Molecular Design Strategy of π‐Conjugated Polymers for Efficient Visible‐Light‐Driven Photoelectrocatalytic O₂ Reduction to H₂O₂ Production

Abstract

Toward sustainable hydrogen peroxide (H₂O₂) production, photo(electro)catalytic oxygen (O₂) reduction/H₂O₂ production has attracted increasing attention. Recently, we have found that a thin film of the π‐conjugated polymer, poly(1,4‐bis(2‐thienyl)benzene) (PBTB), exhibits exceptionally high (photo)electrocatalytic activity for O₂ reduction/H₂O₂ production. To achieve higher photoelectrocatalytic activity and efficient visible‐light‐driven photoelectrocatalytic H₂O₂ production, we investigated the molecular design related to the highest occupied molecular orbital (HOMO) energy level (E _HOMO) of these polymers. We designed and synthesized poly(1,4‐bis(2‐thienyl)naphthalene) (PBTN), in which replacing the phenyl unit of PBTB with a naphthalene unit—a stronger electron‐withdrawing group—and increasing the polymer chain twist angle selectively deepened E _HOMO relative to PBTB. The degree of E _HOMO deepening quantitatively affected the onset potential of PBTN. Under visible‐light irradiation and 0 V vs. Ag/AgCl, the PBTN thin film achieved a high O₂ reduction/H₂O₂ production rate (1.11 × 10³  ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst}), 1.47 times higher than that of PBTB, with excellent Coulombic efficiency (99%) and selectivity (99%). The onset potential of PBTN for visible‐light‐assisted O₂ reduction enabled a photocatalytic H₂O₂ production setup. Upon visible‐light irradiation, this setup achieved a high photocatalytic O₂ reduction/H₂O₂ production rate of 128 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_photocathode. These results clearly demonstrate the tunability of the photoelectrocatalytic activity of π‐conjugated polymers through E _HOMO‐related molecular design.

This study demonstrated the high tunability of the photoelectrocatalytic activity of π‐conjugated polymer thin films for oxygen (O₂) reduction/hydrogen peroxide (H₂O₂) production through molecular design. The degree of HOMO energy level deepening quantitatively affected the onset potential for light‐assisted O₂ reduction toward more positive values, enhancing the photoelectrocatalytic H₂O₂ production rate.

1. Introduction

Hydrogen peroxide (H₂O₂) is a representative oxidizing agent and is environmentally friendly, as it produces only water (H₂O) and oxygen (O₂) after use. H₂O₂ plays an important role in various applications, including pulp bleaching [1], chemical synthesis (such as for propylene oxide) [2], wastewater treatment [3], and disinfection [4]. The global production volume of H₂O₂ has been steadily increasing, from 5.50 Mt per year in 2015 to 6.55 Mt in 2024 [5, 6].

Currently, more than 95% of H₂O₂ is produced by the auto‐oxidation (AO) method, which uses anthraquinone derivatives [7]. The AO method allows for large‐scale production and remains one of the most cost‐effective methods for producing H₂O₂ compared to other currently competing methods [8]. However, it has several drawbacks, such as the use of toxic organic solvents (e.g., benzene), reliance on rare and expensive palladium catalysts, and the generation of large amounts of waste solvents to remove byproducts derived from anthraquinone derivatives [9, 10, 11], making it far from a green and sustainable technology. In addition, the AO method requires large‐scale plants for multiple production steps (hydrogenation, oxidation, extraction, and distillation), usually constructed near oil refineries (commonly far from consumption sites) to utilize low‐cost hydrogen gas [11]. These requirements necessitate additional processes for efficient transport of H₂O₂ aqueous solution, such as concentrating the solution to 35–70 wt% (which poses an explosion risk), adding stabilizers, and performing further purification to remove stabilizers. As most applications require just low‐concentration H₂O₂ aqueous solutions (e.g., 0.1–3 wt%) [12, 13], these processes are inefficient. For a sustainable society, an industrial method that minimizes environmental impact and enables facile H₂O₂ production close to consumption sites is urgently required [14].

