Photoinduced Accumulative Electron-Transfer Dynamics in Perylene Diimide Polymers as Photoredox Catalysts

Abstract

Understanding the mechanisms of light-driven charge separation (CS), accumulative electron transfer, and free carrier (FC) dynamics in polymer semiconductors as photoredox catalysts is essential for artificial photosynthesis and for developing efficient solar energy conversion materials. This study presents a comparative investigation on the photoinduced electron transfer dynamics of three perylene diimide (PDI) polymers, o PDI, m PDI, and p PDI, that differ by their phenylenediamine linkers. Notably, o PDI, which is constructed with ortho-phenylenediamine linkers, exhibits the fastest two-electron CS and the subsequent FC generation among this series. These superior properties are attributed to its large molecular dipole moment and driving force for multielectron accumulation. The high photocatalytic activity of o PDI was further verified through two independent two-electron quenching experiments and the oxygen reduction reaction (ORR). The latter efficiently generates H₂O₂, particularly in the case of o PDI, leading to the rapid degradation of uric acid (UA). This work elucidates the mechanism of photoinduced accumulative electron transfer in PDI-based polymers and provides a theoretical framework for the rational design of efficient multielectron accumulation systems in advanced photocatalytic applications.

Introduction

The primary redox step of natural photosynthesis involves light-induced electron transfer from chlorophyll to pheophytin and subsequently to quinone, resulting in a charge separation (CS) state composed of oxidized chlorophyll and reduced quinone. , Inspired by this natural process, numerous artificial photosynthetic systems have been developed to mimic such photoinduced CS, which is widely recognized as a critical step toward efficient solar energy conversion. − However, key reactions in artificial photosynthesis, particularly the oxygen evolution reaction (OER), typically demand multielectron transfer and must overcome substantial energy barriers, making it the rate-determining step in overall water splitting. − Simulating multielectron transfer in artificial photosynthesis systems remains a major challenge, as the microscopic ultrafast dynamics of electron accumulation and its intrinsic correlation with catalytic performance remain largely unexplored.

Substantial research efforts have been devoted to constructing artificial photosynthetic systems, primarily through the following strategies: (1) intramolecular donor–acceptor (D–A) motifs, such as donor–photosensitizer–acceptor (D–PS–A) or donor–bridge–acceptor (D–B–A) architectures; − (2) multiple virtually identical and symmetric subunits with upon photoexcitation via excited-state symmetry breaking; − (3) three-dimensional D–A frameworks exhibiting high crystallinity and moderate porosity, such as metal–organic frameworks (MOFs), − covalent organic frameworks (COFs), − and hydrogen-bonded frameworks (HOFs); − and (4) conjugated photocatalytic organic polymers with amorphous and moderate porosity. − These strategies share a common goal: to promote rapid exciton dissociation and drive CS via a strong interfacial space charge region, which is achieved by minimizing exciton binding and enhancing the built-in electric field (BIEF). , The strong BIEF depends critically on both structural crystallinity and molecular dipole moment, and the former enables efficient charge transport through ordered pathways, while the latter facilitates exciton dissociation and provides the driving force for CS.

Perylene diimide (PDI) and its derivatives have garnered considerable attention in solar-driven water splitting, particularly in OER applications, owing to their chemical modification ability, excellent photostability, and strong visible-light absorption. , Previously, strategies to enhance the photocatalysis performance of PDI systems have primarily relied on supramolecular assembly or composite formation. − However, such systems often suffer from limited electric dipole moments of molecular fragments, leading to insufficient driving force for CS and poor multielectron accumulation. A significant advance was achieved by Zhu et al., who demonstrated notable OER activities using crystalline PDI polymers (e.g., urea-PDI). Building on this, our earlier work compared three isomeric PDI polymers, o PDI, m PDI, and p PDI, whose molecular orientations are dictated by ortho-, meta-, and para-substitutions on the phenylenediamine linker, respectively (Figure a). We established a static macroscopic structure–property correlation in that work: rational linker engineering can amplify the molecular dipole moment and BIEF strength, boosting photoelectrocatalytic oxygen evolution performance by over 2 orders of magnitude compared to self-assembled monomers and urea-PDI. Despite these promising advances, the fundamental mechanisms that govern ultrafast CS and multielectron accumulation processes remain ambiguous. Specifically, the full charge evolution cascade spanning from initial photoexcitation to long-lived active species has not been resolved, and how molecular-scale dipole alignment modulates the kinetics of CS and accumulative electron transfer to govern multielectron redox photocatalysis efficiency is still not well understood.

