Enhanced H2O2 Production via Photocatalytic O2 Reduction over Structurally-Modified Poly(heptazine imide)

Solar H2O2 produced by O2 reduction provides a green, efficient, and ecological alternative to the industrial anthraquinone process and H2/O2 direct-synthesis. We report efficient photocatalytic H2O2 production at a rate of 73.4 mM h–1 in the presence of a sacrificial donor on a structurally engineered catalyst, alkali metal-halide modulated poly(heptazine imide) (MX → PHI). The reported H2O2 production is nearly 150 and >4250 times higher than triazine structured pristine carbon nitride under UV–visible and visible light (≥400 nm) irradiation, respectively. Furthermore, the solar H2O2 production rate on MX → PHI is higher than most of the previously reported carbon nitride (triazine, tri-s-triazine), metal oxides, metal sulfides, and other metal–organic photocatalysts. A record high AQY of 96% at 365 nm and 21% at 450 nm was observed. We find that structural modulation by alkali metal-halides results in a highly photoactive MX → PHI catalyst which has a broader light absorption range, enhanced light absorption ability, tailored bandgap, and a tunable band edge position. Moreover, this material has a different polymeric structure, high O2 trapping ability, interlayer intercalation, as well as surface decoration of alkali metals. The specific C≡N groups and surface defects, generated by intercalated MX, were also considered as potential contributors to the separation of photoinduced electron–hole pairs, leading to enhanced photocatalytic activity. A synergy of all these factors contributes to a higher H2O2 production rate. Spectroscopic data help us to rationalize the exceptional photochemical performance and structural characteristics of MX → PHI.


INTRODUCTION
As a consequence of the growing demand for clean energy, 1−3 there has been extensive research in recent decades aimed at replacing conventional fuels with carbon-free energy sources. 4−6 A higher energy density H 2 O 2 (3.0 MJ L −1 60% H 2 O 2 ) has been projected as a potential energy carrier that is relatively free from storage and transport issues. 7 Furthermore, H 2 O 2 already has significant importance in medical and industrial uses. The estimated global market for H 2 O 2 was 4.5 million metric tons in 2020, and it is projected to reach 5.7 million metric tons by 2027. 8 Despite its importance as a chemical and ecological oxidant, the production of H 2 O 2 still relies mainly on the energy intensive and waste generating industrial anthraquinone autooxidation process. 9 An alternative approach for H 2 O 2 is the direct electrolysis of H 2 /O 2 over expensive metal catalysts. 10 However, its dependence on H 2 , i.e., the consumption of one carbon-free energy source (H 2 ) to produce another carbonfree energy source (H 2 O 2 ), shows that this process cannot at present be regarded as a suitable alternative. Due to the drawbacks of direct electrolysis, the photocatalytic H 2 O 2 production via direct or indirect utilization of solar power has generated substantial interest as a potential, sustainable route. 11,12 However, even with significant advances in metallic and nonmetallic photocatalysts (PCs), existing photocatalytic systems can only generate low yields of H 2 O 2 . 13−19 Among nonmetallic photocatalysts, carbon−nitrogen (C− N)-based materials have attracted much attention, as they have a suitable conduction-band edge to carry out the two-electron transfer photochemical O 2 reduction reactions (PCORR) for H 2 O 2 generation. It is also possible easily to improve their catalytic efficacy by simple structural modifications, where various strategies including alkali metal doping, 16,20,21 cocatalyst loading, 22 structural/heterostructural engineering, 23−25 band alignment, 26 structural defects/vacancy center creation, 27−29 and surface shielding 15,30,31 have been adopted.
