Bridging the gap between photovoltaics R&D and manufacturing with data-driven optimization

Novel photovoltaics, such as perovskites and perovskite-inspired materials, have shown great promise due to high efficiency and potentially low manufacturing cost. So far, solar cell R&D has mostly focused on achieving record efficiencies, a process that often results in small batches, large variance, and limited understanding of the physical causes of underperformance. This approach is intensive in time and resources, and ignores many relevant factors for industrial production, particularly the need for high reproducibility and high manufacturing yield, and the accompanying need of physical insights. The record-efficiency paradigm is effective in early-stage R&D, but becomes unsuitable for industrial translation, requiring a repetition of the optimization procedure in the industrial setting. This mismatch between optimization objectives, combined with the complexity of physical root-cause analysis, contributes to decade-long timelines to transfer new technologies into the market. Based on recent machine learning and technoeconomic advances, our perspective articulates a data-driven optimization framework to bridge R&D and manufacturing optimization approaches. We extend the maximum-efficiency optimization paradigm by considering two additional dimensions: a technoeconomic figure of merit and scalable physical inference. Our framework naturally aligns different stages of technology development with shared optimization objectives, and accelerates the optimization process by providing physical insights.

[1]  Giuseppe Romano,et al.  Bayesim: A tool for adaptive grid model fitting with Bayesian inference , 2018, Comput. Phys. Commun..

[2]  Peng Gao,et al.  Effect of Annealing Temperature on Film Morphology of Organic–Inorganic Hybrid Pervoskite Solid‐State Solar Cells , 2014 .

[3]  Wei Chen,et al.  Perovskite solar cells - An overview of critical issues , 2017 .

[4]  Christoph J. Brabec,et al.  Physics-guided characterization and optimization of solar cells using surrogate machine learning model , 2019, 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC).

[5]  Christoph J. Brabec,et al.  Impurity Tracking Enables Enhanced Control and Reproducibility of Hybrid Perovskite Vapor Deposition , 2019, ACS applied materials & interfaces.

[6]  Henk J. Bolink,et al.  Quantification of spatial inhomogeneity in perovskite solar cells by hyperspectral luminescence imaging , 2016 .

[7]  Lydia Helena Wong,et al.  Modelling and loss analysis of meso-structured perovskite solar cells , 2017 .

[8]  Kai Zhu,et al.  Top and bottom surfaces limit carrier lifetime in lead iodide perovskite films , 2017, Nature Energy.

[9]  Ilke Celik,et al.  A technoeconomic analysis of perovskite solar module manufacturing with low-cost materials and techniques , 2017 .

[10]  Patrick Huck,et al.  Active learning for accelerated design of layered materials , 2018, npj Computational Materials.

[11]  Alán Aspuru-Guzik,et al.  Phoenics: A Bayesian Optimizer for Chemistry , 2018, ACS central science.

[12]  Ying Shirley Meng,et al.  Homogenized halides and alkali cation segregation in alloyed organic-inorganic perovskites , 2019, Science.

[13]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[14]  Anders Hagfeldt,et al.  Perovskite Solar Cells: From the Atomic Level to Film Quality and Device Performance. , 2018, Angewandte Chemie.

[15]  Mukundan Thelakkat,et al.  Capturing the Sun: A Review of the Challenges and Perspectives of Perovskite Solar Cells , 2017 .

[16]  Qianxiao Li,et al.  Embedding physics domain knowledge into a Bayesian network enables layer-by-layer process innovation for photovoltaics , 2019, npj Computational Materials.

[17]  Tonio Buonassisi,et al.  Precursor Concentration Affects Grain Size, Crystal Orientation, and Local Performance in Mixed-Ion Lead Perovskite Solar Cells , 2018, ACS Applied Energy Materials.

[18]  Steve Albrecht,et al.  How to Make over 20% Efficient Perovskite Solar Cells in Regular (n–i–p) and Inverted (p–i–n) Architectures , 2018, Chemistry of Materials.