1.

(a) Synthesis strategy and chemical structures for o PDI, m PDI, and p PDI; (b) normalized absorption of o PDI, m PDI, p PDI, and ref-PDI in DMF; (c) normalized photoluminescence of o PDI, m PDI, p PDI, and ref-PDI in DMF; (d) a comparison of the photocurrent performance of o PDI, m PDI, and p PDI with previously reported state-of-the-art solar-driven water oxidation.

To address these challenges, we selected o PDI, m PDI, and p PDI as representative model systems to investigate the interplay among molecular dipole moment, multielectron accumulation, and the associated thermodynamics and kinetics. By combining femtosecond and nanosecond transient absorption spectroscopy with theoretical calculations, we resolve the full charge evolution cascade across subpicosecond to microsecond time scales, proceeding sequentially from charge-transfer excitons to two-electron-accumulated charge-separated states (PDI dianions) and finally to long-lived free carriers (FC). We demonstrate that o PDI features the fastest charge transfer and most efficient FC generation in the series; its robust built-in electric field enables spontaneous multielectron accumulation, which is the essential origin of its superior multielectron photocatalytic activity. In addition, the high photocatalytic activity of o PDI was further verified through two independent two-electron quenching experiments. Furthermore, under light irradiation, o PDI efficiently generates hydrogen peroxide (H₂O₂) through a two-electron oxygen reduction reaction (ORR), leading to the rapid degradation of uric acid (UA). These findings provide fundamental insights into photoinduced accumulative electron transfer and establish guiding principles for the rational design of photoinduced accumulative electron-transfer systems for advanced photocatalysis.

Results and Discussion

Structural and Photophysical Properties

The o PDI, m PDI, and p PDI polymers were synthesized according to previously reported methods, and their chemical structures are illustrated in Figure a. Figure b shows the steady-state absorption and emission spectra of o PDI, m PDI, and p PDI dispersed in N,N-dimethylformamide (DMF). Compared with the parent chromophore ref-PDI (the structure shown in Figure S1), the PDI polymers exhibit broad absorption across 400–800 nm, overlapping well with the solar spectrum, which is a desirable characteristic for constructing solar-driven photocatalytic systems. Owing to its ortho-phenylenediamine linkers, which impose a minimal orientation angle between adjacent PDI units, o PDI exhibits a red-shifted absorption edge extending to 700–800 nm. This is attributed to a tendency for π–π stacking along the polymer backbone. In contrast, m PDI and p PDI exhibit absorption profiles that remain relatively close to that of ref-PDI, with well-resolved 0–0, 0–1, and 0–2 vibronic bands at 525, 487, and 458 nm, respectively. , For o PDI, these vibronic bands are red-shifted to 560, 525, and 487 nm, respectively, reflecting stronger electronic coupling between neighboring PDI units in this polymer architecture.

The emission spectra of o PDI, m PDI, p PDI, and ref-PDI in DMF all exhibit similar emission profiles (Figure c). Compared with ref-PDI, which shows a high fluorescence quantum yield (Φ_F) ∼98%, o PDI, m PDI, and p PDI all exhibit significantly lower Φ_F (ca. 10%) in highly polar DMF solutions (Table S1). Notably, the fluorescence spectra show no emission broadening or red shift characteristic of excimer-like states commonly observed in PDI covalent or stacked aggregates, indicating that none of these polymers exhibit prominent CT-resonance-enhanced excimer emission features. , Furthermore, the fluorescence lifetimes of o PDI, m PDI, and p PDI are all shorter than that of ref-PDI (Figure S2). This reduction may be attributed to enhanced nonradiative relaxation resulting from the stabilization of CS states in the polar solvent.

