Skip to main content
NCBI home page
JACS Au logoLink to JACS Au
. 2023 Jul 24;3(8):2237–2246. doi: 10.1021/jacsau.3c00262

Nafion-Integrated Resorcinol-Formaldehyde Resin Photocatalysts for Solar Hydrogen Peroxide Production

Yasuhiro Shiraishi †,‡,*, Masahiro Jio †, Koki Yoshida †, Yoshihiro Nishiyama †, Satoshi Ichikawa ∥, Shunsuke Tanaka §, Takayuki Hirai †
PMCID: PMC10466369  PMID: 37654590

Abstract

graphic file with name au3c00262_0010.jpg

Photocatalytic generation of H2O2 from water and O2 is a promising strategy for liquid solar-fuel production. Previously reported powder photocatalysts promote a subsequent oxidative/reductive decomposition of the H2O2 generated, thereby producing low-H2O2-content solutions. This study reports that Nafion (Nf)-integrated resorcinol-formaldehyde (RF) semiconducting resin powders (RF@Nf), synthesized by polycondensation of resorcinol and formaldehyde with an Nf dispersion solution under high-temperature hydrothermal conditions, exhibit high photocatalytic activities and produce high-H2O2-content solutions. Nf acts as a surface stabilizer and suppresses the growth of RF resins. This generates small Nf-woven resin particles with large surface areas and efficiently catalyze water oxidation and O2 reduction. The Nf-woven resin surface, due to its hydrophobic nature, hinders the access of H2O2 and suppresses its subsequent decomposition. The simulated-sunlight irradiation of the resins in water under atmospheric pressure of O2 stably generates H2O2, producing high-H2O2-content solutions with more than 0.06 wt % H2O2 (16 mM).

Keywords: Photocatalysis, hydrogen peroxide, polymeric semiconductor, artificial photosynthesis, solar fuel

Introduction

Artificial photosynthesis, the transformation of earth-abundant resources into fuels under sunlight, is a challenging issue for the development of a sustainable energy society.1 Several reactions such as water splitting (H2 generation),2,3 CO2 reduction (CO, HCOOH, CH3OH, and CH4 generation),4,5 and N2 reduction (NH3 generation)6,7 have been studied extensively for solar fuel generation. Recently, hydrogen peroxide (H2O2), a storable and transportable liquid that can generate electricity in a fuel cell, has gained prominence as a solar fuel.8 H2O2 can be generated under ambient conditions by the semiconductor photocatalysis of water and O2 (eq 1);9,10 photogenerated valence band holes (VB h+) oxidize water (eq 2), while conduction band electrons (CB e–) promote the two-electron reduction of O2 to generate H2O2 (eq 3),11,12 as follows:

graphic file with name au3c00262_m001.jpg 1
graphic file with name au3c00262_m002.jpg 2
graphic file with name au3c00262_m003.jpg 3

Although several powder photocatalysts have been proposed, they show poor performance.9−12 This can be attributable to their low water-oxidation activity (eq 2), poor selectivity for the two-electron O2 reduction (eq 3), and low long-wavelength light absorption (λ > 500 nm). Thus, it is vital to design photocatalysts that efficiently promote water oxidation and O2 reduction upon absorbing long-wavelength light.

Resorcinol-formaldehyde (RF) resins13 are traditionally labeled insulators;14−16 however, they show the properties of n-type semiconductors when synthesized by polycondensation under high-temperature hydrothermal conditions and show high H2O2-generation activity.17 Acid-catalyzed polycondensation produces RF resin powders comprising the quinoid forms of resorcinol, π-conjugated with the benzenoid forms of resorcinol (Scheme 1a).18−20 The π-stacking interactions of the benzenoid–quinoid donor–acceptor (D-A) units (Scheme 1b) hybridize the energy levels and create a semiconducting band with a low bandgap energy (Ebg = ∼1.7 eV), with the VB and CB composed of D and A units, respectively (Scheme 1c), as confirmed by density functional theory calculations.17 The resins absorb long-wavelength light up to ∼700 nm and efficiently catalyze water oxidation (eq 2) and direct two-electron O2 reduction (eq 3), as confirmed by the photoreactions with isotope-labeled reagents17 and electron spin resonance analysis.20 However, it is difficult to describe the detailed reaction mechanism on the resins at the molecular level, as encountered in all of the reported polymeric semiconductor photocatalysts.21 Simulated-sunlight illumination of the RF resins in water under atmospheric pressure of O2 produces H2O2 at a solar-to-chemical conversion (SCC) efficiency of more than 0.5% after 5 h of photoirradiation,17−20 which is higher than the efficiency of previously reported photocatalysts.9−12 Thus, the fabrication of RF resins using inexpensive reagents may open a new route for sustainable sunlight-driven H2O2 generation.

Scheme 1. (a) Structure, (b) Three-Dimensional Architecture, and (c) Semiconducting Band Image of Pristine RF Resin.

Scheme 1

Open in a new tab

However, for practical applications, it is vital to improve the H2O2 generation efficiency of RF resins. Acid-catalyzed polycondensation proceeds slowly,14−16 producing nonporous and spherical resin particles with large size (∼3 μm) and small specific surface area (∼1 m2 g–1).18−20 Therefore, increasing their surface area by decreasing the particle size could cause activity enhancement. Current research is also focused on stable H2O2 generation under prolonged photoirradiation. As encountered in all of the previously reported photocatalytic systems,9−12 H2O2 generated on the resin undergoes subsequent photocatalytic decomposition due to oxidation by VB h+ (the reverse reaction of eq 3) and reduction by CB e– (eq 4). Although increasing the photoirradiation time increases H2O2 concentration, it also promotes subsequent decomposition of H2O2; thus, a long photoirradiation time decreases the overall H2O2 generation efficiency, producing low-H2O2-content solutions. Therefore, for the production of high-H2O2-content solutions, it is vital to enhance the RF resin activity by enlarging their surface area and to suppress the subsequent decomposition of H2O2.