Hirai and co-workers reported the synthesis of a metal-free pyromellitic diimide-doped carbon nitride (g-C 3 N 4 /PDI) photocatalyst hybridized with reduced graphene oxide (rGO) for photochemical production of H 2 O 2 . They successfully generated nearly 20 mM H 2 O 2 by O 2 reduction in 90% (v) 2propanol/water using a 1.7 g L −1 photocatalyst suspension for 9 h irradiation, 24 which is the highest reported solar H 2 O 2 production yet obtained via O 2 reduction in the presence of a sacrificial agent. Recently, Quan et al. 16 reported that the synergistic effect of Na + , K + dopants and N vacancies on C 3 N 4 resulted in a H 2 O 2 production rate of 10.2 mM h −1 , which is 89.5-fold higher than that of pristine C 3 N 4 . Unfortunately, despite the extensive efforts toward a polymeric structured g-C 3 N 4 (triazine → tri-s-triazine) synthesis, including doping and defect and structural engineering, the highest values obtained for H 2 O 2 production on the g-C 3 N 4 based photocatalysts such as g-C 3 N 4 /PDI/rGO 24 and Na + , K + /N@g-C 3 N 4 16 are similar to those reported earlier for metal oxides 15,32,33 sulfides, 34,35 and molecular 14,36 photocatalysts. Indeed, in all reported studies, the main obstacle for developing solar-driven H 2 O 2 production as a suitable alternative is the low yield of H 2 O 2 generated.
There is, therefore, a pressing need for the development of an effective photocatalyst that could greatly increase photocatalytic H 2 O 2 production. Stimulated by the earlier reported strategies, we have synthesized a new photocatalyst that combines properties including a higher intrinsic surface area, modified electronic structure, reduced band gap, and defect sites for enhanced H 2 O 2 production. To this end, we have successfully synthesized an alkali metal-halide (MX, M = K + ; Li + , X = Cl − ) modulated C−N based poly(heptazine imide) (PHI) molecular photocatalyst, MX → PHI for PCORR to produce a much higher yield of H 2 O 2 (73.4 mM h −1 ) than obtained previously. The present structurally modulated MX → PHI photocatalyst was synthesized by facile polymerization of an environmentally benign precursor, urea, in the presence of alkali metal halides. A combination of microscopic, spectroscopic, and optoelectronic techniques verified the successful intercalation of MXs, found that the 3D-hollow fibers had a lamellar structure, and verified a broadening of the light absorption range and an enhanced light absorption ability of the synthesized catalyst, leading to substantially increased H 2 O 2 production rates. Our work clearly demonstrates the potential of MX → PHI for PCORR generating high yields of H 2 O 2 . (e) HRTEM image of MX → PHI particles with superimposed structure (C, N, and embedded alkali metal halides atoms colored in gray, blue, and red, respectively). (f) HAADF-STEM image and corresponding XEDS maps for carbon, nitrogen, potassium, chlorine, and oxygen. All elements appear to be distributed homogeneously within an agglomerated region.

RESULTS AND DISCUSSION
2.1. MX → PHI Growth and Characteristics Evaluation. Solid-state polymerization of tri-s-triazine structured metal-doped g-C 3 N 4 typically involves two steps: thermal condensation of nitrogen-rich precursors (urea, melamine, etc.) followed by ionothermal polymerization of the C−N based polymer. 37 The present highly photoactive MX → PHI photocatalyst was, however, directly synthesized from urea using a single-step ionothermal polymerization. To analyze the growth of the MX → PHI photocatalyst, two separate sets of experiments were performed. The samples collected at 350 and 500°C during ionothermal polymerization of urea were denoted as MX → PHI 350 and MX → PHI 500 , respectively, and the growth process is illustrated schematically in Figure 1a.
The morphologies of MX → PHI 350 , MX → PHI 500 , and MX → PHI (the final product after 5 h polymerization at 550°C) were characterized by scanning electron microscopy (SEM) to substantiate the growth mechanism (Figures S1−S6). The SEM micrographs of solidified MX → PHI 350 ( Figure S1) revealed coiling of thin polymeric sheets of basic carbon nitride (BCN) to form swirled polymeric hollow fibers/rods at the initial stage of polymerization, which later transformed into highly crystalline hollow fibers/rods ( Figures S3 and S5). Furthermore, the presence of alkali metal-halide ions results in the polymeric sheets folding to achieve energetically favorable hollow fibers, with a self-shaping crystal growth mechanism. 38,39 MX → PHI therefore has a 3D-hollow fiber morphology which consists of macroporous lamellar walls with a higher thickness compared to triazine structured BCN, which has an aggregated sheetlike morphology ( Figures S7 and  S8). High-angle annular dark field (HAADF) scanning transmission electron microscope (STEM) images ( Figure 1b) revealed that the MX → PHI materials have aggregates of nanosized particles and rods. High-resolution TEM images of an MX → PHI particle (Figure 1c,d, and Figure S9) confirm the crystalline nature of MX → PHI. Characteristic distances of 10.57 Å, corresponding to the (110) plane in poly(heptazine imide), and 3.25 Å, corresponding to the (001) plane were found. The poly(heptazine imide) structure can be matched to features in the HRTEM images ( Figure 1e) and the Fourier transform of a simulated HRTEM image contains lowfrequency peaks that match those in the Fourier transform of the experimental images ( Figure S10).
