Perovskite solar cells built to match durability of silicon photovoltaics

Researchers at Oxford University and Exciton Science have demonstrated a new way to create stable perovskite solar cells, with fewer structural and energetic defects and the potential to rival silicon’s durability. By removing the solvent dimethyl-sulfide and introducing dimethylammonium chloride as a crystallisation agent, the researchers were able to better control the intermediate phases of the perovskite crystallisation process, leading to thin films of greater quality, with reduced defects and enhanced stability. Groups of up to 138 sample devices were then subjected to a rigorous accelerated aging and testing process at high temperatures and in real-world conditions.

Artificially synthesised in laboratory conditions, semiconductor thin films made up of perovskite compounders are cheaper to make than silicon solar cells, with greater flexibility and a tuneable band gap. They have also reached power-conversion efficiencies of over 25%. However, too much focus has been placed on creating the most efficient perovskite solar cell, rather than resolving the fundamental problems inhibiting the material from being used in widespread commercial applications.

Compared to silicon, perovskites can degrade rapidly in real-world conditions, with exposure to heat and moisture negatively impacting device performance. Solving these issues is the challenge for perovskites in order to take on, or ‘boost’, silicon via a tandem architecture and take their place in the commercial photovoltaics landscape.

Formamidinium-caesium perovskite solar cells created using the new synthesis process outperformed the control group and demonstrated resistance to thermal, humidity and light degradation. This is a step forward to matching commercial silicon’s stability and makes perovskite-silicon tandem devices a more realistic candidate for becoming the dominant next-generation solar cell.

Led by Professor Henry Snaith (Oxford University) and Professor Udo Bach (Monash University), the work was published in Nature Materials. Oxford University PhD student Philippe Holzhey said it’s important that people realise there is no value in performance if it’s not a stable performance, adding that if the device lasts for a day or a week, there’s not so much value it in. According to Holzhey, it has to last for years. During testing, the best device operated above the T80 threshold for over 1400 hours under simulated sunlight at 65°C. T80 is the time it takes for a solar cell to reduce to 80% of its initial efficiency, a common benchmark within the research field.

Beyond 1600 hours, the control device fabricated using the conventional dimethyl-sulfoxide approach stopped functioning, while devices fabricated with the new design retained 70% of their original efficiency, under accelerated aging conditions. The same degradation study was performed on a group of devices at 85°C, with the new cells again outperforming the control group. Extrapolating from the data, the researchers calculated that the new cells age by a factor of 1.7 for each 10°C increase in the temperature they are exposed to, which is close to the two-fold increase expected of commercial silicon devices.

Dr David McMeekin, the corresponding and joint first author on the paper, said what separates the study from others is that the researchers carried out a lot of accelerated aging, having aged the cells at 65°C and 85°C under the whole light spectrum. The number of devices used in the study is also significant, with other perovskite research projects limited to one or two prototypes. “Most studies only show one curve without any standard deviation or any kind of statistical approach to determine if this design is more stable than the other,” McMeekin said.

The researchers hope the study will encourage a greater focus on the intermediate phase of perovskite crystallisation as an important factor in achieving greater stability and commercial viability.

Image credit: iStock.com/lovelyday12