Ultrafast symmetry-breaking charge separation in thin films of J-aggregated perylenediimide multimers in a nonpolar solid-state environment

Abstract

We demonstrate ultrafast symmetry-breaking charge separation (τ_CS <110 fs) in thin films of J-aggregated perylenediimide (PDI) multimers in a nonpolar polymethyl methacrylate polymer matrix (ε = 2.80–3.20). Theoretical calculations reveal the role of through-space electronic communication and π–π interactions promoting ultrafast charge separation in PDI multimers in the solid-state.

---

Achieving efficient charge separation (CS) with minimal energy loss is a key challenge in advancing organic photovoltaic (OPV) devices.^1,2 Symmetry-breaking charge separation (SB-CS) has recently emerged as a prospective solution for generating hole–electron pairs in chromophoric assemblies without conventional donor–acceptor interfaces.^3,4 SB-CS takes advantage of asymmetry introduced by molecular packing or the surrounding environment to generate long-lived charge separated states (CSS) from photoexcited neutral chromophores, ultimately paving the way for improved open-circuit voltages and better device performance.^5–8 Although SB-CS is well documented in solution-phase systems where solvent polarity helps stabilize the charge separated state, achieving it in the solid-state, without the aid of external dielectric stabilization, presents both a fundamental challenge and a valuable opportunity for technological advancement.^1,9 Among the various π-conjugated systems explored, perylenediimides (PDIs) are outstanding electron-deficient chromophores known for their strong absorption, photostability, and high electron mobility.^10,11 Their excited-state dynamics are highly sensitive to molecular packing, and recent studies have shown that PDIs organized into solid-state multichromophoric assemblies can undergo SB-CS through intermolecular pathways.^12–15 Notably, SB-CS has been observed in the solid-state in systems where PDIs adopt slip-stacked geometries that promote through-space electronic coupling while suppressing excimer formation, otherwise leading to non-radiative losses.¹³

Rational molecular design plays a crucial role in achieving such favourable packing geometries. Introducing bulky or directional substituents, tailoring the rigidity and conjugation of linkers, or using multimeric scaffolds can promote ordered slip-stacked aggregates in the solid-state.^12,15 These arrangements facilitate the electronic coupling necessary for SB-CS while simultaneously mitigating structural motifs that favour excimer formation.¹³ Previously, we have shown that ultrafast intramolecular SB-CS can be achieved under the effect of polar solvents in angular PDI bichromophoric and trichromophoric structures through the modulation of intrachromophore electronic coupling.^16,17 Herein, we investigate the excited-state dynamics of two PDI-based multichromophoric assemblies (PDI₂ and PDI₃) forming J-aggregated thin film structures. We demonstrate that such architectures enable ultrafast SB-CS even in a nonpolar polymethyl methacrylate (PMMA) polymer matrix, thus providing a framework for designing next-generation organic materials capable of efficient solid-state charge generation.

PDI₂ and PDI₃ (Fig. 1a and b) were synthesized and characterized (Scheme S1 and Fig. S1–S8) following the previously reported procedures.^16,17 Key insights into the aggregate stacking were obtained using multi-dimensional NMR analyses in CDCl₃, where ∼12 molecules of PDI₂ and PDI₃ each were estimated to be stacked in the individual aggregate structure (Fig. S9–S18; refer to SI for more details). The PDI₂ optimized structure obtained at the B3LYP-D3/6-311+G(d,p) level of theory in vacuum, employing density functional theory (DFT), exhibited a core-twist of φ = 20.3° in the PDI fragment with bay-substitution (Fig. 1c). Meanwhile, the PDI₃ optimized structure displayed core-twisting of both the terminal PDI fragments with φ = 20.6° (Fig. 1d).

Fig. 1 Molecular structures of (a) PDI₂ and (b) PDI₃. Optimized structures of (c) PDI₂ and (d) PDI₃ showing the core-twist in the PDI multimers.

