A photovoltaic cell composed of a layer of perovskite on another silicon reached a yield of 25.2%. This new record was established thanks to a manufacturing process developed by researchers from the Swiss Federal Institute of Technology in Lausanne and the Swiss Center for Electronics and Microtechnology in Neuchâtel.
The photovoltaic cell based on perovskite continues to rise in the race for performance. The previous record of a two-terminal tandem cell combining a silicon layer with a perovskite material was February 2017: it was 23.6%. Researchers from the PV-Lab of the Swiss Federal Institute of Technology in Lausanne (EPFL) and the PV Center of the Swiss Center for Electronics and Microtechnology (CSEM) in Neuchâtel have managed to do better: the efficiency of their cell has reached 25 , 2%. Their work was published in the journal Nature Materials on June 11, 2018. At the origin of this record performance, an improvement in the manufacturing process.
The process used starts from a conventional silicon layer. Its surface consists of a network of pyramids formed by a chemical attack during its manufacture. Why these pyramids? They limit the reflections of the light and trap it to increase the efficiency of the cell, says Florent Sahli, the first author of the study: “Pyramids are always present in industrial silicon cells to guarantee optimal optical properties. Problem: In the case of a tandem cell, that is to say composed of two layers of superimposed materials, the perovskite is generally deposited above in liquid form and accumulates in the hollows. It happens then that the summits of the pyramids exceed. Result: Short circuits are likely to occur. To remedy this, the pyramids were so far leveled, which made them lose their anti-reflective power.
Towards industry-compatible methods
This is where the innovation imagined by the EPFL and CSEM researchers comes in. It breaks down in two stages. The first is to co-evaporate a layer of lead iodide and cesium bromide to line the pyramids evenly. The second is a “spin-coating” stage: a liquid containing an organic compound is deposited on the substrate in rapid rotation. During annealing at 150 ° C., the elements deposited by evaporation and liquid pathway react and form the active perovskite phase. Finally, the layers that make it possible to create the contacts to extract the current supplied by the cell are deposited using industrializable processes. “Our goal is to use only manufacturing methods compatible with the industry,” says Sahli. This is the case except for spin-coating. Replacing this step is the continuation of our work. ”
The tandem cell based on perovskite and silicon is an interesting way to obtain yields in excess of 30%. “It’s clearly doable experimentally,” says Sahli. The architecture of these cells makes it possible to limit losses by superimposing two layers of materials that absorb light in different wavelengths. In the case of cells combining silicon and perovskite, the latter absorbs the most energetic photons. Those who are less cross it and are absorbed by the silicon layer below.
Increase the size and stability of cells
These two layers can be manufactured separately or together. These are then four or two terminal cells respectively. In the first case, each layer has a higher contact and a lower contact to extract the electrons produced. The record yield achieved in this configuration is 26.7%. “If these cells have the advantage of being able to be manufactured and contacted separately, without disturbing each other, they have disadvantages,” says Sahli. They require more power electronics and different production lines for each cell. ”
Researchers at EPFL and CSEM therefore opted for a two-terminal cell. The layers of silicon and perovskite are made together, glued one on top of the other. There is then only one contact under the silicon and another on the perovskite. “If the connection of the layers in series limits the efficiency of the cell, this two-terminal architecture makes more sense from a commercial point of view, ensures Florent Sahli. Indeed, by adding a few steps to a current production line of silicon cells, a 20% more efficient cell can be obtained at a relatively similar cost. ”
Other improvements considered: improve the stability of the cell and increase its size. It is currently only 1.4 cm². The goal is to be able to deposit it on a classic woofer of 6 inches, or 240 cm². “It’s possible, but it requires considerable developments, concludes Mr. Sahli. We are still on a proof of concept. ”