Data-driven identification of functional additives and solution parameters in mixed Sn-Pb perovskite solar cells via β-VAE augmentation

Abstract

Optimizing perovskite solar cells (PSCs) requires precise control of solution chemistry and functional additives. However, limited experimental data hinder systematic discovery. Here, we integrate 1,540 carefully selected experimental device records with 4,000 synthetic data points generated by a beta-variational autoencoder to investigate solution parameters and organic additives governing device performance. A residual neural network trained on this hybrid dataset achieves strong predictive accuracy with an R² of 0.87 for power conversion efficiency. Even when trained solely on synthetic data, the model attains an R² of 0.785. Within this framework, 733 organic additives with diverse functional groups were evaluated to identify molecular features that enhance absorber quality. High-efficiency PSCs are associated with solution concentrations above 1.3 molar and elevated formamidinium iodide (FAI) ratios, in combination with additives containing benzene rings, methylene, and amine groups. Notably, a composition comprising FAI (1.05), cesium iodide (0.03), methylammonium chloride (0.3), lead(II) iodide (1.5), and a molybdenum trioxide interlayer, combined with 1,3-dihydro-1-[1-(phenylmethyl)-4-piperidinyl]-2Hbenzimidazol-2-one as an additive, yields a PCE of 25.66%. This additive was absent from the training data, demonstrating the capability of the proposed framework to discover novel and effective organic additives for PSC optimization.

Keywords

Perovskite solar cell / Experimental data / Synthetic data / Additive

Acknowledgements

The authors would like to thank Laser Plasma Research Institute for its support.

Funding

None.

Conflict of interest

The authors declare that they have no competing interests.

Author contributions

Conceptualization: All authors

Data curation: Behzad Iranipour

Formal analysis: Behzad Iranipour, Mohammadreza Sadeghian

Funding acquisition: Ezeddin Mohajerani

Investigation: All authors

Methodology: All authors

Project administration: Ezeddin Mohajerani

Software: Behzad Iranipour, Mohammadreza Sadeghian

Supervision: Ezeddin Mohajerani

Validation: All authors

Visualization: Behzad Iranipour, Mohammadreza Sadeghian

Writing-original draft: Behzad Iranipour

Writing-review & editing: All authors

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data

The code and dataset supporting the findings of this study are publicly available in the Beta-Variational- Autoencoder repository on GitHub at https://github.com/Behzadiranipour/Beta-Variational-Autoencoder.

References

[1] Gomez-Peralta JI, Bokhimi X. Discovering new perovskites with artificial intelligence. J Solid State Chem. 2020; 285:121253. doi: 10.1016/j.jssc.2020.121253

[2] Liu M, Cao Z, Wang X, et al. Perovskite material-based memristors for applications in information processing and artificial intelligence. J Mater Chem C. 2023; 11(39):13167-13188. doi: 10.1039/D3TC02309E

[3] Tao Q, Xu P, Li M, Lu W. Machine learning for perovskite materials design and discovery. Npj Comput Mater. 2021; 7(1):23. doi: 10.1038/s41524-021-00495-8

[4] Liu Y, Yan W, Han S, et al. How machine learning predicts and explains the performance of perovskite solar cells. Solar RRL. 2022; 6(6):2101100. doi: 10.1002/solr.202101100

[5] Lu Y, Wei D, Liu W, et al. Predicting the device performance of the perovskite solar cells from the experimental parameters through machine learning of existing experimental results. J Energy Chem. 2023; 77:200-208. doi: 10.1016/j.jechem.2022.10.024

[6] Khan A, Kandel J, Tayara H, Chong KT. Predicting the bandgap and efficiency of perovskite solar cells using machine learning methods. Mol Inform. 2024 ; 43(2):e202300217. doi: 10.1002/minf.202300217

[7] Huang F, Pascoe AR, Wu WQ, et al. Effect of the microstructure of the functional layers on the efficiency of perovskite solar cells. Adv Mater. 2017; 29(20):1601715. doi: 10.1002/adma.201601715

[8] Lakhdar N, Hima A. Electron transport material effect on performance of perovskite solar cells based on CH3NH3GeI3. Opt Mater. 2020; 99:109517. doi: 10.1016/j.optmat.2019.109517

[9] Shao S, Loi MA. The role of the interfaces in perovskite solar cells. Adv Mater Interfaces. 2020; 7(1):1901469. doi: 10.1002/admi.201901469