graphic file with name au3c00262_m004.jpg 4

Nafion (Nf), an anionic perfluorosulfonic polymer, is widely used as a H+-permeable membrane in fuel cells (Scheme 2a, left).22 It consists of a hydrophobic poly(tetrafluoroethylene) (PTFE) backbone (green) and a hydrophilic sulfonate side chain (black); due to its amphiphilic structure, it exhibits both water-impermeable and H+-permeable properties.23 Thus, the integration of Nf (with amphiphilic properties) with RF resins could affect the resin morphology and the H2O2-decomposition activity. This study describes the synthesis of RF resin powders by polycondensation of resorcinol and formaldehyde with a commercially available Nf dispersion solution under the high-temperature hydrothermal conditions. The Nf-integrated RF resins (RF@Nf) consisted of small particles (with an average size of ∼0.6 μm) with a large specific surface area (∼30 m2 g–1) and exhibited high H2O2-generation activity. Additionally, due to the Nf-weaving within the structure, the resins possess a hydrophobic surface; this suppresses the H2O2 access to the resin surface and hinders the decomposition of H2O2. Therefore, the fabricated RF@Nf resins stably generated H2O2 even under prolonged photoirradiation and produced high-H2O2-content solutions.

Scheme 2. (a) Proposed Formation Mechanism, (b) Three-Dimensional Architecture, and (c) Semiconducting Band Image of RF@Nf Resins.

Scheme 2

Open in a new tab

Results and Discussion

Synthesis and Activity of Nf-Integrated Resins

RF@Nf-x powders were synthesized by an acid-catalyzed high-temperature hydrothermal method (see Methods in the Supporting Information), similar to the method used for the synthesis of pristine RF.18 Resorcinol, formaldehyde, oxalic acid, and x mL of a commercially available 5 wt % Nf dispersion solution were added to water (80 mL), and the colloidal suspension was kept at 523 K for 12 h under hydrothermal conditions. Washing of the resulting solids by Soxhlet extraction with acetone followed by drying in vacuo yielded the RF@Nf-x powders. As shown in Figure 1a, pristine RF is a red-black powder;18 this color faded on adding increasing amounts of Nf (x) during resin synthesis (the resin synthesized with x ≥ 1 was a brown powder).

Figure 1.

Figure 1

Open in a new tab

Properties of the resins. (a) Diffuse-reflectance UV–vis spectra of the fresh resins and RF@Nf-1.0 recovered after photoreaction under a solar simulator for 24 h (Figure 5a). (b) Zeta potential of the resins and an Nf dispersion solution. (c) Photographs of the resins (50 mg) when added to water (30 mL). (d) Electronic band structures of the resins. (e) Nyquist plots of the resin-loaded FTO in 0.1 M Na2SO4 solution under visible light at a bias of 1.0 V (vs Ag/AgCl), where the equivalent circuit model indicates the ohmic resistance (RS), double-layer capacitance (CDL), and CT resistance (RCT). (f) Photocurrent of the resin-loaded FTO monitored under photoirradiation in 0.1 M Na2SO4 at a bias of 0.2 V (vs Ag/AgCl).

Photoreactions were performed by visible light irradiation (λ > 420 nm) of the resin (50 mg) suspended in water (30 mL) using a Xe lamp with magnetic stirring under an O2 atmosphere (1 bar) at 298 K. The emission spectra of the light sources are presented in Figure S1. Figure 2a shows the amount of H2O2 generated on the different resins during 6 h of photoirradiation. Pristine RF produced 39 μmol of H2O2, and the resin activity increased on integrating larger amounts of Nf. RF@Nf-1.0 generated the highest amount of H2O2 (57 μmol), which was ∼1.5 times that generated by RF. However, excess integration of Nf (RF@Nf-2.0 and -5.0) decreased the activity. Notably, the photoirradiation of pristine RF with 1.0 mL of an Nf dispersion solution (RF + Nf-1.0) or the photoirradiation of RF after the hydrothermal treatment with 1.0 mL of an Nf dispersion solution (RF/Nf-1.0) did not cause any activity enhancement. Thus, the RF@Nf resins synthesized by polycondensation with an appropriate amount of Nf showed enhanced photocatalytic activity.

Figure 2.

Figure 2

Open in a new tab

Photocatalysis properties of the resins. (a) H2O2 generated during 6 h of photoirradiation [water (30 mL), resin (50 mg), O2 (1 bar), λ > 420 nm (Xe lamp), 298 K]. The RF + Nf-1.0 run was performed using pristine RF with an Nf dispersion solution (1.0 mL), and the RF/Nf-1.0 run was performed using the RF after the hydrothermal treatment with an Nf dispersion solution (1.0 mL). (b) Change in the amount of H2O2 generated with time. Changes in the H2O2 amount during photoirradiation (c) with NaIO3 (electron acceptor) and (d) with benzyl alcohol (electron donor) [H2O2 (200 μmol), water (30 mL), resin (50 mg), NaIO3 (2.5 mM), or benzyl alcohol (350 mM), Ar (1 bar), λ > 420 nm (Xe lamp), 298 K]. (e) Absorption spectra of RF and RF@Nf-1.0, and action spectra for H2O2 generation on the resins.

Figure 2b shows the change in the amount of H2O2 generated with photoirradiation time. For RF (black), the amount of H2O2 increased linearly under <9 h irradiation; however, the rate of H2O2 generation decreased after prolonged irradiation because of the subsequent decomposition of H2O2 on the RF surface. In contrast, RF@Nf-1.0 (red) and -5.0 (blue) maintained higher rates of H2O2 generation even under prolonged photoirradiation, indicating a suppression of H2O2 decomposition. However, the photoirradiation of pristine RF with 1.0 mL of an Nf dispersion solution (RF + Nf-1.0, green) or photoirradiation of the RF after the hydrothermal treatment with 1.0 mL of an Nf dispersion solution (RF/Nf-1.0, purple) decreased the rate of H2O2 generation, as did RF (black). Thus, the RF@Nf resins synthesized by polycondensation with Nf inhibit subsequent H2O2 decomposition and show high H2O2-generation activity even under prolonged photoirradiation.

