pith. sign in

arxiv: 2606.13114 · v1 · pith:XR6N7M3Mnew · submitted 2026-06-11 · ❄️ cond-mat.mtrl-sci · physics.app-ph

Disentangling the origin of degradation in perovskite solar cells via optical imaging and Bayesian inference

Akash Dasgupta , Robert D. J. Oliver , Manuel Kober-Czerny , Charlie H. G. Nicholls , Xueli Cao , Yen-Hung Lin , Alexandra J. Ramadan , Henry J. Snaith This is my paper

Pith reviewed 2026-06-27 06:32 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci physics.app-ph
keywords perovskite solar cellsdegradation mechanismsphotoluminescence imagingBayesian inferencedrift diffusion modelinginterface stabilityoptoelectronic parameters
0
0 comments X

The pith

Bayesian inference on photoluminescence images identifies transport layer interfaces as the main source of degradation in perovskite solar cells.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper introduces a method that combines photoluminescence imaging with computer simulations and statistical inference to create detailed maps of how recombination occurs in complete perovskite solar cells. These maps are tracked over time as the cells age under sunlight and heat, allowing separation of effects at the interfaces with charge transport layers from those in the perovskite layer itself. A reader would care because this identifies exactly where degradation starts, which is key to making longer-lasting solar devices. The work shows that degradation is uneven across the cell and mostly happens at the interfaces, and that a chemical treatment can reduce it by acting at those interfaces.

Core claim

The central claim is that degradation in perovskite solar cells is spatially heterogeneous and primarily originates from the perovskite/transport layer interfaces rather than the bulk, as revealed by tracking inferred recombination parameter maps during aging. The method distinguishes between hole transport layer, electron transport layer, and bulk contributions using Bayesian inference on photoluminescence data. An amino-silane passivation treatment is demonstrated to suppress the interface-specific degradation.

What carries the argument

Inferred parameter maps from Bayesian inference on photoluminescence images, generated using drift-diffusion simulations, which disentangle recombination at the two transport layer interfaces and in the bulk.

If this is right

  • Recombination parameter maps evolve differently in different regions of the cell during aging.
  • The strongest degradation signals come from the interfaces with the hole and electron transporting layers.
  • Amino-silane molecular passivation reduces the observed interface degradation.
  • The approach enables tracking of individual recombination processes over time in operating devices.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The technique could help optimize fabrication processes by identifying which steps affect interface stability most.
  • Similar imaging and inference methods might apply to diagnosing issues in other thin-film solar technologies.
  • Prioritizing interface modifications may lead to more stable perovskite devices than bulk improvements alone.

Load-bearing premise

The drift-diffusion model correctly represents the main ways charge carriers recombine, enabling the inference to accurately separate contributions from interfaces and bulk.

What would settle it

A direct measurement showing that interface recombination rates do not change as predicted by the inferred maps after aging, or that the passivation treatment affects bulk parameters instead.

read the original abstract

Machine learning and computational inference, coupled with experimental data, promise to significantly accelerate our rate of learning in most scientific disciplines. In this study, we develop tools that connect microscopic observations to macroscopic device behaviour, a capability that is essential for accelerating the design of durable energy materials. To this end, we introduce a novel approach that integrates photoluminescence imaging with drift diffusion simulations to understand operation and degradation in fully fabricated perovskite solar cells. By employing Bayesian inference, we generate "inferred maps" of parameters that govern recombination processes present in devices. We track these parameter maps while the devices are aged (70 {\deg}C, full spectrum sunlight) to analyse their temporal evolution during degradation. Notably, our approach allows us to distinguish between degradation occurring at the hole or electron transporting layer interface, or within the bulk. Our analysis reveals pronounced spatially non-uniform degradation, with significant macroscopic heterogeneity observed in the optoelectronic parameter maps. We pinpoint the greatest degradation observed in specific regions to stem from the perovskite/transport layer interfaces. Finally, we demonstrate that an amino-silane molecular passivation treatment suppresses this degradation, highlighting its specific role in enhancing device stability. Our approach offers valuable insights for future device fabrication and is a clear exemplification of how advanced Bayesian inference can significantly increase the value of experimental data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces a method that couples photoluminescence imaging of operating perovskite solar cells with drift-diffusion simulations and Bayesian inference to produce spatially resolved maps of recombination parameters. These maps are tracked during accelerated aging (70°C, full-spectrum sunlight) to claim that degradation can be attributed specifically to the perovskite/ETL interface, perovskite/HTL interface, or bulk, revealing macroscopic spatial heterogeneity with the strongest degradation at the interfaces; an amino-silane passivation treatment is shown to suppress the observed interface degradation.