In this context, H₂O₂ production methods that utilize light and/or electrical energy to reduce O₂ from the air using an appropriate catalyst have attracted significant attention [15, 16, 17]. This approach is green and sustainable because it can produce H₂O₂ from O₂ and H₂O, which are abundant resources on the Earth, while using renewable energy sources such as sunlight [18] and producing almost no waste throughout the process. Photo(electro)catalysts for O₂ reduction/H₂O₂ production, which are essential for this technology, have been extensively investigated to achieve high activity, selectivity, stability, and scalability [17].

As photo(electro)catalysts for H₂O₂ production, inorganic materials such as metal semiconductors (TiO₂, ZnO, BiVO₄) and carbon materials have been primarily reported [19, 20]. In particular, as Y. Shiraishi discovered in 2014 that graphitic carbon nitride (g‐C₃N₄) functions as a photocatalyst for O₂ reduction/H₂O₂ production [21], carbon materials have been intensively studied owing to their composition of earth‐abundant elements (C, N, O, H) and excellent stability [22]. Although the catalytic performance of carbon materials can be improved through the introduction of structural defects and elemental doping [17, 23], their photocatalytic H₂O₂ production rate remains lower than approximately 5.0 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photocatalyst} h. Recently, organic materials with high molecular design flexibility, such as covalent organic frameworks (COFs), have attracted attention, with moderate photocatalytic H₂O₂ production rates of approximately 10 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photocatalyst} h being reported [24, 25, 26]. However, as COFs are cross‐linked, they exhibit low solubility in most solvents and are typically obtained as microcrystalline powders, making film formation and large‐scale applications difficult [27].

More recently, π‐conjugated polymers have emerged as promising (photo)electrocatalysts for O₂ reduction/H₂O₂ production [28, 29, 30]. These π‐conjugated polymers have significant advantages over conventional (photo)electrocatalysts, including diverse molecular design options through organic synthesis, the ability to form films via various coating methods [31, 32, 33, 34], a simplified single‐layer architecture functioning as both light absorber and catalyst, high durability (lasting from several days to weeks) [28, 29], and extremely high Coulombic efficiency and selectivity (>95%) for O₂ reduction/H₂O₂ production as photoelectrocatalysts [28, 29, 30]. For example, we previously demonstrated the high‐purity π‐conjugated copolymer of thiophene and phenylene, poly(1,4‐bis(2‐thienyl)benzene) (PBTB), synthesized by iodine‐vapor‐assisted polymerization method, [29, 31, 35, 36, 37] as a (photo)electrocatalyst with a high photoelectrocatalytic H₂O₂ production rate of around 1.0 × 10³  ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h. Combining a PBTB thin film with an H₂O oxidation/O₂ production electrocatalyst enables the combined setup (photocatalyst) to achieve a high photocatalytic O₂ reduction/H₂O₂ production rate exceeding 100 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photocatalyst} h under visible‐light irradiation [29]. These results demonstrate the high potential of high‐purity π‐conjugated polymers as (photo)electrocatalysts. However, the relationship between the molecular structure of π‐conjugated polymers and their catalytic performance remains unclear.

In this work, we aimed to clearly demonstrate the advantages of π‐conjugated polymers over other materials by developing a molecular design strategy for tuning photoelectrocatalytic activities (e.g., the onset potential) to achieve further enhancement. Specifically, we focused on the highest occupied molecular orbital (HOMO) energy level (E _HOMO) of π‐conjugated polymers, which was expected to influence their photoelectrocatalytic activity (e.g., the onset potential).