The photocurrent performance of o PDI is comparable to that of state-of-the-art organic-based photoanodes for water splitting, while m PDI and p PDI exhibit markedly lower efficiency (Figure d and Table S2). This performance hierarchy indicates that the orientation of phenylenediamine linkers in PDI polymers plays a critical role in determining carrier dynamics through continuous structural units that facilitate both photoinduced CS and directional charge transport along the molecular chain, with an ortho-configuration ( o PDI) showing optimal catalytic activity.

Accumulative Electron-Transfer Dynamics

Efficient photoelectrochemical reactions critically rely on the ultrafast CS and the subsequent multielectron accumulation processes. To elucidate these pathways, we investigate the excited-state dynamics of these PDI polymers using femtosecond (fs-) and nanosecond time-resolved transient absorption (ns-TA) spectroscopy. Figure shows the representative fs-TA spectra of o PDI, m PDI, and p PDI in DMF upon photoexcitation at 480 nm. For o PDI (Figure a,b), the initial spectrum at 0.45 ps exhibits a ground-state bleach (GSB) around 500–600 nm and a broad excited-state absorption (ESA) around 350–450 nm and 750–900 nm. The ESA feature profile probed at 790 nm resembles the characteristic absorption of the PDI anion (PDI^•–), consistent with chemical reduction control experiments (Figure S3) and previously reported spectral positions of PDI^•–, − indicating the ultrafast formation of CT-exciton states between adjacent PDI units upon photoexcitation. As the delay time increased to ca. 2 ps, the PDI^•– features diminish, and a new species emerges with a distinct ESA band centered at 550–650 nm. This new band is assigned to the PDI dianion (PDI^2–), as its peak position and line shape are highly consistent with the steady-state absorption spectrum of chemically generated PDI^2–, as well as reported spectral signatures of PDI^2–. , The presence of a clear isosbestic point near 750 nm indicates an efficient conversion from PDI^•– to PDI^2– in o PDI within 10 ps. Notably, all of the above spectral features are distinct from the broad, featureless ESA bands characteristic of PDI excimers or CT aggregates, which, combined with the steady-state fluorescence results, rule out significant contributions from excimer or aggregate species. ,, The subsequent evolution of the TA signal toward negative values corresponds to exciton dissociation and the generation of delocalized polaron, namely, free carriers (FC). , This direct spectral progression provides clear evidence of an ultrafast intramolecular CS process in o PDI, succeeded by multielectron accumulation and eventual FC generation. The spectral evolution of m PDI and p PDI in DMF solution (Figure e,f) is basically similar overall to that of o PDI, but the characteristics of PDI^2– are significantly weaker than those of o PDI, indicating that the double-electron accumulation channels in them are relatively inefficient. Importantly, all spectral features described above are entirely distinct from those of the reference compound ref-PDI (Figure S5), underscoring the unique photoinduced multielectron transfer dynamics enabled by the polymeric architecture.

2.

2D pseudocolor plots of fs-TA spectra of (a) o PDI, (b) m PDI, and (c) p PDI in DMF dispersion upon 480 nm excitation. The representative fs-TA spectra of (d) o PDI, (e) m PDI, and (f) p PDI. Evolution-associated difference spectra (EADS) of (g) o PDI, (h) m PDI, and (i) p PDI, obtained from the global analysis based on a sequential model.

We performed global spectral analysis using a sequential kinetic model to obtain evolution-associated difference spectra (EADSs) and corresponding rate constants. We emphasize that this sequential global fitting model has intrinsic limitations for the present polymer system, and we employ it primarily to provide a qualitative description of the overall kinetic evolution rather than to forcibly decompose the complex dynamics into a limited number of discrete excited-state species. Specifically, the intrinsic structural heterogeneity of PDI copolymers, including polydisperse chain lengths, diverse conformational distributions, and variable local stacking environments, gives rise to distributed charge evolution pathways rather than an ideal, uniform state-to-state sequential transition across the full time window. Only the early stage CT-exciton generation and two-electron accumulation kinetics approximately follow a sequential state-to-state evolution behavior. The subsequent processes, including exciton dissociation, charge diffusion, delocalized polaron formation, and charge recombination, clearly do not conform to a strict state-to-state sequential model but are more consistent with second-order kinetics of nongeminate recombination. Kinetics traces at selected representative wavelengths are shown in Figure S6 to demonstrate the quality of the global fit.

