Neural Network Surrogate Model for Fast Prediction of Sheet Resistance and Transmittance in Polymer-Coated Silver Nanowire Electrodes

Vijayaprabhu A; Elangovan K; C. Priya; M. Karthikeyan

Journal of Polymer & Composites Original Research

Neural Network Surrogate Model for Fast Prediction of Sheet Resistance and Transmittance in Polymer-Coated Silver Nanowire Electrodes

Vijayaprabhu A Department of Electronics and Communication Engineering, Siddharth Institute of Engineering and Technology, Puttur
Elangovan K Department of Electronics and Communication Engineering, Siddharth Institute of Engineering and Technology, Puttur
C. Priya Department of Electronics and Communication Engineering, Siddharth Institute of Engineering and Technology, Puttur
M. Karthikeyan Department of Computer Science Engineering, Siddharth Institute of Engineering and Technology, Puttur

Abstract

Hybrid transparent electrodes that combine silver nanowire (AgNW) networks with conductive PEDOT:PSS overcoats are promising replacements for indium tin oxide in flexible solar cells, but their design requires balancing sheet resistance (Rs) against transmittance (T). Physics-based percolation simulations capture this trade-off but are too slow for large design searches. We present a neural-network surrogate trained on 3,000 Latin-hypercube designs from a Monte Carlo resistor-network simulator in which the polymer acts both as a parallel conductor and as a junction-bridging medium. On unseen data the network predicted log10Rs with R2 = 0.992 (median error 4.5%) and T with R2 = 0.920, and ran 4×102 to 6×104 times faster than the simulator. In repeated cross-validation it matched XGBoost and LightGBM but was less accurate than support-vector and Gaussian-process regression, so the contribution lies in the workflow rather than in the learner. Coupled with a genetic algorithm, the surrogate produced a 100-design Pareto front; re-simulation of six designs gave Rs errors of 0.6–14% and a best figure of merit of 0.034 Ω−1 (T = 81.2%, Rs = 3.7 Ω/sq), about 10% above the best of the 3,000 sampled designs. This advantage persisted (9–18%) when each uncalibrated simulator parameter was perturbed. Nanowire density, diameter and length dominated Rs, whereas polymer thickness and nanowire density dominated T; the polymer lowered Rs by 9–22% at a junction resistance of 50 Ω but by 36–57% at 1000 Ω. Published experimental electrodes lie above the predicted front, which should therefore be read as an optimistic bound. All results are simulation-based and await experimental calibration.

Keywords

References (41)

