Numerical simulations by Bhattacharya and John (Scientific Reports, 2019) predict a power conversion efficiency of about 31% for a flexible 15-micrometer-thick crystalline-silicon solar cell patterned as an inverted-micro-pyramid photonic crystal with interdigitated back contacts, Gaussian front/back surface-field doping profiles, and above-Lambertian wave-interference light trapping. The prediction rests on idealized contacts with zero resistivity, a perfect-electric-conductor back reflector, and a specific fitted model of silicon dispersion and sub-gap absorption near the band edge. The problem is to determine, experimentally and under realistic modeling, whether the predicted beyond-30% efficiency survives real-world implementation: whether the photonic-crystal texture, passivation, doping profiles, and back-contact geometry can be manufactured and integrated into a working device, how much efficiency margin remains under fabrication-induced deviations from the optimized parameters, and how finite contact resistance, realistic metallization, and alternative band-edge absorption data change the final device performance. Later literature has experimentally confirmed the above-Lambertian optical absorption in thin photonic-crystal silicon slabs and independently corroborated the efficiency prediction in simulation, but no thin-film photonic-crystal silicon cell approaching the predicted efficiency has been realized, and the realistic-contact and tolerance-margin questions remain open.

Background

Silicon is an indirect-band-gap semiconductor, so a thin silicon wafer absorbs near-infrared sunlight poorly; thick wafers absorb well but suffer more bulk recombination and are rigid. Light trapping — increasing the path length of weakly absorbed light inside the cell — is therefore central to high-efficiency silicon photovoltaics. The classical benchmark is the Lambertian limit (also called the 4n24n^2 limit, where nn is the refractive index of silicon): the maximum average path-length enhancement achievable with ray-optics, randomly scattering (Lambertian) surfaces. In a widely accepted analysis by Richter, Hermle and Glunz (2013) that combines Lambertian trapping with intrinsic recombination physics — Shockley-Read-Hall (SRH) recombination via bulk defects, Auger recombination at high carrier density, and bandgap narrowing (BGN) — the limiting efficiency of a single-junction crystalline-silicon solar cell is about 29.4% under the standard AM1.5G spectrum (the terrestrial air-mass-1.5 global reference spectrum conventionally used to rate photovoltaic devices), and the best experimental silicon cells have reached only about 27.4% (Green and Zhou, 2025). A photonic crystal (PhC) is a dielectric structure periodic on the wavelength scale of light; wave interference in such structures can trap light in slow, long-lifetime modes and confine near-band-edge photons in ways that ray optics cannot describe, allowing the absorbed-photon density (summarized as the maximum achievable photocurrent density, MAPD) to exceed the Lambertian value. Building on earlier numerical designs by the same group (parabolic-pore PhCs, 2018; 'Towards 30%' thin-silicon PhC cells, 2019), Bhattacharya and John (2019) combined exact finite-difference time-domain (FDTD) solutions of Maxwell's equations for the optical absorption with a state-of-the-art drift-diffusion and Auger-recombination carrier-transport model. Their optimized architecture is a 15-micrometer-thick silicon sheet patterned as a square-lattice array of inverted micro-pyramids whose lattice spacing is comparable to near-infrared wavelengths, with dual anti-reflection coatings, a 50-nanometer silicon-dioxide passivation layer, an interdigitated back contact (IBC) grid that collects carriers at the rear so that the front surface needs no shading metal grid, and Gaussian front and back surface field (FSF/BSF) doping profiles that passivate the surfaces. They report a predicted efficiency of about 31%, above the accepted single-material silicon limit, enabled by absorption beyond the Lambertian limit including sub-gap absorption (bandgap narrowing plus an exponential Urbach tail of phonon-assisted absorption just below the band edge) modeled through a Lorentz fit to experimental silicon dispersion data. The prediction, however, rests on several idealizations: contacts are assumed ideal with zero resistivity; the back reflector is a perfect electric conductor (PEC) rather than a lossy, finite-conductivity metal; and the long-wavelength absorption depends on a specific fitted data set. Follow-up work has partially validated the ingredients: Hsieh, Kaiser, Bhattacharya, John and Lin (2020) measured broadband solar absorption beyond the Lambertian limit in 10-micrometer inverted-pyramid and 'teepee' PhC silicon slabs; Krügener and co-workers (2021), an independent group, simulated a 31.6% maximum efficiency for a 15-micrometer substrate and demonstrated excellent surface passivation of regular inverted-pyramid PhC textures (saturation current densities of a few femtoamperes per square centimeter); Maksimovic and co-workers (2022) demonstrated large-area PhC patterning by laser direct-write and projection lithography; and Peibst and co-workers (2022) designed PhC cells with passivating poly-silicon-on-oxide (POLO) junctions towards 28%. Yet all device-level evidence, including the 27.4% record silicon cells and the IBC heterojunction architecture of Yoshikawa and co-workers (2017) on which the back-contact geometry builds, remains far below the predicted value, and the 29.4% limit is still described in 2025 as 'almost universally regarded' as the silicon ceiling. (A 2025 reanalysis by Kirk argues on detailed-balance grounds, treating thermal as well as luminescent external emission, for a higher 33.4% radiative limit for silicon; that is a revision of the ceiling debate, not a realized cell above 30%.)

