News

Machine learning boosts the discovery of new perovskite solar cell materials

CEST researchers use machine learning to accelerate computational study of perovskite alloy materials
A solar panel with blue sky and a graphic showing perovskite material
Graphic provided by J. Laakso

Researchers in the CEST group have published a study demonstrating the effectiveness of machine learning methods to identify suitable perovskite solar cell materials. Perovskite solar cells are a novel technology gathering a lot of interest due to their high efficiency and potential for radically lower manufacturing costs when compared to the traditional silicon-based solar cells.

Despite their promising qualities, the commercialization of perovskite solar cells has been held back by their fast degradation under environmental stresses, such as heat and moisture. They also contain toxic substances that can negatively impact the environment. The search for new perovskite materials that do not have these problems is ongoing, but the established experimental and computational research methods have not been able to handle the high number of material candidates that need to be tried and tested.

A person standing before trees
CEST researcher Jarno Laakso

CEST members Jarno Laakso and Patrick Rinke, with collaborators from University of Turku and China, developed new machine learning-based methodology for rapidly predicting perovskite properties. This new approach accelerates computations and can be used to study perovskite alloys. These alloy materials contain many candidates for improved solar cell materials, but studying them has been difficult with conventional computational methods. The researchers demonstrated the effectiveness of the new approach by finding the most stable mixing fractions for an alloy of CsPbCl3 and CsPbBr3 perovskites. Having an efficient method for studying the stability of perovskite alloys is a key step towards engineering solar cells that are more resilient to degradation.

The same methodology that was applied to perovskites in this study can boost the discovery of other new alloy materials. After the initial success with their machine learning approach, Laakso and collaborators are looking into studying more complex perovskite alloys to discover solar cell materials that are highly efficient, nontoxic, and resilient to degradation.

The paper was published in Physical Review Materials doi.org/10.1103/PhysRevMaterials.6.113801.

  • Published:
  • Updated:
Share
URL copied!

Read more news

Image from the conferment ceremony
Cooperation, Research & Art, University Published:

Doctoral education pilot arouses wide interest among applicants and corporate partners

The doctoral education pilot has got off to a fast start.
A man stands against a white background.
Awards and Recognition Published:

Broadband miniaturized spectrometer research receives QTF annual discovery award 2024

The clarity and compelling presentation of the research were one of the reasons why Doctoral Researcher Md Uddin earned the prize for the research paper, which was published in Nature Communications.
 Shankar Deka is an Assistant Professor at the Department of Electrical Engineering and Automation.
Research & Art Published:

Robotics needs safe behavior patterns

Robotics and autonomous systems are developing rapidly. Algorithms that withstand disturbances and uncertainties in the system model and environment are critical for development.
kuva puhelimesta ihmisen kädessä
Press releases Published:

Teaching a computer to type like a human

A new typing model simulates the typing process instead of just predicting words