Size and Morphology of the Resins

The enhanced activity of RF@Nf originates from its enlarged surface area, which facilitates efficient water oxidation and O2 reduction. According to scanning electron microscopy (SEM), pristine RF (Figure 3a) was composed of spheres, 2–5 μm in size.18 In contrast, RF@Nf-x samples were the linkages of smaller particles, where the sizes decreased with increasing the amount of Nf integrated: RF@Nf-1.0 (Figure 3b) and -5.0 (Figure 3c) were composed of ∼0.6 and ∼0.3 μm particles, respectively. Dynamic light scattering analysis (Figure 3d) revealed that the average particle-size of RF (3.1 μm) decreased to 2.3 μm (RF@Nf-1.0), 1.1 μm (RF@Nf-2.0), and 0.5 μm (RF@Nf-5.0), confirming the formation of smaller resin particles on polycondensation with Nf. The N2 adsorption/desorption analysis of RF@Nf-1.0 showed a type-III isotherm, similar to that of RF (Figure S2), indicating the formation of nonporous resins. However, the specific surface area of RF@Nf-1.0 (36.5 m2 g–1) was higher than that of RF (1.0 m2 g–1),18 confirming that the polycondensation with Nf produced small particles with large surface areas.

Figure 3.

Figure 3

Open in a new tab

SEM images and size distributions of the resins. SEM images of (a) RF, (b) RF@Nf-1.0, and (c) RF@Nf-5.0. (d) Size distributions of the resins determined by dynamic light scattering analysis.

A notable feature of RF@Nf observed by SEM was the formation of small clusters on the surface; ∼20 nm clusters were dispersed on RF@Nf-1.0 (Figure 3b, bottom), while ∼100 nm clusters were formed on RF@Nf-5.0 (Figure 3c, bottom). Transmission and scanning transmission electron microscopy (TEM/STEM) were used to analyze the resin-surface structure. RF (Figure 4a) was ∼3.0 μm spheres with a smooth surface,18 RF@Nf-1.0 (Figure 4b) contained clusters dispersed on the linkages of ∼0.6 μm particles, and RF@Nf-5.0 contained large clusters on the linkages of ∼0.3 μm particles (Figure 4c). These features are consistent with those observed by SEM (Figure 3a–c). As shown in Figure 4d, the bright-field (BF) image of RF@Nf-1.0 contained dark-colored surface clusters, ∼20 nm in size, and the annular dark-field (ADF) image contained brightly colored clusters; thus, the surface clusters of the resins mainly consisted of F elements (heavier than C and O).24 Other RF@Nf-1.0 particles showed similar surface clusters (Figures S3–S5). As shown in Figure S6, TEM images of an Nf dispersion solution (dropped onto a sample grid and dried) showed ∼10 nm clusters, indicating that the surface clusters were mainly composed of Nf. As shown in Figures S7–S9, RF@Nf-5.0 contained large Nf clusters that covered the linkages of ∼0.3 μm resin particles, which are consistent with SEM observations (Figure 3c).

Figure 4.

Figure 4

Open in a new tab

TEM/STEM images and EDS maps of the resins. TEM images of (a) RF, (b) RF@Nf-1.0, and (c) RF@Nf-5.0. (d) BF- and ADF-STEM images of RF@Nf-1.0 and EDS maps of C (red), O (blue), and F (green).

The surface Nf clusters were created during the polycondensation of resorcinol and formaldehyde with Nf. As shown in Figure S10, when RF was treated with an Nf dispersion solution under hydrothermal conditions (RF/Nf-1.0), it showed a smooth surface, similar to that of pristine RF (Figures 3a and 4a); thus, Nf scarcely interacted with RF by the post-treatment. The infrared spectra of RF@Nf-x (Figure S11) did not show a new band, indicating that RF was not chemically bonded with Nf. Thus, parts of Nf were “physically woven” within the resin particles during the polycondensation of resorcinol with formaldehyde, and the exposed Nf parts formed surface clusters. The weaving of Nf within RF particles was confirmed by STEM–energy-dispersive X-ray spectroscopy (EDS). As shown in Figure 4d and Figures S3–S5, the EDS maps of RF@Nf-1.0 indicate that C, O, and F were distributed over the entire particle; F was not localized on the surface clusters. A similar entire distribution of F was observed for RF@Nf-5.0 (Figures S7–S9), confirming that Nf was woven within the resin.

