Quantitative mechanism of polarization-induced carrier-lifetime enhancement in ferroelectric photoabsorbers and the causal role of ferroelectricity in organohalide perovskites

Morris et al. (2016) showed by transient absorption spectroscopy that single-crystal BaTiO3, a prototypical ferroelectric, sustains photogenerated carriers for ~0.12 s in the absence of bias or scavengers, about $10^4$ times longer than non-ferroelectric (cubic) BaTiO3, and interpreted this via a thermodynamic model in which temperature-dependent spontaneous-polarization-induced band bending creates a thermal barrier $E_b \approx 0.4$ eV to electron-hole recombination. The quantitative mechanism behind this effect remains unresolved: the deep-trap states and tunneling pathways tentatively responsible for the slow, temperature-independent logarithmic decay phase are unidentified; the actual barrier height under realistic, illuminated, adsorbate-screened conditions is unknown (literature values are vacuum values); the crossover between polarization-limited and trap-limited recombination above the Curie temperature is unquantified; the barrier model and its TAS-based extraction are unvalidated beyond single-crystal BaTiO3; and the hypothesis that ferroelectric domains cause the long, non-Langevin carrier lifetimes of organohalide lead perovskites is still untested directly and increasingly contested. The problem is to establish, quantitatively and mechanistically, how spontaneous-polarization-induced internal fields control recombination in ferroelectric semiconductors and whether this mechanism extends to the perovskites.

Background

A ferroelectric material possesses a permanent spontaneous polarization $P_s$ below its Curie temperature $T_\mathrm{C}$, above which it becomes non-polar. BaTiO3 is the prototypical case: tetragonal and ferroelectric at room temperature, cubic and non-ferroelectric above $T_\mathrm{C}\approx 120\,^\circ$C. At a free surface the polarization charge is incompletely compensated, producing a depolarization field and band bending of up to several hundred meV (reported values near 1.3 V in total, or about 0.65 V per surface, measured under vacuum) that drives electrons and holes toward opposite surfaces — the same physical scale as the junction fields used in photovoltaic and photoelectrochemical cells. Under ambient conditions the bending is reduced by external screening through adsorbed molecules, and illumination can partially remove that screening.

Transient absorption spectroscopy (TAS) monitors the decay of the photoinduced optical absorption of photogenerated electrons and holes after a pulsed excitation, and thereby reports recombination and trapping kinetics on microsecond-to-second timescales. Morris, Pendlebury, Hong, Dunn and Durrant (Adv. Mater. 2016) applied microsecond-to-second TAS to single-crystal BaTiO3 (355 nm excitation) and found a half-decay time $t_{50\%}=0.12$ s — extraordinarily long for an unmodified metal oxide without applied bias or sacrificial reagents — while cubic BaTiO3 nanoparticle films decayed in under 100 microseconds. Heating the crystal to 152 °C (above $T_\mathrm{C}$) accelerated the decay to the cubic-film rate, reversibly, linking the slow recombination to ferroelectricity rather than to the single-crystal form.

Quantitatively, the decays are biphasic and were fitted by a sum of a stretched exponential (a decay $\exp[-(t/\tau)^{\beta}]$ with stretch exponent $\beta$; $\beta<1$ indicates a distribution of rates) and a weak, temperature-independent logarithmic tail:

$\Delta OD(t)=A\,e^{-(t/\tau)^{\beta}}+B\,\log_{10}(C/t)$

The dominant stretched-exponential phase accelerates by four orders of magnitude between 298 and 425 K, following an Arrhenius law with apparent activation energy $E_A=0.85$ eV (compared with 0.30 eV in the non-ferroelectric oxide $\alpha$-Fe2O3, where detrapping limits recombination). The logarithmic phase was only tentatively assigned to tunneling recombination of deeply trapped carriers and was not analyzed further. The authors model the polarization-induced band bending as a thermal barrier $E_b(T)$ to recombination that decreases approximately linearly with temperature below $T_\mathrm{C}$ in step with the known decrease of $P_s$ (from about 26 to 20 $\mu\mathrm{C\,cm^{-2}}$ between 30 and 100 °C, then to zero above $T_\mathrm{C}$); this model predicts an effective Arrhenius activation energy $E_a^{\mathrm{eff}}=E_b(30\,^\circ\mathrm{C})+4.3\,\Delta E_b$, which equals about 0.83 eV for literature-derived values $E_b(30\,^\circ\mathrm{C})\approx 0.4$ eV and $\Delta E_b\approx 0.1$ eV, matching the measured 0.85 eV. Above $T_\mathrm{C}$ no abrupt further lifetime reduction occurs, which the authors ascribe to trap-limited recombination taking over the timescale — but this crossover is not quantified.