Problem Statement

Determine whether the numerically predicted beyond-30% (approximately 31%) power conversion efficiency of the thin-film crystalline-silicon photonic-crystal interdigitated-back-contact solar cell of Bhattacharya and John (2019) is realizable and correct under realistic, non-idealized conditions. This has one common objective — validating the predicted efficiency of this specific architecture end to end — and comprises three parts: (1) experimentally fabricate and characterize the device — the 15-micrometer-thick silicon sheet patterned as a square-lattice inverted-micro-pyramid photonic crystal with lattice spacing comparable to near-infrared wavelengths, dual anti-reflection coatings, passivated surfaces, Gaussian front and back surface-field doping profiles, and an interdigitated back-contact grid — and measure its power conversion efficiency under standard AM1.5G illumination, establishing how close the measured efficiency comes to the prediction and where the losses arise; (2) quantify the efficiency margin under fabrication-induced deviations from the optimized parameters, i.e. how the efficiency degrades with departures from the prescribed photonic-crystal geometry, Gaussian doping profiles, surface recombination velocities (the 2019 analysis delineated rapid degradation above roughly 100 cm/s at contacts), and bulk Shockley-Read-Hall lifetime, over the tolerances achievable in real large-area processing; and (3) establish the effect of replacing the idealized elements with realistic ones — finite contact resistivity and series resistance from real metallization instead of ideal zero-resistance contacts, and a lossy finite-conductivity back reflector instead of the perfect-electric-conductor approximation — together with the robustness of the predicted efficiency when the long-wavelength, near-band-edge silicon absorption is computed from alternative dispersion data or alternative treatments of sub-gap absorption (bandgap narrowing and Urbach tail). The realistic determination of the achievable efficiency and its tolerance margins for this architecture is anchored to, and reproduces within stated tolerances, the experimentally established reference behaviors the prediction relies on — the above-Lambertian photonic-crystal absorption of Hsieh and co-workers (2020) and the photonic-crystal surface passivation of Krügener and co-workers (2021) — rather than merely reproducing the paper's idealized 31% efficiency.

