Mechanism-resolved, quantitative prediction of annihilation-driven intrinsic luminance degradation in phosphorescent organic light-emitting devices

Giebink et al. (J. Appl. Phys. 103, 044509, 2008) showed that the intrinsic luminance loss of a blue phosphorescent organic light-emitting diode (PHOLED) is consistent with permanent defect formation dominated by exciton-polaron annihilation, fitting a coupled rate-equation model with a defect-formation coefficient $K_X=(7\pm2)\times10^{-24}\ \mathrm{cm^3\,s^{-1}}$ and estimating that only $\sim2\times10^{-9}$ of each annihilation encounter produces a permanent defect. The paper itself acknowledged four gaps that jointly define the present problem: defect growth is not coupled back into charge balance; the microscopic branching of the hot polaron created in an annihilation event (dissociation versus ultrafast internal conversion) is unmeasured, leaving the bond-breaking pathway conjectural; the dependence of $K_X$ on guest exciton energy, intermolecular spacing, and host chemical structure is not quantitatively established; and the quantitative roles of competing mechanisms (exciton localization, exciton-exciton annihilation) across architectures remain undelineated. The problem is to close these gaps as one objective: a quantitative, mechanism-resolved account connecting molecular-scale degradation chemistry to device lifetime, validated across host-guest systems and device designs. Prior work has made substantial partial progress - measuring the competition between radical-ion-pair recombination and bond scission for specific dopant classes, establishing a Marcus-theory account of the exciton-energy-lifetime relation, and building self-consistent drift-diffusion and kinetic Monte Carlo degradation models - but the objective as a whole remains unresolved in the literature to date.

Background

An organic light-emitting diode (OLED) is a thin-film diode in which electrons and holes recombine within an organic emissive layer (EML) to form excitons (bound electron-hole pairs) that decay radiatively. In a phosphorescent OLED (PHOLED), the EML is a guest-host blend: a few weight percent of a heavy-metal complex, typically an octahedral Ir(III) phosphor, is doped into a wide-bandgap host; strong spin-orbit coupling on the metal lets the guest harvest both spin channels and emit phosphorescence from its triplet state, giving near-unity internal quantum efficiency. A polaron is a molecule carrying a radical charge (cation or anion) in the film; under current flow, polarons and long-lived ($\sim1\ \mu\mathrm{s}$) triplet excitons coexist in the EML. Their coexistence enables bimolecular annihilation reactions: triplet-triplet annihilation (TTA, in which two excitons interact and leave one molecule in a highly excited state) and exciton-polaron (triplet-polaron) annihilation (TPA, in which an exciton transfers its energy to a charged molecule, producing a 'hot' polaron - a radical ion in a highly excited electronic state). Such hot states can carry more energy than chemical bond dissociation thresholds, so a small fraction of encounters produce permanent molecular defects that act as deep charge traps, nonradiative recombination centers, and luminescence quenchers; this intrinsic (non-extrinsic, e.g. not moisture- or impurity-driven) defect accumulation is what limits operational lifetime, measured as LT80, the time for the luminance to fall to 80% of its initial value under fixed drive.

Giebink et al. (2008) studied a blue PHOLED with an mCBP host (a high-triplet-energy carbazole-based host) doped with an Ir complex. They wrote coupled rate equations for the hole, electron, exciton, and defect densities, separating fast transport/optical dynamics from slow ($t'$) defect growth, and modeled the EML recombination zone as an exponential profile of characteristic length $d_{\rm rec}$. Recombination itself was taken to be Langevin-like, i.e., with a rate $\gamma=q(\mu_p+\mu_n)/(\varepsilon\varepsilon_0)$ set by the carrier mobilities and dielectric constant. Comparing the candidate defect-formation routes - single-polaron ($K_X n$ or $K_X p$), exciton localization ($K_X N$), exciton-exciton annihilation ($K_X N^2$), and exciton-polaron annihilation ($K_X Nn$ or $K_X Np$), where $N$, $n$, $p$ are exciton, electron, and hole densities - they found that only exciton-polaron annihilation simultaneously reproduces the measured luminance decay, voltage rise, and EML photoluminescence transients at all four drive current densities. The fits gave $d_{\rm rec}=8\pm2\ \mathrm{nm}$, a defect-driven exciton quenching rate $K_{\rm DR}=(5\pm3)\times10^{-12}\ \mathrm{cm^3\,s^{-1}}$, and $K_X=(7\pm2)\times10^{-24}\ \mathrm{cm^3\,s^{-1}}$; a defect density of order $10^{18}\ \mathrm{cm^{-3}}$ (about 0.1% of molecules) already costs more than 50% of the luminance. From $K_X$ and the exciton lifetime $\tau=1.1\ \mu\mathrm{s}$ they estimated the probability $P\approx2\times10^{-9}$ that a single exciton-polaron encounter within a $\sim1$ nm encounter radius produces a permanent defect, concluding that most hot polarons relax harmlessly by fast internal conversion (nonradiative conversion of electronic energy into molecular vibrations). Model extrapolations showed that making recombination uniform across the EML roughly doubles LT80 and that reducing $K_X$ severalfold extends it further - design levers (broadening the recombination zone, shortening $\tau$, adding steric bulk to increase intermolecular separation) that later device experiments vindicated.