Composition of the Resins

The combustion method was used for the elemental analysis of the resins (Table S1), where RF consists of C, O, and H, and Nf consists of C, O, H, F, and S. Table 1 (left) summarizes the total C, O, F, and S mole compositions of the resins; the O compositions were determined by subtracting the sum of the C, H, F, and S amounts from the total amounts. RF@Nf-1.0 and -5.0 also contained F and S, with the latter containing larger amounts of both elements. X-ray photoelectron spectroscopy (XPS) was used for characterization of the resin surface (Figure S12). The O 1s spectra of RF@Nf-1.0 and -5.0 (Figure S13) showed C=O (531.4 eV) and C–O (533.0 eV) peaks corresponding to resin components, while the S=O peak for Nf (537.3 eV)25 was not observed, possibly due to the relatively low amount of S=O. In contrast, the S 2p spectra of RF@Nf-x (Figure S14) showed an S=O peak (168.1 eV),25 confirming the integration of Nf. The C 1s spectrum of Nf (Figure S15) contained two peaks assigned to the hydrophobic −CF2– backbone (293.5 eV) and hydrophilic −OCF2– chain (291.6 eV).25 The RF spectrum showed C–C (284.6 eV), C–O (286.3 eV), and C=O (288.9 eV) peaks corresponding to resin components.17,18 The RF@Nf-1.0 and -5.0 spectra also contained the −OCF2– peak; however, they did not contain the −CF2– peak. The F 1s spectrum of Nf (Figure S16) showed −CF2– (690.3 eV) and −OCF2– (688.8 eV) peaks,26 whereas the RF@Nf-1.0 and -5.0 spectra contained only the −OCF2– peak. These C 1s and F 1s XPS results indicate that, as shown in Scheme 2a (right), the RF@Nf-x has a structure, in which the hydrophobic Nf parts (green) are woven within the resin body, with the hydrophilic Nf parts (black) facing outward. The surface elemental compositions of the resins determined by XPS peak areas with the atomic sensitivity factor27 (Table 1, right) revealed that the surface F and S compositions of RF@Nf-1.0 and -5.0 are larger than the total F and S compositions (left); this confirms the presence of outward hydrophilic Nf parts. Moreover, the total and surface compositions indicate the F/S ratio of Nf to be ∼36, which is consistent with the reported composition of Nf (F: 60.0 mol %, S: 1.5 mol %).28 In contrast, the surface F/S ratios of RF@Nf-1.0 and -5.0 are 14 and 22, respectively, indicating the surface-exposure of S-rich hydrophilic Nf parts. This validates the proposed RF@Nf-x structure (Scheme 2a, right), consisting of a hydrophobic body that weaves the hydrophobic Nf parts, with the hydrophilic Nf chains exposed on the resin surface.

Table 1. Total and Surface Elemental Compositions of the Resinsa.

  combustion method (mol %)b
 XPS (mol %)c
catalyst C O F S C O F S
RF 75.77 24.23     81.03 18.97    
RF@Nf-1.0 74.27 25.37 0.35 0.01 76.45 17.85 5.38 0.32
RF@Nf-1.0_usedd 71.17 28.46 0.35 0.01 76.15 18.30 5.19 0.36
RF@Nf-5.0 75.49 22.78 1.68 0.05 75.21 13.12 14.04 0.63
Nf 28.19 15.44 54.85 1.52 25.45 14.03 59.06 1.46
Open in a new tab
a

H atoms are omitted from the compositions.

b

Full compositions (wt %) are summarized in Table S1.

c

Determined from the XPS peak areas of the respective elements (Figures S13–16) with atomic sensitivity factors (C 1s, 1.00; O 1s, 2.85; F 1s, 4.26; S 2p, 0.59).27

d

RF@Nf-1.0 resin recovered after 24 h of photoreaction using a solar simulator (Figure 5a).

The exposed hydrophilic Nf parts were confirmed by a zeta potential analysis (Figure 1b). The isoelectric point of RF was observed at a weakly acidic pH (∼3.0), owing to the presence of phenolic −OH groups. This point was shifted to ∼2.5 for RF@Nf-1.0, while RF@Nf-5.0 did not show the point at pH > 2 owing to acidic −SO3– groups of Nf, similar to an Nf dispersion solution;29 this confirms the outward-facing hydrophilic Nf parts. As shown in Figure 1c, on addition to water, the RF and RF@Nf-1.0 powders were highly dispersed, while RF@Nf-5.0 floated on water. Thus, the resin body became hydrophobic owing to the weaving of hydrophobic Nf parts. These data further validate the proposed RF@Nf-x structure (Scheme 2a). Moreover, the mechanism for the formation of RF@Nf-x can be explained as follows (Scheme 2a): Nf dispersed in aqueous media forms an aggregate with the hydrophilic parts facing outward,30 which is disassembled by hydrothermal treatment, forming an unraveled structure.31 The polycondensation of resorcinol and formaldehyde creates a π-conjugated and π-stacked hydrophobic D-A network (Scheme 1a,b), within which the hydrophobic Nf backbone is woven via hydrophobic interactions, forming small RF particles with outward-faced hydrophilic Nf parts. The hydrophilic parts suppress the growth of RF particles due to electrostatic repulsion, facilitating the formation of small resin particles.

Semiconducting Properties of the Resins

The carbon compositions of the resins were determined using solid-state dipolar-decoupling magic-angle spinning 13C nuclear magnetic resonance (DD/MAS/13C NMR) spectroscopy (Figure S17). The spectra of RF@Nf-1.0 and -5.0 were similar to those of RF and could be deconvoluted into 14 components.18−20Table S2 summarizes the carbon compositions of the resins. Notably, the C–O and C=O ratios are consistent with those determined from the O 1s XPS spectra (Table S3), confirming the accuracy of the determined compositions. RF, RF@Nf-1.0, and -5.0 showed similar carbon compositions; they consisted of aromatics (∼66%), linkers (∼24%), and residual groups (∼10%). In addition, the ratio of the number of linkers to the number of aromatic rings was ∼2 for all the resins, indicating similar cross-linking degrees. Further, the benzenoid/quinoid (D/A) ratios of the resins were similar (50/50),18 indicating similar numbers of D/A units even after the Nf integration.