The steady-state photophysical properties of PDI₂ and PDI₃ were investigated in thin films in a nonpolar PMMA matrix (ε = 2.80–3.20) in comparison to the photophysical properties at the monomer state in 1,4-dioxane (DIOX) at room temperature. The UV-vis absorption and fluorescence excitation spectra of PDI₂ and PDI₃ thin films in PMMA reveal significantly broad red-shifted features compared to the PDI₂ and PDI₃ monomeric optical properties in DIOX ( Fig. S19 and S20a). PDI₂ thin film showed red-shifted broadened maxima in the excitation spectra (Fig. S19a) with λ^Exc_max ≈ 528 nm (521 nm in DIOX), while PDI₃ thin film exhibited broad maxima (Fig. S19b) at λ^Exc_max ≈ 521 nm (516 nm in DIOX). The red-shifted excitation spectra of PDI₂ and PDI₃ thin films compared to the monomer excitation spectra in DIOX indicate the formation of J-type self-assembly in the aggregated state of PDI₂ and PDI₃.^18,19 Moreover, the fluorescence emission spectra recorded for the PDI₂ and PDI₃ thin films in PMMA at room temperature exhibited red-shifted featureless emission profiles compared to the well-defined monomeric emission observed in DIOX ( Fig. S19c, d and S20b).

Further, we have estimated the fluorescence quantum yields (ϕ_FL) of PDI₂ and PDI₃ to probe the emissive nature of the PDI dimers in a closely packed solid-state arrangement. PDI₂ and PDI₃ showed reduced ϕ_FL of 0.88% and 0.53% in the solid-state, respectively, compared to the ϕ_FL estimated in the solution-state in DIOX (16.20% for PDI₂ and 3.77% for PDI₃). The lower ϕ_FL values of PDI₂ and PDI₃ in the solid-state indicate the presence of nonradiative decay channels involved in the excited-state dynamics of the PDI dimers in thin films.¹⁶

To obtain critical insights into the nonradiative decay pathways governing the excited-state dynamics of the PDI multichromophoric assemblies in the solid-state, we performed femtosecond transient absorption (fsTA) spectroscopic measurements of PDI₂ and PDI₃ in thin films in PMMA (Fig. 2). Upon photoexcitation at 480 nm with a 100 fs laser pulse, the fsTA spectra of PDI₂ thin film depict negative ground-state bleach (GSB) and/or stimulated emission (SE) at ∼540 to 610 nm and two distinct positive excited-state absorption (ESA) peaks centered at ∼627 nm and ∼707 nm (Fig. 2a). The observed twin ESA peaks at ∼627 nm and ∼707 nm correspond to the characteristic spectral signature of the perylenediimide radical cation (PDI˙⁺) and radical anion (PDI˙⁻), respectively.^12,13 The simultaneous occurrence of PDI˙⁺ and PDI˙⁻ transient features represent the CSS, confirming the SB-CS process in thin films of PDI₂ in a nonpolar PMMA polymer matrix (ε = 2.80–3.20). At longer time delays of ∼150 ps, the CSS state decays to a relaxed CSS state (CSS_rel) with non-negligible blue-shifted ESA features. Similar excited-state dynamics were observed for PDI₃ thin film in PMMA upon photoexcitation at 480 nm. The fsTA spectra of PDI₃ thin film depict GSB and/or SE at ∼500 to 555 nm and ESA peaks at ∼619 nm and ∼698 nm (Fig. 2d). The two distinct ESA peaks characterize the unique spectral feature of PDI˙⁺ and PDI˙⁻, confirming SB-CS in thin films of PDI₃ in PMMA.^12,13 The CSS state in PDI₃ decays to a CSS_rel state with blue-shifted ESA peaks at higher time delays of ∼150 ps.

Fig. 2 fsTA spectra of (a) PDI₂ and (d) PDI₃ in thin films in PMMA showing the excited-state dynamics upon photoexcitation. EAS reconstructed from global analysis of the fsTA data of (b) PDI₂ and (e) PDI₃ with the A → B → GS kinetic model, where A is the CSS, B is the CSS_rel, and GS is the ground state (λ_ex = 480 nm). Relative population profile of the excited-states fitted using the above kinetic models in (c) PDI₂ and (f) PDI₃.

