Furfuryl Alcohol‐Driven Proton Supply Enables Efficient Photocatalytic H2O2 Production Beyond Water‐Based Systems Authors: Pengfei Bai, Ning Li, Quan Zhou, Bin Liu, Xiangqian Fan, Lei Liu, Jingkai Lin, Weijie Ren, Shengliang Hu, Huayang Zhang Published: 2026 ABSTRACT ABSTRACTPhotocatalytic H2O2 production via the two‐electron oxygen reduction reaction is often limited by rapid charge recombination and insufficient proton supply in aqueous systems. Here, we construct a covalently linked ZnIn2S4‐carbon dots (ZIS‐CDs) develop a coupled system that integrates H2O2 generation with selective oxidation of furfuryl alcohol (FFA). The built‐in electric field in the ZIS‐CDs heterojunction enables efficient charge separation. FFA acts simultaneously as a sacrificial electron donor, oxidation substrate, and proton source in the ultra‐dry acetonitrile (ACN)‐FFA system. During its oxidation, FFA supplies protons for H2O2 generation and is selectively converted into furoic acid (FA). This dual regulation of charge transfer and proton supply leads to an H2O2 production rate of 24 mmol·g− 1·h− 1, an apparent quantum yield of 14.57% at 400 nm, and a solar‐to‐chemical conversion efficiency of 1.78%, with 100% selectivity and 98% conversion to furoic acid. This work introduces a new strategy that couples organic transformation with proton management to enhance photocatalytic H2O2 synthesis.Keywords: built‐in electric field, charge separation, furfuryl alcohol oxidation, photocatalytic H2O2 production, proton supply, ZnIn2S4/carbon dots Introduction Hydrogen peroxide (H2O2) is a crucial green oxidant extensively employed in fuel processing, wastewater treatment, bleaching, and disinfection [1, 2, 3, 4, 5]. Industrial production mainly relies on the anthraquinone process, which involves sequential hydrogenation and oxidation steps under energy‐intensive conditions. The complexity and environmental concerns associated with this process have driven the search for sustainable alternatives. Photocatalytic H2O2 production, which directly converts solar energy into chemical energy, is considered a promising approach [6, 7, 8, 9, 10]. In typical photocatalytic systems, semiconductors such as ZnIn2S4, In2S3, MoS2, ZnO, C3N4, covalent organic frameworks, metal‐organic frameworks, and carbon dots(CDs) drive two half‐reactions: the two‐electron oxygen reduction reaction (2e− ORR) and the water oxidation reaction (WOR) [4, 5, 6, 7, 8, 11, 12, 13, 14]. Photogenerated electrons reduce O2 to H2O2 (O2+ 2H++ 2e− → H2O2), while photogenerated holes promote the WOR to oxidize H2O to H2O2 (2 H2O + 2h+ → H2O2 + 2H+) [8]. The two‐electron WOR pathway (H2O→H2O2, 1.76 V vs. NHE) typically demands a higher redox potential than the two‐electron ORR pathway (O2→H2O2, 0.695 V vs. NHE), and its sluggish reaction kinetics lead to excessive accumulation of photogenerated holes, further exacerbating charge recombination and decreasing the photocatalytic performance [8, 12]. To improve performance, most studies have focused on enhancing charge separation through heterojunction construction, defect engineering, and cocatalyst modification [6, 7, 8, 15, 16, 17, 18, 19]. Nevertheless, photocatalytic H2O2 production fundamentally proceeds through proton‐coupled electron transfer (PCET) [20, 21]. Besides electrons, interfacial protons also play a pivotal role in the kinetics of PCET. In existing aqueous catalytic systems, the protons required for the 2e− ORR to generate H2O2 are mainly derived from water dissociation. This process is kinetically sluggish due to its high energy barrier and the low proton diffusion coefficient in water (∼2.39 × 10− 9 m2/s at 300 K) [22], which results in a prevalent proton‐deficient microenvironment around the catalyst and directly limits the total H2O2 yield. To address these issues, strategies including the protonation of functional units and the introduction of strongly polar oxygen‐containing functional groups (e.g., ─SO3H, ─COOH) have been developed to increase the proton concentration in the system [23]. For example, Zhu et al. prepared protonated hydrogen‐bonded organic frameworks (HCOFs) via sulfuric acid acidification. Protonation significantly extended the light absorption range to ∼700 nm, thus improving the photocatalytic H2O2 production and achieving a high apparent quantum efficiency (AQY) of 2.05% at 600 nm [24]. Li et al. uniformly immobilized ─SO3H or ─COOH groups onto highly crystalline triazine‐based COFs via a topology‐guided synthesis strategy [25]. These groups acted as “dual‐functional electron/proton extractors”, which not only enhances the electron extraction capacity but also facilitates proton transport through hydrogen‐bonding networks, thereby realizing the synergistic acceleration of the PCET process. Despite the advances achieved, each approach has its own inherent limitations. For instance, organic materials tend to hydrolyze under protonation conditions, and introducing oxygen‐containing functional groups often requires cumbersome procedures, which hampers the overall