Significance. If validated, the approach would enable non-destructive, spatially resolved identification of degradation sources in complete devices, directly informing fabrication strategies for improved stability. The explicit use of Bayesian inference on physics-based forward models to extract more information from experimental images is a clear methodological strength.

major comments (2)
  1. [Methods (Bayesian inference workflow)] The manuscript contains no description of a synthetic-data recovery test (forward simulation of known interface vs. bulk recombination maps to PL images, followed by inference recovery) in the section describing the Bayesian workflow. This test is required to establish that the drift-diffusion model produces distinguishable signatures under the experimental geometry, which is load-bearing for the abstract claim that degradation can be pinpointed to specific interfaces rather than the bulk.
  2. [Results (parameter maps during aging)] No quantitative validation metrics, posterior uncertainties, or comparison against independent measurements (e.g., electrical JV parameters or other spatially resolved probes) are reported for the inferred recombination maps in the results on aged devices. Without these, the attribution of greatest degradation to the transport-layer interfaces remains unsupported by visible data.
minor comments (1)
  1. [Abstract] The abstract contains the unrendered LaTeX fragment '70 {°}C'; this should be corrected to 70 °C for clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which highlight important aspects of methodological validation. We address each major comment below and will revise the manuscript accordingly to strengthen the presentation.

read point-by-point responses

  1. Referee: [Methods (Bayesian inference workflow)] The manuscript contains no description of a synthetic-data recovery test (forward simulation of known interface vs. bulk recombination maps to PL images, followed by inference recovery) in the section describing the Bayesian workflow. This test is required to establish that the drift-diffusion model produces distinguishable signatures under the experimental geometry, which is load-bearing for the abstract claim that degradation can be pinpointed to specific interfaces rather than the bulk.

    Authors: We agree that a synthetic-data recovery test is necessary to confirm that the model can distinguish interface versus bulk recombination under the experimental conditions. Although the manuscript emphasizes application to measured devices, we will add a new subsection to the Methods describing forward simulations of known recombination maps (interface and bulk) to generate synthetic PL images, followed by Bayesian recovery. This will explicitly demonstrate distinguishability and support the claims. revision: yes

  2. Referee: [Results (parameter maps during aging)] No quantitative validation metrics, posterior uncertainties, or comparison against independent measurements (e.g., electrical JV parameters or other spatially resolved probes) are reported for the inferred recombination maps in the results on aged devices. Without these, the attribution of greatest degradation to the transport-layer interfaces remains unsupported by visible data.

    Authors: We acknowledge that quantitative validation metrics and uncertainty quantification would provide stronger support. In the revised manuscript we will add posterior uncertainty maps, goodness-of-fit metrics between simulated and experimental PL images, and direct comparisons to measured JV parameters before and after aging. These additions will make the interface-degradation attribution more robust while preserving the observed spatial heterogeneity and temporal trends already shown. revision: yes

Circularity Check

0 steps flagged

No circularity: inference uses external PL images and independent physics model

full rationale

The workflow takes measured photoluminescence images as input and employs a drift-diffusion forward model (with fixed assumptions on mobilities, lifetimes, and boundary conditions) inside Bayesian inference to produce posterior maps of interface vs. bulk recombination parameters. No equation defines the output maps in terms of the fitted parameters themselves, nor renames a fitted quantity as a prediction. The central claim that degradation is interface-dominated rests on the model's ability to produce distinguishable signatures, which is an external modeling assumption rather than a self-referential definition or self-citation chain. The paper is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the physical model and the ability of the imaging data to constrain the inferred parameters; no new particles or forces are introduced.

free parameters (1)
  • recombination rate parameters
    Inferred via Bayesian inference; their values are not known a priori and are updated from the photoluminescence data.
axioms (1)
  • domain assumption Drift-diffusion equations accurately describe charge transport and recombination throughout the device stack.
    Invoked to generate the simulated photoluminescence images used in the inference step.

pith-pipeline@v0.9.1-grok · 5802 in / 1240 out tokens · 23774 ms · 2026-06-27T06:32:04.315867+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

8 extracted references · 3 canonical work pages

  1. [1]