The stakes extend to organohalide lead perovskites such as CH3NH3PbI3 (MAPbI3), the leading thin-film photovoltaic absorbers, whose carrier recombination is strongly 'non-Langevin', i.e. orders of magnitude slower than the Langevin rate expected for bimolecular capture of oppositely charged carriers in a medium of given dielectric constant and mobility (Wehrenfennig et al. 2014). Frost et al. (2014) proposed that ferroelectric domains in MAPbI3 form internal junctions that separate charges and suppress recombination, and Morris et al. noted that the magnitude of their BaTiO3 effects 'may indeed be sufficient' to explain the perovskite lifetimes. Whether halide perovskites are ferroelectric at all, however, remains one of the field's longest-running controversies: piezoresponse force microscopy (PFM, an atomic-force technique imaging local piezoelectric response) has been interpreted both as evidence for domains and as artifact, and the question of a causal link between polar order and recombination has never been settled by direct, material-specific experiments. Reviews of polarization-enhanced photocatalysis and photoferroelectric photovoltaics (Chen et al. 2019; Li et al. 2021) treat the qualitative picture as established while leaving exactly these quantitative questions open.

Problem Statement

Establish a quantitative, mechanism-level account of how spontaneous-polarization-induced internal fields control photogenerated charge-carrier recombination in ferroelectric semiconductors, and determine whether this mechanism extends to organohalide lead perovskites. Concretely, starting from the state reached by Morris et al. (2016) in single-crystal BaTiO3, determine: (1) the identity, energies, densities, and kinetics of the deep trap states responsible for the slow, temperature-independent logarithmic decay phase, and whether the tentative assignment to tunneling recombination of deeply trapped carriers is correct; (2) the actual magnitude of the polarization-induced surface band bending (the recombination barrier $E_b$, an energy) under the realistic, near-atmospheric, illuminated, and where relevant electrolyte-screened conditions in which the carrier dynamics are measured, rather than either of the vacuum values available in the literature (the measured band bending of about 0.65 V per surface and the model-fitted barrier $E_b(30\,^\circ\mathrm{C})\approx 0.4$ eV), including how adsorbate screening and its illumination-driven modification evolve during the measurement; (3) the location and quantitative description of the crossover between polarization-limited recombination below $T_\mathrm{C}$ and trap-limited recombination above it, as a function of trap density, trap energetics, and trapping/detrapping kinetics, explaining why no abrupt lifetime reduction accompanies the loss of polarization at $T_\mathrm{C}$; (4) whether the simple thermodynamic barrier model, and the extraction of $E_b$ from variable-temperature transient absorption data, remain valid for ferroelectric materials other than single-crystal BaTiO3 and for other morphologies such as thin films, nanostructures, and domain-engineered samples, with the barrier heights so obtained checked against independent measurements; and (5) by direct, material-specific experiments, whether ferroelectric or polar order is causally responsible for the long non-Langevin carrier lifetimes of organohalide lead perovskites, or whether those lifetimes are attributable to other mechanisms. These sub-questions are the jointly necessary components of one objective: a validated causal chain from spontaneous polarization, through the screened depolarization field and its band bending, to the measured recombination kinetics — with the trap landscape and the polarization barrier separated and their competition located — validated beyond single-crystal BaTiO3 and tested for whether it extends to the organohalide lead perovskites.