efficiency of photosynthetic H2O2 production. Therefore, the construction of novel synergistic catalytic coupling systems is of great research significance. In recent years, researchers have incorporated selective organic oxidation reactions to rapidly scavenge photogenerated holes, while utilizing photogenerated electrons for the 2e−‐ORR process, thereby achieving the co‐production of H2O2 and high‐value organic compounds [26, 27, 28, 29, 30, 31]. Furoic acid (FA) is a critical intermediate for the synthesis of plasticizers, thermosetting resins, preservatives, coating additives, and other products [32]. The oxidation of furfural alcohol (FFA) to FA releases a considerable number of protons. Directly utilizing these protons to drive 2e−‐ORR for H2O2 generation synergistically accelerates the PCET process. Thus, the photocatalytic production of H2O2 coupled with the selective oxidation of FFA to FA exhibits substantial potential. In this work, we coupled FFA conversion with photocatalytic H2O2 synthesis in an acetonitrile (ACN)/FFA catalytic system, attaining a maximum H2O2 production rate of 24 mmol·g−1·h−1. This study innovatively proposes a novel system of ‘photocatalytic oxygen reduction coupled with FFA photooxidation’. By using the ZIS‐CDs heterojunction to promote charge separation and interfacial electron transfer, FFA oxidation was coupled with photocatalytic H2O2 production to form an integrated redox system (Scheme 1). In conventional aqueous photocatalytic H2O2 production, protons are mainly derived from water. In contrast, in the ACN‐FFA system, FFA functions as both a sacrificial electron donor and a major proton source. FFA reacts with photogenerated holes to release protons for H2O2 formation, while being selectively converted into FA. This strategy integrates proton supply with target product synthesis, achieving 100% selectivity and 98% conversion toward FA. Theoretical calculations demonstrate that the efficient electron transfer channels constructed by ZIS‐CDs facilitate charge separation and transfer, substrate activation, and the dehydrogenation process of FFA. These effects further accelerate the reaction kinetics of both H2O2 production and FFA photooxidation, thereby enhancing the overall reaction efficiency. This work offers a promising and innovative research direction for advancing sustainable photocatalytic H2O2 generation technology. Schematic diagram of photocatalytic H2O2 production assisted by different proton sources originated from water dissociation and FFA photooxidation. Results and Discussion Catalyst Synthesis, Structure, Morphology and Composition Characterization Figure 1a shows the fabrication of the ZIS‐CDs heterojunction. In brief, carbon dots were covalently anchored onto ZnIn2S4(ZIS) nanobelts through APTES‐assisted Schiff‐base coupling, forming a tightly bonded heterostructure. The microstructures of CDs, ZIS, and ZIS‐CDs photocatalysts were observed by using transmission electron microscopy (TEM). As displayed in Figure 1b and Figure S1, CDs are uniformly dispersed, with a lattice spacing of 0.20 nm, corresponding to the (101) crystal plane of the graphite phase [33]. Both ZIS and the ZIS‐CDs hybrid material exhibit typical 1D nanobelt morphologies with consistent structural regularity (Figure 1c,d). This observation indicates that the introduction of CDs molecules only modifies the surface of ZIS without altering its intrinsic morphology, demonstrating good structural compatibility during the hybridization process. In Figure 1e, two distinct lattice spacings can be observed. The lattice fringe spacing of 0.32 nm can be indexed to the (102) crystal plane of ZnIn2S4, whereas a measured interplanar distance of 0.20 nm is characteristic of the (101) crystal plane of CDs, consistent with previous literature reports [33, 34]. Furthermore, the energy dispersive spectroscopy (EDS) elemental mapping (Figure 1f) reveals a uniform distribution of Mo, Zn, In, S, C, O, Si, and N elements in the ZIS‐CDs heterostructure. This further verifies that the CDs are uniformly dispersed on the surface of the ZIS nanobelts, laying a structural foundation for the highly efficient photocatalytic performance of this catalyst. X‐ray diffraction (XRD) patterns were performed to analyze the crystallographic information of the samples (Figure 1g). The characteristic diffraction peaks of both ZIS and ZIS‐CDs samples at 21.6°, 27.7°, 30.4°, 39.8°, 47.2°, 52.4°, and 55.6°can all be attributed to the hexagonal phase of ZIS (JCPDS: 65–2023) [35, 36], indicating that the introduction of CDs does not alter the integrity of the ZIS crystal structure. The characteristic diffraction signals of CDs are not observed, primarily due to the low content and their highly uniform dispersion on the ZIS surface [37]. As a result, their diffraction signals are masked by the strong characteristic peaks of ZIS. (a) Schematic diagram of covalently grafting CDs molecules onto ZnIn2S4 nanobelts via Schiff base polycondensation. TEM and high‐resolution TEM images of (b) CDs (c) ZIS, and (d, e) ZIS‐CDs. (f) Element mapping images of C, N, O, Si, In, Zn, and S for ZIS‐CDs sample measured by STEM‐EDS. (g) XRD patterns of ZIS, ZIS‐CDs, and CDs molecules. (h) FTIR spectra of ZIS, ZIS‐NH2, ZIS‐CDs, and CDs molecules. To verify the covalent bonding between ZIS and CDs, Fourier transform infrared (FTIR) spectra of ZIS, ZIS‐NH2, CDs, and ZIS‐CDs were measured (Figure 1h, Figure S2). The ZIS nanobelts exhibit a broad absorption peak in the range of 3200–3600 cm−1, indicative of the ─OH stretching vibration, and a peak at 1401 cm−1 is attributed to the ─OH bending vibration [38], which provides the necessary reactive sites for APTES functionalization. Compared to pure ZIS, the FTIR spectrum of ZIS‐NH2 shows several new characteristic peaks. The peaks at 2925 and 1385 cm−1 are assigned to the stretching and bending vibrations of the Si─CH2 group in APTES molecules, respectively [39]. Peaks in the 600–800 nm−1 range and at 1116 cm−1 belong to the asymmetric stretching and bending vibrations of the siloxane group (Si─O─Si) [40]. The peak at 890 cm−1 is ascribed to the stretching vibration of the Si─O─S bond [41]. These characteristic peaks confirm the successful grafting of APTES onto the ZIS surface via chemical bonding, achieving ZIS‐NH2. Compared to the ZIS‐NH2 sample, the ZIS‐CDs sample retains its key characteristic peaks while displaying additional peaks belonging to CDs. Importantly, C═O vibrational peak of the ZIS‐CDs is significantly reduced compared to that of the CDs [42], which indicates CDs molecules successfully graft onto the ZIS surface via Schiff‐base reaction. X‐ray photoelectron spectroscopy (XPS) was employed to systematically investigate the elemental composition and chemical states of the as‐prepared samples. As presented in the XPS survey spectrum (Figure S3a), the characteristic signals of C, N, O, S, In, and Zn elements are clearly distinguishable, which is in good agreement with the results obtained from EDS element mapping. This consistency further corroborates the successful incorporation of APTES and CDs into the ZIS‐CDs hybrid system. The high‐resolution N 1s spectrum (Figure S3b) was deconvoluted into two distinct peaks centered at 399.8 and 401.70 eV, which can be assigned to the C═N bond and other nitrogen‐containing species, respectively [35]. Notably, the presence of the C═N bond provides direct evidence that CDs are covalently anchored to ZIS‐NH2 via a Schiff base reaction, which is crucial for constructing a stable interfacial contact between the two components. The high‐resolution XPS spectra of C 1s, Zn 2p, In 3d, and S 2p are displayed in Figure S3c–f, respectively, to gain deeper insights into the chemical bonding configurations of the ZIS‐CDs heterojunction. For the C 1s spectrum, three characteristic peaks at 284.71, 286.30, and 288.20 eV are observed, corresponding to the C─C, C─O/C─N, and C═O bonds in the ZIS‐CDs sample [43]. Intriguingly, a noticeable shift of the C 1s peaks toward lower binding energies is observed when compared with the pure CDs sample. In the Zn 2p spectrum of ZIS‐CDs, two well‐resolved peaks at 1045.05 and 1022.01 eV are identified, which are characteristic of Zn 2p1/2 and Zn 2p3/2 orbitals, respectively [44]. Similarly, the In 3d spectrum exhibits two distinct peaks at 452.55 and 445.01 eV, which are attributed to In 3d3/2 and In 3d5/2, respectively. The S 2p spectrum can be deconvoluted into a spin‐orbit doublet with binding energies of 162.75 and 161.55 eV, corresponding to S 2p1/2 and S 2p3/2 in the ZIS‐CDs sample, respectively. In sharp contrast to the ZIS‐NH2 sample, the binding energies of Zn 2p, In 3d, and S 2p of ZIS‐CDs exhibit significant positive shifts, which is contrary to that of the C 1s. This phenomenon demonstrates that the ZIS‐CDs interface possesses strong interfacial interaction, with electrons spontaneously migrating from ZIS‐NH2 to CDs due to the differences in Fermi energy levels (E f) [40], resulting in the generation of a built‐in electric field at their interface [35]. The Brunauer–Emmett–Teller (BET) specific surface area and pore structure of the samples were characterized via N2 adsorption‐desorption (Figure S4). The specific surface area of pristine ZIS was measured to be 111.14 m2/g, while that of the ZIS‐CDs heterojunction decreased to 75.28 m2/g. The reduction in surface area can be attributed to the close contact formed between CD molecules and the ZIS surface via covalent bonding, which partially blocks the pores on ZIS and reduces its effective adsorption area [44]. Photocatalytic Performance Evaluation As depicted in Figure S5a, the pristine ZIS photocatalyst exhibits an H2O2 production rate of 1.1 mmol·g−1·h−1 in aqueous solution. After compositing ZIS with CDs at varying mass ratios, the ZIS‐CDs composite with a mass ratio of 5:1 (denoted as ZIS‐CDs‐5:1) demonstrates the optimal photocatalytic performance for H2O2 production. To clarify the reaction pathway, the catalytic performance of the ZIS‐CDs‐5:1 sample was further compared under an O2‐saturated