The same paper stated precisely what the model could not do, and these acknowledged limitations define the unresolved problem. First, defect growth was not fed back into charge balance: as defects accumulate they trap charge, the voltage rises, and the carrier balance (and hence polaron and exciton densities) shifts, a positive feedback that the model omitted and that was invoked to explain its deviation from measurement once the luminance fell below roughly $0.4L_0$, where $L_0$ denotes the initial (unaged) luminance. Second, the microscopic fate of the hot polaron - the branching between relaxation channels such as charge dissociation/separation and ultrafast internal conversion versus a bond-breaking chemical event - was not measured, so the pathway from annihilation energy transfer to a broken chemical bond remained conjectural. Third, the phenomenological suggestion that $K_X$ falls with decreasing guest exciton energy (consistent with red PHOLEDs outliving green and blue ones) was not quantitatively established: no functional dependence of $K_X$ on guest triplet energy, intermolecular spacing, or host chemical structure existed. Fourth, small contributions from exciton localization and exciton-exciton annihilation were not excluded, and the conclusions were established only for one host-guest combination and one recombination-zone geometry, so their generalization awaited further study.

Problem Statement

Establish a quantitative, mechanism-resolved, and predictive account of intrinsic luminance degradation in phosphorescent guest-host organic light-emitting devices, in which the microscopic annihilation chemistry is measured and connected to device lifetime, generalizing beyond the single blue mCBP:Ir device of Giebink et al. (2008). Concretely, the problem as posed by that paper's acknowledged limitations comprises four jointly supporting questions: (1) couple defect growth self-consistently into the evolving charge balance and carrier transport of the device, so that the positive feedback by which accumulating traps raise polaron densities and accelerate degradation (invoked to explain the deviation of the model from measurement once the luminance falls below roughly $0.4L_0$, i.e. below 40% of the initial luminance $L_0$) is captured rather than omitted, and the predicted luminance and voltage transients remain valid at advanced aging; (2) measure or compute the microscopic fate of the hot polaron produced in an exciton-polaron annihilation event - the branching between ultrafast internal conversion (or other non-destructive relaxation, including charge separation and recovery by charge recombination) and a bond-breaking chemical event - so that the defect-formation probability per encounter (estimated there as $\sim2\times10^{-9}$) becomes a derived rather than fitted quantity and the chemical pathway from annihilation energy transfer to a specific broken bond is identified; (3) quantitatively determine the functional dependence of the defect-formation coefficient $K_X$ on guest exciton (triplet) energy, intermolecular spacing, and host chemical structure, and thereby whether and why lower-energy (red) emitters form defects more slowly than green and blue ones and whether the color ordering of operational lifetimes follows from this dependence; and (4) delineate the quantitative contributions of exciton localization and exciton-exciton annihilation relative to exciton-polaron annihilation across device architectures and operating regimes, and establish the validity domain of the mechanism across host-guest combinations and recombination-zone designs. Completion means that the operational lifetime (e.g., LT80) of a phosphorescent guest-host OLED can be predicted from molecular-level degradation rates and device structure, not only fitted after aging. It requires resolving all four sub-questions above and demonstrating the prediction for at least two host-guest systems that differ in both host and emitter and for more than one device architecture and recombination-zone design, so that the account generalizes beyond the single blue mCBP:Ir device of Giebink et al. (2008).