For o PDI (Figure g), the first EADS corresponds to the promptly formed CT-exciton state. The second EADS exhibits the characteristic absorption feature of PDI^2– and is therefore assigned to the two-electron accumulated CS state. The electron accumulation forms with time constants of 0.76 ps and exhibits a multistage decay, since the third and fourth EADSs both exhibit the dianion features and should be further determined by ns-TA measurements. For m PDI and p PDI (Figure h,i), the first and second EADSs are both assigned to the CT-exciton state. The third and fourth EADSs are assigned to the two-electron accumulated CS state. For all three polymers, the absorption intensity decreases significantly on the subpicosecond time scale upon photoexcitation, and this behavior is attributed to exciton–exciton annihilation (EEA). This assignment is further corroborated by the pump-fluence-dependent transient kinetics (Figure S8). Remarkably, the two-electron accumulated CS state in o PDI forms within 0.76 ps. This rate is substantially faster than those observed for m PDI and p PDI, highlighting a markedly more efficient two-electron transfer process in the ortho-substituted isomer than in its meta- and para-substituted counterparts. The dynamics of FC generation and recombination are critical to multielectron redox catalysis. These processes can be tracked by monitoring the decay of the two-electron accumulated CS state, which manifests as the characteristic ESA of PDI^2– in ns-TA measurements (Figure ). Upon 480 nm excitation in DMF solution, the structured absorption feature of the dianion, which is superimposed on the GSB and appears as a negative net signal, gradually vanishes within ∼30 ns, evolving into a featureless GSB profile that decays to baseline within ∼20 μs (Figure a,d). Similar decay profiles were observed for m PDI and p PDI (Figure e,f). Consistent with the fs-TA results, the PDI^2– spectral signature in m PDI and p PDI is markedly weaker than that of o PDI.

3.

2D pseudocolor plots of ns-TA spectra of (a) o PDI, (b) m PDI, and (c) p PDI in DMF dispersion upon 480 nm excitation. The representative ns-TA spectra of (d) o PDI, (e) m PDI, and (f) p PDI. Evolution-associated difference spectra (EADS) of (g) o PDI, (h) m PDI, and (i) p PDI, obtained from the global analysis based on a sequential model. Note that EADS corresponds to kinetically resolved evolution intermediates, rather than strictly distinct chemical species.

To quantify the charge recombination dynamics, we applied the same sequential kinetic model for global analysis and obtained the corresponding EADS and rate constants. Kinetics traces at selected wavelengths demonstrate the quality of the global fit (Figure S7). Overall, all three polymers exhibit multistage decay dynamics spanning from nanosecond to microsecond time scales, and their corresponding EADS spectra are dominated by negative GSB features. On the nanosecond and longer time scales, no distinct positive ESA features are detected across the entire probe wavelength range, which is inconsistent with the spectral signatures of bound charge pairs or trapped charge states. This observation, combined with the insensitivity of the decay kinetics to oxygen (Figure S9), confirms that the long-lived species are predominantly fully dissociated FC rather than triplet states or localized trapped charges. This indicates that the apparent lifetimes of FC in all three compounds reach the microsecond scale. Notably, in o PDI, both the first and second EADS show faint yet discernible absorption features attributable to PDI^2–, overlapping with the broad GSB bands, whereas in m PDI and p PDI, only the first EADS displays this characteristic (Figure g–i). This distinction suggests that exciton dissociation in o PDI likely occurs to a greater extent through an accumulative electron-transfer CS state, rather than a CT-exciton, which also explains its superior performance in multielectron redox reactions.

