Impact of a doping-induced space-charge region on the collection of photo-generated charge carriers in thin-film solar cells based on low-mobility semiconductors

Dorothea Scheunemann; Mathias Nyman; Oskar J. Sandberg; Ronald \"Osterbacka; Sebastian Wilken; Staffan Dahlstr\"om

REVIEW 2 major objections 5 minor 1 references

Impact of a doping-induced space-charge region on the collection of photo-generated charge carriers in thin-film solar cells based on low-mobility semiconductors

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Doping-induced space-charge regions, not field-dependent charge generation, explain the voltage-dependent photocurrents seen in low-mobility solar cells.

desk verdict Clean closed-form photocurrent for doped low-mobility solar cells; theoretical core is solid, experimental fit is suggestive but not fully independent. read the letter →

arxiv 1908.07815 v2 pith:WM46I54K submitted 2019-08-21 physics.app-ph cond-mat.mtrl-sci

Oskar J. Sandberg , Staffan Dahlström , Mathias Nyman , Sebastian Wilken , Dorothea Scheunemann , Ronald Österbacka This is my paper · ORCID

classification physics.app-phcond-mat.mtrl-sci

keywords organicsolarcellsunintentionaldopingspace-chargeregionphotocurrentchargecollectiondiffusionlengthdrift-diffusionsimulationlow-mobilitysemiconductors

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

This paper argues that unintentional p-type doping in low-mobility thin-film solar cells creates a space-charge region near one contact, and that this region, not a change in how light generates charges, is what makes the photocurrent depend on voltage. The authors derive a closed-form expression for the photocurrent: only electrons photogenerated within the depletion width plus a diffusion length are collected, and the rest recombine with dopant-induced holes. The result predicts a collection efficiency that does not change with light intensity at low intensities and a photocurrent that grows with the square root of the reverse voltage. They verify this with drift-diffusion simulations and on solution-processed organic solar cells, and they give a simple plot-based test to tell doping-induced collection limits apart from field-dependent charge generation.

What carries the argument

The load-bearing object is the effective collection length $w + L_n \tanh((d-w)/L_n)$, the distance from the cathode within which photo-generated electrons are gathered. Here $w$ is the efficiently extracting part of the depletion region, $L_n = \sqrt{\mu_n kT \tau/q}$ is the diffusion length of electrons in the neutral region, and $\tau = 1/(\beta N_p)$ is the recombination lifetime against dopant holes. The depletion width is $w_0 = \sqrt{2\varepsilon\varepsilon_0(V_0 - kT/q - V)/(qN_p)}$, and the correction $\Delta w_R = L_D \sqrt{2\ln(1 + L_D^2/L_n^2)}$, with $L_D = \sqrt{\varepsilon\varepsilon_0 kT/(q^2N_p)}$, accounts for recombination at the edge of the depletion zone. This single length ties the voltage dependence, the light-intensity independence, and the apparent electric-field dependence together.

What would settle it

Measure $J_{ph}$ versus reverse voltage at several low light intensities in a doped, low-mobility cell and plot $J_{ph}$ against $\sqrt{V_0 - kT/q - V}$ using independently determined $V_0$: the model predicts the same straight line at every intensity. A slope that changes with light intensity, or a voltage dependence that persists when the hole mobility is high enough that $\mu_p N_p kT/d \gg |J_{ph}|$, would rule out the diffusion-limited depletion model.

Watch

Extended reading notes

Core claim

The central claim is that, in a p-doped active layer with a partially depleted space-charge region of thickness $w_0$, the net photocurrent is $J_{ph} = -qG_L [w + L_n \tanh((d-w)/L_n)]$, with $w = w_0 - \Delta w_R$. Here $w_0 = \sqrt{2\varepsilon\varepsilon_0(V_0 - kT/q - V)/(qN_p)}$, $L_n$ is the electron diffusion length in the neutral region, and $\Delta w_R = L_D \sqrt{2 \ln(1 + L_D^2/L_n^2)}$ corrects for recombination near the edge of the space-charge region. As a result, the collection efficiency is independent of light intensity in the low-intensity regime and the photocurrent has an apparent $\sqrt{V_0 - kT/q - V}$ voltage dependence, which can be mistaken for field-dependent charge generation. The paper shows that this expression matches numerical drift-diffusion simulations and fits the measured photocurrent of a P3HT:PCBM cell once $N_p$ and $V_0$ are known from capacitive extraction measurements.