atmosphere versus evacuated conditions. The H2O2 production rate increased significantly in the O2‐saturated environment, whereas it dropped to nearly zero under vacuum. These results confirm that the photocatalytic H2O2 generation over the ZIS‐CDs heterojunction follows a sole ORR pathway, and the participation of molecular oxygen is an essential prerequisite for H2O2 formation in this system. Subsequently, this study designed and conducted coupled reactions for photocatalytic H2O2 production and FFA photo‑oxidation under varied conditions. Using 5 mg of ZIS‑CDs as the photocatalyst dispersed in 30 mL of different reaction solutions, the reactions were carried out under continuous O2 bubbling, visible light (λ > 420 nm) and 15°C, with a focus on the effects of solution type and FFA concentration on the H2O2 yield. As shown in Figure 2a, H2O2 was generated only at a low rate of 2.4 mmol·g−1·h−1 in the aqueous system alone. To investigate the role of oxygen in the photocatalytic production of H2O2, H2O2 photosynthesis experiments were carried out using oxygen‐saturated aqueous solution. The results demonstrated that when ZIS‐CDs served as the catalyst, the H2O2 generation rate was enhanced up to 3.3 mmol·g−1·h−1 (Figure S5b). In sharp contrast, the H2O2 yield decreased significantly under vacuum conditions, which fully confirmed that oxygen is an essential reactant for the H2O2 generation reaction in this photocatalytic system. When 10 vol% FFA was added to the water as a scavenger, the H2O2 production rate increased to 3.5 mmol·g−1·h−1. In aqueous systems, water molecules as proton donors exhibit a relatively low proton release rate, which is insufficient to meet the continuous proton demand for photocatalytic H2O2 synthesis, thereby limiting efficient H2O2 production [45, 46, 47]. To address this issue, FFA is employed in this work to act simultaneously as a sacrificial electron donor and oxidation substrate. To eliminate the interference of water, ACN was introduced as the reaction solvent. Experimental results confirm that no H2O2 is detected in pure ACN (Figure 2a), indicating that ACN neither supports H2O2 production over ZIS‐CDs nor provides protons. This excludes the contribution of ACN to proton supply and confirms that ACN mainly serves as a non‐protic solvent and diluent for preparing the ACN/FFA system. In Figure 2a, when ACN was used as the diluent, the H2O2 production rate reached 19 mmol g−1 h−1 at an FFA concentration of 1.1 m (approximately 10 vol%), while the maximum activity of 24 mmol g−1 h−1 was achieved in neat FFA. To clarify the role of FFA concentration, H2O2 evolution was first evaluated within 1 h using FFA concentration as the only variable. As presented in Figure 2b, the H2O2 production rate increased with FFA concentration. Time‐dependent measurements were then performed over 5 h at different FFA concentrations. Low‐concentration FFA was gradually consumed during the reaction (Figure 2c), leading to a clear decrease in H2O2 generation over time (Figure 2d). Further analysis using 16 mm FFA showed that FFA was almost completely consumed after 5 h, with 98% conversion and nearly 100% selectivity toward FA (Figure 2e and Figure S6). Because dilute FFA cannot support stable long‐term H2O2 production, 1.1 m FFA/ACN solution was used for the stability test. Under this condition, the H2O2 evolution rate remained nearly constant over 10 h (Figure 2f). These results manifest that FFA can not only effectively accelerate the photosynthetic production of hydrogen peroxide, but also undergo highly selective conversion into FA. By analyzing the coupled reaction data at different FFA concentrations, the experimental molar ratio between H2O2 evolution and FFA oxidation was determined to be approximately 0.7:1 across all tested systems. This ratio confirms the close correlation between FFA oxidation and H2O2 formation, while the deviation from the theoretical 1:1 ratio may arise from side reactions, incomplete charge utilization, and competitive consumption of reactive intermediates during the coupled photocatalytic process. (a) H2O2 yield of ZIS‐CDs under different solvents. (b) Photocatalytic H2O2 production rate of ZIS‐CDs at different FFA concentrations within 1 h. (c) Time‐dependent FFA concentration profiles over ZIS‐CDs at various FFA concentrations. (d) Time‐dependent H2O2 concentration profiles over ZIS‐CDs at various FFA concentrations. (e) Temporal evolution of FFA and FA concentrations over ZIS‐CDs in 16 mm FFA solution. (f) Photocatalytic H2O2 evolution stability of ZIS‐CDs in 1.1 m FFA aqueous solution. (g) Photocatalytic H2O2 production cycle test over ZIS‐CDs under 1.1 m FFA. (h, i) Active species trapping experiments for ZIS‐CDs. The apparent quantum yields (AQY) of ZIS‐CDs were determined to be 14.57%, 13.8%, and 12.5% at 400, 420, and 500 nm, respectively (Figure S7a, Table S1), which is in good agreement with the UV–vis absorption spectrum of the ZIS‐CDs composite. Solar‐to‐chemical conversion (SCC) efficiency was calculated as 1.78% (Figure