Pith. sign in

REVIEW 3 major objections 5 minor 82 references

An Investigation into the Effect of Mobile Ions on the Steady State Performance of Perovskite Solar Cells

T0 review · 3 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

Pith's one-line read Mobile ions have only a small effect on the steady-state efficiency of efficient perovskite solar cells, and the effect can be positive or negative.

desk verdict A serious paired simulation study arguing that mobile ions have little steady-state effect in efficient perovskite cells; the headline claim is plausible but rests on a non-random missing-data mechanism that the paper does not model. read the letter →

arxiv 2607.25951 v1 pith:2MKC2CYY submitted 2026-07-28 physics.chem-ph physics.comp-ph

classification physics.chem-phphysics.comp-ph
keywords mobileionsperovskitesolarcellssteady-stateperformancedrift-diffusionsimulationdesignofexperimentsfieldscreeningbuilt-involtageholelifetime
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Using drift-diffusion simulations, the paper compares 601 pairs of perovskite solar cells that are identical in every material parameter except that in one member of each pair the ions are mobile and in the other they are fixed and uniformly distributed. The headline result is that the efficiency distributions of the two populations cannot be statistically distinguished (p = 0.10), and that for efficient cells the presence of mobile ions changes steady-state efficiency by only a few percent relative, with roughly equal numbers of devices helped and hurt. The paper concludes that ion migration is not automatically the main cause of steady-state efficiency loss, and identifies built-in voltage and hole lifetime as the parameters that decide whether mobile ions are neutral, beneficial, or harmful.

What carries the argument

The central tool is paired-device drift-diffusion simulation: for each parameter set, one cell has mobile ions (DEI) and an otherwise identical cell has fixed uniformly distributed ions (SUI). The ion-normalized ratio of each J-V metric isolates the ion effect. A two-level fractional factorial design samples the 32-parameter space, with convergence checks and a non-parametric heteroscedastic test supporting the statistical claims. The mechanism is field screening: ions accumulate at interfaces and screen the built-in field, quantified by ΔV = Vbi − Vmpp, and the paper fits a rational function to predict when screening is harmful or helpful.

What would settle it

Solve the full 1024-pair design with a fixed-ion solver that does not fail on the previously excluded cases and check whether any efficient (SUI PCE > 20%) and high-ion-density pair shows a DEI/SUI PCE ratio outside the paper's observed range; alternatively, in experiment, build two otherwise identical cell sets differing only in ion density (e.g., by halide-vacancy concentration) and measure steady-state MPP; a >10% systematic difference would refute the claim.

Watch

Extended reading notes

Core claim

The paper argues that, at steady state, mobile ions are not inherently harmful in efficient perovskite solar cells. Across 601 pairs of simulated devices identical except for whether ions can move, the power-conversion efficiency distributions of the mobile-ion and immobile-ion sets cannot be distinguished statistically (p = 0.10). The same pairing shows that most devices fall within ±10% efficiency change, poor devices are often made worse, some devices are improved, and the key determinants of harm are a large built-in voltage relative to the maximum-power-point voltage and a short hole lifetime.

Load-bearing premise

The load-bearing premise is that the 423 device pairs excluded because the fixed-ion simulation did not solve are missing at random; if those cases systematically fall in the regimes where mobile ions matter most, the 'no significant difference' conclusion would not hold.

Editorial extensions

If this is right

  • Efficient devices (SUI PCE > 20%) are minimally affected by mobile ions even at ion densities of 1e19 cm^-3, so high ion concentrations do not necessarily preclude high steady-state efficiency.
  • Devices with poor steady-state performance (<15% PCE) are made worse by mobile ions in rough proportion to their underperformance.
  • A large built-in voltage relative to the maximum-power-point voltage (ΔV = Vbi − Vmpp) is the main predictor of ion-induced harm; keeping ΔV below about 0.27 V should make mobile ions neutral or mildly beneficial.
  • Longer hole lifetimes reduce the impact of ion field screening, so improving bulk recombination resilience is a design lever against ion effects.
  • Because the no-difference result is a failure to reject the null, it does not prove equivalence; the paper frames it as showing the effect is not statistically detectable within the studied parameter range.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: If the conclusion transfers to real cells, then efforts to suppress ion migration (e.g., via additives or grain-boundary passivation) may not improve steady-state efficiency as much as assumed; their benefit may lie mainly in stabilising dynamic behaviour and preventing chemical degradation.
  • Editorial inference: The paired-design methodology could be applied to other slow processes (e.g., trap filling or thermal effects) that also alter steady-state J-V curves, using the same ratio statistics.
  • Editorial inference: A direct experimental test would compare steady-state MPP efficiency of two cell types with identical optoelectronic parameters but differing ionic mobility—for example by temperature or composition—and check whether PCE differences track the paper's ΔV threshold.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper uses the open-source drift-diffusion simulator IonMonger with a two-level fractional factorial design over 32 parameters to study how mobile ions affect steady-state solar-cell performance. For each parameter combination, the authors simulate a pair of devices—one with mobile ions (DEI) and one with immobile, uniformly distributed ions (SUI)—and compare their J-V characteristics. Of the 1024 parameter combinations sampled, only 601 yield solvable SUI cases, so the main analysis is based on those 601 complete pairs. The authors report that the DEI and SUI power-conversion-efficiency distributions are not significantly different (Brunner-Munzel p=0.10), that mobile ions tend to slightly reduce Jsc and increase Voc, and that a few poorly performing devices are strongly degraded by ions. They identify the built-in voltage Vbi and hole lifetime τp as the main moderators of ionic impact and propose a rational relationship between ΔV=Vbi−Vmpp and the ion-normalized PCE, from which they derive a design guideline (ΔV<0.27 V) for avoiding negative ionic effects.