Load-bearing premise

The derivation assumes the hole conductivity in the neutral region is high enough that the electric field there is negligible, making electron transport purely diffusive; if $|J_{ph}| > \mu_p N_p kT/d$, drift-assisted collection takes over and the simple formula no longer applies.

Editorial extensions

If this is right

In the linear-intensity regime ($n \ll N_p$), the collection efficiency is independent of light intensity, so the photocurrent scales linearly with $G_L$ while its voltage shape remains fixed.
At reverse bias, the photocurrent grows as $\sqrt{V_0 - kT/q - V}$, so a plot of $J_{ph}$ against the depletion width $w_0$ is a straight line, providing a practical fingerprint of doping-limited collection.
Doping-induced collection losses become negligible when the electron diffusion length spans the neutral region, equivalent to $N_p < \mu_n kT/(q\beta d^2)[1 - w_0/d]^{-2}$, a design target for materials with low recombination.
A modest reduction of the majority-carrier (hole) mobility can initially improve collection, because once $|J_{ph}| > \mu_p N_p kT/d$ the neutral-region field assists extraction.
The model gives a way to separate mechanisms: if a field-dependent generation rate is present, the $J_{ph}$ versus $w_0$ plot is nonlinear rather than linear.

Reading between the lines

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

The same square-root diagnostic could be transferred to cells where ionic migration screens the built-in field rather than chemical doping, if the depletion width is measured independently by capacitance methods.
The paper does not push the design implication of Eq. (21): deliberately lowering majority-carrier conductivity could be used as a controlled knob to assist minority-carrier extraction, up to the point where sluggish majority carriers create space-charge-limited photocurrent.
For optically thick cells, the uniform-generation assumption breaks down, so the effective collection length should be weighted by the local generation profile; comparing the predicted voltage law in cells of different thickness would test this extension.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Short version: this is a solid analytical paper. The key deliverable is Eq. (14) — a closed form for the photocurrent in a p-doped low-mobility solar cell — with the recombination correction Eq. (18). The derivation is clean, the parameter range is spelled out, and the comparison with numerical drift-diffusion simulations is convincing. The paper earns its place in the OPV literature even if the experimental section is the least tight part.

What's new: previous work (Kirchartz, Dibb, Deledalle, Gärtner) treated doping-induced space-charge effects numerically or experimentally, but not this closed form. The J_ph vs w0 diagnostic is a practical tool for distinguishing doping-limited collection from field-dependent generation. The discussion around Eq. (21) — when the neutral-region field is negligible — is genuinely helpful; the paper is honest that outside that regime Eq. (14) stops being valid and drift-assisted collection takes over.

Where I have reservations: the experimental demonstration is a fit, not a prediction. L_n and G_L are free, and there are no error bars. The check that the 0.06-sun data lie in the diffusion regime leans on a hole mobility from the literature (Ref. [76]) rather than measured on the same device. The stress-test point is valid: if the actual mobility were a factor of ~20 lower, the data would sit in the drift-assisted regime and the extracted L_n = 80 nm would not be a literal diffusion length. That doesn't threaten the analytic core, which is independently supported by the simulations, but it does mean the experiment should be read as a consistency check, not a quantitative extraction.

Minor: the derivation assumes perfectly selective contacts and uniform generation; the paper addresses both in the supplement, so this is not a hidden assumption.

Would I cite it? Yes. The analytical expression and the diagnostic are directly useful. Would I send it to a serious referee? Yes. It deserves peer review and will probably be accepted; the experimental caveats should be acknowledged in the final text.

Referee Report

2 major / 5 minor

Summary. This paper presents an analytical model for the steady-state photocurrent in a thin-film solar cell with a p-doped active layer. For a device with a space-charge region of width w0 at the cathode and a quasi-neutral region where the majority-carrier conductivity is large, the authors derive J_ph = -q G_L [w + L_n tanh((d-w)/L_n)] with w = w0 - Δw_R, where Δw_R accounts for recombination near the space-charge-region edge. The expression is verified against 1D drift-diffusion simulations over varying doping concentration, mobility, and light intensity, and is used to fit experimental J-V curves of a P3HT:PCBM solar cell at 0.06 suns. The paper further derives a condition for the validity of the diffusion-limited regime and discusses criteria for avoiding doping-induced collection losses.