Global analysis was applied to the fsTA data of PDI₂ and PDI₃ thin films to extract the evolution-associated spectra (EAS) and the corresponding population dynamics of the excited-states involved in the above-mentioned transformations. Representative kinetic traces at selected wavelengths, along with the globally fitted curves, are shown in Fig. S21a and b to illustrate the accuracy of the fit. The EAS and the relative population profile of PDI₂ demonstrated two principal components (Fig. 2b and c). The first component (A) is assigned to the CSS state formed due to SB-CS. The CSS state undergoes rapid vibrational relaxation (τ_VR = 91.3 ps) to form the second principal component (B). The component (B) is assigned to a CSS_rel state, which decays to the ground-state via charge recombination (CR) with a time constant of τ_CR = 1.10 ns. Likewise, the EAS and the relative population profile of PDI₃ exhibited two principal components (Fig. 2e and f). The first principal component (A) is attributed to the CSS state. The CSS state in PDI₃ decays via vibrational relaxation (τ_VR = 100.5 ps) to form the CSS_rel. Further, the CSS_rel decays to the ground-state via CR within τ_CR = 1.07 ns. The immediate appearance of the CSS feature upon photoexcitation shows that ultrafast CS occurs for PDI₂ and PDI₃ in the solid-state within the ∼110 fs instrument response.^12,13,15 Recently, we have reported SB-CS in monomeric solutions of PDI₂ and PDI₃ in a polar solvent like acetone.^16,17 The experimentally determined CS time constants in acetone were τ_CS = 6.3 ps and 3.7 ps for PDI₂ and PDI₃, respectively. Notably, the SB-CS dynamics observed in the solid-state for both molecules in this work are approximately an order of magnitude faster than those measured in solution. This could be due to the angular molecular design of PDI₂ and PDI₃, with core-twisted PDI units, which compels the molecules to arrange in a J-aggregate type solid-state packing, promoting the long-range quadrupolar effects^13,15,20 and short-range intermolecular π–π interactions^12,21,22 necessary for efficient charge separation in the solid-state. Additionally, to understand the matrix effects modulating the aggregation process, favouring ultrafast SB-CS in the solid-state, we probed the feasibility of SB-CS in PDI₂ and PDI₃ in another nonpolar (ε = 2.60) polymer matrix, polystyrene (PS). The deconvoluted fsTA spectra showed the population of a broad singlet excited-state with partial charge transfer character (Fig. S22).¹³ The absence of the distinct ESA features of the CSS in the fsTA spectra confirmed the infeasibility of SB-CS in PDI₂ and PDI₃ thin films in PS (refer to SI for more details). The above result sheds light on the crucial role played by PMMA in providing an optimum dielectric environment and promoting favourable solid-state packing necessary for ultrafast SB-CS.

Further, we measured the fluorescence lifetimes of PDI₂ and PDI₃ thin films in PMMA (Fig. S23 and S24). The major fluorescence lifetime component in PDI₂ ( = 2.63 ns (71%)) and PDI₃ ( = 2.60 ns (73%)) is in reasonable agreement with the decay constants for the CSS_rel state in PDI₂ and PDI₃ obtained from the global analysis fits of the solid-state fsTA data in PMMA (refer to SI for more details).

To understand the arrangement and packing of the individual monomers in the aggregate structure contributing to the experimentally observed SB-CS dynamics, we optimized the PDI₂ and PDI₃ aggregate ground-state geometries using DFT at the B3LYP-D3/6-31G level of theory in Gaussian 16. The optimizations were carried out employing a dimer model in vacuum. The optimized structure of PDI₂ aggregate exhibited a slip-stacked orientation between the monomers, where the two planar PDI fragments are forced to arrange in a head-to-tail fashion due to the ineffective stacking of the core-twisted PDI fragments (Fig. S25a). Comparably, the optimized structure of PDI₃ aggregate showed a slip-stacked arrangement between the central PDI fragments (Fig. S25b). The interplanar distance between the slip-stacked PDI fragments was computed to be d_interplanar = 3.30 Å and 3.66 Å in PDI₂ and PDI₃ aggregates, respectively (Fig. S25a and b).