Significance. If its central claim holds, the paper is a valuable counterweight to the common assumption that mobile ions necessarily degrade steady-state performance in perovskite solar cells. The study benefits from a well-motivated paired DEI/SUI design, literature-based parameter ranges, and a careful verification that the ultra-slow J-V scans approximate true steady state (reported deviation at MPP is 2.0×10−6% on average). The use of a published open-source model and transparent statistical reporting (Brunner-Munzel test, Bonferroni correction) are additional strengths. However, the headline population comparison and the abstract's more specific claim about efficient devices are both conditional on a substantial, potentially informative missing-data mechanism: 423 of 1024 SUI simulations failed to solve. The paper's own convergence diagnostic is computed only on the included pairs, so it cannot detect bias from deleting entire regions of parameter space. The efficient-device subclaim is also not supported by a dedicated statistical test. For these reasons, the significance of the paper depends on whether the missingness is addressed.

major comments (3)
  1. [Numerical limitations; Figure .5; Figure 4A] The central comparison relies on 601 of 1024 parameter pairs because 423 SUI simulations failed, almost all failures being SUI. The paper argues the remaining pairs are sufficient because correlation coefficients converge (Fig. .1) and because excluded DEI devices have a similar PCE distribution, with a slightly higher mean and smaller variance (Fig. .5). Neither of these checks addresses the missingness mechanism. If SUI non-convergence is more likely for large Vbi and short τp—the parameters that Fig. 4A identifies as the strongest moderators of ion-normalized PCE—then the complete-case sample underrepresents precisely the devices in which mobile ions have the largest negative effect. The convergence of correlations on the included pairs cannot detect this selection bias, and Fig. .5 reports only DEI PCE, not the paired DEI/SUI ratio. The paper should analyze the missingness as a funct
  2. [Impact of Mobile Ions on Steady State Power Conversion; Figure 2A] The abstract's headline claim is that 'in efficient devices, mobile ions have only a small impact on steady-state performance.' The statistical test reported (Brunner-Munzel p=0.10) is applied to the entire 601-device population, not to the efficient-device subset. Figure 2A shows visually that points with SUI PCE > 15% cluster near the y=x line, but no quantitative analysis of this subset is given (e.g., the fraction of efficient devices with |η~−1| > 10%, or a test comparing efficient vs. inefficient devices). Given the paper's emphasis on efficient devices, the authors should either restrict the B-M test (or an equivalent paired analysis) to the efficient-device group, or explicitly state that the population-level result is what supports the conclusion, with the efficient-device statement being a qualitative observation.
  3. [Bias at MPP strongly influences mobile ion impact; Eqs. 8–10] Equations (9) and (10) are presented as predictions, but they are algebraic rearrangements of the four fitted constants a–d in Eq. (8). Specifically, Eq. (9) solves Eq. (8) for ΔV at η~=1, and Eq. (10) is the limit a/c. These do not constitute independent predictions; their validity is entirely inherited from the quality of the fit, which has R²=0.65 and for which no confidence intervals are reported. The statement that a device with ΔV<0.27 V will have 'neutral or positive impact' is therefore an extrapolation from a moderate-correlation fit, not a derived design rule. The authors should present uncertainty bounds on a/c and the ΔV|η~=1 crossing, and ideally validate the rational function on a holdout sample of the same parameter space.
minor comments (5)
  1. [Conclusions] The sentence 'Our results suggest (§ & § ) it is for less efficient devices...' contains unresolved section references ('§ & §'). Please replace with actual section numbers or remove.
  2. [Figure .2 caption] The caption says the VMPP is held static for '1×10−5s', while the main text says '1×10^5 s'. These differ by ten orders of magnitude; please correct the typo.
  3. [Throughout] There are several typos and minor grammatical errors, e.g., 'a abd physics audience' in the Results section, 'through literature review' in the Conclusions, and inconsistent use of SI figure labels ('Figure .1', etc.). A careful proofread is needed.
  4. [Table 1] Some parameters in Table 1 (β, Auger coefficient) are listed with 'N/A' high/low values, meaning they are fixed rather than varied. This is acceptable, but the table would be clearer if a separate column indicated which parameters are varied in the factorial design and which are fixed.
  5. [Statistical reporting] The B-M p-value is reported in the text but not in Figure 2C. Adding the p-value and the test name directly to the figure panel would improve transparency.