Significance. The analytical result is a useful and compact description of an important loss mechanism in low-mobility solar cells. It explains the apparent sqrt(V) dependence of the photocurrent without invoking field-dependent generation, and it offers a diagnostic plot (J_ph versus w0) to distinguish doping-induced collection losses from other mechanisms. The numerical verification is thorough and the derivation is internally consistent. The experimental section is a reasonable demonstration, though the validity check of the diffusion-limited regime relies on a literature value for the hole mobility and on an optically-thin assumption that is not explicitly justified for the measured device.

major comments (2)

[Section III.4 (Experimental comparison), Eq. (21)] The condition |J_ph| < μ_p N_p kT/d is used to justify that the experimental data at 0.06 suns are in the diffusion-limited regime, but μ_p = 3 x 10^-4 cm^2/Vs is taken from Ref. [76] rather than measured on the device under test. Because hole mobilities in P3HT:PCBM can vary by more than an order of magnitude with processing conditions, a factor-of-20 lower mobility would reduce the threshold μ_p N_p kT/d to roughly 0.35 mA/cm^2, comparable to the measured J_SC = 0.32 mA/cm^2, placing the fitted data outside the validity range of Eq. (14). The authors should either measure μ_p on the fitted device, perform a sensitivity analysis of the fitted L_n and G_L to plausible μ_p values, or explicitly phrase the experimental verification as conditional on the assumed mobility. As it stands, the experimental demonstration does not independently establish the regime in which Eq. (14) applies.
[Section II and Section III.4 (model assumptions and experiment)] The analytical expression (14) is derived under the assumption of a uniform generation rate inside an optically thin active layer. The experimental device has a 250 nm thick P3HT:PCBM layer, which is not optically thin at the peak absorption wavelengths (absorption length is of order 100 nm or less). The authors do not discuss how a non-uniform generation profile affects the validity of fitting Eq. (14) to the measured photocurrent, nor do they correct for it. Because the fitted G_L and L_n depend on the assumed generation profile, the experimental fit is questionable unless a uniform-generation profile is demonstrated or a non-uniform generalization is applied.

minor comments (5)

[Supplemental Material, Eq. (S1) caption] The phrase 'q is the the elementary charge' contains a typo; it should read 'q is the elementary charge'.
[Fig. 7(b) and surrounding text] The experimental photocurrent data in Fig. 7(b) are normalized to the fitted J_G; showing absolute current densities (or reporting the fitted J_G value) would allow the reader to assess the quality of the fit quantitatively.
[Section III.1, Eq. (16)] The expression for the dark saturation current J_0 = q G_th [w + L_n tanh((d-w)/L_n)] is introduced abruptly; a sentence explaining that it follows from evaluating |J_ph| in Eq. (14) at G_L = G_th would improve clarity.
[Section III.2, discussion of high mobilities] When stating that the current saturates to J_ph = -J_G at high mobilities, the authors should note that this presumes selective contacts; for non-selective contacts Eq. (S2) gives a lower saturation value, which the text mentions only later.
[Introduction and Section I] The introduction would benefit from a more explicit statement of how the present analytical result differs from or extends the earlier numerical and experimental studies of doping in organic solar cells (Refs. [34,47-49]), to help the reader identify the new contribution.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: Eq. (14) is derived from the drift-diffusion equations with stated assumptions; the experimental fit of L_n and G_L is explicitly declared, and the self-citations are peripheral.

full rationale

The central analytical result, Eq. (14), is derived from the stated drift-diffusion and continuity equations: in the neutral region the electron current is taken as diffusive, recombination is linearized as R ≈ β N_p n, and the boundary conditions are stated. The depletion width w0 comes from the standard Poisson/electrostatics expression, and the correction Δw_R is obtained from an approximate boundary-matching argument. No fitted parameter is introduced into the derivation; L_n and τ are defined from material parameters. The numerical verification uses the same drift-diffusion model, so it is a consistency check rather than an independent benchmark, but that is not circularity. In the experimental section, the paper explicitly says the J-V curve at 0.06 suns is "fitted using the analytical expression in Eq. (14) with L_n and G_L as fitting parameters," so the experimental demonstration is an honest fit, not a disguised prediction. The self-cited CELIV method and the hole-mobility value from Ref. [76] are used only to support the claim that the fitting data lie in the diffusion-dominated regime; they are external inputs, not outputs of the model, and the theoretical derivation does not depend on them. No load-bearing circular step is present.

