Abstract of M.Sc Thesis (Oshri Blank)-Materials Science and Engineering
M.Sc Thesis

M.Sc StudentOshri Blank
SubjectCu2O Thin Film Solar Cells
DepartmentDepartment of Materials Science and Engineering
Supervisor Professor Rothschild Avner
Full Thesis text - in Hebrew Full thesis text - Hebrew Version


One of the most promising renewable energy technologies is photovoltaics (PV), which is dominated by Si panels. The knowledge that has been collected makes it difficult to achieve further progress with Si alone. Tandem cells with two or more semiconductor absorbers could potentially boost the power conversion efficiency (PCE) beyond that of single absorber PV cells. Calculations suggest that the best match for Si is an absorber with bandgap energy of ~2 eV. This promise has motivated us to investigate Cu2O, an oxide semiconductor with bandgap energy of 2.1 eV, and to explore its photovoltaic properties. Toward this end, it was necessary to construct Cu2O PV cell prototypes and study their properties .

Using reactive magnetron sputtering we deposited Cu2O thin films on various substrates such as Si wafers, Corning’s Eagle glass, and FTO coated glass substrates (TEC). Various deposition conditions were examined to find optimal conditions for growing phase-pure Cu2O films. An extensive microstructural characterization was carried out to examine the effect of deposition conditions on the morphology and phase composition.

After finding optimal deposition conditions, we fabricated PV cell prototypes using the Cu2O films as the photoactive layers, and examined their electrical and optical properties. The simplest TEC/Cu2O/Metal prototypes showed Ohmic I-V curves with low resistance and no PV effect. The hypothesis was that the cell is symmetric for charge transport and therefore it shows no PV effect. In order to rectify this flaw we deposited on one of the two electrodes a ZnO film which is known to serve as hole blocking layer. The TEC/ZnO/Cu2O/Metal cells showed slight improvement in their I-V characteristics, displaying some diode-like rectification, but there was no PV effect. This is, most likely, due to shunts arising from microscopic voids and cracks due to the high roughness of the TEC substrate .

To rectify this flaw the TEC substrates were replaced by ITO coated Eagle glass substrates, deposited by sputtering, with less roughness compared to the TEC substrates. Using the ITO films we prepared ITO/ZnO/Cu2O/Ag PV cell prototypes. However, the new cells didn't show any PV effect or diode-like rectification. HRSEM images showed no cracks, voids, holes or any other suspects for shunts. In the thicker Cu2O layers we saw CuO formations close to the ITO substrate, which could have given rise to recombination .

We explored different routes to improve the PV cell prototypes. We tried separating the contacts, which did not show the expected improvement. Reducing the size of the contacts did show a significant improvement in rectification, but the series resistance was too high, diminishing the PV effect. Thus, we conclude that sputtering is probably not the best tool for metallization. Finally, instead of sputtering the contacts, we attached Cu tapes to some of the early prototype cells. We found that the series resistance was too high, due to lack of intimate contact to the films. The bottom line is that good contacts that do not give rise to shunts through the cells are crucial for making PV cells.