Circularity Check

1 steps flagged · score 4.0 of 10

Auxiliary 'predictions' from the fitted rational function are algebraically forced by the fit; the headline simulation comparison is not circular.

  1. fitted input called prediction [Section 'Bias at MPP strongly influences mobile ion impact', Eqs. (8)-(10)]
    "Fitting this relationship finds an R2 of 0.65 and allows us to make several predictions about the relationships between ΔV and η̃ for the devices in this subset. By rearranging our rational function for ΔV when η̃=1... we predict... Finally, we can calculate the negative limit of this function via... giving a value of η̃=1.21."

    The constants a-d in Eq. (8) are least-squares fitted to the subset's (ΔV, η̃) data. Eq. (9) is obtained by setting η̃=1 in Eq. (8) and solving for ΔV; Eq. (10) is the ΔV→−∞ limit a/c. Both are algebraically determined by the already-fitted constants; no new data, independent model, or out-of-sample test is involved. Calling these 'predictions' misrepresents in-sample rearrangements of the fitted curve as forecast tests. Therefore the numerical claims (ΔV<0.27 V, upper bound η̃=1.21) reduce by construction to the same data used to fit the parameters, rather than being independent derivations.

full rationale

The central result—that 601 DEI/SUI paired simulations are statistically indistinguishable (Brunner–Munzel p=0.10)—is a direct output of the drift-diffusion simulations and the factorial design; it is not derived from the assumptions by definition. The DEI/SUI pairing and ion-normalised ratio are legitimate operationalisations, and the use of IonMonger is an external, published code rather than a self-citation forcing the conclusion. The one circular element is in the 'Bias at MPP' section: Eq. (8) is fitted to the subset data, and Eqs. (9)-(10) are then presented as 'predictions' but are simply rearrangements/limits of Eq. (8). Those values (ΔV=0.27 V and asymptotic η̃=1.21) are in-sample consequences of the fitted constants, so they cannot serve as independent tests. This is a partial circularity in auxiliary predictive claims, not in the headline no-significant-difference result. The 'Numerical limitations' discussion also flags a real non-circular weakness: nearly all excluded pairs are SUI solver failures, and the convergence checks are computed only on included pairs, so the missingness assumption is untested. That is a robustness concern, not a circularity.

Assumptions & free parameters 1 free parameters · 7 assumptions · 0 invented entities

The central claim is an output of a simulation code (IonMonger) and a statistical design, so the ledger is dominated by domain assumptions about model fidelity, the meaning of the SUI control, and the representativeness of the sampled parameter ranges. The only fitted quantities are the four constants a, b, c, d of the rational function used for the ΔV design rule; the 32 high/low parameter settings are literature-based inputs, not fitted values.

free parameters (1)
  • a, b, c, d (rational fit constants in Eq. 8) = not reported (R²=0.65)
    Fitted to η̃ vs ΔV for the subset of devices with large Vbi and short τp; Eqs. 9 and 10 are algebraic rearrangements of this fit, so the resulting 'predictions' inherit the fitted constants.
assumptions (7)
  • domain assumption IonMonger's drift-diffusion model (Poisson + continuity for electrons, holes, and ions) accurately describes steady-state PSC operation.
    The entire study is a simulation campaign on this model; no experimental validation is provided in this paper.
  • domain assumption An ultra-slow J-V scan at 1e-5 V/s is equivalent to true steady state for all bias points and all devices.
    Checked by holding at MPP for 1e5 s with deviations below 1e-4%, but the check is itself model-based and performed at one bias point.
  • domain assumption A uniform, immobile ion distribution (SUI) is a valid 'no mobile ions' counterfactual.
    SUI removes all ion redistribution, not just mobility; this is the intended control but is a modeling construct rather than a physically realized device.
  • domain assumption Beer-Lambert absorption with a single wavelength-independent photon flux depending on bandgap (Eq. 3).
    IonMonger's default absorption model; ignores wavelength-dependent absorption profiles and interference effects.
  • domain assumption Built-in voltage is given by Vbi = E_c^el - E_v^hl, and the flat-band potential is close to Vbi.
    Used in Eq. 6 and in defining ΔV (Eq. 7); the authors explicitly flag the flat-band approximation as a caveat in the Conclusions.
  • domain assumption All devices have Ohmic contacts, and the chosen two-level ranges bracket physically relevant PSC parameters.
    Methods state 'all of the simulated devices have Ohmic contacts'; parameter ranges in Table 1 are literature-based but restrict the scope of the claims.
  • standard math The fractional factorial design of resolution 9 allows main effects to be estimated independently of up to 8-factor interactions.
    Standard design-of-experiments theory (ref. 78); needed to interpret the correlation/volcano results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Investigation into the Effect of Mobile Ions on the Steady State Performance of Perovskite Solar Cells." pith.science (2026). https://pith.science/paper/2MKC2CYY