S7b), indicating that ZIS‐CDs possess efficient photon‐to‐chemical energy conversion capability in the UV‐visible region for photocatalytic H2O2 production. Cycling tests of ZIS‐CDs over five consecutive cycles in 1.1 m FFA/CAN solution showed no obvious decrease in H2O2 production activity or FFA conversion efficiency (Figure 2g and Figure S8a). TEM images confirmed that the morphology of ZIS‐CDs remained well preserved after the reaction (Figure S8b). The characteristic peaks in FTIR spectra and XRD patterns showed no obvious changes after cycling, confirming the stable chemical bonding and crystal structures of the catalyst (Figure S8c,d). These results demonstrate the good cycling stability of the ZIS‐CDs catalyst. Comparative analysis shows that the ZIS‐CDs heterojunction exhibits superior photocatalytic H2O2 production and FFA conversion performances over numerous other catalysts (Figure S7b, Tables S2 and S3) [3, 4, 6, 7, 8, 9, 11, 12, 13, 15, 16, 17, 18, 19, 26, 29, 30, 35, 43]. Intrinsic Mechanism Underlying the Efficient Photocatalytic Performance To elucidate the cooperative mechanism between photocatalytic H2O2 production and FFA‐to‐FA conversion, this study systematically investigated the roles of key reactive intermediates through active species trapping experiments and electron paramagnetic resonance (EPR) spectroscopy. First, specific scavengers were introduced into the reaction system to selectively quench potential active species generated during photocatalysis. The effects on both H2O2 yield and FA production were examined to clarify the contribution of each species (Figure 2h,i). The scavengers and their targets were as follows: AgNO3 (to trap photogenerated electrons), benzoquinone (BQ, to quench superoxide radicals ·O2 −), methanol (MeOH, to scavenge photogenerated holes), and dimethyl anthracene (DMA, to quench singlet oxygen 1O2). The addition of AgNO3 or BQ significantly suppressed both H2O2 and FA productions, confirming that photogenerated electrons and ·O2 − are key active radicals driving the coupled synthesis of H2O2 and oxidation of FFA [28, 29, 30, 31]. When MeOH was introduced, the H2O2 generation rate increased, while FA yield decreased, indicating that holes participated in FFA oxidation. Interestingly, the incorporation of DMA induced a remarkable enhancement in the generation rates of H2O2 and FA, with the values increasing to 33 and 72 mmol·g− 1·h− 1, respectively. Subsequent EPR analysis of the ACN+FFA system showed no characteristic signal of 1O2 (Figure S9a). A control experiment using benzylamine instead of FFA yielded highly consistent results upon DMA addition (Figure S10). Combining these observations, DMA does not function as a 1O2 quencher in this system. Instead, its primary role is to promote the efficient transfer of photogenerated electrons [29], thereby accelerating the generation and transport kinetics of ·O2 − and holes. This enhancement strengthens the synergy between H2O2 synthesis and organic substrate oxidation, ultimately driving the overall photocatalytic cooperative reaction with high efficiency. Electron paramagnetic resonance (EPR) spectroscopy was employed to discern the existence and transformation of active species and intermediates [48, 49]. Using 5, 5‐dimethyl‐1‐pyrroline N‐oxide (DMPO) and 2,2,6,6‐tetramethylpiperidine (TEMP) as trapping agents, red under light irradiation. As shown in Figure S9b, no signals corresponding to ·OH were detected for the ZIS‐CDs sample under illumination. Importantly, characteristic signals of ·O2 − and the intermediate radical ·C5H5O2 were clearly observed after 10 min of light exposure. Concurrently (Figure 3a,b), signals of hydroperoxyl radicals (·OOH) were also detected (Figure 3c).·OOH and ·C5H5O2 are the key intermediates formed via the protonation of ·O2 − and the conversion of FFA. These observations indicate that the ZIS‐CDs heterojunction facilitates the photocatalytic reduction of O2 to H2O2 alongside the conversion of FFA to FA. (a) EPR signals of DMPO‐·O2 on ZIS‐CDs in O2‐saturated furfuryl alcohol. (b) EPR signals of DMPO‐·C5H5O2 over ZIS‐CDs in FFA solution in dark and under light irradiation. (c) EPR signals of DMPO‐ ·OOH on ZIS‐CDs in O2‐saturated furfuryl alcohol; In situ DRIFTS spectra of ZIS‐CDs exposed to ACN+FFA system (d) in the dark (e) under illumination. To evaluate the ORR tendency of ZIS‐CDs, the average electron transfer number during O2 reduction was estimated from linear sweep voltammetry (LSV) curves obtained using rotating ring‐disk electrode (RRDE) measurements and the Koutecky–Levich (K‐L) equation (Figure S11a–c) [50]. The electron transfer number increased from 1.25 to 1.43 after adding FFA, suggesting an enhanced tendency toward the 2e− ORR pathway. Because RRDE measurements do not directly represent the photocatalytic reaction environment, gas‐controlled photocatalytic experiments were further performed in 1.1 m FFA/ACN solution under different atmospheres at the same flow rate (Figure S11d). The H2O2 production rate was highest under