| Work Detail |
An international team of researchers has demonstrated how to accelerate the research and development of scalable, high-performance organic solar cells using a high-throughput, automated platform. It is based on digital twin technology and roll-to-roll (R2R) printing in a closed-loop system.
An international team of researchers has created a high-throughput platform to discover high-performance organic solar cells that can be manufactured on a large scale. They used digital twin technology and roll-to-roll (R2R) printing in a closed-loop system.
The new platform, dubbed MicroFactory, was used to fabricate, characterize, and analyze 11,800 non-fullerene acceptor (NFA) organic photovoltaic devices over a 24-hour period. After analyzing the initial devices, the team used a large dataset of fabrication and characterization parameters to train and optimize machine learning models.
In a subsequent iteration, 1,200 devices were developed with improved PCEs based on “machine learning for inverse parameter generation,” according to Leonard Ng Wei Tat, a co-author of the research, who noted that champion efficiencies of up to 9.35% were recorded, representing a “1% improvement in a single cycle.”
Commercially available equipment had to be modified to fabricate the processing and characterization systems required for the study. “R2R equipment is available on the market, but most of it is designed for graphic printing and is mostly not suitable for printing solar cells,” Ng told pv magazine .
“What we were doing was applying a century-old, mature technology to the manufacturing of high-performance photovoltaic cells.” “The concept of printing solar cells is simple: deposit layer after layer of functional material until you build up a heterostructure that can act as the various components needed for a solar cell,” explains Ng.
The team consists of slotted mold coating and annealing subsystems with integrated sensors. “We fabricated the photovoltaic cells by depositing functional layers on a polyethylene terephthalate (PET) strip with a patterned transparent conductive electrode (TCE),” the research team explains. The functional layers consisted of a conductive polymer layer, PEDOT:PSS, a silver grid layer, and zinc oxide nanoparticles.
Multiple sensors collected data on 36 process parameters, which are stored in a database on a remote data server for use in digital twin models. “These models suggested specific alterations to vital manufacturing parameters, especially the donor-to-acceptor (D:A) ratio, and also allowed for the incorporation of newly disclosed scientific knowledge including the introduction of new interface layers,” the research team stated.
The ability to collect a large amount of data allowed for analysis and identification of trends and performance factors. As an example, Ng pointed to the discovery that humidity control was much more important than temperature control in ensuring good device quality. “This has a lot to do with the observed trend that our manufactured solar cells perform better in the low humidity conditions of winter than in summer, despite producing them in the same air-conditioned environment during both seasons,” Ng says.
The scientists stressed that the iterative approach, based on machine learning, represents a strategic optimization as an alternative to traditional design of experiments. “For example, large-scale solar module manufacturers can quickly create simple digital twins of their processes in order to create large data sets to identify the factors that really move the needle to improve productivity and performance,” said Ng.
The research team needed a cross-disciplinary set of expertise that included materials science, hardware and software development, and machine learning. “Most researchers are familiar with just one field, and it takes a lot of coordination and effort to put things into context,” Ng explains.
Details of the study are outlined in “ A printing-inspired digital twin for the self-driving, high-throughput, closed-loop optimization of roll-to-roll printed photovoltaics , ” published in Cell Reports Physical Science . Members of the research team are from the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia, Pukyong National University in South Korea, and Nanyang Technological University in Singapore.
Looking ahead, researchers are investigating new materials and device architectures for more efficient flexible solar cells, as well as continuing to apply artificial intelligence (AI) technologies, digital twins, and parameter reverse generation capabilities to other processes, such as batch processing and traditional solar manufacturing. “Over time, we hope to develop a unified system capable of connecting multiple machines, factories, and laboratories around the world, enabling more advanced AI,” Ng said. |