According to diffuse-reflectance UV–vis spectra (Figure 1a), RF showed a broad absorption band extending to ∼700 nm owing to charge transfer (CT) transitions of the π-conjugated and π-stacked D-A units.17 The integration of Nf caused a blue shift of the spectra. Tauc (Figure S18) and Mott–Schottky (Figure S19) plots were used to determine the semiconducting band structures of the resins (Figure 1d). Ebg increased with the amount of Nf integrated, where the CB and VB shifted to more negative and positive levels, respectively. The increase in Ebg could be attributed to the weaving of the insulator Nf within the resin (Scheme 2b), which suppressed the hybridization of the D-A units and created dissipated VB and CB bands (Scheme 2c).32 Powder X-ray diffraction (XRD) spectrum of RF showed a broad peak at 2θ = ∼21.0° (d002 = ∼4.2 Å), assigned to π-stacked D-A units (Figure S20). The peak intensity decreased with the amount of Nf integrated, indicating that the Nf-weaving within the resin suppressed D-A π-stacking (Scheme 2b). Electrochemical impedance spectroscopy analysis was performed using resin-loaded fluorinated tin oxide (FTO) electrodes under visible light. As indicated by Nyquist plots (Figure 1e), RF@Nf-1.0 showed a CT resistance (RCT) lower than that of RF because the smaller resin particles of the former enhanced the electron migration within particles.33 However, RF@Nf-5.0 showed a larger RCT than RF@Nf-1.0 because the dissipation of semiconducting bands by the Nf-weaving (Scheme 2c) decreased the electron conductivity. The photocurrent response of RF@Nf-1.0 and -5.0 (Figure 1f) showed a trend similar to that of their EIS data; the current density of RF@Nf-1.0 was larger than that of RF, while the current density of RF@Nf-5.0 was smaller than that of RF@Nf-1.0. Thus, the integration of an optimum amount of Nf is vital for producing small resin particles with high electron conductivity.

Role of Nf Integration

The hydrophobic surface of RF@Nf-x inhibited the photocatalytic decomposition of H2O2 (Scheme 3); this was confirmed by half-photoreactions using sacrificial reagents. The photoirradiation of the resins in a H2O2 solution with NaIO3 as an electron acceptor under Ar (Figure 2c) decomposed H2O2 by VB h+ (the reverse reaction of eq 3); the rate of decomposition decreased with the amount of Nf integrated. In addition, the photoirradiation of the resins in a H2O2 solution with benzyl alcohol as an electron donor under Ar (Figure 2d) also decomposed H2O2 by CB e– (eq 4); here, the decomposition rate also decreased with the amount of Nf integrated. Electrochemical linear sweep voltammetry (LSV) using resin-loaded FTO in a H2O2 solution (Figure S21) showed a cathodic current at <0.6 V (vs RHE) for H2O2 reduction and an anodic current at >1.0 V for H2O2 oxidation; both currents decreased with the amount of Nf integrated. Thus, as shown in Scheme 3b,c, the H2O2 access to the Nf-woven hydrophobic resin surface was severely limited as compared to that to the pristine RF surface (Scheme 3a); this inhibited the photocatalytic decomposition of H2O2.

Scheme 3. Proposed Photocatalysis Images on (a) RF, (b) RF@Nf-1.0, (c) RF@Nf-5.0 Resin Surfaces.

Scheme 3

Open in a new tab

The Nf-weaving within RF also affected its water-oxidation (eq 2) and O2-reduction (eq 3) activity (Scheme 3). Photoirradiation of RF@Nf-1.0 in water with NaIO3 as an electron acceptor (Figure S22) generated O2 with an activity higher than that of RF, owing to its larger surface area. However, the activity of RF@Nf-5.0 was lower than that of RF@Nf-1.0. In addition, photoirradiation of RF@Nf-1.0 in water with benzyl alcohol as an electron donor and O2 (Figure S23) generated H2O2 with higher activity than that of RF; however, RF@Nf-5.0 showed lower activity than RF@Nf-1.0. The enlarged surface area provides a larger number of reactive sites for water oxidation and O2 reduction and shorter carrier diffusion lengths to suppress the charge recombination,34 resulting in photocatalytic activity enhancement (Scheme 3b). The decreased activity of RF@Nf-5.0 could be attributed to the suppressed access of water and O2 to its highly hydrophobic surface (Scheme 3c),23 similar to the restricted access of H2O2, and its decreased electron conductivity (Figure 1e). The LSV of resin-loaded FTO in water (Figure S24) exhibited a cathodic current at <0.6 V (vs RHE) for O2 reduction and an anodic current at >1.6 V (vs RHE) for water oxidation; both currents obtained with RF@Nf-5.0 are lower than those obtained with RF@Nf-1.0. These results are consistent with the half-photoreaction results. Figure 2e shows the action spectra for H2O2 generation on RF and RF@Nf-1.0 obtained by irradiation of monochromatic light. The apparent quantum yields (ΦAQY) for RF@Nf-1.0 were larger than those for RF over the entire wavelength range. This confirms that the enlarged surface area of the synthesized resin efficiently promoted water oxidation and O2 reduction, while its hydrophobic surface suppressed the subsequent decomposition of H2O2, enhancing the overall generation of H2O2 (Scheme 3b).

Artificial Photosynthesis

The artificial photosynthesis performance of the RF@Nf-1.0 resin was evaluated at 298 K under AM1.5G simulated sunlight (1 sun) (Figure S1).35Figure 5a shows changes in the amount of H2O2 generated by RF and RF@Nf-1.0 with time. Initially, the rate of H2O2 generation on RF@Nf-1.0 was higher than that on RF, confirming that water oxidation and O2 reduction were enhanced by the increased surface area of the Nf-integrated resin. For RF, prolonged photoirradiation suppressed H2O2 generation, owing to the subsequent decomposition of H2O2; a H2O2 concentration of 10 mM was recorded after 24 h. In contrast, RF@Nf-1.0 significantly inhibited H2O2 decomposition and produced 16 mM of H2O2 after 24 h. Notably, as shown in Figure S25, 16 mM H2O2 is the highest concentration of H2O2 obtained by previously reported photocatalytic systems using water and O2. As shown in Figure 5a, the solar-to-chemical conversion (SCC) efficiency for H2O2 generation on RF was ∼0.5% at 5 h, which decreased to ∼0.2% after 24 h of irradiation. In contrast, the SCC efficiency of RF@Nf-1.0 remained at ∼0.35% even after 24 h of photoirradiation, confirming the inhibition of subsequent H2O2 decomposition; moreover, this SCC efficiency (0.35%) is the highest SCC efficiency during a long time photoirradiation (>24 h) among the photocatalytic systems reported to date (Table S4). The above results indicate that the suppression of subsequent H2O2 decomposition is important for efficient H2O2 generation.