Mechanism of Multielectron Transfer in PDI Polymers

To elucidate the structure–property relationship and the driving force for multielectron transfer during exciton dissociation in PDI polymers, density functional theory (DFT) calculations were performed at the B3LYP/6–31g­(d,p) level. Models consisting of two, three, four, and five PDI units, denoted as dimer, trimer, tetramer, and pentamer, respectively, were employed. Figure presents the molecular surface color-mapped according to the calculated electrostatic potential (ESP). For all systems ( o PDI, m PDI, p PDI, and ref-PDI), the conjugated carbon rings reside in regions of positive ESP, whereas the carbonyl groups (–CO) are located in areas of negative ESP. This indicates that the conjugated rings carry a partial positive charge and can serve as a donor site for photogenerated holes, while the electron-withdrawing –CO groups tend to stabilize and accumulate photogenerated electrons.

4.

Electrostatic potential and molecular dipole of the oligomers of o PDI, m PDI, and p PDI, and their parent chromophore ref-PDI.

The charge-polarization characteristics within the molecular fragments were evaluated by optimizing the oligomeric repeat unit structures and calculating the ground-state electric dipole moment under the SMD implicit solvation model (DMF solvent) to match the solution-phase experimental conditions. Compared with ref-PDI (ground-state electric dipole moment: 0.000695 D), the introduction of ortho- and meta-phenylenediamine linkers significantly enhances the dipole moment in o PDI and m PDI fragments, whereas the linearly structured p PDI fragment shows no significant increase (Figure ). Notably, the dimer configuration of o PDI exhibits a large dipole moment of 7.21 D, and a similarly high value is observed for m PDI. It should be noted that the total dipole moment of oligomers does not increase monotonically with polymerization degree: as the chain length increases, conformational adjustment and improved local structural symmetry induce partial cancellation of dipole vectors between adjacent repeating units, leading to irregular variations in the overall dipole moment of longer oligomers. Nevertheless, o PDI consistently exhibits the largest segmental dipole moment among the three isomeric polymers across all of the tested oligomer lengths.

Meanwhile, the calculated excited-state (S₁ state) dipole moments follow the same trend as the ground-state values (Table S3). For o PDI, the slightly amplified excited-state dipole further strengthens the local electric field during charge transfer, forming a synergistic effect with the BIEF derived from the ground-state permanent dipoles. The small-angle folded conformation of o PDI enhances its molecular dipole, promoting the superposition of surface electrostatic vectors and thereby facilitating the formation of a strong BIEF. Consistently, o PDI shows the highest surface voltage and ζ-potential among the PDI polymers, confirming its strong BIEF effect, in agreement with the large dipole moment predicted theoretically. The enhanced dipole moment strengthens the BIEF, thereby increasing the driving force for CS and accelerating the CS and two-electron accumulation process.

The thermodynamics of the two-step electron transfer process in PDI polymers are critical for understanding the mechanism of the electron accumulation observed in time-resolved spectroscopy. The first step corresponds to photoinduced electron transfer between two adjacent PDI units within the polymer, generating an anion–cation pair (i.e., PDI* + PDI → PDI^•+ + PDI^•–). The Gibbs free energy changes for CS (ΔG _CS1) of this step can be evaluated independently using the Rehm–Weller equation. The photoinduced electron transfer of o PDI, m PDI, and p PDI is calculated as ΔG _CS1< 0 (largely negative) in DMF (Table S4), confirming the thermodynamic feasibility of the intramolecular/intermolecular electron transfer under illumination. Subsequently, the driving force for the formation of a PDI dianion (PDI^2–) via a two-electron accumulated CS reaction was assessed using ΔG _CS2. This second step involves two adjacent PDI^•‑ species undergoing electron transfer to yield neutral PDI and PDI^2– (i.e., PDI^•– + PDI^•– → PDI + PDI^2–). The ΔG _CS2 values in o PDI, m PDI, and p PDI dimers and ref-PDI were calculated by DFT in solvents of varying dielectric constants (toluene, , acetonitrile (MeCN), and DMF). As summarized in Table , m PDI dimer, p PDI dimer, and ref-PDI exhibit positive ΔG _CS2 in all solvents, whereas ΔG _CS2 for o PDI dimer decreases with increasing solvent polarity, becoming negative in MeCN and DMF. These results indicate that, compared with other systems, the formation of PDI^2– in o PDI dimer is more thermodynamically favorable and more stable in DMF. The thermodynamically downhill nature of the second electron transfer step, together with strong π–π stacking along the polymer backbone in the ortho-configured backbone, explains the subpicosecond two-electron accumulation time observed in the fs-TA experiments.

