High-efficiency thin film solar cells

The Laboratory for Thin Films and Photovoltaics is working on different absorber materials like chalcopyrites, CdTe, kesterites and perovskites. Additionally, the use in tandem applications are studied.

CuIn1-xGaxSe2 (CIGS) is one of the most promising materials for highly efficient thin film solar cells. By changing the gallium to indium ration, the bandgap can be shifted between 1.0 and 1.7 eV to adjust the devices to various needs. This makes those cells also interesting for application in multijunction solar cells.

In our group, efficiencies up to 18.8% on glass substrates were achieved. Flexible thin film solar cells on polymer film with a new record efficiency of 20.4% have been independently certified by Fraunhofer Institute for Solar Energy Systems (ISE) in Freiburg, Germany.

The 20.4% efficiency solar cell is an independently certified highest efficiency record for any type of flexible solar cell on polymer film reported up to now. Furthermore, it represents the highest reported efficiency of a CIGS solar cell on any type of substrate. This solar cell processing can be adapted for roll-to-roll maufacturing of monolithically interconnected solar modules on polymer films.

The research in our group focuses on alternative materials and processes for buffer layers and back contacts, as well as the development of thin film tandem solar cells.

An In2S3 buffer layer can increase the efficiency by better utilization of the light in the visible range of the solar spectrum which leads to an increased current density. Here, different deposition methods like PVD, flash evaporation, and ultrasonic spray pyrolysis, are investigated.

For back contacts, transparent conductive oxides like ITO or ZnO:Al are investigated, which can be used for bifacial illumination using a mirror system (higher current with same are of absorber, thus potential to decrease cost) as well as for production of tandem cells using the combination of CGS with CIS. Furthermore, or group works on highly reflective back contacts like TiN, which should enable a decrease of the absorber thickness while maintaining the current density, and thus reduce cost per watt.


Cadmium telluride is a compound semicounductor consisting of the metal cadmium and the semiconductor tellurium, in a ratio of 1:1. The band gap of CdTe is about 1.5 eV, which is close to the optimum for conversion of sunlight to electricity. Additionally, the material is very stable both against heat and chemically. The excellent absorption characteristics of this material make it an excellent candidate for the use in photovoltaic applications. The material deposition is very easy with varying technologies and thus allows high throughput in producition.

CdTe was first used as an absorber material in solar cells in the 1960's. In 1980, the 10% mark in efficiency on lab-scale cells was broken, 2005 the still valid efficiency world record of 16.5% was established by the National Renewable Energy Laboratories (NREL) in the United States. Optical losses were minimized in that record cell. The challenge for the future is the improvement of the electronic properties for improving the cell voltage and thus the efficiency.
The annual production capacity of commercial solar modules exceeded 1 GWp in 2009. The largest producer is the U.S. company First Solar, the largest part of the production takes place in Malaysia. Its CdTe solar modules are currently the cheapest on a price per watt basis.
In our lab CdTe solar cells with efficiencies of up to 15.6% (March 2010) on glass substrates and up to 13.5% on polyimide foil could be produced. In contrast to CIGS, for all substrates the same deposition process with similar layer quality can be used. The difference in efficiency between glass and polymer substrates is caused by the lower transparency of polyimide compared to glass.
The development CdTe solar cells on flexible substrates is one of our focus points. The production process used in our lab uses temperatures of 420 °C or lower, allowing the use of polyimide as a substrate (can only be used in processes using up to 450 °C).
New back contacts are also developed in our lab. Up to now, the back contact is one of the crucial points in a CdTe solar cell. For high efficiencies, it has to be deposited as last layer, thus prohibiting the use of opaque substrates.
Doping of the semiconductor material CdTe is a new focus point in our research under a project started in December 2009. The increase of charge carrier concentration in polycrystalline CdTe is an important step for further improvements in efficiency.
Solar cells made from Cu2ZnSn(Se,S)4 are a promising alternative to other thin film technologies using non-toxic and readily available materials. Using two different precursor synthesis routes (non-vacuum, 11.7% and vacuum, 8.0%), we investigate the origin of the VOC-deficit of kesterite absorbers.
The perovskite is grown on a thin interlayer made of the substance abbreviated as PCBM (phenyl-C61-butyric acid methyl ester) is used . Each PCBM molecule contains 61 carbon atoms interconnected in the shape of a soccer ball. The perovskite film is prepared by a combination of vapour deposition and spin coating onto this layer, which has tiny football like structure,  followed by an annealing at a “lukewarm” temperature. This magic perovskite crystal absorbs blue and yellow spectrum of visible light and converts these into electricity. By contrast, red light and infrared radiation simply pass through the crystal. As a result, the researchers can attach a further solar cell underneath the semi-transparent perovskite cell in order to convert the remaining light into electricity.
Tandem cells

The maximum achievable efficiency of solar cells is theoretically described by the Shockley Queisser limit. This model takes as assumption that all photons with higher energy than the bandgap will be absorbed and converted to an electron hole pair, and all recombination processes are neglected. For solar cells with a bandgap of 1.1 eV, this limit is about 30%.

However, the photon energy exceeding the bandgap will be lost as the excited electron hole pair will instantly thermalize with the lattice and drop to the conduction or valence band edge for electrons and holes, respectively. To minimize these losses, several solar cells with varying bandgaps can be stacked on top of each other. Thus, the photons with higher energy will be absorbed in a wide gap absorber, while those with lower energy will be transmitted to a lower cell which has a smaller bandgap. This results in a better utilization of the photon energy.

With a triple junction device using monocrystalline III-V semiconductors, efficiencies exceeding 40% could be achieved with this approach under concentrated light. However, those epitactically grown cells are expensive in production. Thus, we investigate the possiblity to implement this approach in thin film solar cells.

In our group, we investigate the possibility of using CIGS solar cells in tandem devices. The investigated or planned-to-investigate structures include CGS/CIS, dye sensitized cell (DSC)/CIGS, a-Si/CIGS, as well as (DSC?)/CdTe(CI(G)S tandem cells.
Structure of a 2-terminal tandem solar cell