References

1. Madeleine R. Morris, Stephanie R. Pendlebury, Jongin Hong, Steve Dunn, James R. Durrant, 'Effect of Internal Electric Fields on Charge Carrier Dynamics in a Ferroelectric Material for Solar Energy Conversion', Advanced Materials 28, 7123-7128 (2016). DOI: 10.1002/adma.201601238, https://doi.org/10.1002/adma.201601238
2. Jarvist M. Frost, Keith T. Butler, Federico Brivio, Christopher H. Hendon, Mark van Schilfgaarde, Aron Walsh, 'Atomistic Origins of High-Performance in Hybrid Halide Perovskite Solar Cells', Nano Letters 14, 2584-2590 (2014). DOI: 10.1021/nl500390f, https://doi.org/10.1021/nl500390f
3. Christian Wehrenfennig, Giles E. Eperon, Michael B. Johnston, Henry J. Snaith, Laura M. Herz, 'High Charge Carrier Mobilities and Lifetimes in Organolead Trihalide Perovskites', Advanced Materials 26, 1584-1589 (2014). DOI: 10.1002/adma.201305172, https://doi.org/10.1002/adma.201305172
4. Fang Chen, Hongwei Huang, Lin Guo, Yihe Zhang, Tianyi Ma, 'The Role of Polarization in Photocatalysis', Angewandte Chemie International Edition 58, 10061-10073 (2019). DOI: 10.1002/anie.201901361, https://doi.org/10.1002/anie.201901361
5. Andrés Gómez, Qiong Wang, Alejandro R. Goñi, Mariano Campoy-Quiles, Antonio Abate, 'Ferroelectricity-free lead halide perovskites', Energy & Environmental Science 12, 2537-2547 (2019). DOI: 10.1039/c9ee00884e, https://doi.org/10.1039/c9ee00884e
6. Huilin Li, Fumin Li, Zhitao Shen, Su-Ting Han, Junwei Chen, Chao Dong, Chong Chen, Ye Zhou, Mingtai Wang, 'Photoferroelectric perovskite solar cells: Principles, advances and insights', Nano Today 37, 101062 (2021). DOI: 10.1016/j.nantod.2020.101062, https://doi.org/10.1016/j.nantod.2020.101062
7. Adriana Augurio, Alberto Alvarez-Fernandez, Vishal Panchal, Bede Pittenger, Peter De Wolf, Stefan Guldin, Joe Briscoe, 'Controlled Porosity in Ferroelectric BaTiO3 Photoanodes', ACS Applied Materials & Interfaces 14, 13147-13157 (2022). DOI: 10.1021/acsami.1c17419, https://doi.org/10.1021/acsami.1c17419
8. Samutr Assavachin, Frank E. Osterloh, 'Ferroelectric Polarization in BaTiO3 Nanocrystals Controls Photoelectrochemical Water Oxidation and Photocatalytic Hydrogen Evolution', Journal of the American Chemical Society 145, 18825-18833 (2023). DOI: 10.1021/jacs.3c03762, https://doi.org/10.1021/jacs.3c03762
9. Xianlong Li, Zhiliang Wang, Wenzhong Ji, Teng Lu, Jiakang You, Kai Wang, Gang Liu, Yun Liu, Lianzhou Wang, 'Polarization Alignment in Polycrystalline BiFeO3 Photoelectrodes for Tunable Band Bending', ACS Nano 17, 22944-22951 (2023). DOI: 10.1021/acsnano.3c08081, https://doi.org/10.1021/acsnano.3c08081
10. Qi Qian, Zhong Wan, Hiroyuki Takenaka, Jong K. Keum, Tyler J. Smart, Laiyuan Wang, Peiqi Wang, Jingyuan Zhou, Huaying Ren, Dong Xu, Yu Huang, Yuan Ping, Xiangfeng Duan, 'Photocarrier-induced persistent structural polarization in soft-lattice lead halide perovskites', Nature Nanotechnology 18, 357-364 (2023). DOI: 10.1038/s41565-022-01306-x, https://doi.org/10.1038/s41565-022-01306-x
11. Alexei A. Bokov, Maryam Bari, Zuo-Guang Ye, 'Incipient ferroelectricity in methylammonium lead halide perovskites', Communications Materials 6, 80 (2025). DOI: 10.1038/s43246-025-00791-6, https://doi.org/10.1038/s43246-025-00791-6
12. Chloe Forrester, Adriana Augurio, Charlie Henderson, Ji-Seon Kim, James R. Durrant, Joe Briscoe, 'Illuminating the Origins of Ferroelectric-Induced Photocurrent and Hole Lifetime Enhancement in BaTiO3 Photoanodes', Advanced Functional Materials 35, 2503119 (2025). DOI: 10.1002/adfm.202503119, https://doi.org/10.1002/adfm.202503119