Assumptions & free parameters 2 free parameters · 7 assumptions · 0 invented entities

The analytical derivation introduces no free parameters; its inputs are material parameters (mu, beta, N_p, d, epsilon). The experimental demonstration fits two parameters (L_n, G_L) to one J-V curve. The derivation rests on standard drift-diffusion and depletion approximations, plus one ad-hoc matching approximation for the recombination correction. No new physical entities are postulated.

free parameters (2)

L_n (electron diffusion length in neutral region) = 80 nm
Fitted to the experimental J-V curve of the P3HT:PCBM device at 0.06 suns using Eq. (14); used to validate the analytical model. No uncertainty reported.
G_L (photo-generation rate) = 1.6e20 cm-3/s at 0.06 suns
Fitted as a scale factor to the same experimental J-V curve; no independent absorption measurement is given.

assumptions (7)

domain assumption Drift-diffusion equations (Eqs. 3-7) and the Einstein relation govern carrier transport in the active layer.
Starting point for the analytical derivation and the numerical model (Sec. II).
domain assumption The photo-generation rate G_L is uniform and independent of the electric field.
Assumed throughout (Sec. II); the authors note field-dependent generation is weak in P3HT:PCBM.
domain assumption The depletion approximation with an abrupt space-charge-region boundary and w0 given by Eq. (2).
Standard Schottky-junction depletion width; assumes uniform ionized doping and 0 < w0 < d.
domain assumption In the neutral region, recombination is first-order with R = beta N_p n, i.e., the hole density is pinned at N_p and n << N_p.
Yields Eq. (12) and the intensity-independent collection efficiency; valid only at low light intensities.
ad hoc to paper The diffusive hole tail in the SCR is approximated by p(x) almost equal to N_p exp(-(x-d+w0)^2/(2 L_D^2)), with the matching condition at x=d-w used to derive Delta_w_R.
A specific functional form and matching condition (Eqs. 17-19) introduced for tractability, not derived from a more fundamental principle.
domain assumption Contacts are perfectly selective: J_n(0)=0 and J_p(d)=0.
Used to solve Eq. (12) and obtain Eq. (14); the supplemental addresses the non-selective case (Eq. S2).
domain assumption Majority-carrier conductivity in the neutral region is high enough that the electric field there is negligible (|F| << kT/(q d)).
Required for the diffusion-only transport in Eq. (12); the paper gives the validity condition Eq. (21) and shows deviations in Fig. 5 when it is violated.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Impact of a doping-induced space-charge region on the collection of photo-generated charge carriers in thin-film solar cells based on low-mobility semiconductors." pith.science (2026). https://pith.science/paper/WM46I54K

@misc{pith2026190807815,
  author       = {Pith},
  title        = {Pith review of: Impact of a doping-induced space-charge region on the collection of photo-generated charge carriers in thin-film solar cells based on low-mobility semiconductors},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WM46I54K}},
  note         = {Machine review of arXiv:1908.07815}
}

read the original abstract

Unintentional doping of the active layer is a source for lowered device performance in organic solar cells. The effect of doping is to induce a space-charge region within the active layer, generally resulting in increased recombination losses. In this work, the impact of a doping-induced space-charge region on the current-voltage characteristics of low-mobility solar cell devices has been clarified by means of analytical derivations and numerical device simulations. It is found that, in case of a doped active layer, the collection efficiency of photo-generated charge carriers is independent of the light intensity and exhibits a distinct voltage dependence, resulting in an apparent electric-field dependence of the photocurrent. Furthermore, an analytical expression describing the behavior of the photocurrent is derived. The validity of the analytical model is verified by numerical drift-diffusion simulations and demonstrated experimentally on solution-processed organic solar cells. Based on the theoretical results, conditions of how to overcome charge collection losses caused by doping are discussed. Furthermore, the presented analytical framework provides tools to distinguish between different mechanisms leading to voltage dependent photocurrents.

Figures

Figures reproduced from arXiv: 1908.07815 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

1 extracted references · 1 canonical work pages

[1]

[S1] O. J. Sandberg, M. Nyman, R. Österbacka, Direct determination of doping concentration and built-in voltage from extraction current transients, Org. Electron. 15, 3413 (2014)

work page 2014

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

[S1] O. J. Sandberg, M. Nyman, R. Österbacka, Direct determination of doping concentration and built-in voltage from extraction current transients, Org. Electron. 15, 3413 (2014)

work page 2014