@misc{pith2026260725951,
  author       = {Pith},
  title        = {Pith review of: An Investigation into the Effect of Mobile Ions on the Steady State Performance of Perovskite Solar Cells},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2MKC2CYY}},
  note         = {Machine review of arXiv:2607.25951}
}
read the original abstract

In perovskite solar cells, the interplay between mobile ions and photo-excited charge carriers is complex,with recent studies suggesting mobile ions can be either beneficial or detrimental to cell efficiency depending on the cell properties. In this study we use drift diffusion modelling and factorial analysis to simulate 601 pairs of perovskite solar cells (1202 total cells) across a broad range of physically relevant materials parameters. In each pair of devices, one cell contains mobile ions that can move freely. The second paired cell is identical but has no mobile ions. This approach allows us to systematically investigate the impact of mobile ions on cell performance. We deconvolve the contributions of key cell parameters including ion density, recombination rate and band offsets, for n-i-p and p-i-n devices with both organic and inorganic contact layers. Importantly, we find that in efficient devices, mobile ions have only a small impact on steady-state performance.

Figures

Figures reproduced from arXiv: 2607.25951 by the authors.

Figure 1
Figure 1. An efficient simulation study to deconvolve the influence of parameters and condi￾tions. (A) Schematic showing the distribution of investigated parameters across the layers that comprise a perovskite solar cell, including the Electron Transport Layer, ETL and Hole Transport Layer, HTL. (B) Table illustrating the two-level factorial design of the study. (C) Schematic plots showing the two paradigms investigated: mobi… view at source ↗
Figure 2
Figure 2. The effect of mobile ions is minimal compared to the physically relevant parameter range. (A) Scatter plot of power-conversion efficiencies (PCE) of parameter pairs of devices with mobile ions (DEI) and those without (SUI). The dashed line is a y = x line for visual aid, and r is Pearson’s correlation coefficient. (B) Histogram and kernel density estimation of the PCE of DEI devices, stratified by equilibrium ion de… view at source ↗
Figure 3
Figure 3. Key statistics of solar cell quality are affected by ions. Histograms of ion-normalised (mobile ion simulation / static ion simulation) (A) power-conversion efficiencies (PCE), (B) short-circuit current (Jsc), (C) open-circuit voltage (Voc), and (D) fill-factor (FF). 10 [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Engineering a small built-in voltage and a long hole lifetime is protective against the presence of mobile ions. (A) Scatter plot of the results from calculating Pearson’s correlation coefficient between each of the 32 model variables and the η˜. The Bonferroni correct…
Figure 5
Figure 5. Figure 5: Severe instances of negative mobile ion impact are due to a large built-in voltage relative to the maximum power point. (A) The normalised vacancy density at the electron transport layer interface (ETL) plotted against the operational potential drop ∆V . The blue dashe…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

82 extracted references

  1. [1]

    M.; Barnes, P

    Eames, C.; Frost, J. M.; Barnes, P. R. F.; O’Regan, B. C.; Walsh, A.; Islam, M. S.; O’Regan, B. C.; Walsh, A.; Islam, M. S.Nature Communications2015,6, 7497

  2. [2]

    V.; Deibel, C.; Garnett, E

    Thiesbrummel, J.; Milić, J. V.; Deibel, C.; Garnett, E. C.; Tao, S.; Kirchartz, T.; Guerrero, A.; Cameron, P.; Tress, W.; Saiful Islam, M.; Ehrler, B.Nature Reviews Chemistry2026,10, 179–195

  3. [3]

    J.; Cowley, M.; Valadez-Villalobos, K.; Oskam, G.; Bennett, L

    García-Rodríguez, R.; Riquelme, A. J.; Cowley, M.; Valadez-Villalobos, K.; Oskam, G.; Bennett, L. J.; Wolf, M. J.; Contreras-Bernal, L.; Cameron, P. J.; Walker, A. B.; Anta, J. A.Energy Technology2022, 10, 2200507

  4. [4]

    J.; Walker, A.; Richardson, G.J

    Clarke, W.; Wolf, M. J.; Walker, A.; Richardson, G.J. Phys. Energy2023,5, 025007

  5. [5]