References

1. N. C. Giebink, B. W. D'Andrade, M. S. Weaver, P. B. Mackenzie, J. J. Brown, M. E. Thompson, and S. R. Forrest, 'Intrinsic luminance loss in phosphorescent small-molecule organic light emitting devices due to bimolecular annihilation reactions', Journal of Applied Physics 103, 044509 (2008), DOI 10.1063/1.2884530, https://doi.org/10.1063/1.2884530
2. N. C. Giebink, B. W. D'Andrade, M. S. Weaver, J. J. Brown, and S. R. Forrest, 'Direct evidence for degradation of polaron excited states in organic light emitting diodes', Journal of Applied Physics 105, 124514 (2009), DOI 10.1063/1.3151689, https://doi.org/10.1063/1.3151689
3. Hossein Zamani Siboni, Yichun Luo, and Hany Aziz, 'Luminescence degradation in phosphorescent organic light-emitting devices by hole space charges', Journal of Applied Physics 109, 044501 (2011), DOI 10.1063/1.3549128, https://doi.org/10.1063/1.3549128
4. Yifan Zhang, Jaesang Lee, and Stephen R. Forrest, 'Tenfold increase in the lifetime of blue phosphorescent organic light-emitting diodes', Nature Communications 5 (2014), DOI 10.1038/ncomms6008, https://doi.org/10.1038/ncomms6008
5. Sebastian Scholz, Denis Kondakov, Bjorn Lussem, and Karl Leo, 'Degradation Mechanisms and Reactions in Organic Light-Emitting Devices', Chemical Reviews 115, 8449-8503 (2015), DOI 10.1021/cr400704v, https://doi.org/10.1021/cr400704v
6. Alexandra Ya. Freidzon, Andrey A. Safonov, Alexander A. Bagaturyants, Dmitry N. Krasikov, Boris V. Potapkin, Alexey A. Osipov, Alexander V. Yakubovich, and Ohyun Kwon, 'Predicting the Operational Stability of Phosphorescent OLED Host Molecules from First Principles: A Case Study', The Journal of Physical Chemistry C 121, 22422-22433 (2017), DOI 10.1021/acs.jpcc.7b05761, https://doi.org/10.1021/acs.jpcc.7b05761
7. Jaesang Lee, Changyeong Jeong, Thilini Batagoda, Caleb Coburn, Mark E. Thompson, and Stephen R. Forrest, 'Hot excited state management for long-lived blue phosphorescent organic light-emitting diodes', Nature Communications 8, 15566 (2017), DOI 10.1038/ncomms15566, https://doi.org/10.1038/ncomms15566
8. Caleb Coburn and Stephen R. Forrest, 'Effects of Charge Balance and Exciton Confinement on the Operational Lifetime of Blue Phosphorescent Organic Light-Emitting Diodes', Physical Review Applied 7, 041002 (2017), DOI 10.1103/PhysRevApplied.7.041002, https://doi.org/10.1103/PhysRevApplied.7.041002
9. Sinheui Kim, Hye Jin Bae, Sangho Park, Wook Kim, Joonghyuk Kim, Jong Soo Kim, Yongsik Jung, Soohwan Sul, Soo-Ghang Ihn, Changho Noh, Sunghan Kim, and Youngmin You, 'Degradation of blue-phosphorescent organic light-emitting devices involves exciton-induced generation of polaron pair within emitting layers', Nature Communications 9, 1211 (2018), DOI 10.1038/s41467-018-03602-4, https://doi.org/10.1038/s41467-018-03602-4
10. Dong-Gwang Ha, Jan Tiepelt, Michael A. Fusella, Michael S. Weaver, Julie J. Brown, Markus Einzinger, Michelle C. Sherrott, Troy Van Voorhis, Nicholas J. Thompson, and Marc A. Baldo, 'Dominance of Exciton Lifetime in the Stability of Phosphorescent Dyes', Advanced Optical Materials 7, 1901048 (2019), DOI 10.1002/adom.201901048, https://doi.org/10.1002/adom.201901048