References

  1. Sayak Bhattacharya & Sajeev John (2019). Beyond 30% Conversion Efficiency in Silicon Solar Cells: A Numerical Demonstration. Scientific Reports 9, 12482. DOI: 10.1038/s41598-019-48981-w. https://doi.org/10.1038/s41598-019-48981-w
  2. Sayak Bhattacharya & Sajeev John (2018). Designing High-Efficiency Thin Silicon Solar Cells Using Parabolic-Pore Photonic Crystals. Physical Review Applied 9, 044009. DOI: 10.1103/PhysRevApplied.9.044009. https://doi.org/10.1103/PhysRevApplied.9.044009
  3. Sayak Bhattacharya, Ibrahim Baydoun, Mi Lin & Sajeev John (2019). Towards 30% Power Conversion Efficiency in Thin-Silicon Photonic-Crystal Solar Cells. Physical Review Applied 11, 014005. DOI: 10.1103/PhysRevApplied.11.014005. https://doi.org/10.1103/PhysRevApplied.11.014005
  4. Sayak Bhattacharya & Sajeev John (2020). Photonic crystal light trapping: Beyond 30% conversion efficiency for silicon photovoltaics. APL Photonics 5, 020902. DOI: 10.1063/1.5128664. https://doi.org/10.1063/1.5128664
  5. Mei-Li Hsieh, Alex Kaiser, Sayak Bhattacharya, Sajeev John & Shawn-Yu Lin (2020). Experimental demonstration of broadband solar absorption beyond the Lambertian limit in certain thin silicon photonic crystals. Scientific Reports 10, 11857. DOI: 10.1038/s41598-020-68704-w. https://doi.org/10.1038/s41598-020-68704-w
  6. J. Krügener, M. Rienäcker, S. Schäfer, M. Sanchez, S. Wolter, R. Brendel, S. John, H.J. Osten & R. Peibst (2021). Photonic crystals for highly efficient silicon single junction solar cells. Solar Energy Materials and Solar Cells 233, 111337. DOI: 10.1016/j.solmat.2021.111337. https://doi.org/10.1016/j.solmat.2021.111337
  7. R. Peibst, M. Rienäcker, Y. Larionova, N. Folchert, F. Haase, C. Hollemann, S. Wolter, J. Krügener, P. Bayerl, J. Bayer, M. Dzinnik, R.J. Haug & R. Brendel (2022). Towards 28%-efficient Si single-junction solar cells with better passivating POLO junctions and photonic crystals. Solar Energy Materials and Solar Cells 238, 111560. DOI: 10.1016/j.solmat.2021.111560. https://doi.org/10.1016/j.solmat.2021.111560
  8. Jovan Maksimovic, Jingwen Hu, Soon Hock Ng, Tomas Katkus, Gediminas Seniutinas, Tatiana Pinedo Rivera, Michael Stuiber, Yoshiaki Nishijima, Sajeev John & Saulius Juodkazis (2022). Beyond Lambertian light trapping for large-area silicon solar cells: fabrication methods. Opto-Electronic Advances 5, 210086. DOI: 10.29026/oea.2022.210086. https://doi.org/10.29026/oea.2022.210086
  9. Sayak Bhattacharya & Sajeev John (2024). Light-trapping by wave interference in intermediate-thickness silicon solar cells. Optics Express 32, 29795. DOI: 10.1364/oe.530718. https://doi.org/10.1364/oe.530718
  10. Martin A. Green & Zibo Zhou (2025). Improved silicon solar cells by tuning angular response to solar trajectory. Nature Communications 16, 251. DOI: 10.1038/s41467-024-55681-1. https://doi.org/10.1038/s41467-024-55681-1
  11. D. Akira Engelbrecht & Thomas Tiedje (2021). Temperature and Intensity Dependence of the Limiting Efficiency of Silicon Solar Cells. IEEE Journal of Photovoltaics 11, 73-84. DOI: 10.1109/JPHOTOV.2020.3035115. https://doi.org/10.1109/JPHOTOV.2020.3035115
  12. Armin Richter, Martin Hermle & Stefan W. Glunz (2013). Reassessment of the Limiting Efficiency for Crystalline Silicon Solar Cells. IEEE Journal of Photovoltaics 3, 1184-1191. DOI: 10.1109/JPHOTOV.2013.2270351. https://doi.org/10.1109/JPHOTOV.2013.2270351
  13. Alexander P. Kirk (2025). What is the efficiency limit of a single-junction silicon solar cell? Applied Physics Letters 127, 183504. DOI: 10.1063/5.0299237. https://doi.org/10.1063/5.0299237
  14. Kunta Yoshikawa, Hayato Kawasaki, Wataru Yoshida, Toru Irie, Katsunori Konishi, Kunihiro Nakano, Toshihiko Uto, Daisuke Adachi, Masanori Kanematsu, Hisashi Uzu & Kenji Yamamoto (2017). Silicon heterojunction solar cell with interdigitated back contacts for a photoconversion efficiency over 26%. Nature Energy 2, 17032. DOI: 10.1038/nenergy.2017.32. https://doi.org/10.1038/nenergy.2017.32
  15. Sara M. Almenabawy, Rajiv Prinja & Nazir P. Kherani (2023). Comparison of random upright pyramids and inverted pyramid photonic crystals in thin crystalline silicon solar cells: An optical and morphological study. Solar Energy 256, 88-95. DOI: 10.1016/j.solener.2023.03.045. https://doi.org/10.1016/j.solener.2023.03.045