[10] Bag A, Radhakrishnan R, Nekovei R, Jeyakumar R. Effect of absorber layer, hole transport layer thicknesses, and its doping density on the performance of perovskite solar cells by device simulation. Sol Energy 2020; 196:177-182. doi: 10.1016/j.solener.2019.12.014

[11] Li Z, Xiao C, Yang Y, et al. Extrinsic ion migration in perovskite solar cells. Energy Environ Sci. 2017; 10(5):1234-1242. doi: 10.1039/C7EE00358G

[12] Valencia A, Liu F, Zhang X, et al. Auto-generating a database on the fabrication details of perovskite solar devices. Sci Data 2025; 12(1):270. doi: 10.1038/s41597-025-04566-z

[13] Kenfack AK, Mashamba DR, Thantsha NM, Msimanga M. Prediction of band gap and optimum electrical parameters of a thin homojunction perovskite solar cell based on FA1- xCsxSnyPb1- yI 3 through a combination of SCAPS-1D and machine learning based modelling. Mater Today Commun. 2023; 37:107318. doi: 10.1016/j.mtcomm.2023.107318

[14] Mammeri M, Dehimi L, Bencherif H, Pezzimenti FJ. Paths towards high perovskite solar cells stability using machine learning techniques. Sol Energy. 2023; 249:651-660. doi: 10.1016/j.solener.2022.12.002

[15] Li W, Hu J, Chen Z, et al. Performance prediction and optimization of perovskite solar cells based on the Bayesian approach. Sol Energy. 2023; 262:111853. doi: 10.1016/j.solener.2023.111853

[16] Yan W, Liu Y, Zang Y, et al. Machine learning enabled development of unexplored perovskite solar cells with high efficiency. Nano Energy. 2022; 99:107394. doi: 10.1016/j.nanoen.2022.107394

[17] Calvo ME. Materials chemistry approaches to the control of the optical features of perovskite solar cells. J Mater. Chem. A. 2017; 5(39):20561-20578. doi: 10.1039/C7TA05666D

[18] Yang B, Suo J, Di Giacomo F, et al. Interfacial passivation engineering of perovskite solar cells with fill factor over 82%and outstanding operational stability on nip architecture. ACS Energy Lett. 2021; 6(11):3916-3923. doi: 10.1021/acsenergylett.1c01811

[19] Foster JM, Snaith HJ, Leijtens T, Richardson G. A model for the operation of perovskite-based hybrid solar cells: Formulation, analysis, and comparison to experiment. SIAM J Appl Math. 2014; 74(6):1935-1966. doi: 10.1137/130934258

[20] Hernandez-Balaguera E, Arredondo B, del Pozo G, Romero B. Exploring the impact of fractional-order capacitive behavior on the hysteresis effects of perovskite solar cells: A theoretical perspective. Commun Nonlinear Sci Numer Simul. 2020; 90:105371. doi: 10.1016/j.cnsns.2020.105371

[21] Hunde BR, Woldeyohannes AD. Performance analysis and optimization of perovskite solar cell using SCAPS-1D and genetic algorithm. Mater Today Commun. 2023; 34:105420. doi: 10.1016/j.mtcomm.2023.105420

[22] Danladi E, Gyuk PM, Tasie NN, et al. Impact of hole transport material on perovskite solar cells with different metal electrode: a SCAPS-1D simulation insight. Heliyon. 2023; 9(6). doi: 10.1016/j.heliyon.2023.e16838

[23] Wang Y, Wu J, Zhang P, et al. Stitching triple cation perovskite by a mixed anti-solvent process for high performance perovskite solar cells. Nano Energy. 2017; 39:616-625. doi: 10.1016/j.nanoen.2017.07.046

[24] Slimi B, Mollar M, Assaker IB, et al. Perovskite FA1-xMAxPbI3 for solar cells: films formation and properties. Energy Procedia. 2016; 102:87-95. doi: 10.1016/j.egypro.2016.11.322

[25] Abedini-Ahangarkola H, Soleimani-Amiri S, Rudi SG. Modeling and numerical simulation of high efficiency perovskite solar cell with three active layers. Sol Energy. 2022; 236:724-732. doi: 10.1016/j.solener.2022.03.055