11. J. Vijaya sundar, V. Subramanian, and B. Rajakumar, 'Excited state C-N bond dissociation and cyclization of tri-aryl amine-based OLED materials: a theoretical investigation', Physical Chemistry Chemical Physics 21, 438-447 (2019), DOI 10.1039/c8cp06314a, https://doi.org/10.1039/c8cp06314a
12. Bomi Sim, Jong Soo Kim, Hyejin Bae, Sungho Nam, Eunsuk Kwon, Ji Whan Kim, Hwa-Young Cho, Sunghan Kim, and Jang-Joo Kim, 'Comprehensive Model of the Degradation of Organic Light-Emitting Diodes and Application for Efficient, Stable Blue Phosphorescent Devices with Reduced Influence of Polarons', Physical Review Applied 14, 024002 (2020), DOI 10.1103/PhysRevApplied.14.024002, https://doi.org/10.1103/PhysRevApplied.14.024002
13. Yu Kyung Moon, Ho Jin Jang, Sanju Hwang, Seongsoo Kang, Sinheui Kim, Juwon Oh, Sangheon Lee, Dongho Kim, Jun Yeob Lee, and Youngmin You, 'Modeling Electron-Transfer Degradation of Organic Light-Emitting Devices', Advanced Materials 33, 2003832 (2021), DOI 10.1002/adma.202003832, https://doi.org/10.1002/adma.202003832
14. Christoph Hauenstein, Stefano Gottardi, Engin Torun, Reinder Coehoorn, and Harm van Eersel, 'Identification of OLED Degradation Scenarios by Kinetic Monte Carlo Simulations of Lifetime Experiments', Frontiers in Chemistry 9, 823210 (2022), DOI 10.3389/fchem.2021.823210, https://doi.org/10.3389/fchem.2021.823210
15. Lorenz Graf von Reventlow, Manikandan Koodalingam, Christian Siebert, Philipp Marlow, Emma V. Puttock, Paul L. Burn, and Alexander Colsmann, 'Influence of chromophore spacing on the stability and efficiency of host-free sky-blue dendrimer organic light emitting diodes', Journal of Materials Chemistry C 10, 8278-8287 (2022), DOI 10.1039/d2tc00602b, https://doi.org/10.1039/d2tc00602b
16. Jaewook Kim, Joonghyuk Kim, Yongjun Kim, Youngmok Son, Youngsik Shin, Hye Jin Bae, Ji Whan Kim, Sungho Nam, Yongsik Jung, Hyeonsu Kim, Sungwoo Kang, Yoonsoo Jung, Kyunghoon Lee, Hyeonho Choi, and Woo Youn Kim, 'Critical role of electrons in the short lifetime of blue OLEDs', Nature Communications 14, 7508 (2023), DOI 10.1038/s41467-023-43408-7, https://doi.org/10.1038/s41467-023-43408-7
17. Haonan Zhao, Boning Qu, and Stephen R. Forrest, 'Understanding and Controlling the Formation of Nonradiative Defects in Blue Organic Triplet Emitters', Physical Review X 14, 041044 (2024), DOI 10.1103/PhysRevX.14.041044, https://doi.org/10.1103/PhysRevX.14.041044
18. Adrian Pizano, Raju Lampande, Robert Cawthorn, and Noel C. Giebink, 'Drift-diffusion modeling of blue OLED degradation', Synthetic Metals 311, 117797 (2025), DOI 10.1016/j.synthmet.2024.117797, https://doi.org/10.1016/j.synthmet.2024.117797
19. Stan E. A. Jaspars, Hiroki Tomita, Clint van Hoesel, Nicolas Daub, René A. J. Janssen, and Reinder Coehoorn, 'Spectroelectrochemical Determination of Förster Radii for Triplet-Polaron Quenching in Phosphorescent Organic Light-Emitting Diodes', Advanced Functional Materials 35, 2425548 (2025), DOI 10.1002/adfm.202425548, https://doi.org/10.1002/adfm.202425548
20. Byung Hak Jhun, Yerin Park, Hwang Suk Kim, Ji Hye Baek, Joonghyuk Kim, Eunji Lee, Hyejin Moon, Changjin Oh, Yongsik Jung, Seunghee Choi, Mu-Hyun Baik, and Youngmin You, 'The degradation mechanism of multi-resonance thermally activated delayed fluorescence materials', Nature Communications 16, 392 (2025), DOI 10.1038/s41467-024-55620-0, https://doi.org/10.1038/s41467-024-55620-0