The insertion of alkali metal halides was confirmed by X-ray energy dispersive spectroscopy (XEDS) in the STEM. The HAADF-STEM image and corresponding XEDS maps of individual elements (carbon, nitrogen, oxygen, potassium, and chlorine) (Figure 1f) reveal successful and uniform distributions of each element into MX → PHI, which is consistent with the SEM-XEDS data ( Figures S2, S4, and S6). The homogeneous distribution of K and Cl throughout the sample demonstrates that ionothermal polymerization of urea results in the diffusion of the alkali metal halide into the growing polymeric unit of the heptazine imide.
X-ray photoelectron spectroscopy (XPS) was used to characterize the surface composition and inductively coupled showed the presence of K in MX → PHI framework. The XPS spectra of Cl 2p and Li 1s were also measured and the peak assignments confirmed the presence of Cl and Li ( Figure 2a). Overall, these results from XPS, STEM-XEDS, and ICP-MS (Table S1) showed the uniform and successful insertion of MX into the PHI framework.
The crystal structure of MX → PHI was characterized by powder XRD measurements. A comparison of XRD patterns of MX → PHI 350 , MX → PHI 500 , and MX → PHI ( Figure S13a) showed the appearance of additional diffraction peaks and peak shifts in MX → PHI, while some peaks, initially observed in MX → PHI 350 , disappeared with increased temperature. The high-intensity diffraction peaks at 8.3°(10.57 Å) and 27.5° The XRD patterns demonstrated that, relative to the 3.20 Å interplanar stacking in the triazine structured BCN ( Figure  S13b), there is a slightly wider interplanar stacking (3.25 Å) of poly(heptazine imide) units in the perpendicular direction and heptazine unit stacking with about 10.57 Å in-plane periodicity, which are driven by the insertion of MX. The results show that the triazine phase is further polymerized into the polyheptazine phase through a controlled ionothermal polymerization process in the presence of MX under an Ar atmosphere. Thus, the potential changes in interplanar stacking, together with the possible adjustments in the electron-rich π conjugated framework, the in-plane lattice packing, and the edge defects resulting in −CN and −NO x functionalization are affirmed upon intercalation of alkalimetals and halide ions. 40−42 Attenuated total reflectance coupled Fourier transform infrared (ATR-FTIR) (Figure 3a and Figure S14), and Raman ( Figure 3b and Figure S15) spectroscopic techniques were used for the characterization of BCN, MX → PHI 350 , MX → PHI 500 , and MX → PHI, so that the thermal transformation of urea to MX → PHI in the presence of alkali metals halides could be confirmed (as discussed in Supporting Note S1). To explore further the textural properties and to provide confirmation of the porous geometry of the 3D hollow fibers/ rods in MX → PHI particles, as sketched in Figure 1, N 2 adsorption−desorption measurements were performed at 77 K and isotherms have been reported and discussed in Figure 3c and Figure S16 and Supporting Note S2. The results of SEM, TEM, XPS, XRD, FTIR, and N 2 adsorption−desorption together demonstrate the successful preparation of an alkali metal halide incorporated poly(heptazine imide) photocatalyst. To test further the suitability of MX → PHI as an efficient photocatalyst for increased photocatalytic H 2 O 2 production, the optical properties and charge separation ability of the material were analyzed.