pure O2 and gradually decreased when O2 was diluted with air or Ar. Only limited H2O2 production was observed under pure Ar or N2, confirming that molecular O2 is the primary electron acceptor and that photocatalytic H2O2 generation mainly proceeds through the ORR pathway. To investigate the surface‐adsorbed species and reaction intermediates, in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was performed. When the ZIS‐CDs sample was exposed to a vapor mixture of FFA, acetonitrile, and O2 in the dark for 1 h, characteristic absorption peaks corresponding to FFA and ACN were detected, with their intensities gradually increasing over time. As shown in Figure 3d, the absorption peak at 3270 cm−1 can be assigned to the O─H stretching vibration of FFA molecules [27], confirming the successful adsorption of FFA on the ZIS‐CDs surface. Under continuous illumination (0–60 min), the intensity of the O─H peak belonging to FFA decreased (Figure 3e), consistent with the consumption of FFA [26]. Simultaneously, two new peaks emerged at 1740 and 3500 cm−1, corresponding to the C═O and O─H stretching vibrations [38], respectively, which together confirm the continuous formation of FA. The conversion of FFA to FA was further verified by 1H and 13C nuclear magnetic resonance (NMR) spectroscopy (Figure S12). A characteristic peak near 1070 cm−1 (Figure 3e), whose intensity increased progressively with irradiation time, can be attributed to ·OOH [28]. Notably, the generated ·OOH can be further reduced by photogenerated electrons (e−) to yield the final product H2O2, providing direct spectroscopic evidence for the H2O2 formation pathway. To further unravel the intrinsic mechanism governing the efficient photocatalytic H2O2 production and FFA conversion over the ZIS‐CDs heterojunction, this study systematically interrogated the separation and transfer dynamics of photogenerated charge carriers through a suite of complementary characterization techniques, including UV–vis–NIR absorption spectroscopy, transient photocurrent measurements, electrochemical impedance spectroscopy (EIS), photoluminescence (PL) spectroscopy, and band structure analysis. Figure 4a presents the UV–vis–NIR absorption spectra of the as‐prepared samples. It is evident that pristine ZIS exhibits predominant absorption in the ultraviolet (UV) region, whereas the ZIS‐CDs heterojunction displays remarkably enhanced and broadened light absorption spanning the 200–1500 nm range. This extended light‐harvesting capability facilitates the generation of a greater number of photogenerated electron‐hole pairs, thereby contributing to the improved photocatalytic performance [37]. Figure 4b depicts the transient photocurrent responses of ZIS‐NH2 and ZIS‐CDs upon intermittent light irradiation. Notably, the ZIS‐CDs heterojunction exhibits a substantially higher photocurrent density compared to pristine ZIS‐NH2. A higher photocurrent intensity is typically indicative of more efficient separation and faster migration of photoinduced charge carriers [51], which further supports the superior charge dynamics of the ZIS‐CDs heterojunction. Electrochemical impedance spectroscopy (EIS) measurements yielded Nyquist plots where the semicircle radius of ZIS‐NH2 was smaller than that of ZIS‐CDs [52]. This observation demonstrates that ZIS‐NH2 has enhanced charge transfer kinetics, which is one of the essential properties for high‐performance photocatalysts (Figure 4c). PL spectroscopy was employed to assess the recombination efficiency of photogenerated charge carriers. As shown in Figure 4d, the ZIS‐CDs sample exhibits a significantly lower PL intensity compared to the ZIS‐NH2 sample, which implies a substantially reduced recombination rate of photogenerated electron‐hole pairs. To quantitatively analyze the charge carrier lifetime, the PL decay curves were fitted using a double‐exponential model, and the average PL lifetime (τ a) was calculated in accordance with the following equation [2, 44]: where A 1 and A 2 correspond to the amplitudes, while τ 1 and τ 2 represent the short lifetime and long lifetime, respectively. The average exciton lifetime of ZIS‐CDs is longer than that of ZIS‐NH2 (Figure 4e,f), which indicates that efficient separation of photogenerated charge carriers is achieved in ZIS‐CDs through interfacial interactions, thereby suppressing charge recombination and enhancing photocatalytic efficiency [6, 8]. (a) UV–vis–NIR absorption spectra. (b) Photocurrent response curves of different samples and (c) EIS plots. (d) PL spectrum (e, f) time‐resolved photoluminescence (TRPL) decay curve of ZIS‐NH2 and ZIS‐CDs, with fitted lifetime parameters. To elucidate the dynamic process of interfacial charge transfer, the band structures were systematically characterized via multiple techniques. First, the optical bandgaps (E g) of ZIS, ZIS‐NH2, ZIS‐CDs and CDs were calculated from UV–vis absorption spectra using the Tauc plots, yielding values of 2.33, 2.38, 2.44, and 2.72 eV (Figures S13 and S14), respectively. Subsequently, XPS valence