratios of In, Zn and Sn were very uniform in accordance with those as designed. Meanwhile, Al was deposited onto glass by sputtering in a vacuum chamber for mesh architecture. The Al was well deposited onto the glass, and no pore was observed from the Al surface. The sheet resistance of Al on glass was about $0.3{\Omega}/{\square}$ with small deviation of $0.025{\Omega}/{\square}$, and adhesion was good on the glass substrate since no pelt-off part of Al was observed by tape test. If the Al mesh is combined with ink coated layer which is consistent of IZTO nanoparticles, it is expected that the good and reliable metal mesh architecture for TCE will be formed.

Dye-sensitized solar cells (DSSCs), developed two decades ago, are considered to be an attractive technology among various photovoltaic devices because of their low cost, accessible dye chemistry, ease of fabrication, high power conversion efficiency, and environmentally friendly nature. A typical DSSCs consists of a dye-coated $TiO_2$ photoanode, a redox electrolyte, and a platinum (Pt)-coated fluorine-doped tin oxide (FTO) counter electrode. Among them, redox electrolytes have proven to be extremely important in improving the performance of DSSCs. Due to many drawbacks of iodide electrolytes, many research groups have paid more attention to seeking other alternative electrolyte systems. With regard to this, one-electron outer sphere redox shuttles based on cobalt complexes have shown promising results: In 2014, porphyrin dye (SM315) with the cobalt (II/III) redox couple exhibited a power conversion efficiency of 13% in DSSCs. In this review, we will provide an overview and perspectives of cobalt redox electrolytes in DSSCs.

High-efficiency in $Cu(In,Ga)Se_2$ (CIGS) solar cells were usually achieved on soda-lime glass substrates due to Na incorporation that reduces deep-level defects. However, this supply of sodium from sodalime glass to CIGS through Mo back electrode could be limited at low deposition temperature. Na content could be more precisely controlled by supplying Na from known amount of an outside source. For the purpose, an $Na_2S$ layer was deposited on Mo electrode prior to CIGS film deposition and supplied to CIGS during CIGS film. With the $Na_2S$ underlayer a more uniform component distribution was possible at $350^{\circ}C$ and efficiency was improved compared to the cell without $Na_2S$ layer. With more precise control of bulk and surface component profile, CIGS film can be deposited at low temperature and could be useful for flexible CIGS solar cells.

Formation mechanism of residual carbon layer, frequently observed in the $CuInSe_2$ (CIS) thin film prepared by direct solution coating routes, was investigated in order to find a way to eliminate it. As a model system, a methanol solution with dissolved Cu and In salts, whose viscosity was adjusted by adding ethylcellulose (EC), was chosen. It was found that a double layer, a top metal ion-derived film and bottom EC-derived layer, formed during an air drying step presumably due to different solubility between metal salts and EC in methanol. Consequently, the top metal ion-derived film acts as a barrier layer inhibiting further thermal decomposition of underlying EC, resulting a formation of bottom carbon residue layer.