    A.; Wu, Y.; Shen, H.; Barugkin, C.; Beck, F

    Jacobs, D. A.; Wu, Y.; Shen, H.; Barugkin, C.; Beck, F. J.; White, T. P.; Weber, K.; Catchpole, K. R. Physical Chemistry Chemical Physics2017,19, 3094–3103

  6. [6]

    A.; Tress, W.ACS Energy Letters2025,18, 4849

    Torre Cachafeiro, M. A.; Tress, W.ACS Energy Letters2025,18, 4849

  7. [7]

    Hart, L. J. F.; Angus, F. J.; Li, Y.; Khaleed, A.; Calado, P.; Durrant, J. R.; Djurišić, A. B.; Docampo, P.; Barnes, P. R. F.Energy Environ. Sci.2024,17, 7107

  8. [8]

    Richardson, G.; O’Kane, S. E. J.; Niemann, R. G.; Peltola, T. A.; Foster, J. M.; Cameron, P. J.; Walker, A. B.Energy & Environmental Science2016,9, 1476–1485. 14

Show all 82 references
  1. [9]

    E.; Foster, J

    Courtier, N. E.; Foster, J. M.; O’Kane, S. E. J.; Walker, A. B.; Richardson, G.Eur. J. Appl. Math 2019,30, 427–457

  2. [10]

    V.; Wolf, M

    Clarke, W.; Cowley, M. V.; Wolf, M. J.; Cameron, P.; Walker, A.; Richardson, G.Journal of Applied Physics2023,133, 095001

  3. [11]

    M.; Courtier, N

    Cave, J. M.; Courtier, N. E.; Blakborn, I. A.; Jones, T. W.; Ghosh, D.; Anderson, K. F.; Lin, L.; Dijkhoff, A. A.; Wilson, G. J.; Feron, K.; Saiful Islam, M.; Foster, J. M.; Richardson, G.; Walker, A. B. Journal of Applied Physics2020,128, 184501

  4. [12]

    Wu, N.; Walter, D.; Fell, A.; Wu, Y.; Weber, K.The Journal of Physical Chemistry C2020,124, 219–229

  5. [13]

    et al.Advanced Energy Materials2021,11, 2101447

    Thiesbrummel, J. et al.Advanced Energy Materials2021,11, 2101447

  6. [14]

    et al.Nature Energy2024,9, 664

    Thiesbrummel, J. et al.Nature Energy2024,9, 664

  7. [15]

    S.; Cowley, M

    Hill, N. S.; Cowley, M. V.; Gluck, N.; Fsadni, M. H.; Clarke, W.; Hu, Y.; Wolf, M. J.; Healy, N.; Freitag, M.; Penfold, T. J.; Richardson, G.; Walker, A. B.; Cameron, P. J.; Docampo, P.Advanced Materials2023,

  8. [16]

    N.; Noel, N

    Habisreutinger, S. N.; Noel, N. K.; Snaith, H.ACS Energy Letters2018,3, 2472

  9. [17]

    Gong, J.Optik2021,232, 166382

  10. [18]

    M.; Sherkar, T

    Le Corre, V. M.; Sherkar, T. S.; Koopmans, M.; Koster, L. J. A.Cell Reports Physical Science2021, 2, 100346

  11. [19]

    W.; Forfar, L.; Liang, X.; Cameron, P

    Baker, R. W.; Forfar, L.; Liang, X.; Cameron, P. J.Reaction Chemistry & Engineering2021,6, 709

  12. [20]

    M.; Bellany, F.; Benhamou, L.; Bučar, D.-K.; Tabor, A

    Murray, P. M.; Bellany, F.; Benhamou, L.; Bučar, D.-K.; Tabor, A. B.; Sheppard, T. D.Org. Biomol. Chem.2016,14, 2373–2384

  13. [21]

    E.; Cave, J

    Courtier, N. E.; Cave, J. M.; Walker, A. B.; Richardson, G.; Foster, J. M.Journal of Computational Electronics2019,18, 1435–1449

  14. [22]

    J.; Grudeva, Y.; Foster, J

    Clarke, W.; Bennett, L. J.; Grudeva, Y.; Foster, J. M.; Richardson, G.; Courtier, N. E.J Comput Electron2022,

  15. [23]

    W.; Forfar, L.; Liang, X.; Cameron, P

    Baker, R. W.; Forfar, L.; Liang, X.; Cameron, P. J.React. Chem. Eng.2021,6, 709–719

  16. [24]

    C.; Alvarez, A

    Schmidt, M. C.; Alvarez, A. O.; Seid, B. A.; de Boer, J. J.; Lang, F.; Ehrler, B.PRX Energy2025,4, 033017

  17. [25]