Wang J, Fan J, Wan T, Hu L, Li Z, Chu D-W. Recent progress in silver nanowire-based transparent conductive electrodes. Adv Energy Sustain Res. 2025;6. doi:10.1002/aesr.202500033.
Oh MJ, Son G-C, Kim M, Jeon J, Kim Y, Son M. An aqueous process for preparing flexible transparent electrodes using non-oxidized graphene/single-walled carbon nanotube hybrid solution. Nanomaterials. 2023;13(15):2249. doi:10.3390/nano13152249.
Maurya S, Labeyrie L, Zimny K, Zheng B, Schumacher S, Muñoz-Rojas D, et al. Recent advances in metallic nanowire based transparent electrodes: from chemistry of metallic nanowires to physics behind the conducting networks. Adv Phys X. 2025;10. doi:10.1080/23746149.2025.2573818.
Wu X, Zhou Z, Wang Y, Li J. Syntheses of silver nanowires ink and printable flexible transparent conductive film: a review. Coatings. 2020;10(9):865. doi:10.3390/coatings10090865.
Manning HG, Gomes da Rocha C, O'Callaghan C, Ferreira MS, Boland JJ. The electro-optical performance of silver nanowire networks. Sci Rep. 2019;9:11550. doi:10.1038/s41598-019-47777-2.
Liu W, Hu Y, Chen Y, Hu Z, Zhou K, Min Z, et al. Improvement of electrical properties of silver nanowires transparent conductive by metal oxide nanoparticles modification. Coatings. 2022;12(12):1816. doi:10.3390/coatings12121816.
Zhang Z, Hu L, Wang R, Zhang S-J, Fu L, Li M-X, et al. Advances in Monte Carlo method for simulating the electrical percolation behavior of conductive polymer composites with a carbon-based filling. Polymers. 2024;16(4):545. doi:10.3390/polym16040545.
Ehsani M, Prosheva M, Heise K, Gilev JB, Tomovska R, Joseph Y. Effects of carbon fillers on electrical and mechanical properties of water-based polymer nanocomposites. Nanomaterials. 2026;16(5):294. doi:10.3390/nano16050294.
Fegade R, Chaudhari R, Nehete R, et al. Comprehensive review of composite materials: classification, manufacturing methods, mechanical behavior, failure modes, and emerging applications. J Polym Compos. 2026;14(1):923–39.
Fegade R, Chaudhari R, Nehete R, et al. Development and mechanical characterization of hybrid natural fiber-reinforced polymer matrix composites for structural applications. J Polym Compos. 2026;14(1):1181–9.
Yadav S, Fegade R, et al. Evaluating fiber volume fraction and defects in fiber reinforced polymer composites for T joint applications. J Polym Compos. 2026;14(1):1222–35.
Berg C, Sahimi M. Percolation and conductivity in evolving disordered media. Phys Rev E. 2023;108(2-1):024132. doi:10.1103/physreve.108.024132.
Keum K, Park SK, Kim Y-H. Predictive modeling of optical and electrical coupling in silver nanowire networks for stretchable transparent electrodes. Nanoscale. 2026;18:4199–206. doi:10.1039/d5nr04505c.
Grazioli D, Dadduzio AC, Roso M, Simone A. Quantitative electrical homogeneity assessment of nanowire transparent electrodes. Nanoscale. 2023;15:6770–84. doi:10.1039/d2nr06564a.
Yeboah D, Asenso EG, Fosuah PB. Exploring new dimensions beyond ITO: a comprehensive study of transparent conductive materials using a novel figure of merit for photovoltaics. Int J Ambient Energy. 2024. doi:10.1080/01430750.2024.2367110.
Lee M, Piper RT, Bhandari B, Hsu JWP. Multiobjective optimization of silver-nanowire deposition for flexible transparent conducting electrodes. ACS Appl Nano Mater. 2023;6(19):17364–8. doi:10.1021/acsanm.3c03599.
Han Y, Ndikumana J, Choi S, Lim D, Kim I, Kim JY, et al. Nondestructive sheet resistance prediction of silver nanowire transparent electrode with convolutional neural network. Sci Rep. 2026;16:11028. doi:10.1038/s41598-026-40528-0.
Xiong W, He X, Huang Z, Wang H, Zeng X, Liu X, et al. Physics-informed graph diffusion model for cross-scale performance prediction of silver nanowire percolation networks. ACS Omega. 2026;11:54465–75.
Choi E, Choi S, An K, Kang K-T. Image-based impedance spectroscopy for printed electronics. npj Flex Electron. 2025;9:11. doi:10.1038/s41528-025-00382-y.
Naftaly M, Das S, Gallop J, Pan K, Alkhalil F, Kariyapperuma D, et al. Sheet resistance measurements of conductive thin films: a comparison of techniques. Electronics. 2021;10(8):960. doi:10.3390/electronics10080960.
Tike RH, Gautam NK, Fegade R, et al. Design and computational analysis of hybrid carbon fiber-enhanced metal matrix composite spur gears. J Polym Compos. 2026;14(1):158–72.