1. Calculation of ΔG _CS2 for Two-Electron Accumulation Reactions in PDI Dimers and ref-PDI in Solvents.

compound	solvent	ΔG CS2/kJ/mol
o PDI dimer	toluene	98.2
	MeCN	–2.47
	DMF	–30.1
m PDI dimer	toluene	121
	MeCN	15.0
	DMF	41.7
p PDI dimer	toluene	115
	MeCN	43.4
	DMF	56.5
ref-PDI	toluene	157
	MeCN	37.6
	DMF	37.6

^aAll values are corrected with zero-point energy and thermal entropic contributions at 298 K, calculated at the B3LYP/6–31G­(d,p) level with the SMD solvation model.

Mechanism Validation via Two-Electron Quenching Experiments

To further validate the proposed accumulative electron transfer mechanism, we investigated the photocatalytic oxidation of 1,2,3,4-tetrahydroisoquinoline (THIQ) to 3,4-dihydroisoquinoline (DHIQ), a prototypical two-electron-transfer photocatalytic reaction, using o PDI, m PDI, and p PDI as catalysts (Figure a). , The concentration of THIQ was systematically varied from 0 to 20 mM (Figure S11). As the PDI polymer forms a homogeneous aggregate, its exciton dissociation and exciton recombination can be viewed as a competitive process. Consequently, in the photocatalytic oxidation of THIQ, an increase in substrate concentration leads to the quenching of the PDI polymer’s fluorescence. As shown in Figure b, among these polymers, o PDI exhibited the most pronounced fluorescence quenching. Furthermore, the photocatalytic reduction was conducted with trans-4,5-dihydroxy-1,2-dithiane (DTT), which is the light-driven two-electron transfer reaction of an aliphatic disulfide to its dithiolate form. (Figure a). , In this quenching experiment, o PDI again demonstrated a significant advantage (Figures c and S12). These two independent two-electron quenching experimental results consistently align with the trend of efficient electron accumulation and FC generation in o PDI, as validated by methods such as time-resolved spectroscopy.

5.

(a) Reaction mechanism for photocatalytic oxidation of THIQ to DHIQ and photocatalytic reduction of DTT to DTT^2–; Stern–Volmer plot obtained from the fluorescence intensity of o PDI, m PDI, and p PDI with different concentrations of (b) THIQ and (c) DTT.

Photodegradation Based on Two-Electron ORR

The ability to accumulate multiple electrons significantly enhances the practical potential of o PDI. Specifically, efficient multiple-charge accumulation endows o PDI with enhanced performance in two-electron ORR, which is the core pathway for reactive oxygen species (ROS) generation in PDI polymer systems. The main ROS involved (O₂ ^•–, H₂O₂, and ¹O₂) follow the reaction pathways described by eq –: ,

$${\mathrm{O}}_2+{\mathrm{e}}^{-}\to {\mathrm{O}}_2^{\mathrm{•}-}$$ 1

$${\mathrm{O}}_2^{\mathrm{•}-}+{\mathrm{e}}^{-}+{2\mathrm{H}}^{+}\to {\mathrm{H}}_2{\mathrm{O}}_2$$ 2

$${\mathrm{O}}_2^{\mathrm{•}-}+{\mathrm{h}}^{+}\to {}^1\mathrm{O}_2$$ 3

To systematically evaluate the ROS generation capability of PDI polymers and verify the two-electron transfer photocatalytic mechanism, we employed dihydrorhodamine 123 (DHR123), hydrogen peroxide test strips, and 9,10-anthracenediyl-bis­(methylene)­dimalonic acid (ABDA) as specific probes for O₂ ^•–, H₂O₂, and ¹O₂, respectively. To further confirm the identity of each ROS species and exclude interference from other pathways, we performed complementary ROS scavenger control experiments: Tiron was used as a selective O₂ ^•– scavenger, and KN₃ was used as a ¹O₂ quencher. As shown in Figure a–c and Figures S13–S14, o PDI and p PDI generate all three types of ROS (O₂ ^•–, H₂O₂, and ¹O₂), whereas m PDI produces only O₂ ^•– and H₂O₂, with o PDI consistently exhibiting the highest ROS generation efficiency among the series. For H₂O₂ quantification, semiquantitative test strips were adopted. This activity hierarchy is fully consistent with the time-resolved spectroscopic results, demonstrating that the strong accumulative electron transfer ability and efficient FC generation in o PDI collectively account for its superior two-electron ORR performance.