    C.; Ehrler, B.ACS Energy Letters2025,10, 2457–2460

    Schmidt, M. C.; Ehrler, B.ACS Energy Letters2025,10, 2457–2460

  18. [26]

    Nelson, J.The Physics of Solar Cells; Imperial College Press, 2003

  19. [27]

    E.; Richardson, G.; Foster, J

    Courtier, N. E.; Richardson, G.; Foster, J. M.Applied Mathematical Modelling2018,63, 329–348

  20. [28]

    A.; Adhikari, R

    Bag, M.; Renna, L. A.; Adhikari, R. Y.; Karak, S.; Liu, F.; Lahti, P. M.; Russell, T. P.; Tuominen, M. T.; Venkataraman, D.J. Am. Chem. Soc.2015,137, 13130–13137

  21. [29]

    J.; Chen, X.; Wen, T

    Tang, L. J.; Chen, X.; Wen, T. Y.; Yang, S.; Zhao, J. J.; Qiao, H. W.; Hou, Y.; Yang, H. G.Chemistry - A European Journal2018,24, 2845–2849

  22. [30]

    T.; Walsh, A.; van Schilfgaarde, M.Physical Review B2014,89, 155204

    Brivio, F.; Butler, K. T.; Walsh, A.; van Schilfgaarde, M.Physical Review B2014,89, 155204

  23. [31]

    H.; Lee, J

    Futscher, M. H.; Lee, J. M.; McGovern, L.; Muscarella, L. A.; Wang, T.; Haider, M. I.; Fakharuddin, A.; Schmidt-Mende, L.; Ehrler, B.Materials Horizons2019,6, 1497–1503. 15

  24. [32]

    Löper, P.; Stuckelberger, M.; Niesen, B.; Werner, J.; Filipič, M.; Moon, S.-J.; Yum, J.-H.; Topič, M.; De Wolf, S.; Ballif, C.Journal Phys. Chem. Letts.2015,6, 66–71

  25. [33]

    Schulz, P.; Edri, E.; Kirmayer, S.; Hodes, G.; Cahen, D.; Kahn, A.Energy & Environmental Science 2014,7, 1377

  26. [34]

    D.; Eperon, G

    Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J.Science2013,342, 341–344

  27. [35]

    A.; Anaya, M.; Chahbazian, R.; Bakr, O

    Stavrakas, C.; Delport, G.; Zhumekenov, A. A.; Anaya, M.; Chahbazian, R.; Bakr, O. M.; Barnard, E. S.; Stranks, S. D.ACS Energy Letters2020,5, 117–123

  28. [36]

    Bou, A.; Abolin,š, H.; Ashoka, A.; Cruanyes, H.; Guerrero, A.; Deschler, F.; Bisquert, J.ACS Energy Letters2021,6, 2248–2255

  29. [37]

    O.; Chen, S.; Gong, X

    Walsh, A.; Scanlon, D. O.; Chen, S.; Gong, X. G.; Wei, S.Angewandte Chemie International Edition 2015,54, 1791–1794

  30. [38]

    C.; Rolston, N.; Xu, J.; Prasanna, R.; O’Regan, B

    Bertoluzzi, L.; Boyd, C. C.; Rolston, N.; Xu, J.; Prasanna, R.; O’Regan, B. C.; McGehee, M. D.Joule 2020,4, 109–127

  31. [39]

    A.; Quarti, C.; An, Q.; Bitton, S.; Tessler, N.; Beljonne, D.; Vaynzof, Y.ACS Energy Letters 2022,7, 3302–3310

    Kress, J. A.; Quarti, C.; An, Q.; Bitton, S.; Tessler, N.; Beljonne, D.; Vaynzof, Y.ACS Energy Letters 2022,7, 3302–3310

  32. [40]

    W.; Koch, S.; Burke, S.; Paranji, R

    deQuilettes, D. W.; Koch, S.; Burke, S.; Paranji, R. K.; Shropshire, A. J.; Ziffer, M. E.; Ginger, D. S. Science2015,348, 683–686

  33. [41]

    S.; Momblona, C.; Gil-Escrig, L.; Koster, L

    Sherkar, T. S.; Momblona, C.; Gil-Escrig, L.; Koster, L. J. A.ACS Energy Letters2017,2, 1214–1222

  34. [42]

    Kirchartz,T.Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences2019,377, 20180286

  35. [43]

    A.; Ball, J

    Duijnstee, E. A.; Ball, J. M.; Le Corre, V. M.; Koster, L. J. A.; Snaith, H. J.; Lim, J.ACS Energy Letters2020,5, 376–384

  36. [44]

    Q.; Chen, S.; Dai, X.; Chen, B.; Hartweg, B.; Yu, Z.; Holman, Z.; Huang, J.Science2020,367, 1352–1358