2. Results and Discussion

2.1. Preparation and Characterization of PBTN Thin Film

We aimed to synthesize π‐conjugated polymers with a selectively deepened E _HOMO of PBTB, which has been reported to exhibit a high (photo)electrocatalytic activity [29], and therefore explored suitable monomers. First, we sought to deepen only E _HOMO by introducing electron‐withdrawing groups into 1,4‐bis(2‐thienyl)benzene (BTB), a monomer of PBTB. However, introducing electron‐withdrawing groups typically deepens both E _HOMO and the lowest unoccupied molecular orbital (LUMO) level (E _LUMO) [38]. As shown in Table S1, quantum chemical calculations also indicated similar trends for BTB derivatives containing common electron‐withdrawing groups (e.g., −COOH and −F) on the phenylene unit of BTB. Then, we focused on the fact that as the twist angle of the polythiophene chain increases, the π‐orbital overlap within the polymer decreases, resulting in a deeper E _HOMO and a shallower E _LUMO [39]. We hypothesized that both introducing electron‐withdrawing groups and increasing the twist angle of the π‐conjugated polymer chain would allow us to deepen only the E _HOMO without significantly altering E _LUMO. Based on this hypothesis, we designed poly(1,4‐bis(2‐thienyl)naphthalene) (PBTN), in which the phenylene unit in PBTB was replaced with a naphthalene unit. The naphthalene unit is a stronger electron‐withdrawing group than that of the phenylene unit [40, 41], and the repulsion between hydrogen atoms on the naphthalene and thiophene units, as shown in Figure 1a, was expected to increase the twist angle of the polymer chain (Figures 1a and S1).

FIGURE 1.

(a) Chemical structure of PBTB and PBTN. (b) Schematic illustration of the iodine‐vapor‐assisted polymerization procedure for the PBTN thin film. The image shows the PBTN thin film formed on a glass plate. (c) MALDI‐TOF MS spectrum of PBTN. (d) FT‐IR spectra of PBTN (red trace) and BTN (black trace) normalized by the peaks at 1380 cm⁻¹, corresponding to the C—C stretching vibration of naphthalene units [42]. (e) UV‐Vis absorption spectra of PBTN thin film and PBTB film normalized to film thickness. (f) HOMO/LUMO energy levels and energy gaps of PBTB and PBTN.

As shown in Figure 1b, the PBTN thin film was prepared by iodine‐vapor‐assisted polymerization of 1,4‐bis(2‐thienyl)naphthalene (BTN) (detailed procedures are summarized in the Supporting Information). After polymerization, the PBTN thin film was soaked in ethanol to remove residual iodine and monomers, resulting in a high‐purity and homogeneous PBTN thin film with a thickness of 8–30 nm (confirmed by energy‐dispersive X‐ray spectroscopy (EDX), X‐ray photoelectron spectroscopy (XPS), and scanning electron microscope (SEM), as shown in Figures S2–S4). The chemical structure of PBTN was confirmed by Raman spectroscopy (Figure S5 and Table S2). PBTB was also prepared by iodine‐vapor‐assisted polymerization, and its chemical structure was confirmed by XPS and Raman spectroscopy (Figures S3 and S5 and Table S3). The film thickness could be controlled by adjusting the concentration of the monomer solution and the spin‐coating speed.

The progress of polymerization was confirmed by matrix‐assisted laser desorption ionization‐time of flight mass spectroscopy (MALDI‐TOF MS) and Fourier transform infrared (FT‐IR) spectroscopy (Figure 1c,d). As shown in Figure 1c, the MALDI‐TOF MS spectrum indicated the formation of polymers comprising ≥20 BTN units. As shown in Figure 1d, the FT‐IR measurements demonstrated differences between the spectra of PBTN and BTN in the 600–800 cm⁻¹ region, which corresponded to the C–H out‐of‐plane bending vibration (Table S4). In BTN, a strong absorption peak appeared at 696 cm⁻¹, corresponding to the 2‐monosubstituted thiophene ring [43]. In contrast, in PBTN, the absorbance at 696 cm⁻¹ decreased, and a new peak appeared at 796 cm⁻¹, corresponding to the 2,5‐disubstituted thiophene ring. These results indicated the formation of PBTN.

The ultraviolet–visible spectroscopy (UV‐Vis) spectra (Figure 1e) showed that the HOMO/LUMO energy gap (E _g), which was determined from the onset of absorption, was 2.42 and 2.19 eV for PBTN and PBTB. In addition, while PBTB exhibited multiple absorption peaks at 450, 490, and 540 nm due to intermolecular π‐stacking, PBTN exhibited only a single peak at 407 nm. This result supports the conclusion that the twist angle increased when the phenylene unit was replaced with a naphthalene unit, leading to weaker intermolecular π‐stacking interactions in PBTN than those in PBTB. The UV‐Vis spectra (Figure 1e) and atmospheric photoelectron spectroscopy results (Figure S6) showed that E _HOMO and E _LUMO of each π‐conjugated polymer were −3.19 and −5.61 eV for PBTN and −3.23 and −5.42 eV for PBTB, respectively (Figure 1f). Therefore, by replacing the phenyl unit of PBTB with a naphthalene unit, we succeeded in selectively deepening E _HOMO by 0.19 eV (Figure 1f).