Tike RH, Gautam NK, Fegade R, et al. A systematic review of advanced composite materials for high-performance spur gear applications. J Polym Compos. 2026;14(1):209–20.
Wadate PR, Fegade RS, Gaikwad MN, et al. Design and mathematical quantification of a compressed-air vehicle. J Polym Compos. 2026;14(1):1026–35.
Liu S, Wang H, Peng W, Yao W-J. Surrogate-assisted evolutionary algorithms for expensive combinatorial optimization: a survey. Complex Intell Syst. 2024;10:5933–49. doi:10.1007/s40747-024-01465-5.
Ismailov V. Universality of shallow and deep neural networks on non-Euclidean spaces. arXiv. 2026;abs/2602.23381. doi:10.48550/arxiv.2602.23381.
Yang F, Lin D, Zhou Y-D, He Y. Doubly coupled designs for computer experiments with both qualitative and quantitative factors. Stat Sin. 2022. doi:10.5705/ss.202020.0317.
Salman HA, Kalakech A, Steiti A. Random forest algorithm overview. Babylonian J Mach Learn. 2024;2024:69–79. doi:10.58496/BJML/2024/007.
Rapp F, Roth M. Quantum Gaussian process regression for Bayesian optimization. Quantum Mach Intell. 2023;6. doi:10.1007/s42484-023-00138-9.
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825–30.
Chen T, Guestrin C. XGBoost: a scalable tree boosting system. In: Proc 22nd ACM SIGKDD Int Conf Knowl Discov Data Min. 2016. p. 785–94. doi:10.1145/2939672.2939785.
Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, et al. LightGBM: a highly efficient gradient boosting decision tree. Adv Neural Inf Process Syst. 2017;30:3146–54.
Herman J, Usher W. SALib: an open-source Python library for sensitivity analysis. J Open Source Softw. 2017;2(9):97. doi:10.21105/joss.00097.
Lakshminarayanan B, Pritzel A, Blundell C. Simple and scalable predictive uncertainty estimation using deep ensembles. Adv Neural Inf Process Syst. 2017;30:6402–13.
Smola AJ, Schölkopf B. A tutorial on support vector regression. Stat Comput. 2004;14(3):199–222. doi:10.1023/B:STCO.0000035301.49549.88.
Saltelli A, Annoni P, Azzini I, Campolongo F, Ratto M, Tarantola S. Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index. Comput Phys Commun. 2010;181(2):259–70. doi:10.1016/j.cpc.2009.09.018.
Daniels RK. Finite-size scaling in quasi-three-dimensional stick percolation. Phys Rev E. 2026;114(1-1):014102. doi:10.1103/nr7x-t2kn.
De S, Higgins TM, Lyons PE, Doherty EM, Nirmalraj PN, Blau WJ, et al. Silver nanowire networks as flexible, transparent, conducting films: extremely high DC to optical conductivity ratios. ACS Nano. 2009;3(7):1767–74. doi:10.1021/nn900348c.
Choi DY, Kang HW, Sung HJ, Kim SS. Annealing-free, flexible silver nanowire–polymer composite electrodes via a continuous two-step spray-coating method. Nanoscale. 2013;5(3):977–83. doi:10.1039/C2NR32221H.
Jung ED, Nam YS, Seo H, Lee BR, Yu JC, Lee SY, et al. Highly efficient flexible optoelectronic devices using metal nanowire-conducting polymer composite transparent electrode. Electron Mater Lett. 2015;11(5):906–14. doi:10.1007/s13391-015-5120-z.
Kim YU, Park SH, Nhan NT, Hoang MH, Cho MJ, Choi DH. Optimal design of PEDOT:PSS polymer-based silver nanowire electrodes for realization of flexible polymer solar cells. Macromol Res. 2021;29(1):75–81. doi:10.1007/s13233-021-9005-8.
Guo Z, Li X-L, Li N, Liu X, Li H, Li X-Z, et al. Multilayer directionally arranged silver nanowire networks for flexible transparent conductive films. Phys Chem Chem Phys. 2023. doi:10.1039/d3cp01591b.

Neural Network Surrogate Model for Fast Prediction of Sheet Resistance and Transmittance in Polymer-Coated Silver Nanowire Electrodes

Abstract

Keywords

References (41)

Read more of it

Publish your article

Abstract

Keywords

References (41)

Related articles

Graph Neural Networks for Molecular Scale Property Prediction and Inverse Design of Thermoset Polymer Nanocomposites: A Computational Framework

AI-Driven Inverse Design of Functionally Graded Bio-Nanocomposites for Sustainable High-Barrier Packaging

Polymer-Nanowire Hybrid Composites Based on PEDOT: PSS and AgNWs for High-Performance Conductive Films in Flexible Devices

Polymer-Based Transparent Conductive Composites for Flexible Solar Panel Coatings

A Clinical Study on Comparing the Efficacy of Areca Catechu Mother Tincture with Areca Tea in Control of Blood Sugar Level Among Healthy Individuals

A Comparative Pharmaceutical Analysis on Different Extracts of Areca Catechu by Maceration & Percolation with Areca Tea

Read more of it

Publish your article