To comprehend the role of intermolecular excitonic interactions intrinsic to the spatial chromophore arrangement and packing in the aggregate structure, we theoretically computed the long-range Coulombic coupling (J_Coul) and the short-range charge transfer-mediated coupling (J_CT) interactions in PDI₂ and PDI₃ aggregates (Table S1). The J_Coul interactions were computed for the singlet excited-state with highest oscillator strength (f) at the CAM-B3LYP-D3/DEF2-TZVP level of theory using excitation energy transfer (EET) method in Gaussian 16 (Fig. S26a, b and Table S2).^23,24PDI₂ and PDI₃ exhibited comparable J_Coul values of −732.76 cm⁻¹ and −710.30 cm⁻¹, respectively. The negative J_Coul values reinstate the J-type dipole–dipole excitonic interactions operative in the self-assembled aggregate structures of PDI₂ and PDI₃.^19,25,26 In molecular aggregates, charge transfer (CT) plays a critical role in excitonic coupling and dictating the optical properties of the aggregates.^27,28J_CT was calculated according to the equation S3 given in supporting information. PDI₂ aggregate showed a significant J_CT of −85.22 cm⁻¹, whereas PDI₃ aggregate exhibited weaker CT interactions, amounting to a J_CT of 37.92 cm^-1. The smaller d_interplanar of 3.30 Å (Fig. S25a) and the effective HOMO-HOMO/LUMO-LUMO overlap (Fig. S27) between the J-type dimer stacks in PDI₂ aggregate structure contribute to the theoretically observed higher strength of CT interactions. On the contrary, the larger d_interplanar of 3.66 Å (Fig. S25b) and the poor HOMO–HOMO/LUMO–LUMO overlap (Fig. S28) between the J-type dimer stacks in PDI₃ aggregate structure could be responsible for the lower strength of CT interactions.¹⁹ According to equation S4, the total excitonic coupling (J_Total), was estimated to be −817.98 cm⁻¹ and −672.38 cm⁻¹ in PDI₂ and PDI₃ aggregates, respectively.

A plausible schematic Jablonski diagram summarizing the mechanism of the SB-CS dynamics in PDI₂ and PDI₃ upon photoexcitation is shown in Fig. 3. Compared to the S₁* state energy in solution, PDI₂ and PDI₃ exhibit a J-type excitonic energy splitting upon aggregation in the solid-state. The initially populated excited-state shows an order of magnitude faster charge separation in the solid-state compared to solution, populating the CSS via SB-CS. In the solid-state, the CSS relaxes to a lower vibrational energy level (CSS_rel), which recombines radiatively to repopulate the ground-state, whereas in solution, triplet excited-states are populated.^16,17

Fig. 3 Schematic energy profile diagram summarizing the excited-state dynamics of PDI₂ and PDI₃ in (left) solution as monomer and (right) thin films in PMMA; S₀ = ground-state, S₁* = singlet excited-state; CSS = charge separated state; CSS_rel = relaxed charge separated state; T_n = triplet excited-state. N. B.- The solution-state photophysics is adapted from the previously reported energy profile diagrams.^16,17

In summary, we have demonstrated ultrafast symmetry-breaking charge separation (SB-CS) in angularly oriented PDI multimers in the solid-state. A perylenediimide dimer (PDI₂) and a trimer (PDI₃) were prepared with systematic enhancement of core-twisted PDI fragments. Excitation and emission spectra of PDI₂ and PDI₃ thin films in a nonpolar polymethyl methacrylate (PMMA) polymer matrix exhibited broad red-shifted excitation and emission bands compared to the monomer photophysical properties in DIOX solvent. The red-shifted excitation spectra of PDI₂ and PDI₃ thin films relative to their monomeric counterparts in DIOX, suggest the emergence of J-type aggregates in the solid-state assemblies of these chromophores. Solid-state fsTA measurements of PDI₂ and PDI₃ in thin films in PMMA matrix exhibited ultrafast charge separation within the instrument response function (τ_CS <110 fs). The optimized dimer structures of the PDI aggregates corroborated the formation of J-type slip-stacked architectures. Theoretical calculations substantiated the role of through-space long-range intermolecular excitonic communication and short-range π–π interactions in the solid-state molecular packing driving the ultrafast charge separation process. Therefore, this investigation could be beneficial for designing SB-CS materials capable of solid-state charge separation for advanced optoelectronic devices.

M. H. acknowledges the Science and Engineering Research Board (CRG/2023/005859), Department of Science and Technology, Govt. of India, for financial support. We greatly acknowledge the support for high-performance computing time at the Padmanabha cluster, IISER TVM, India. A. M., A. T., and P. D. M. thank IISER TVM for the financial support. K. V. and A. R. acknowledge CSIR and UGC, respectively, for financial assistance.

Conflicts of interest

There are no conflicts to declare.

Data availability

All the experimental and theoretical data are provided in the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5cc04382d.