    Ni, Z.; Bao, C.; Liu, Y.; Jiang, Q.; Wu, W. Q.; Chen, S.; Dai, X.; Chen, B.; Hartweg, B.; Yu, Z.; Holman, Z.; Huang, J.Science2020,367, 1352–1358

  37. [45]

    Yuan, Y.; Yan, G.; Akel, S.; Rau, U.; Kirchartz, T.Science Advances2025,11, eadt1171

  38. [46]

    C.; Taddei, M.; Wang, J.; Eperon, G

    Jariwala, S.; Burke, S.; Dunfield, S.; Shallcross, R. C.; Taddei, M.; Wang, J.; Eperon, G. E.; Arm- strong, N. R.; Berry, J. J.; Ginger, D. S.Chemistry of Materials2021,33, 5035–5044

  39. [47]

    T.; Yan, Y.; Miller, E

    Yang, Y.; Yang, M.; Moore, D. T.; Yan, Y.; Miller, E. M.; Zhu, K.; Beard, M. C.Nature Energy2017, 2, 16207

  40. [48]

    W.; Koch, S.; Burke, S.; Paranji, R

    deQuilettes, D. W.; Koch, S.; Burke, S.; Paranji, R. K.; Shropshire, A. J.; Ziffer, M. E.; Ginger, D. S. ACS Energy Letters2016,1, 438–444

  41. [49]

    W.; Milot, R

    Crothers, T. W.; Milot, R. L.; Patel, J. B.; Parrott, E. S.; Schlipf, J.; Müller-Buschbaum, P.; John- ston, M. B.; Herz, L. M.Nano Letters2017,17, 5782–5789

  42. [50]

    Salim, K. M. M.; Koh, T. M.; Bahulayan, D.; Harikesh, P. C.; Jamaludin, N. F.; Febriansyah, B.; Bruno, A.; Mhaisalkar, S.; Mathews, N.ACS Energy Letters2018,3, 1068–1076

  43. [51]

    Li, M.; Zhao, C.; Wang, Z.; Zhang, C.; Lee, H. K. H.; Pockett, A.; Barbé, J.; Tsoi, W. C.; Yang, Y.; Carnie, M. J.; Gao, X.; Yang, W.; Durrant, J. R.; Liao, L.; Jain, S. M.Advanced Energy Materials 2018,8, 1801509

  44. [52]

    Mukametkali, T.; Ilyassov, B.; Aimukhanov, A.; Serikov, T.; Baltabekov, A.; Aldasheva, L.; Zeinide- nov, A.Physica B: Condensed Matter2023,659, 414784. 16

  45. [53]

    Luo, W.; Wu, C.; Wang, D.; Zhang, Z.; Qi, X.; Guo, X.; Qu, B.; Xiao, L.; Chen, Z.Organic Electronics 2019,74, 7–12

  46. [54]

    C.; Seebauer, E

    Sellers, M. C.; Seebauer, E. G.Thin Solid Films2011,519, 2103–2110

  47. [55]

    S.; Sánchez, J

    Balderrama, V. S.; Sánchez, J. G.; Estrada, M.; Ferré-Borrull, J.; Pallarès, J.; Marsal, L. F.IEEE Journal of Photovoltaics2015,5, 1093

  48. [56]

    M.; Bisquert, J.; Ugarte, I.; Pacios, R.Organic Electronics 2008,9, 847–851

    Garcia-Belmonte, G.; Munar, A.; Barea, E. M.; Bisquert, J.; Ugarte, I.; Pacios, R.Organic Electronics 2008,9, 847–851

  49. [57]

    Liu, Q.; Wei, L.; Yuan, S.; Ren, X.; Zhao, Y.; Wang, Z.; Zhang, M.; Shi, L.; Li, D.; Li, A.RSC Advances 2015,5, 71778–71784

  50. [58]

    F.; Naab, B

    Shang, Z.; Heumueller, T.; Prasanna, R.; Burkhard, G. F.; Naab, B. D.; Bao, Z.; McGehee, M. D.; Salleo, A.Advanced Energy Materials2016,6, 1601149

  51. [59]

    Hochgesang, A.; Biberger, S.; Grüne, J.; Mohanraj, J.; Kahle, F.; Dyakonov, V.; Köhler, A.; The- lakkat, M.Adv Elect Materials2022,8, 2200113

  52. [60]

    R.; Hollman, D

    Abate, A.; Staff, D. R.; Hollman, D. J.; Snaith, H. J.; Walker, A. B.Phys. Chem. Chem. Phys.2014, 16, 1132–1138

  53. [61]

    Zhang, J.; Zhou, P.; Liu, J.; Yu, J.Physical Chemistry Chemical Physics2014,16, 20382–20386

  54. [62]

    Garcia-Belmonte, G.; Boix, P.; Bisquert, J.; Sessolo, M.; Bolink, H.Solar Energy Materials and Solar Cells2010,94, 366–375