2.2. Photo and Electrochemical Properties of PBTN Thin Film

First, we investigated the electrocatalytic ability of the PBTN thin film under dark conditions. As shown in Figure S7, similar to the PBTB thin films which we reported previously [29], PBTN thin films exhibited high electrocatalytic ability for the O₂ reduction/H₂O₂ (HO₂ ⁻) production reaction (Equation (2)) in alkaline aqueous solution at pH 12. In addition, as shown in Figure S8, this current almost disappeared under Ar bubbling, indicating that the PBTN thin film was reacting selectively with O₂.

$${\mathrm{O}}_{\text{2}}+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\rightleftarrows {\mathrm{H}}_2{\mathrm{O}}_2\mathit{\text{E}}°=0.695\hspace{0.17em}\text{V vs}.\hspace{0.17em}\text{SHE}\text{pH}\le 11.6$$ (1)

$$\begin{array}{c}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}+2{\mathrm{e}}^{-}\rightleftarrows {\text{HO}}_{2^{-}}+{\text{OH}}^{-}\mathit{\text{E}}°=0.358\hspace{0.17em}\text{V vs}.\hspace{0.17em}\text{SHE}\\ \text{\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}pH}\hspace{0.17em}>\hspace{0.17em}11.6\hspace{0.17em}({\mathit{\text{E}}}_{\text{eq}}=-0.190\hspace{0.17em}\text{V vs}.\hspace{0.17em}\text{Ag}/\text{AgCl}\\ \text{\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{0.17em}\hspace{0.5em}at pH}12)\end{array}$$ (2)

Next, using the photoelectrochemical cell setup shown in Figure 2a,b, we demonstrated the photoelectrocatalytic ability of the PBTN thin films. The PBTN thin films are considered to produce H₂O₂ (HO₂ ⁻ in the alkaline aqueous solution at pH 12) under visible‐light irradiation through the three steps as shown in Figure 2c. Specifically, (1) the PBTN thin films absorb visible‐light (Figure 1e) and generate a hole (h⁺)‐electron (e⁻) pair in the HOMO and LUMO. (2) The hole‐electron pair is separated; the electron diffuses to the surface of the PBTN thin film, while the hole diffuses to the glassy carbon (GC) substrate. (3) The electron is donated to O₂, facilitating the reaction in Equation (2). The hole in the HOMO is neutralized by electron donation from the external circuit, and with an appropriate catalyst, the H₂O oxidation/O₂ production reaction in (Equation (3)) occurs at the counter electrode. For the external circuit to donate electrons to the HOMO, the potential applied to the PBTN thin films must be more negative than E _HOMO. Therefore, E _HOMO is expected to significantly influence the onset potential.

FIGURE 2.

(a) Schematic of the photoelectrochemical cell setup. (b) Image of the photoelectrochemical cell setup. (c) Mechanism of photoelectrochemical H₂O₂ production by the PBTN thin film. (d) Linear sweep voltammograms (LSVs) recorded for the PBTN thin film (28 nm) as a cathode at 10 mV/s and pH 12 under air bubbling (4.0 mL/min). Black trace: under dark conditions; red trace: under visible‐light. The onset potential under visible‐light irradiation was defined as the point where the photocurrent curve intersects the x‐axis. Under dark conditions, the onset potential was defined as the falling edge of the reduction current in the LSV. (e) LSVs recorded for PBTN (28 nm, red trace) and PBTB (30 nm, black trace) thin films as cathodes under visible‐light at 10 mV/s and pH 12 under air bubbling (4.0 mL/min). (f) CA recorded for PBTN (28 nm, red trace) and PBTB (30 nm, black trace) thin films under 0 V vs. Ag/AgCl, air bubbling (200 mL/min) and visible‐light irradiation at pH 12. (g) Photoelectrocatalytic O₂ reduction/H₂O₂ production rates of PBTN (8.0 nm) and PBTB (8.0 nm) thin films measured under 0 V vs. Ag/AgCl, air bubbling (200 mL/min) and visible‐light irradiation at pH 12.