[19]  Anders Hagfeldt,et al.  Identifying and suppressing interfacial recombination to achieve high open-circuit voltage in perovskite solar cells , 2017 .

[20]  Laura Méndez,et al.  Mass Production Test of Solar Cells and Modules Made of 100% UMG Silicon. 20.76% Record Efficiency , 2019, Energies.

[21]  Tonio Buonassisi,et al.  Economic viability of thin-film tandem solar modules in the United States , 2018 .

[22]  Martin A. Green,et al.  Manufacturing cost and market potential analysis of demonstrated roll-to-roll perovskite photovoltaic cell processes , 2018 .

[23]  Ian Marius Peters,et al.  The Value of Efficiency in Photovoltaics , 2019, Joule.

[24]  Anders Hagfeldt,et al.  Unbroken Perovskite: Interplay of Morphology, Electro‐optical Properties, and Ionic Movement , 2016, Advanced materials.

[25]  Lawrence A. Adutwum,et al.  How To Optimize Materials and Devices via Design of Experiments and Machine Learning: Demonstration Using Organic Photovoltaics. , 2018, ACS nano.

[26]  Matthias Meixner,et al.  Influence of the Grain Size on the Properties of CH3NH3PbI3 Thin Films. , 2017, ACS applied materials & interfaces.

[27]  Haohui Liu,et al.  The GaAs/GaAs/Si solar cell – Towards current matching in an integrated two terminal tandem , 2017 .

[28]  Michael Woodhouse,et al.  Correction: Revisiting thin silicon for photovoltaics: a technoeconomic perspective , 2019, Energy & Environmental Science.

[29]  Sang Il Seok,et al.  Solvent engineering for high-performance inorganic-organic hybrid perovskite solar cells. , 2014, Nature materials.

[30]  Peter J. Holliman,et al.  Quantitative chemical analysis of perovskite deposition using spin coating , 2019, Materials Letters: X.

[31]  Anders Hagfeldt,et al.  Cesium-containing triple cation perovskite solar cells: improved stability, reproducibility and high efficiency† †Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ee03874j Click here for additional data file. , 2016, Energy & environmental science.

[32]  Anders Hagfeldt,et al.  Not All That Glitters Is Gold: Metal-Migration-Induced Degradation in Perovskite Solar Cells. , 2016, ACS nano.

[33]  David Cahen,et al.  Halide Perovskites: Is It All about the Interfaces? , 2018, Chemical reviews.

[34]  Nando de Freitas,et al.  Taking the Human Out of the Loop: A Review of Bayesian Optimization , 2016, Proceedings of the IEEE.

[35]  D. M. Powell,et al.  Modeling the Cost and Minimum Sustainable Price of Crystalline Silicon Photovoltaic Manufacturing in the United States , 2013, IEEE Journal of Photovoltaics.

[36]  Heming Wang,et al.  Highly reproducible perovskite solar cells via controlling the morphologies of the perovskite thin films by the solution-processed two-step method , 2018, Journal of Materials Science: Materials in Electronics.

[37]  Shri Kant Machine Learning and Pattern Recognition , 2010 .

[38]  Savitha Ramasamy,et al.  Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks , 2018, npj Computational Materials.

[39]  Inho Kim,et al.  Optimization of device design for low cost and high efficiency planar monolithic perovskite/silicon tandem solar cells , 2019, Nano Energy.

[40]  Michael Woodhouse,et al.  Technology advances needed for photovoltaics to achieve widespread grid price parity , 2016 .

[41]  Anubhav Jain,et al.  Harnessing the Materials Project for machine-learning and accelerated discovery , 2018, MRS Bulletin.

[42]  C. Brabec,et al.  Analyzing the efficiency, stability and cost potential for fullerene-free organic photovoltaics in one figure of merit , 2018 .

[43]  Christoph J. Brabec,et al.  Evaluation of Electron Donor Materials for Solution‐Processed Organic Solar Cells via a Novel Figure of Merit , 2017 .

[44]  Gerbrand Ceder,et al.  Rapid Photovoltaic Device Characterization through Bayesian Parameter Estimation , 2017 .