band spectra were employed to directly determine the valence band maximum (E VBM) relative to the normal hydrogen electrode (NHE), with E VBM values of 1.62 eV (ZIS), 1.42 eV (ZIS‐NH2),1.54 eV (ZIS‐CDs) and 2.43 eV (CDs). The conduction band minimum (E CBM) was further derived by the equation E CBM = E VBM‐E g, resulting in values of −0.71 eV (ZIS), −0.96 eV (ZIS‐NH2), −0.9 eV (ZIS‐CDs), and −0.29 eV (CDs). Furthermore, ultraviolet photoelectron spectroscopy (UPS) measurements were performed with He I excitation (21.22 eV) [7]. As depicted in Figure S15, the cut‐off binding energies (E cutoff) of ZIS‐NH2 and CDs were determined as values of 17.27 and 16.67 eV, respectively. The work function (Φ) was calculated using the formula Φ = 21.22 eV—E cutoff, yielding Φ values of 3.95 eV for ZIS‐NH2 and 4.55 eV for CDs. Based on these results, the energy‐level alignment between ZIS‐NH2 and CDs is illustrated in Figure 5a. A staggered band alignment is observed, wherein both the conduction band (CB) and valence band (VB) of CDs are positioned at lower energy levels than those of ZIS‐NH2. Moreover, the Fermi level (E f) of ZIS‐NH2 is higher than that of CDs. Upon formation of intimate interfacial contact, electrons are spontaneously injected from ZIS‐NH2 with a relatively higher Fermi level toward CDs until their Fermi levels become aligned. This induces positive charging and upward band bending on the ZIS side, alongside negative charging and downward band bending on the CDs side, thus establishing a built‐in electric field directed from ZIS to CDs. To further verify the interfacial charge transfer behavior, density functional theory (DFT) simulations were carried out to calculate the differential charge density of the ZIS‐CDs heterojunction (Figure 5b). In the differential charge density, blue and yellow regions correspond to electron depletion and accumulation, respectively. Notably, electron accumulation is observed on the CDs side, whereas electron depletion occurs on the ZIS side, accompanying a net charge transfer of 0.622 e from ZIS to CDs. These results are highly consistent with the band structure analysis as well as XPS measurements, collectively confirming the directional charge transfer at the heterogeneous interface. (a) Mechanism of Heterojunction Photocatalytic Hydrogen Peroxide Production by ZIS‐CDs. (b) Differential charge density map of ZIS‐CDs. (c)Free energy diagram of ZIS‐CDs for photocatalytic H2O2 synthesis by the 2e− ORR path; The structure within the green rectangular box is the local structure of the O2 reduction intermediate adsorbed on the surface of ZIS‐CDs (e, f) without *C5H5O2, (g, h) with *C5H5O2. (d) Energy profile of the FFA oxidation process with O2 assistance; The structures within the orange rectangular frame are the local structures of the intermediates in the FFA oxidation process optimally adsorbed on the ZIS‐CDs surface (i–l) without *OOH, (m–p) with *OOH. Based on the above analyses, ZIS and CDs are interconnected via a Schiff‐base reaction between the amino groups of APTES and the aldehyde groups on CDs. Upon light excitation, the built‐in electric field across the as‐formed heterojunction significantly facilitates the separation of photogenerated electron–hole pairs. Furthermore, photogenerated electrons rapidly transfer from the conduction band (CB) of CDs to the valence band (VB) of ZIS through the covalent‐bond linkage of APTES. This unique charge transfer pathway thus retains highly reductive electrons in the CB of ZIS and highly oxidative holes in the VB of CDs. Specifically, the photogenerated electrons migrate to the ZIS surface to participate in the ORR for H2O2 generation, while the photogenerated holes migrate to the CDs surface to drive FFA conversion. The overall coupled reaction can be summarized as follows: As illustrated in the reaction scheme, hydrogen ions generated during FFA conversion exactly act as the proton source for the oxygen reduction reaction (ORR), while the reactive intermediate ·O2 − involved in H2O2 generation also participates in FA formation. The coupling of these two reactions is the key to achieving highly efficient synergy between photocatalytic H2O2 generation and FFA photooxidation. Subsequently, we investigated the kinetic process of the coupled reaction via density functional theory (DFT) calculations. As shown in Figure 5c,e–h, in the presence of the ·C5H5O2 intermediate (corresponding to the C4H3O‐·CH‐OH species), the energy barrier for the rate‐determining step (RDS) in the ZIS‐CDs catalytic system is significantly lower than that in the system without this intermediate. This result directly confirms that the C4H3O‐·CH‐OH intermediate effectively promotes the reduction of O2 to H2O2. Detailed mechanistic studies revealed that under light excitation, photogenerated holes from the ZIS‐CDs heterojunction first activate FFA molecules (C4H3O‐CH2‐OH) adsorbed at the C═N‐linked active sites formed via Schiff base coupling, cleaving the α─C─H bond to generate the C4H3O‐·CH‐OH intermediate. The hydrogen