Figure 5.

Figure 5

Open in a new tab

Artificial photosynthesis performance of the resins. (a) Amounts of H2O2 generated on RF and RF@Nf-1.0, and the SCC efficiency under AM1.5G simulated sunlight (1 sun) [water (50 mL), catalyst (150 mg), O2 (1 bar), λ > 300 nm, 298 K]. (b) Results for repeated photoreactions with RF@Nf-1.0 under AM1.5G simulated sunlight. The resin after 24 h of photoreaction (a) was recovered by centrifugation and used for sequential reactions, with water being replaced every 1 h of photoirradiation.

Similar to RF,18 RF@Nf-1.0 underwent self-oxidation under photoirradiation. According to NMR analysis (Table S2), photoirradiation led to a decrease in the numbers of methylene and methylol carbons, while increasing the numbers of aldehyde and ketone carbons. Thus, some of the methylene and methylol units of the resin were oxidized by VB h+ to the corresponding aldehyde and ketone units, respectively. However, the benzenoid/quinoid ratio (50/50) of the resin did not change even after the reaction. The diffuse-reflectance UV–vis spectrum of the recovered resin (Figure 1a) was similar to that of the fresh resin. TEM images of the recovered resin indicate the linkages of ∼0.6-μm particles with surface clusters (Figure S26), similar to those observed in the fresh resin (Figure 4b). In addition, the recovered and fresh resins showed similar total and surface elemental compositions (Table 1). The STEM-EDS maps of the recovered resin (Figures S27–S29) show elemental distributions similar to those of the fresh resin (Figure 4d, Figures S3–S5), indicating that the woven Nf remained stable even after the photoreaction. Furthermore, as shown in Figure 5b, even after 10 reuse cycles, the recovered RF@Nf-1.0 maintained its high activity. These data demonstrate that the structure and activity of the resin did not change during the reaction, facilitating stable H2O2 generation. In addition, RF@Nf-1.0 efficiently generates H2O2 even at higher temperatures up to 333 K (Figure S30), which is considered to be a practical operating temperature under sunlight illumination.36 Therefore, the property of the resin is advantageous for sunlight-driven photocatalysis.

Conclusions

This study described the synthesis of Nf-integrated RF resin (RF@Nf) by polycondensation of resorcinol and formaldehyde with an Nf dispersion solution under high-temperature hydrothermal conditions. The resin efficiently generates H2O2 from water and O2 under sunlight even under photoirradiation for long durations of time. The RF@Nf powders are composed of Nf-woven small-particle linkages with large surface areas; Nf behaves as a surface stabilizer to suppress the growth of RF particles. The powders, with large surface areas, efficiently catalyze water oxidation and O2 reduction, thereby enhancing H2O2 generation. The hydrophobic surface of RF@Nf showed limited H2O2 access and inhibited the subsequent photocatalytic decomposition of H2O2. The RF@Nf powder synthesized with an optimum amount of Nf produced a high-H2O2-content (16 mM, 0.06 wt %) solution, which is the highest concentration of H2O2 produced by photocatalytic H2O2-generation systems reported to date. The simple Nf-based resin design described here could facilitate photocatalytic H2O2 generation using metal-free powder photocatalysts, opening new frontiers in catalysis and green energy research.

Experimental Section

Synthesis of Resins

RF@Nf-x samples were synthesized by the high-temperature hydrothermal method:18 resorcinol (0.8 g), formaldehyde (33 wt % solution, 540 μL), a (COOH)2 solution (1.0 M, 360 μL), and x mL of a 5 wt % Nf dispersion (Sigma-Aldrich, 510,211-100ML) were added to pure water (80 mL) and stirred for 5 min. The mixture was left in an autoclave at 523 K for 12 h. The solids produced were washed by Soxhlet extraction with acetone for 12 h and dried in vacuo at room temperature for 12 h. The RF/Nf-1.0 resin was prepared as follows: RF (0.8 g) and an Nf dispersion solution (1.0 mL) were added to pure water (80 mL) and treated hydrothermally in a manner similar to that for the RF@Nf-x synthesis.

Acknowledgments

We thank Ryuta Sato and Takumi Hagi for their assistance with the experiments. This study was supported by a Grant-in-Aid for Scientific Research (22H01867) and the Nanotechnology Platform Project (Nanotechnology Open Facilities in Osaka University) (JPMXP09A20OS0032 and JPMXP09A21OS0005) from the Ministry of Education, Culture, Sports, Science and Technology, Japan (MEXT).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.3c00262.

  • Methods, emission spectra of light sources (Figure S1), N2 adsorption/desorption data (S2), STEM-EDS of RF@Nf-1.0 (S3–S5), TEM of an Nf dispersion solution (S6), STEM-EDS of RF@Nf-5.0 (S7–S9), SEM and TEM of RF/Nf-1.0 (S10), FT-IR spectra (S11), XPS spectra (S12–S16), DD/MAS/13C NMR charts (S17), Tauc, Mott–Schottky, and XRD data (S18–S20), LSV and half-photoreaction results (S21–S24), comparison of H2O2 concentrations (S25), TEM and STEM-EDS of RF@Nf-1.0 after photoreaction (S26–S29), effect of reaction temperature (S30), total elemental composition (Table S1), carbon composition (S2), C=O and C–O compositions (S3), catalytic performance of reported systems (S4), and references (PDF)

Author Contributions

All authors equally contributed. CRediT: Yasuhiro Shiraishi conceptualization, funding acquisition, methodology, project administration, supervision, validation, visualization, writing-review & editing; Masahiro Jio data curation, formal analysis, investigation, writing-original draft; Koki Yoshida formal analysis, investigation; Yoshihiro Nishiyama formal analysis, investigation; Satoshi Ichikawa formal analysis, investigation; Shunsuke Tanaka formal analysis, investigation; Takayuki Hirai funding acquisition, investigation, supervision, writing-review & editing.