$${\mathrm{O}}_{\text{2}}+2{\mathrm{H}}_2\mathrm{O}+4{\mathrm{e}}^{-}\rightleftarrows 4{\text{OH}}^{-}\mathit{\text{E}}°=0.401\hspace{0.17em}\text{V vs}.\hspace{0.17em}\text{SHE}$$ (3)

As shown in Figure 2d,e, the onset potential for O₂ reduction/H₂O₂ production of PBTN thin films under visible‐light irradiation was +0.67 V vs. Ag/AgCl (+1.56 V vs. RHE at pH 12), which was 0.21 V more positive than that of PBTB thin films (Table 1). Therefore, the difference in onset potential for light‐assisted O₂ reduction/H₂O₂ production between PBTN and PBTB thin films (0.21 V, both approximately 30 nm thickness) closely matches the difference in their E _HOMO values (0.19 eV). This result indicates a strong correlation between E _HOMO of the π‐conjugated polymers and their onset potential for O₂ reduction/H₂O₂ production under visible‐light irradiation. In addition, as shown in Figure 2f, chronoamperometry (CA) measurements were performed on PBTN and PBTB thin films with nearly identical thicknesses (30 nm) under visible‐light irradiation at 0 V vs. Ag/AgCl. The amount of H₂O₂ produced after the CA measurements was determined by spectrophotometric titration (Figure S9) [44]. The photoelectrocatalytic O₂ reduction/H₂O₂ production rates for the PBTN and PBTB thin films were 468 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h and 318 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h, respectively (Table 1). Therefore, the photoelectrocatalytic O₂ reduction/H₂O₂ production rate of the PBTN thin film was 1.47 times higher than that of the PBTB thin film. This is presumably because the more positive onset potential of the PBTN thin film compared with that of the PBTB thin film (Figure 2e) enables higher current density under the same potential of 0 V vs. Ag/AgCl. In addition, when the film thickness was adjusted to 8 nm, the PBTN thin films exhibited a high rate of O₂ reduction/H₂O₂ production of 1.11 × 10³  ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h (37.7 ${\mathrm{g}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h) with a remarkably high Coulombic efficiency (99%) and selectivity (99%) under visible‐light irradiation, air bubbling, and 0 V vs. Ag/AgCl. This photoelectrocatalytic O₂ reduction/H₂O₂ production rate was 1.58 times higher than that of the PBTB thin film (8 nm), which produced 703 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h (23.9 ${\mathrm{g}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} h), as shown in Figure 2g and Table 1. Furthermore, the photoelectrocatalytic O₂ reduction/H₂O₂ production rate of PBTN thin film was higher than that reported in previous works [30, 45]. The higher photoelectrocatalytic performance observed in the thinner film is attributed to the homogeneity of the films prepared in this work, ensuring that the surface area did not decrease even when the film thickness was reduced. This was supported by electrochemical double‐layer capacitance measurements (Figure S10). In addition, SEM and Raman spectroscopy confirmed that the PBTN thin films did not degrade after the CA measurements (Figures S11 and S12). The energy levels and photoelectrocatalytic performances of PBTN and PBTB are summarized in Table 1.

TABLE 1.

HOMO/LUMO energy levels and photoelectrocatalytic O₂ reduction/H₂O₂ production rates of PBTN and PBTB.

Polymer
E HOMO (eV)	−5.42	−5.61
E LUMO (eV)	−3.23	−3.19
E g (eV)	2.19	2.42
Onset potential (V vs. RHE at pH 12)	+1.35	+1.56
Photoelectrocatalytic O2 reduction/H2O2 production rate (30 nm) (${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/gphotoelectrocatalyst h)	318	468
Photoelectrocatalytic O2 reduction/H2O2 production rate (8 nm) (${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/gphotoelectrocatalyst h)	703	1.11 × 103