[26] Xiao Z, Song Z, Yan Y. From lead halide perovskites to leadfree metal halide perovskites and perovskite derivatives. Adv Mater. 2019; 31(47):1803792. doi: 10.1002/adma.201803792

[27] Tang Y, Li Z, Nellikkal MA, et al. Improving data and prediction quality of high-throughput perovskite synthesis with model fusion. J Chem Inf Model. 2021; 61(4):1593-1602. doi: 10.1021/acs.jcim.0c01307

[28] Marchenko EI, Fateev SA, Petrov AA, et al. Database of two-dimensional hybrid perovskite materials: open-access collection of crystal structures, band gaps, and atomic partial charges predicted by machine learning. Chem Mater. 2020; 32(17):7383-7388. doi: 10.1021/acs.chemmater.0c02290

[29] Zhou L, Pan S, Wang J, Vasilakos AV. Machine learning on big data: Opportunities and challenges. Neurocomputing. 2017; 237:350-361. doi: 10.1016/j.neucom.2017.01.026

[30] Tufail S, Riggs H, Tariq M, Sarwat AI. Advancements and challenges in machine learning: A comprehensive review of models, libraries, applications, and algorithms. Electronics. 2023; 12(8):1789. doi: 10.3390/electronics12081789

[31] Jacobs R, Liu J, Abernathy H, Morgan D. Machine learning design of perovskite catalytic properties. Adv Energy Mater. 2024; 14(12):2303684. doi: 10.1002/aenm.202303684

[32] Jacobsson TJ, Hultqvist A, Garcia-Fernandez A, et al. An open-access database and analysis tool for perovskite solar cells based on the FAIR data principles. Nat Energy. 2022; 7(1):107-115. doi: 10.1038/s41560-021-00941-3

[33] Kusuma FJ, Widianto E, Santoso I, et al. Optimizing novel device configurations for perovskite solar cells: Enhancing stability and efficiency through machine learning on a large dataset. Renew Energy. 2025; 247:122947. doi: 10.1016/j.renene.2025.122947

[34] Zhao S, Wang J, Guo Z, et al. Exploring device physics of Perovskite solar cell via machine learning with limited samples. J Energy Chem. 2024; 94:441-448. doi: 10.1016/j.jechem.2024.03.003

[35] Iranipour B, Sadeghian M, Mohajerani E. Artificial data generation: A strategy to improve efficiency predictions in mixed Sn-Pb perovskite solar cells. Mater Today Commun. 2025; 43:111625. doi: 10.1016/j.mtcomm.2025.111625

[36] Boubchir M, Boubchir R, Aourag H. The Principal Component Analysis as a tool for predicting the mechanical properties of Perovskites and Inverse Perovskites. Chem Phys Lett. 2022; 798:139615. doi: 10.1016/j.cplett.2022.139615

[37] Li Y, Scheel KR, Clevenger RG, et al. Highly efficient and stable perovskite solar cells using a dopant‐free inexpensive small molecule as the hole‐transporting material. Adv Energy Mater. 2018; 8(23):1801248. doi: 10.1002/aenm.201801248

[38] Fukasawa R, Asahi T, Taniguchi T. Effectiveness and limitation of the performance prediction of perovskite solar cells by process informatics. Energy Adv. 2024; 3(4):812-820. doi: 10.1039/D3YA00617D

[39] Anowar F, Sadaoui S, Selim B. Conceptual and empirical comparison of dimensionality reduction algorithms (pca, kpca, lda, mds, svd, lle, isomap, le, ica, t-sne). Comput Sci Rev. 2021; 40:100378. doi: 10.1016/j.cosrev.2021.100378

[40] Srivastava M, Howard JM, Gong T, et al. Machine learning roadmap for perovskite photovoltaics. J Phys Chem Lett. 2021; 12(32):7866-7877. doi: 10.1021/acs.jpclett.1c01961

[41] Li X, Dan Y, Dong R, et al. Computational screening of new perovskite materials using transfer learning and deep learning. Appl Sci. 2019; 9(24):5510. doi: 10.3390/app9245510

[42] Zhang R, Motes B, Tan S, et al. Machine Learning Prediction of Organic-Inorganic Halide Perovskite Solar Cell Performance from Optical Properties. ACS Energy Lett. 2025; 10(4):1714-1724. doi: 10.1021/acsenergylett.4c03592

