Photon upconversion for thin film solar cells

Abstract

In this research one of the many possible methods to increase the efficiency of solar cells is described. The method investigated is based on adapting the solar light in such a way that the solar cell can convert more light into electricity. The part of the solar spectrum that is adapted is the part that cannot be absorbed by the solar cells, because the photon energy is too low. This conversion of light is done by so called upconversion, which means that lower energy photons are converted into higher energy photons that can be absorbed by the solar cell. The upconverters used in this thesis are those based on lanthanide ions doped in crystalline hosts. Lanthanide ions have very specific absorption and emission lines, which means that by choosing an appropriate ion one can convert any arbitrary wavelength. One of the most important aspects when one wants to apply upconverters onto solar cells is the light intensity necessary for efficient conversion. Because the upconversion process requires two photons to make a new, higher energy photon, the conversion process is non-linearly dependent on the light intensity. This is the main limitation for practical applications. Therefore, next to applying upconverters onto solar cells also more fundamental questions are addressed in this thesis, for instance, the question what determines efficient conversion. At first the upconverter materials in different hosts are characterized and investigated. The host material influences non-radiative decays, the absorption strength, the lifetime and the energy transfer rate between the lanthanide ions. By investigating two upconverter hosts with small differences (α and β-NaYF4 doped with Er3+ and Yb3+), we have tried to investigate the origin of the difference in upconversion efficiency. For this, emission and absorption spectra are measured under the same conditions and concentrations of the lanthanide ions. Also the absorption strength on the upconverter efficiency is investigated. For this, another host material (Gd2O2S:Er3+, Yb3+) was characterized and coupling of β-NaYF4 upconverter with a plasmon resonance is investigated. The second part of the research is concerned with application of the upconverter onto solar cells. In this research the β-NaYF4 and Gd2O2S upconverters were applied onto amorphous silicon solar cells (a-Si:H). At first, proof of principle experiments on solar cells with β-NaYF4 upconverter were performed with laser light. I-V curves were measured and an increased response was determined in the upconverter solar cells. Finally, to proof viability of the concept further a set-up was made to concentrate simulated solar light. Concentration of solar light is not uncommon and more real life than laser light. All wavelengths longer than 900 nm were concentrated, which means that the range of the spectrum was much broader than the part that is absorbed by the upconverter. As upconverter material Gd2O2S was applied. Though a large part of the response is due to sub band gap defect absorption an increased response due to the upconverter was measured as well. The upconverter efficiency is thus high enough under moderate concentration of sunlight.