2.2. Optical Properties and Electronic Band Structure. The high light absorption efficiency of MX → PHI was confirmed by diffuse reflectance UV−visible (DR-UV vis) absorption spectroscopy, reported in Figure 3d. Compared with BCN, the MX → PHI hollow fibers/rods show significantly higher light absorption, both in the UV and visible regions ( Figure 3d) as well as a red-shift ( Figure 3d, highlighted by the red arrow). The red-shift in the absorption spectrum of the MX → PHI photocatalyst suggests extended π conjugation, and a delocalized aromatic π conjugated system. 43,44 The high light absorption efficiency of MX → PHI is probably due to multiple diffuse reflectance inside the nanoarchitecture, leading to trapping and deep penetration of solar radiation. (Figure 1 and Figure S5).
The inset digital images (Figure 3d) of BCN and MX → PHI samples show an apparent color change from white to greenish yellow, which suggests that the bandgap is altered in MX → PHI resulting in extended solar spectrum absorption efficacy. The UV−vis absorption spectra also highlighted an intense band between 350 and 450 nm assigned to a π → π* transition in the s-triazine unit of the C−N based polymers. 38,45 The visible region absorption edge and steep UV−vis absorption spectrum for MX → PHI (Figure 3d) demonstrate the high purity of the light absorber and the UV− vis absorption results from the band gap transition. These transitions are attributed mainly to charge transfer from the filled valence band (VB) of the N 2p orbital to the conduction band (CB) of the C 2p orbital. Furthermore, the band gap energy (E BG ) calculated using the Kubelka−Munk function: (F(R)hν) 1/2 = hν (Figure 3d, inset) for MX → PHI (2.67 eV) is lower than that for BCN (2.98 eV), which is consistent with their band edge wavelengths and demonstrates that the MX → PHI is viable as a visible light adsorber photocatalyst. These results indicate that structure modulation by alkali metal halides can reduce the band gap and increase the light harvesting ability of MX → PHI.
In addition to bandgap engineering, the band edge positions (VB and CB) also have great significance for the efficient use of photogenerated charge carriers to perform specific redox reactions. The estimated valence band energies (E VB ) for BCN and MX → PHI photocatalysts, measured by UPS analysis, are 6.72, and 7.02 eV, respectively ( Figure S17). 46 After determining the E VB , the conduction band energy (E CB ) for respective photocatalysts is estimated by the relation E CB = E VB − E BG ; and the E VB , E CB , and E BG values are schematically illustrated in Figure S18. This band structure is evidence of a positive shift in the band positions for MX → PHI relative to BCN, with better-aligned energy levels for PCORR to produce H 2 O 2 and water oxidation or organic molecule oxidation ( Figure S18). In particular, the more positive CB value possibly enhanced the photochemical O 2 reduction capability of MX → PHI to generate significant amounts of H 2 O 2 .
Room-temperature PL emission spectra were recorded under excitation at 350 nm for BCN and MX → PHI ( Figure  3e) to probe the separation and recombination of photogenerated charge carriers. As shown in Figure 3e, BCN exhibited an intense emission peak centered around 450 nm, which highlighted the higher recombination rate of the photogenerated charge carriers. In contrast, a marked drop in the peak intensity and a flat emission spectrum were observed for MX → PHI (Figure 3e), which indicate a suppressed electron−hole pair recombination rate and enhanced charge carrier separation efficiency. Thus, the MX → PHI photocatalyst clearly inhibits the different radiative charge carriers' recombination pathways, associated with aromatic structured photocatalysts.
The accumulation of O 2 gas molecules around the active sites on the photocatalyst surface can facilitate the 2e − pathway of PCORR to , and is possibly one of the primary causes of the exceptional photochemical performance of MX → PHI. Therefore, preliminary thermal studies were conducted to gain information about the interaction or encapsulation of O 2 into the porous structured, polymerized heptazine units, and the interplanar stacking of the ion MX → PHI photocatalyst. In addition, we may compare with the BCN samples to highlight the superior photoactivity of MX → PHI toward PCORR. The temperature-programmed deoxygenation (O 2 TPD) and TGA profiles for both the materials are reported in Figure 3f and Figure (Figure 3f, inset), respectively. As well as a significant peak shift to higher temperature for MX → PHI photocatalyst, the amount of desorbed O 2 , based on the peak area, is also ∼4 times higher than for the reference BCN photocatalyst. The positively charged alkali metal encapsulated into the C−N based PHI framework can play a key role in the interaction of O 2 with the surface of the molecular photocatalyst. The TGA thermograms ( Figure S19) further complemented the O 2 TPD results, as a steady weight loss in the temperature range of 25−550°C was observed for MX → PHI, possibly because of adsorbed water molecules and atmospheric gases.