[43] Tetko IV, Livingstone DJ, Luik AI. Neural network studies. 1. Comparison of overfitting and overtraining. J Chem Inf Comput Sci. 1995; 35(5):826-833. doi: 10.1021/ci00027a006

[44] Salehin I, Kang DK. A review on dropout regularization approaches for deep neural networks within the scholarlydomain. Electronics. 2023; 12(14):3106. doi: 10.3390/electronics12143106

[45] Dias Da Cruz S, Taetz B, Stifter T, Stricker D. Autoencoder and partially impossible reconstruction losses. Sensors. 2022; 22(13):4862. doi: 10.3390/s22134862

[46] Asperti A, Trentin M. Balancing reconstruction error and kullback-leibler divergence in variational autoencoders. IEEE Access. 2020; 8:199440-199448. doi: 10.1109/ACCESS.2020.3034828

[47] Chen J, Zhan Y, Yang Z, et al. Predicting and analyzing stability in perovskite solar cells: Insights from machine learning models and SHAP analysis. Mater Today Energy. 2025; 48:101769. doi: 10.1016/j.mtener.2024.101769

[48] Priyanga GS, Sampath S, Shravan PV, et al. Advanced prediction of perovskite stability for solar energy using machine learning. Sol Energy. 2024; 278:112782. doi: 10.1016/j.solener.2024.112782

[49] Kumari N, Patel SR, Gohel JV. Enhanced stability and efficiency of Sn containing perovskite solar cell with SnCl2 and SnI 2 precursors. J MaterSci Mater Electron. 2018; 29(21):18144-18150. doi: 10.1007/s10854-018-9926-y

[50] Wang L, Shahiduzzaman M, Muslih EY, et al. Double-layer CsI intercalation into an MAPbI3 framework for efficient and stable perovskite solar cells. Nano Energy. 2021; 86:106135. doi: 10.1016/j.nanoen.2021.106135

[51] Foo S, Thambidurai M, Senthil Kumar P, et al. Recent review on electron transport layers in perovskite solar cells. Int J Energy Res. 2022; 46(15):21441-21451. doi: 10.1002/er.7958

[52] Noh MF, Teh CH, Daik R, et al. The architecture of the electron transport layer for a perovskite solar cell. J Mater Chem C. 2018; 6(4):682-712. doi: 10.1039/C7TC04649A

[53] Sinha NK, Ghosh DS, Khare A. Role of built-in potential over ETL/perovskite interface on the performance of HTL-free perovskite solar cells. Opt Mater. 2022; 129:112517. doi: 10.1016/j.optmat.2022.112517

[54] Szmytkowski J, Galagan Y, Glowienka D. Exploring the interfacial effects at the ETL/perovskite boundary in the semi-transparent perovskite solar cells. Sol Energy. 2023; 266:112176. doi: 10.1016/j.solener.2023.112176

[55] Zhang F, Zhu K. Additive engineering for efficient and stable perovskite solar cells. Adv Energy Mater. 2020; 10(13):1902579. doi: 10.1002/aenm.201902579

[56] Zhang Y, Li Y, Zhang L, et al. Propylammonium chloride additive for efficient and stable FAPbI3 perovskite solar cells. Adv Energy Mater. 2021; 11(47):2102538. doi: 10.1002/aenm.202102538

[57] Zheng Z, Xia M, Chen X, et al. Enhancing the performance of Fa‐based printable mesoscopic perovskite solar cells via the polymer additive. Adv Energy Mater. 2023; 13(23):2204335. doi: 10.1002/aenm.202204335

[58] Gao Y, Wu Y, Lu H, et al. CsPbBr3 perovskite nanoparticles as additive for environmentally stable perovskite solar cells with 20.46% efficiency. Nano Energy. 2019; 59:517-526. doi: 10.1016/j.nanoen.2019.02.070

[59] Gong X, Li M, Shi XB, et al. Controllable perovskite crystallization by water additive for high‐performance solar cells. Adv Funct Mater. 2015; 25(42):6671-6678. doi: 10.1002/adfm.201503559

[60] Wang J, Bi L, Fu Q, Jen AK. Methods for passivating defects of perovskite for inverted perovskite solar cells and modules. Adv Energy Mater. 2024; 14(35):2401414. doi: 10.1002/aenm.202401414