6.

ROS generation by o PDI, m PDI, and p PDI upon white light irradiation using (a) DHR123, (b) hydrogen peroxide test strips (the solution containing 200 μM ascorbic acid (Vc); Vc served as an electron and proton sacrificial agent in these processes), and (c) ABDA as O₂ ^•–, H₂O₂, and ¹O₂ indicators, respectively. (d) The photocatalytic degradation activity of PDI polymer with UA under visible light irradiation in 30 min. Condition: white LED light (30 mW/cm²). (e) The mechanism of photocatalytic degradation by polymer o PDI in uric acid (UA).

To directly assess the practical effect of multielectron accumulation on photocatalytic performance, uric acid (UA) was selected as a model pollutant, which can be efficiently degraded by ROS. After 30 min of visible light irradiation, o PDI exhibits the highest UA degradation (∼13.1%) among the PDI polymer and reference sample (ref-PDI) (Figures d and S15). Dark control experiments confirm that no significant UA degradation occurs without light irradiation, verifying that the observed degradation is driven by the photocatalytic process (Figure S16). Based on the above experiments, a possible photocatalytic mechanism for UA degradation over o PDI under visible light irradiation is proposed in Figure e. Analogous to inorganic semiconductor photocatalytic systems, photoexcitation of o PDI generates free holes (h⁺) and electrons (e^–) via efficient accumulative electron transfer and exciton dissociation. The photogenerated h⁺ can directly oxidize pollutant molecules, while e^– reduces dissolved O₂ to form O₂ ^•–, which further converts to H₂O₂. Additionally, O₂ ^•– can be oxidized by h⁺ to yield ¹O₂. Eventually, UA is targeted by abundant ROS and degraded to small-molecule products.

Conclusion

In summary, we have systematically elucidated the photoinduced accumulative electron-transfer mechanisms in a series of PDI polymers differentiated by their phenylenediamine linkers (o PDI, m PDI, and p PDI). Comparative analysis reveals that o PDI, constructed with ortho-phenylenediamine linkers, exhibits the fastest two-electron CS and the most efficient subsequent FC generation within the series. These superior properties are attributed to its large molecular dipole moment and strong driving force for multielectron transfer. The efficient electron accumulation is also proved by independent two-electron quenching experiments and enables o PDI to serve as an excellent multielectron accumulation model, facilitating the efficient generation of H₂O₂ through two-electron ORR and leading to the rapid degradation of UA. This work thus establishes a clear structure–property relationship and demonstrates a structure-driven strategy for regulating charge-transfer energetics, offering a blueprint for the rational design of efficient accumulative electron-transfer systems for advanced photocatalysis.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.6c00868.

• Experimental details of general methods, Rehm–Weller equations for estimating ΔG _CS and theoretical calculation for two-electron accumulated reactions, transient absorption spectra, two-electron quenching experiment, and photodegradation based on two-electron oxygen reduction reaction results (PDF)

⊥.

Z.S., X.L., and Y.-X.Q contributed equally to this work. Z.S. conceived the study, performed the experiments, conducted data analysis, and drafted the manuscript. X.L. performed the experiments and conducted data analysis. Y.-X.Q. synthesized the compounds. C.Y., Q.C., and M.T. contributed to data curation, formal analysis, and validation. Y.L. and Y.W. contributed to the key methodology. H.S., Q.-X.T., J.-X.J., and Z.K. supervised the research and secured funding. Z.K. oversaw the project and finalized the manuscript. All authors discussed the results and reviewed the manuscript.

The authors declare no competing financial interest.