2.3. Photocatalytic H₂O₂ Production by Combining PBTN Thin Film and Ni Foam

Finally, as shown in Figure 3a, a full‐cell setup was fabricated using the PBTN thin film (19 nm) as the photocathode, a common H₂O oxidation/O₂ production electrocatalyst (Ni foam) as the anode [46], and 0.01 M NaCl aqueous solution (pH 12) as the electrolyte. Air bubbling and visible‐light irradiation on the photocathode enabled this setup to generate a reduction current (Figure 3b) and achieve a high photocatalytic O₂ reduction/H₂O₂ production rate of 128 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_photocathode h (4.40 ${\mathrm{g}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_photocathode h), which was 4.9 times higher than that obtained with PBTB thin film (22 nm). As shown in Figure 3c, the highly positive onset potential of the PBTN thin film for visible‐light‐assisted O₂ reduction/H₂O₂ production (+1.56 V vs. RHE at pH 12) enabled efficient photocatalytic O₂ reduction/H₂O₂ production without the application of a bias potential.

FIGURE 3.

(a) Schematic of the full‐cell setup in the absence of a bias potential at pH 12 using a PBTN thin film (under visible‐light) and Ni foam as the cathode and anode. (b) CA recorded for the PBTN thin film (19 nm) in the absence of a bias potential under air bubbling (200 mL/min) and visible‐light at pH 12. (c) Mechanism of photocatalytic H₂O₂ production by PBTN.

3. Conclusion

In this study, we found that replacing the phenyl unit of PBTB, which has previously been reported to exhibit high photoelectrocatalytic activity [29], with a naphthalene unit (a stronger electron‐withdrawing group) and increasing the twist angle of the polymer chain enabled E _HOMO to become selectively deeper than that of PBTB. PBTN thin films exhibited a more positive onset potential than that of PBTB thin film and achieved a high O₂ reduction/H₂O₂ production rate of 1.11 × 10³  ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_{photoelectrocatalyst} with a remarkably high Coulombic efficiency (99%) and selectivity (99%) under visible‐light irradiation and 0 V vs. Ag/AgCl. In addition, visible‐light irradiation enabled a full‐cell setup combining the PBTN thin film with a common H₂O oxidation/O₂ production electrocatalyst (Ni foam) to achieve a high photocatalytic O₂ reduction/H₂O₂ production rate of 128 ${\mathrm{mmol}}_{{\mathrm{H}}_2{\mathrm{O}}_2}$/g_photocathode h. These results demonstrate the high tunability of the photoelectrocatalytic ability of π‐conjugated polymers through appropriate molecular design. We expect that π‐conjugated polymers exhibiting high H₂O₂ production rates, suitability for device fabrication, and high durability in previous reports [28, 29] will significantly advance the development of green and sustainable O₂ reduction/H₂O₂ production technologies. In our ongoing work, we are planning to further enhance the photoelectrocatalytic activities of π‐conjugated polymers based on the design strategy presented in this study and to demonstrate continuous H₂O₂ production over extended periods, such as several months, using the device (or photocatalyst).