Since light harvesting, energy band structure, surface area, and charge carrier separation efficiency are the main factors affecting the performance of photocatalysts, from the factors discussed above, we would expect a significantly enhanced solar H 2 O 2 production for MX → PHI. The mesoporous character, higher surface area, lamellar hollow fiber structure of MX → PHI, and the presence of alkali metal halides contributed to improving the light absorbance efficiency, the charge carrier separation, and the O 2 gas molecule confinement, which in turn should improve the photochemical performance of MX → PHI for the O 2 reduction reaction.  Chemistry of Materials pubs.acs.org/cm Article Thus, MX → PHI resulted in nearly 150 times higher solar H 2 O 2 production than that of BCN under UV−visible irradiation and >4250 times higher solar H 2 O 2 production as compared to BCN under visible light irradiation. As discussed earlier (Figure 1), the significantly enhanced photocatalytic performance of MX → PHI is a consequence of a combination of factors: the synergistic effect of the morphology and optical and electronic properties induced by the structure-modulation of poly(heptazine imide) with alkali metal halides through controlled ionothermal polymerization.
The pH of the reaction solution may also have a significant effect on the proton-coupled electron transfer-assisted solar H 2 O 2 production. Therefore, the photocatalytic production of solar H 2 O 2 was also carried out at pH 4 and pH 10 ( Figure  4e). The MX → PHI showed a significant decrease in solar H 2 O 2 production with increased pH (pH 10) whereas an insignificant difference was observed in the H 2 O 2 production profile for PCORR carried out at pH 4 and neutral pH solution (without maintaining the pH using acid or base). The results show that no additional pH adjustment steps are required to maximize the performance.
The photocatalytic performance of the as-synthesized MX → PHI for reductive solar H 2 O 2 generation from O 2 saturated deionized water (DIW) without using any electron and proton donor sacrificial agent was also evaluated to corroborate the greater possibilities and high potential of MX → PHI for unassisted solar fuel production. A significant amount of solar H 2 O 2 production (74.0 μM) in the initial 15 min of light irradiation over bare MX → PHI was observed under UV− visible light irradiation (Figure 4f), which is also comparable to some of the most recently reported photocatalytic systems. 15,18,32 A relatively low solar H 2 O 2 production and SCC efficiency in the absence of a sacrificial agent is probably due to the consecutive decomposition of photogenerated H 2 O 2 on the MX → PHI surfaces during the photochemical reaction, which explains why the self-oxidation of photogenerated H 2 O 2 resulted in the saturation of the H 2 O 2 production after 30 min of irradiation (Figure 4f). The cyclic photocatalytic performance of MX → PHI was examined under the same reaction conditions for three repeated runs (Figure 5a). The linear increase in solar H 2 O 2 production for each run demonstrated a sustained photoactivity of MX → PHI. To substantiate further the unchanged surface structure and intact optical properties of MX → PHI during the PCORR recyclability tests, the samples collected after each run were analyzed by DR-UV−vis ( Figure 5b) and FTIR (Figure 5c) spectroscopy. The FTIR spectra for each collected sample at the end of PCORR did not show any significant change in the vibration peak positioning and their intensities (Figure 5c). However, relative to the original MX → PHI sample, the absorption spectra for MX → PHI collected after the first and second runs displayed a slight improvement in light absorption, with a red-shift and extended tailing (Figure 5b). The extended tail may correspond to minor changes in the surface functionality of the polymeric structure of MX → PHI, as a result of photoactivation during PCORR. The cyclic photochemical performance (Figure 5a), spectroscopic examinations (Figure 5b,c) and N 2 adsorption− Chemistry of Materials pubs.acs.org/cm Article desorption studies ( Figure S22) showed that the photoactivity, chemical structure, and texture properties of the MX → PHI photocatalyst remained largely unchanged during the repeated experiments. The apparent quantum yield (AQY) of MX → PHI for solar H 2 O 2 production was also measured using 365 and 450 nm light irradiation. The AQY values obtained for H 2 O 2 production at 365 nm (UV light) and 450 nm (visible light) are ∼96% and 21%, respectively, for MX → PHI with a catalyst dosage of 1 g L −1 in 10 M ethanol ( Figure S23). These values are higher than those reported for previous photocatalysts for peroxide production, indicating that MX → PHI is a highly efficient molecular photocatalyst for sustainable solar H 2 O 2 production via a 2e − pathway. AQY values for PCORR to H 2 O 2 matched well with the DR-UV visible spectrum of MX → PHI proving that the PCORR is via a 2e − process.