The authors declare no competing financial interest.

Supplementary Material

au3c00262_si_001.pdf (4MB, pdf)

References

  1. Hisatomi T.; Kubota J.; Domen K. Recent advances in semiconductors for photocatalytic and photoelectrochemical water splitting. Chem. Soc. Rev. 2014, 43, 7520–7535. 10.1039/C3CS60378D. [DOI] [PubMed] [Google Scholar]
  2. Hisatomi T.; Domen K. Reaction systems for solar hydrogen production via water splitting with particulate semiconductor photocatalysts. Nat. Catal. 2019, 2, 387–399. 10.1038/s41929-019-0242-6. [DOI] [Google Scholar]
  3. Kudo A.; Miseki Y. Heterogeneous photocatalyst materials for water splitting. Chem. Soc. Rev. 2009, 38, 253–278. 10.1039/B800489G. [DOI] [PubMed] [Google Scholar]
  4. White J. L.; Baruch M. F.; Pander J. E. III; Hu Y.; Fortmeyer I. C.; Park J. E.; Zhang T.; Liao K.; Gu J.; Yan Y.; Shaw T. W.; Abelev E.; Bocarsly A. B. Light-driven heterogeneous reduction of carbon dioxide: photocatalysts and photoelectrodes. Chem. Rev. 2015, 115, 12888–12935. 10.1021/acs.chemrev.5b00370. [DOI] [PubMed] [Google Scholar]
  5. Kondratenko E. V.; Mul G.; Baltrusaitis J.; Larrazábal G. O.; Pérez-Ramírez J. Status and perspectives of CO2 conversion into fuels and chemicals by catalytic, photocatalytic and electrocatalytic processes. Energy Environ. Sci. 2013, 6, 3112–3135. 10.1039/c3ee41272e. [DOI] [Google Scholar]
  6. Chen X.; Li N.; Kong Z.; Ong W.-J.; Zhao X. Photocatalytic fixation of nitrogen to ammonia: state-of-the-art advancements and future prospects. Mater. Holiz. 2018, 5, 9–27. 10.1039/C7MH00557A. [DOI] [Google Scholar]
  7. Li M.; Huang H.; Low J.; Gao C.; Long R.; Xiong Y. Recent Progress on Electrocatalyst and Photocatalyst Design for Nitrogen Reduction. Small Methods 2019, 1800388. 10.1002/smtd.201800388. [DOI] [Google Scholar]
  8. Fukuzumi S. Production of Liquid Solar Fuels and Their Use in Fuel Cells. Joule 2017, 1, 689–738. 10.1016/j.joule.2017.07.007. [DOI] [Google Scholar]
  9. Fukuzumi S.; Lee Y. M.; Nam W. Solar-Driven Production of Hydrogen Peroxide from Water and Dioxygen. Chem. – Eur. J. 2018, 24, 5016–5031. 10.1002/chem.201704512. [DOI] [PubMed] [Google Scholar]
  10. Hou H.; Zeng X.; Zhang X. Production of Hydrogen Peroxide by Photocatalytic Processes. Angew. Chem., Int. Ed. 2020, 59, 17356–17376. 10.1002/anie.201911609. [DOI] [PubMed] [Google Scholar]
  11. Hu X.; Zeng X.; Liu Y.; Lu J.; Zhang X. Carbon-based materials for photo- and electrocatalytic synthesis of hydrogen peroxide. Nanoscale 2020, 12, 16008–16027. 10.1039/D0NR03178J. [DOI] [PubMed] [Google Scholar]
  12. Wang L.; Zhang J.; Zhang Y.; Yu H.; Qu Y.; Yu J. Inorganic Metal-Oxide Photocatalyst for H2O2 Production. Small 2022, 18, 2104561. 10.1002/smll.202104561. [DOI] [PubMed] [Google Scholar]
  13. Pekala R. W. Organic aerogels from the polycondensation of resorcinol with formaldehyde. J. Mater. Sci. 1989, 24, 3221–3227. 10.1007/BF01139044. [DOI] [Google Scholar]
  14. Durairaj R. B., Ed. Resorcinol, Chemistry, Technology and Applications; Springer-Verlag Berlin Heidelberg: New York, 2005; p 255. [Google Scholar]
  15. Al-Muhtaseb S. A.; Ritter J. A. Preparation and properties of resorcinol-formaldehyde organic and carbon gels. Adv. Mater. 2003, 15, 101–114. 10.1002/adma.200390020. [DOI] [PubMed] [Google Scholar]
  16. ElKhatat A. M.; Al-Muhtaseb S. A. Advances in Tailoring Resorcinol-Formaldehyde Organic and Carbon Gels. Adv. Mater. 2011, 23, 2887–2903. 10.1002/adma.201100283. [DOI] [PubMed] [Google Scholar]
  17. Shiraishi Y.; Takii T.; Hagi T.; Mori S.; Kofuji Y.; Kitagawa Y.; Tanaka S.; Ichikawa S.; Hirai T. Resorcinol–formaldehyde resins as metal-free semiconductor photocatalysts for solar-to-hydrogen peroxide energy conversion. Nat. Mater. 2019, 18, 985–993. 10.1038/s41563-019-0398-0. [DOI] [PubMed] [Google Scholar]
  18. Shiraishi Y.; Hagi T.; Matsumoto M.; Tanaka S.; Ichikawa S.; Hirai T. Solar-to-hydrogen peroxide energy conversion on resorcinol–formaldehyde resins photocatalysts prepared by acid-catalysed polycondensation. Commun. Chem. 2020, 3, 169. 10.1038/s42004-020-00421-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Shiraishi Y.; Matsumoto M.; Ichikawa S.; Tanaka S.; Hirai T. Polythiophene-Doped Resorcinol-Formaldehyde Resin Photocatalysts for Solar-to-Hydrogen Peroxide Energy Conversion. J. Am. Chem. Soc. 2021, 143, 12590–12599. 10.1021/jacs.1c04622. [DOI] [PubMed] [Google Scholar]