References

1. E. Sebastian and M. Hariharan, ACS Energy Lett., 2022, 7, 696–711.
2. A. N. Bartynski, M. Gruber, S. Das, S. Rangan, S. Mollinger, C. Trinh, S. E. Bradforth, K. Vandewal, A. Salleo, R. A. Bartynski, W. Bruetting and M. E. Thompson, J. Am. Chem. Soc., 2015, 137, 5397–5405.
3. C. Lin, T. Kim, J. D. Schultz, R. M. Young and M. R. Wasielewski, Nat. Chem., 2022, 14, 786–793.
4. Y. Guo, Z. Ma, X. Niu, W. Zhang, M. Tao, Q. Guo, Z. Wang and A. Xia, J. Am. Chem. Soc., 2019, 141, 12789–12796.
5. E. Vauthey, Chem. Phys. Chem., 2012, 13, 2001–2011.
6. M. J. Álvaro-Martins, C. Billiaux, P. Godard, R. Oda, G. Raffy and D. M. Bassani, Chem. Commun., 2023, 59, 7963–7966.
7. A. I. Ivanov, J. Photochem. Photobiol. C, 2024, 58, 100651.
8. F. C. Spano, J. Phys. Chem. C, 2024, 128, 248–260.
9. R. M. Young and M. R. Wasielewski, Acc. Chem. Res., 2020, 53, 1957–1968.
10. A. Nowak-Król and F. Würthner, Org. Chem. Front., 2019, 6, 1272–1318.
11. J. Kong, W. Zhang, G. Li, D. Huo, Y. Guo, X. Niu, Y. Wan, B. Tang and A. Xia, J. Phys. Chem. Lett., 2020, 11, 10329–10339.
12. J. Mao, Q. Fan, Z. Yan, X. Chen, S. Zhao, Y. Lu, S. Li, W. Jiang, Z. Xu, Z. Wang and J. Wang, J. Am. Chem. Soc., 2025, 147, 12730–12739.
13. C. E. Ramirez, S. Chen, N. E. Powers-Riggs, I. Schlesinger, R. M. Young and M. R. Wasielewski, J. Am. Chem. Soc., 2020, 142, 18243–18250.
14. G. Ran, J. Zeb, Y. Song, P. A. Denis, U. Ghani and W. Zhang, J. Phys. Chem. C, 2022, 126, 3872–3880.
15. A. Khan, N. A. Tcyrulnikov, R. Roy, R. M. Young, M. R. Wasielewski and A. L. Koner, J. Phys. Chem. C, 2024, 128, 10474–10482.
16. A. Mazumder, K. Vinod, A. C. Thomas and M. Hariharan, J. Phys. Chem. Lett., 2025, 16, 4819–4827.
17. A. Mazumder, K. Vinod, P. D. Maret, P. P. Das and M. Hariharan, J. Phys. Chem. Lett., 2024, 15, 5896–5904.
18. M. Hecht and F. Würthner, Acc. Chem. Res., 2021, 54, 18.
19. N. J. Hestand and F. C. Spano, Chem. Rev., 2018, 118, 7069–7163.
20. S. Seetharaman, N. Zink-Lorre, D. Gutiérrez-Moreno, P. A. Karr, F. Fernández-Lázaro and F. D’Souza, Chem. – Eur. J., 2022, 28, e202104574.
21. P. J. Brown, M. L. Williams, S. B. Tyndall, Y. Qi, R. M. Young and M. R. Wasielewski, J. Phys. Chem. C, 2024, 128, 14185–14194.
22. R. R. Kaswan, D. Molina, L. Ferrer-López, J. Ortiz, P. A. Karr, Á. Sastre-Santos and F. D’Souza, Angew. Chem., Int. Ed., 2025, 64, e202502516.
23. G. D. Scholes and K. P. Ghiggino, J. Phys. Chem., 1994, 98, 4580–4590.
24. Z. Q. You and C. P. Hsu, Int. J. Quantum Chem., 2014, 114, 102–115.
25. M. Kasha, H. R. Rawls and M. A. El-Bayoumi, Pure Appl. Chem., 1965, 11, 371–392.
26. A. S. Davydov, Phys.-Usp., 1964, 7, 145–178.
27. M. Mandal, S. Mardanya, A. Saha, M. Singh, S. Ghosh, T. Chatterjee, R. Patra, S. Bhunia, S. Mandal, S. Mukherjee, R. Debnath, C. M. Reddy, M. Das and P. K. Mandal, Chem. Sci., 2025, 16, 901–909.
28. X. Chang, M. Balooch Qarai and F. C. Spano, J. Phys. Chem. C, 2022, 126, 18784–18795.

---