proton is then captured by ·O2 −, detaches from the α‐carbon, and undergoes reaction to form ·OOH [4]. Subsequently, photogenerated holes act again on the α─C─H bond of C4H3O‐·CH‐OH producing C4H3O‐·C·‐OH with a triplet carbene structure. Simultaneously, the hydrogen proton is captured by ·OOH, departs from the α‐carbon, and reacts to form *HOOH, which then desorbs to afford free H2O2. Two molecules of C4H3O‐·C·‐OH react with abundant O2 in the system to generate two molecules of FA, which ultimately desorb from the ZIS‐CDs catalyst surface, completing the entire reaction cycle. To validate this proposed reaction pathway, DFT calculations were performed to map the complete reaction route for ZIS‐CDs‐catalyzed FFA oxidation (Figure 5d,i‐p). The results indicate that the RDS is the deprotonation of the α─C─H bond, with a reaction barrier of 1.380 eV, while all other steps proceed spontaneously with a decrease in free energy. Importantly, in the presence of the *OOH intermediate, the barrier for the RDS is significantly reduced to 0.699 eV (Figure 5d), demonstrating that the *OOH intermediate markedly accelerates the conversion of FFA to FA. On the whole, the synergistic interplay between the ·C5H5O2 and ·OOH intermediates significantly enhance the redox activity of the system. This finding offers direct theoretical evidence for the efficient separation and utilization of photogenerated carriers, as well as the optimized supply of hydrogen ions, in the ZIS‐CDs‐catalyzed acetonitrile‐furfuryl alcohol system. Conclusions In summary, we have successfully developed a ZIS‐CDs heterogeneous photocatalyst for efficient H2O2 production coupled with photosynthesis of FA. The ZIS nanobelts and CDs are covalently linked via 3‐aminopropyltriethoxysilane through a Schiff‐base reaction between the amino groups of ZIS‐NH2 and carbonyl groups of CDs. This robust covalent connection, together with the built‐in electric field in the interface of the ZIS‐CDs heterojunction, constructs an efficient electron transfer channel, effectively promoting the separation, directional migration, and respective reactions of photogenerated electron‐hole. In contrast to conventional aqueous photocatalytic H2O2 production systems, this work couples FFA oxidation with H2O2 generation in an ACN‐based medium. FFA serves as a sacrificial electron donor, oxidation substrate, and proton source, enabling efficient H2O2 formation while being selectively converted into FA. The system achieves nearly 100% selectivity toward FA and 98% FFA conversion. Meanwhile, the photoinduced electron and reacts with the hydrogen proton and O2 to produce H2O2, reaching a maximum rate of 24 mmol·g−1·h−1, the AQY value of 14.5% at 400 nm and the SCC efficiency of 1.78%. Theoretical calculations confirm that ZIS‐CDs can effectively enhance substrate activation and the dehydrogenation process of FFA, thereby accelerating the reaction kinetics of both H2O2 production and FFA photooxidation and significantly improving the overall reaction efficiency. Overall, this work provides a feasible solution to the proton supply challenge in efficient H2O2 synthesis, offers new insights for the development of synergistic conversion strategies. Author Contributions P. B. and N. L. contributed equally to this work. P. B.: methodology, investigation, data curation, writing – original draft; N. L.: conceptualization, software, funding acquisition, writing – original draft; Q. Z.: investigation, data curation; B. L., X. F. and L. L.: formal analysis, visualization; J. L.: data curation, formal analysis; W. J.: investigation, writing – review and editing; S. H.: supervision, project administration, writing – review and editing; H. Z.: conceptualization, funding acquisition, writing – review and editing. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting File: advs76085‐sup‐0001‐SuppMat.docx. Acknowledgements This work was supported by the Fundamental Research Program of Shanxi Province (202503021211098, 202403021211219, 202403021221126); The National Natural Science Foundation of China (52503327); The Postdoctoral Fellowship Program of CPSF (GZC20241570); The Science and Technology Cooperation and Exchange Program of Shanxi Province (202404041101044); Project of the Central Govermment's Funds for Guiding Local Science and Technology Development in Shanxi Province (YDZJSX2025D035); The special fund for Science and Technology Innovation Teams of Shanxi Province (202304051001014); Subject Reserved and Young Talent Program of North University of China. Open access publishing facilitated by Adelaide University, as part of the Wiley ‐ Adelaide University agreement via the Council of Australasian University Librarians. Contributor Information Ning Li, Email: lning@nuc.edu.cn. Weijie Ren, Email: 2023006@tyust.edu.cn. Shengliang Hu, Email: husl@nuc.edu.cn. Huayang Zhang, Email: huayang.zhang@adelaide.edu.au. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Supporting File: advs76085‐sup‐0001‐SuppMat.docx. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.