  55. [63]

    E.; Tannhauser, D

    Spear, W. E.; Tannhauser, D. S.Physical Review B1973,7, 831–833

  56. [64]

    H.; Zhumagali, S.; T

    Isikgor, F. H.; Zhumagali, S.; T. Merino, L. V.; De Bastiani, M.; McCulloch, I.; De Wolf, S.Nature Reviews Materials2022,8, 89–108

  57. [65]

    A.; Kabanakis, A

    Mikroyannidis, J. A.; Kabanakis, A. N.; Sharma, S. S.; Sharma, G. D.Adv. Funct. Mater.2011,21, 746–755

  58. [66]

    He, Y.; Chen, H.-Y.; Hou, J.; Li, Y.J. Am. Chem. Soc.2010,132, 1377–1382

  59. [67]

    Mi, Y.; Weng, Y.Scientific Reports2015,5, 11482

  60. [68]

    G.; Lin, F.Advanced Energy and Sustainability Research2022,3, 2200045

    Nakka, L.; Cheng, Y.; Aberle, A. G.; Lin, F.Advanced Energy and Sustainability Research2022,3, 2200045

  61. [69]

    L.; Meredith, P.npj Flex Electron2017,1, 13

    Armin, A.; Shoaee, S.; Lin, Q.; Burn, P. L.; Meredith, P.npj Flex Electron2017,1, 13

  62. [70]

    Vacuum2012,86, 2044–2047

    Wang, H.; Wu, G.; Cai, X.; Zhao, Y.; Shi, Z.; Wang, J.; Xia, X.; Dong, X.; Zhang, B.; Ma, Y.; Du, G. Vacuum2012,86, 2044–2047

  63. [71]

    S.; Fukumura, T.; Hirose, Y.; Hasegawa, T.Jpn

    Krasienapibal, T. S.; Fukumura, T.; Hirose, Y.; Hasegawa, T.Jpn. J. Appl. Phys.2014,53, 090305

  64. [72]

    Wypych, A.; Bobowska, I.; Tracz, M.; Opasinska, A.; Kadlubowski, S.; Krzywania-Kaliszewska, A.; Grobelny, J.; Wojciechowski, P.Journal of Nanomaterials2014,2014, 1–9

  65. [73]

    V.; Smakula, A.Journal of Applied Physics1965,36, 2031–2038

    Rao, K. V.; Smakula, A.Journal of Applied Physics1965,36, 2031–2038

  66. [74]

    D.Applied Surface Science2003,218, 318–323

    Stamate, M. D.Applied Surface Science2003,218, 318–323

  67. [75]

    P.; Benko, F

    Koffyberg, F. P.; Benko, F. A.J Electrochem. Soc.1981,128, 2476

  68. [76]

    Nakaoka, K.; Ueyama, J.; Ogura, K.Journal of Electroanalytical Chemistry2004,571, 93–99

  69. [77]

    L.; Meyer, J.; Steirer, K

    Ratcliff, E. L.; Meyer, J.; Steirer, K. X.; Garcia, A.; Berry, J. J.; Ginley, D. S.; Olson, D. C.; Kahn, A.; Armstrong, N. R.Chem. Mater.2011,23, 4988–5000. 17

  70. [78]

    A.Design of Experiments in Chemical Engineering: A Practical Guide; John Wiley & Sons, Ltd, 2004; Chapter 2, pp 262–367

    Lazic, R. A.Design of Experiments in Chemical Engineering: A Practical Guide; John Wiley & Sons, Ltd, 2004; Chapter 2, pp 262–367

  71. [79]

    Wang, X.; Wang, W.; Liu, J.; Qi, J.; He, Y.; Wang, Y.; Hu, W.; Cheng, Y.; Chen, K.; Hu, Y.; Mei, A.; Han, H.Advanced Functional Materials2022,32, 2203872

  72. [80]

    Brunner, E.; Munzel, U.Biometrical Journal2000,42, 17–25

  73. [81]

    A.; Wu, Y.; Duong, T.; Peng, J.; Wen, X.; Fu, X.; Karuturi, S

    Shen, H.; Jacobs, D. A.; Wu, Y.; Duong, T.; Peng, J.; Wen, X.; Fu, X.; Karuturi, S. K.; White, T. P.; Weber, K.; Catchpole, K. R.Journal of Physical Chemistry Letters2017,8, 2672–2680

  74. [82]

    Zhang, D.; Li, D.; Hu, Y.; Mei, A.; Han, H.Communications Materials2022,3, 58. 18 Supplementary Information Figures 0 200 400 600 Number of Devices Sampled 0.8 0.6 0.4 0.2 0.0 0.2 0.4 0.6 Correlation Coefficient Figure .1: Line plot showing the convergence of Pearson’s correla...

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.