    1 M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647. 2 H.-S. Kim, C.-R. Lee, J.-H. Im, K.-B. Lee, T. Moehl, A. Marchioro, S. -J. Moon, R. Humphry -Baker, J.-H. Yum, J. E. Moser, M. Grätzel and N.-G. Park, Sci. Rep., 2012, 2,

    2012

  2. [2]

    4 Best Research-Cell Efficiency Chart, https://www.nrel.gov/pv/cell-efficiency.html. 5 C. Kamaraki, M. T. Klug, T. Green, L. Miranda Perez and C. Case, Appl. Phys. Lett. , DOI:10.1063/5.0054086. 19 6 N. M. Haegel, P. Verlinden, M. Victoria, P. Altermatt, H. Atwater, T. Barnes, C. Breyer, C. Case, S. De Wolf, C. Deline, M. Dharmrin, B. Dimmler, M. Gloeckle...

    work page doi:10.1063/5.0054086 2023

  3. [3]

    13 H. Zhu, S. Teale, M. N. Lintangpradipto, S. Mahesh, B. Chen, M. D. McGehee, E. H. Sargent and O. M. Bakr, Nature Reviews Materials, 2023, 8, 569–586. 14 D. Zhang, D. Li, Y. Hu, A. Mei and H. Han, Communications Materials, 2022, 3, 1–14. 15 C. C. Boyd, R. Cheacharoen, T. Leijtens and M. D. McGehee, Chem. Rev., 2019, 119, 3418–3451. 16 A. J. Ramadan, R. ...

    2023

  4. [4]

    25 A. M. Soufiani, Z. Hameiri, S. Meyer, S. Lim, M. J. Y. Tayebjee, J. S. Yun, A. Ho -Baillie, G. J. Conibeer, L. Spiccia and M. A. Green, Adv. Energy Mater., 2017, 7, 1602111. 26 T. Trupke, E. Pink, R. A. Bardos and M. D. Abbott, Appl. Phys. Lett., 2007, 90, 093506. 27 K. Frohna, M. Anaya, S. Macpherson, J. Sung, T. A. S. Doherty, Y. -H. Chiang, A. J. Wi...

    2017

  5. [5]

    35 D. W. DeQuilettes, S. M. Vorpahl, S. D. Stranks, H. Nagaoka, G. E. Eperon, M. E. Ziffer, H. J. Snaith and D. S. Ginger, Science, 2015, 348, 683–686. 36 F. Akrami, F. Jiang, R. Giridharagopal and D. S. Ginger, arXiv [physics.app-ph],

    2015

  6. [6]

    Lin, Vikram, F

    37 Y.-H. Lin, Vikram, F. Yang, X. -L. Cao, A. Dasgupta, R. D. J. Oliver, A. M. Ulatowski, M. M. McCarthy, X. Shen, Q. Yuan, M. G. Christoforo, F. S. Y. Yeung, M. B. Johnston, N. K. Noel, L. M. Herz, M. S. Islam and H. J. Snaith, Science, 2024, 384, 767–775. 38 G. L. Bretthorst, in Maximum-Entropy and Bayesian Methods in Science and Engineering: Foundation...

    2024

  7. [7]

    50 N. E. Courtier, J. M. Cave, J. M. Foster, A. B. Walker and G. Richardson, Energy Environ. Sci., 2019, 12, 396–409. 51 N. E. Courtier, Phys. Rev. Applied, 2020, 14, 024031. 52 L. J. Bennett, A. J. Riquelme, J. A. Anta, N. E. Courtier and G. Richardson, Phys. Rev. Appl., 2023, 19, 014061. 53 P. Calado, D. Burkitt, J. Yao, J. Troughton, T. M. Watson, M. J...

    work page doi:10.1002/aenm.202000502 2019

  8. [8]

    64 S. Zhou, B. M. Gallant, J. Zhang, Y. Shi, J. Smith, J. N. Drysdale, P. Therdkatanyuphong, M. Taddei, D. P. McCarthy, S. Barlow, R. C. Kilbride, A. Dasgupta, A. R. Marshall, J. Wang, D. J. Kubicki, D. S. Ginger, S. R. Marder and H. J. Snaith, J. Am. Chem. Soc., DOI:10.1021/jacs.4c06659. 65 D. Nguyen Minh, M. A. Haque, F. Yang, S. P. Harvey, R. A. Kerner...

    work page doi:10.1021/jacs.4c06659 2025