Supporting Information

Additional supporting information can be found online in the Supporting Information section. The authors have cited additional references within the Supporting Information [31, 36, 43, 44, 47, 48]. Supporting Scheme S1: Synthesis of 1,4‐bis(2‐thienyl)naphthalene. Supporting Fig. S1: Most stable structures of BTN and BTB were calculated using DFT B3LYP/6‐31G. Supporting Fig. S2: EDX elemental analysis of PBTN thin film. EDX measurements gave only peaks assignable to C, S, and O with no peaks ascribable to residual oxidant (i.e., iodine) in the PBTN thin film (below the detection limit). Supporting Fig. S3: XPS measurements of (a) PBTN thin film and (b) PBTB thin film coated on glassy carbon. XPS measurements gave only peaks assignable to C, S, and O with no peaks ascribable to residual oxidant (i.e., iodine) in the PBTN and PBTB thin films (below the detection limit). Supporting Fig. S4: SEM images of PBTN film. SEM was taken on the PBTN film formed on the GC plate. The PBTN film exhibited a homogeneous surface structure on a 100 nm scale. Supporting Fig. S5: Raman spectra of (a) PBTN and (b) PBTB thin films. Raman laser wavelength is 785 nm. Detailed assignments are summarized in Table S2 and Table S3. Supporting Fig. S6: Photoelectron spectra of (a) PBTN and (b) PBTB measured by photoemission yield spectroscopy in air. The ionization potential was calculated by selecting a linearly arranged plot on the analysis software. In this work, the ionization potential was approximated as the HOMO energy level (E _HOMO). Supporting Fig. S7: a, b) Linear sweep voltammograms (LSVs) recorded for PBTN as a cathode at 10 mV/s and different pHs. The electrocatalytic ability of PBTN was investigated under dark conditions and air bubbling. The PBTN thin film formed on GC plate was electrochemically tested at pH 2–12. The electrochemical response at pH 12 was clearly different from that at lower pH. Supporting Fig. S8: LSV recorded for PBTN as a cathode under dark conditions at 10 mV/s and pH 12. Dash trace: under Ar bubbling. Black trace: under air bubbling (4.0 mL/min). Supporting Fig. S9: Calibration plots for the determination of H₂O₂ concentration. (a) The calibration curve was created by plotting the absorption at 456 nm in the UV‐vis spectra. (b) UV‐vis spectra of solutions at different H₂O₂ concentrations. Supporting Fig. S10: Linear relationship between the scan rate and current density of PBTN thin film. Black: 8 nm, Gray: 30 nm. The capacitive currents were measured at 2.0 V vs. Ag/AgCl from cyclic voltammograms recorded in the non‐Faradaic potential range of 1.5 to 2.5 V vs. Ag/AgCl. The data points represent the average of the absolute values of anodic and cathodic current densities ((|J _a| + |J _c|)/2) at each scan rate. The electrochemical double‐layer capacitance (C _DL) was determined from the slope of the linear fit. CDL is proportional to the electrochemically active surface area (ECSA) [4]. These results indicate that the PBTN thin film has a comparable or larger ECSA at 8 nm compared to that at 30 nm. This is presumably because thinner films are more susceptible to the minute irregularities of the substrate (glassy carbon) and this effect is thought to be the reason why the PBTN thin film (8 nm) has a larger ECSA than that of 30 nm. Supporting Fig. S11: SEM images of the PBTN thin film from the (a) the untested area and (b) area after CA measurement (under the same conditions as Figure 2f). The SEM images revealed no significant changes in the surface morphology of the PBTN thin film after the CA measurement. Supporting Fig. S12: Raman spectra of PBTN thin films before (black trace) and after (red trace) the CA measurement (under the same conditions as Figure 2f). The Raman spectra showed that the chemical structure of PBTN was maintained before and after the CA measurement. Supporting Table S1: HOMO/LUMO energy level calculated by DFT B3LYP/6‐31G. Supporting Table S2: Assignments of the Raman spectrum of PBTN [6, 7, 8, 9]. Supporting Table S3 : Assignments of the Raman spectrum of PBTB [6, 9]. Supporting Table S4: Assignments of the IR spectra of PBTN and BTN [5, 6]. Supporting Table S5: Summary of solutions used for density tests. Supporting Table S6: Cartesian coordinates of the entry 1 monomer in Table S1. Supporting Table S7: Cartesian coordinates of the entry 2 monomer in Table S1. Supporting Table S8: Cartesian coordinates of the entry 3 monomer in Table S1. Supporting Table S9: Cartesian coordinates of the entry 4 monomer in Table S1. Supporting Table S10: Cartesian coordinates of the entry 5 monomer in Table S1. Supporting Table S11: Cartesian coordinates of the entry 6 monomer in Table S1. Supporting Table S12: Cartesian coordinates of the entry 7 monomer in Table S1. Supporting Table S13: Cartesian coordinates of the entry 8 monomer in Table S1. Supporting Table S14: Cartesian coordinates of the entry 9 monomer in Table S1. Supporting Table S15: Cartesian coordinates of the entry 10 monomer in Table S1.

Funding

This study was supported by the Grants‐in‐Aid for Scientific Research (JP23K17945 (K. Oka), JP23H03827 (K. Oka), JP24K01552 (K. Oka), and JP25K21722 (K. Oka)) and Environment Research and Technology Development Fund (JPMEERF20241RA4, (K. Oka)).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.