[61] Saidaminov MI, Williams K, Wei M, et al. Multi-cation perovskites prevent carrier reflection from grain surfaces. Nat Mater. 2020; 19(4):412-418. doi: 10.1038/s41563-019-0602-2

[62] Le Z, Liu A, Reo Y, et al. Ion migration in tin-halide perovskites. ACS Energy Lett. 2024; 9(4):1639-1644. doi: 10.1021/acsenergylett.4c00198

[63] Afroz MA, Garai R, Gupta RK, Iyer PK. Additive-assisted defect passivation for minimization of open-circuit voltage loss and improved perovskite solar cell performance. ACS Appl Energy Mater. 2021; 4(10):10468-10476. doi: 10.1021/acsaem.1c01205

[64] Azam M, Liu K, Sun Y, et al. Recent advances in defect passivation of perovskite active layer via additive engineering: A review. J Phys D Appl Phys. 2020; 53(18):183002. doi: 10.1088/1361-6463/ab6f8d

[65] Liu S, Guan Y, Sheng Y, et al. A review on additives for halide perovskite solar cells. Adv Energy Mater. 2020; 10(13):1902492. doi: 10.1002/aenm.201902492

[66] Mahapatra A, Prochowicz D, Tavakoli MM, et al. A review of aspects of additive engineering in perovskite solar cells. J Mater Chem A. 2020; 8(1):27-54. doi: 10.1039/C9TA07657C

[67] Han G, Hadi HD, Bruno A, et al. Additive selection strategy for high performance perovskite photovoltaics. J Phys Chem C. 2018; 122(25):13884-13893. doi: 10.1021/acs.jpcc.8b00980

[68] Bai Y, Xing D, Luo H, et al. Facilitating the formation of SnO2 film via hydroxyl groups for efficient perovskite solar cells. Appl Surf Sci. 2021; 552:149459. doi: 10.1016/j.apsusc.2021.149459

[69] Fu C, Gu Z, Tang Y, et al. From structural design to functional construction: amine molecules in high‐performance formamidinium‐based perovskite solar cells. Angew Chem Int Ed. 2022; 61(19):e202117067. doi: 10.1002/anie.202117067

[70] Lewinska G, Kanak J, Danel KS, et al. Effect of benzene-based dyes on optothermal properties of active layers for ternary organic solar cells. Appl Surf Sci. 2023; 641:158535. doi: 10.1016/j.apsusc.2023.158535

[71] Wang Z, Ma T, Wang J, et al. Surface passivation for efficient and stable perovskite solar cells in ambient air: The structural effect of amine molecules. Ceram Int. 2024; 50(5):7528-7537. doi: 10.1016/j.ceramint.2023.12.058

[72] Rasool S, Khan N, Jahankhan M, et al. Amine-based interfacial engineering in solution-processed organic and perovskite solar cells. ACS Appl Mater Interfaces 2019; 11(18):16785-16794. doi: 10.1021/acsami.9b03298

[73] Akin S, Dong B, Pfeifer L, et al. Organic ammonium halide modulators as effective strategy for enhanced perovskite photovoltaic performance. Adv Sci. 2021; 8(10):2004593. doi: 10.1002/advs.202004593

[74] Yu Y, Wang C, Grice CR, et al. Improving the performance of formamidinium and cesium lead triiodide perovskite solar cells using lead thiocyanate additives. ChemSusChem. 2016; 9(23):3288-3297. doi: 10.1002/cssc.201601027

[75] Zhang Y, Xie J, Tao L, et al. Passivation strategies of Perovskite film defects for solar cells by bifunctional amides with various molecular structures. Org Electron. 2022; 108:106597. doi: 10.1016/j.orgel.2022.106597

[76] Yang S, Wang Y, Liu P, et al. Functionalization of perovskite thin films with moisture-tolerant molecules. Nat Energy. 2016; 1(2):1-7. doi: 10.1038/nenergy.2015.16

[77] Feng W, Tan Y, Yang M, et al. Small amines bring big benefits to perovskite-based solar cells and light-emitting diodes. Chem. 2022; 8(2):351-383. doi: 10.1016/j.chempr.2021.11.010

[78] Wu R, Ding B, Xiao S, et al. Eco-friendly small molecule with polyhydroxyl ketone as buried interface chelator for enhanced carrier dynamics toward high-performance perovskite solar cells. Sci China Mater. 2025; 68(4):1249-1258. doi: 10.1007/s40843-024-3228-1