Enhancing the stability of organic photovoltaics through machine learning

David, Tudur Wyn; Anizelli, Helder Scapin; Jacobsson, T. Jesper; Gray, Cameron; Teahan, William J.; Kettle, Jeff

doi:10.1016/j.nanoen.2020.105342

Cited by 44 publications

(27 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, OPDs made from NFAs have shown significantly improved stability due to a stable morphology of photoactive later. [ 198 ] Besides, machine‐learning approaches, e.g., kernel ridge regression, [ 199 ] Random Forest regression, [ 200,201 ] decision tree, [ 202,203 ] and sequential minimal optimization regression, [ 204 ] have been demonstrated for predicting efficiency, stability, and significant attributes of devices. The machine‐learning methods can be applied to assist the time‐consuming experimentation and optimization in the development of stable OPDs.…”

Section: Discussionmentioning

confidence: 99%

Recent Advances in Solution‐Processable Organic Photodetectors and Applications in Flexible Electronics

Lan

Lee

Zhu

2021

Advanced Intelligent Systems

View full text Add to dashboard Cite

Organic photodetectors (OPDs) are promising for applications in flexible electronics due to their advantages of excellent photodetection performance, cost-effective solution-fabrication capability, flexible device design, and adaptivity to manufacturability. This review outlines the recent advances in the development of high-performance OPDs and their applications in flexible electronics. The approaches to developing different noise reduction methods, filter-free spectral selective detection, flexible OPDs, and scale-up production of flexible OPDs through solution-fabrication processes are discussed. Applications of the OPD technology ultimately result in the materialization of wearable units, flexible and compact information sensors at commercially viable costs, including wearable health self-monitoring devices, flexible optical communication systems, and flexible large-area image sensors.

show abstract

Section: Discussionmentioning

confidence: 99%

Recent Advances in Solution‐Processable Organic Photodetectors and Applications in Flexible Electronics

Lan

Lee

Zhu

2021

Advanced Intelligent Systems

View full text Add to dashboard Cite

show abstract

“…The large scaling gap between laboratory cells and commercial modules can be attributed to successive performance loss during fabrication, integration, and installation. [ 43 ] New interdisciplinary and dedicated approaches from other research areas, for example, machine learning, [ 126 ] may shed light on the complex issues related to performance loss during the scaling‐up of OPV fabrication.…”

Section: Discussionmentioning

confidence: 99%

Organic Photovoltaics: Where Are We Headed?

2021

View full text Add to dashboard Cite

Organic photovoltaics (OPV) is an emerging technology that combines semi‐transparency and flexibility in lightweight, ultrathin solar modules. The record power conversion efficiencies for OPV are approaching 20%, with reported lifetimes ranging from months to several years. Despite these attributes, OPV has not yet been commercialized on a large scale. Here, we critically examine the commercial potential of OPV. We begin by surveying feasibility studies in the literature to identify threshold values for performance that would enable OPV to be commercially competitive with conventional electricity production. We discuss how these parameters alone are not sufficient to predict the success of a photovoltaic technology, as there is still significant discrepancy between performance in the laboratory and in the field. Even if this gap can be closed, it is unlikely that OPV will ever compete in the conventional electricity market. Instead, the unique properties may make OPV an interesting candidate for niche applications with less rigid demands on performance, but stricter requirements on aesthetics, flexibility, weight, and transparency. In this case, application‐specific requirements on efficiency and stability must first be established. Ultimately, even the most optimistic scenarios for OPV identify poor stability as the biggest challenge for general market entry.

show abstract

“…The complexity of the OPV not only comes from various molecular designs including the choice of the donor and acceptor, but also from the device technologies such as chemical synthesis, photocurrent composition, film forming, and photostability. [ 71 ] Although in theory novel materials could be generated through ML models, it is challenging to characterize their synthesizability and experimental process. Hence, some of the latest ML approaches tackle experimental process optimization, and a brief summary ( Table 3 ) in this trend is included in given review.…”

Section: Ml‐assisted Opv Device Optimizationsmentioning

confidence: 99%

“…In fact, besides PCE, other device properties like V OC , FF , J SC , and stability [ 80 ] can be also modeled and optimized via ML methods. David et al [ 71 ] used supervised learning in a sequential minimal optimization regression model (SMOreg) training on a dataset containing 1,850 entries of device properties (e.g., substrate type, environmental conditions, light type, temperature, and relative humidity) to predict stability and the initial PCE of OPV devices with r > 0.70. They provide methods for material identifications in terms of improved stability and top performance.…”

Section: Ml‐assisted Opv Device Optimizationsmentioning

confidence: 99%

Performance Prediction and Experimental Optimization Assisted by Machine Learning for Organic Photovoltaics

Zhao

Geng

Troisi

et al. 2022

Advanced Intelligent Systems

View full text Add to dashboard Cite

Organic photovoltaic (OPV) based on πconjugated molecules or polymers converts the solar energy to electricity utilizing the electron donor (D)/acceptor (A) bulk heterojunction structure to efficiently separate the bound excitons generated by light absorption into free charge carriers. [1] It has attracted immense research interest due to their numerous technology advantages such as inherent flexibility, portability, lightweight, low cost, and low toxicity, [2] as well as recently reported a high power conversion efficiency (PCE) of 19%. [3] However, to commercialize these technologies, further efficiency improvements are desired. This can be hardly accomplished by tedious and expensive trial-and-error methods based on chemical intuitions, labor-intensive synthesis, and characterization, especially when considering the large chemical space available for designing new organic donor or acceptor molecules and the numerous experimental parameters that can be tuned in the fabrication of the devices. [4] It is important to realize that the construction of a quantitative structure-property relationship (QSPR) model [5] would greatly benefit the search of new functional molecules and the optimization of experimental conditions, and, accordingly, it can become a key factor to improve the photovoltaic performance of OPVs. [5b] Furthermore, an interpretable QSPR model can be also used to unravel the fundamental working mechanism for a complex experimental behavior through its statistical correlation analysis with various microscopic physical parameters, which can hardly be achieved by blackbox predictions. Unfortunately, deriving such quantitative and interpretable QSPR models is highly nontrivial due to our current poor understanding of the complex photophysical phenomena in OPVs such as photoabsorption, exciton diffusion, exciton dissociation, charge generation and charge transport, as well as charge collection at electrodes. [1c,6] Earlier empirical QSPR models such as Scharber's model [7] for fullerene-based OPVs and the modified Scharber's model [8] for nonfullerene-based OPVs have been proposed for predicting OPV performance, yet, their accuracy for predicting PCE is unsatisfactory (with a linear Pearson correlation coefficient (r) between Scharber's PCE and experimental PCE at around only 0.1-0.2). [9] Driven by the development of artificial intelligence (AI) [10] algorithms, as well as the increased availability of a high-quality large dataset along with efficient data-mining techniques, [11]

show abstract

Enhancing the stability of organic photovoltaics through machine learning

Cited by 44 publications

References 24 publications

Recent Advances in Solution‐Processable Organic Photodetectors and Applications in Flexible Electronics

Recent Advances in Solution‐Processable Organic Photodetectors and Applications in Flexible Electronics

Organic Photovoltaics: Where Are We Headed?

Performance Prediction and Experimental Optimization Assisted by Machine Learning for Organic Photovoltaics

Contact Info

Product

Resources

About