  20. Shiraishi Y.; Miura K.; Jio M.; Tanaka S.; Ichikawa S.; Hirai T. Solar-Driven Generation of Hydrogen Peroxide on Phenol-Resorcinol-Formaldehyde Resin Photocatalysts. ACS Mater. Au 2022, 2, 709–718. 10.1021/acsmaterialsau.2c00041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Dai C.; Liu B. Conjugated Polymers for Visible-Light-Driven Photocatalysis. Energy Environ. Sci. 2020, 13, 24–52. 10.1039/C9EE01935A. [DOI] [Google Scholar]
  22. Sondheimer S. J.; Bunce N. J.; Fyfe C. A. Structure and Chemistry of Nafion-H: A Fluorinated Sulfonic Acid Polymer. J. Macromol. Sci., Rev. Macromol. Chem. Phys. 1986, 26, 353–413. 10.1080/07366578608081976. [DOI] [Google Scholar]
  23. Mauritz K. A.; Moore R. B. State of understanding of Nafion. Chem. Rev. 2004, 104, 4535–4586. 10.1021/cr0207123. [DOI] [PubMed] [Google Scholar]
  24. Midgley P. A.; Weyland M. 3D electron microscopy in the physical sciences: The development of Z-contrast and EFTEM tomography. Ultramicroscopy 2003, 96, 413–431. 10.1016/S0304-3991(03)00105-0. [DOI] [PubMed] [Google Scholar]
  25. Chen C.; Levitin G.; Hess D. W.; Fuller T. F. XPS investigation of Nafion® membrane degradation. J. Power Sources 2007, 169, 288–295. 10.1016/j.jpowsour.2007.03.037. [DOI] [Google Scholar]
  26. Kaspar P.; Sobola D.; Částková K.; Dallaev R.; Št’astná E.; Sedlák P.; Knápek A.; Trčka T.; Holcman V. Case study of polyvinylidene fluoride doping by carbon nanotubes. Materials 2021, 14, 1428. 10.3390/ma14061428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Scofield J. H. Hartree-Slater subshell photoionization cross-sections at 1254 and 1487 eV. J. Electron Spectros. Relat. Phenom. 1976, 8, 129–137. 10.1016/0368-2048(76)80015-1. [DOI] [Google Scholar]
  28. Schulze M.; Lorenz M.; Wagner N.; Gülzow E. XPS analysis of the degradation of Nafion. Fresenius’ J. Anal. Chem. 1999, 365, 106–113. 10.1007/s002160051454. [DOI] [Google Scholar]
  29. Zhang H.; Pan J.; He X.; Pan M. Zeta Potential of Nafion Molecules in Isopropanol-Water Mixture Solvent. J. Appl. Polym. Sci. 2008, 107, 3306–3309. 10.1002/app.27473. [DOI] [Google Scholar]
  30. Koestner R.; Roiter Y.; Kozhinova I.; Minko S. AFM imaging of adsorbed nafion polymer on mica and graphite at molecular level. Langmuir 2011, 27, 10157–10166. 10.1021/la201283a. [DOI] [PubMed] [Google Scholar]
  31. Yamaguchi M.; Terao T.; Ohira A.; Hasegawa N.; Shinohara K. Size and shape of Nafion particles in water after high-temperature treatment. J. Polym. Sci., Part B: Polym. Phys. 2019, 57, 813–818. 10.1002/polb.24833. [DOI] [Google Scholar]
  32. García de Arquer F. P.; Talapin D. V.; Klimov V. I.; Arakawa Y.; Bayer M.; Sargent E. H. Semiconductor quantum dots: Technological progress and future challenges. Science 2021, 373, 640. 10.1126/science.aaz8541. [DOI] [PubMed] [Google Scholar]
  33. Xue L.; Li X.; Liao Y.; Xing L.; Xu M.; Li W. Effect of particle size on rate capability and cyclic stability of LiNi0.5Mn1.5O4 cathode for high-voltage lithium ion battery. J. Solid State Electrochem. 2015, 19, 569–576. 10.1007/s10008-014-2635-4. [DOI] [Google Scholar]
  34. Wang Q.; Domen K. Particulate Photocatalysts for Light-Driven Water Splitting: Mechanisms, Challenges, and Design Strategies. Chem. Rev. 2020, 120, 919–985. 10.1021/acs.chemrev.9b00201. [DOI] [PubMed] [Google Scholar]
  35. Gordon I.; Krebs F. C.; Mathew X.; Lampert C. M.; Rougier A.; Smestad G. P.; Subrahmanyam A. Editorial. Sol. Energy Mater. Sol. Cells 2015, 133, A1–A6. 10.1016/j.solmat.2014.10.021. [DOI] [Google Scholar]
  36. Wang Q.; Hisatomi T.; Jia Q.; Tokudome H.; Zhong M.; Wang C.; Pan Z.; Takata T.; Nakabayashi M.; Shibata N.; et al. Scalable Water Splitting on Particulate Photocatalyst Sheets with a Solar-to-Hydrogen Energy Conversion Efficiency Exceeding 1%. Nat. Mater. 2016, 15, 611–615. 10.1038/nmat4589. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

au3c00262_si_001.pdf (4MB, pdf)

Articles from JACS Au are provided here courtesy of American Chemical Society

RESOURCES