The reaction kinetics of 2e − PCORR to H 2 O 2 over the surface of irradiated BCN and MX → PHI photocatalysts was investigated using the kinetic model of photochemical H 2 O 2 generation at the initial phase of reaction reported by Hoffmann and co-workers as follows: 48,49 Here, k F and k D are the rate constants for photochemical H 2 O 2 formation and decomposition reactions, respectively. Following the reaction kinetics, the H 2 O 2 formation rate is determined by zero-order kinetics because the reaction solution is continuously purged with O 2 , while the decomposition reaction rate with fixed initial H 2 O 2 concentration follows first order kinetics. The k D value for MX → PHI (0.00208 min −1 ) (Figure 5d, inset), obtained after fitting the H 2 O 2 photodecomposition profile (Figure 5d) to first-order reaction kinetics, was slightly greater than that of triazine structured BCN (0.00183 min −1 ). However, a large difference between k F values of MX → PHI (1.2233 mM min −1 ) and BCN (0.0085 mM min −1 ) was observed. The kinetic data demonstrate that solar H 2 O 2 production is primarily governed by the formation kinetics. Furthermore, electrochemical rotating disc electrode (RDE) analysis confirms the 2e − O 2 reduction pathway to H 2 O 2 generation rather than 4e − (H 2 O formation) over MX → PHI ( Figure  5e,f). The calculated electron transfer number from the slopes of the linearly fitted Koutecky−Levich (K−L) plots at the different potentials (Figure 5f) was around 2.06.
Furthermore, to validate the generation of O 2 •− (superoxide anion radical) intermediate reaction species during photochemical H 2 O 2 production over the MX → PHI surface, the in situ coloration of XTT (2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide) when reacted with photogenerated O 2 •− into orange colored XTT-formazan ( Figure S24a) has been used. The appearance of the dark orange color ( Figure S24b) and absorbance λ max around 481 nm ( Figure S24c Considering the exceptional solar H 2 O 2 production performance of MX → PHI via the 2e − PCORR pathway in an organic solvent, a comparison was drawn with previously reported photocatalysts for similar reaction systems. The present MX → PHI photocatalyst exhibited a higher solar H 2 O 2 production rate than that of most of the carbon nitride, metal oxide, metal sulfide, and metal organic-based photocatalysts, respectively (Table S2).

SUMMARY AND CONCLUSIONS
We successfully achieved the highest ever solar H 2 O 2 production rate (73.4 mM h −1 ) via the 2e − PCORR pathway on an alkali metal-halide modulated poly(heptazine imide) (MX → PHI). Compared to the triazine structured pristine carbon nitride, there is an increase of nearly 150 and >4250 times in H 2 O 2 production on MX → PHI under UV−visible and visible light (≥400 nm) irradiation, respectively, which reflects the effect of the basic structure of poly(heptazine imide) and the engineering of its morphological, optical, and electronic properties via alkali metal-halides. In particular, combining effective light absorption, charge separation, and O 2 trapping in MX → PHI makes it an exceptionally highly photoactive molecular catalyst. Our study provides insight for potential materials based on poly(heptazine imide) for sustainable H 2 O 2 production by utilizing natural resources (sun, water, and air).

EXPERIMENTAL SECTION
4.1. Synthesis of Bulk Triazine Structured Carbon Nitride (BCN). The BCN was synthesized via thermal pyrolysis of urea at 550°C for 3 h in a muffle furnace. After the completion of thermal polymerization of urea to triazine structured g-C 3 N 4 , the product was washed with deionized water and collected by filtration followed by vacuum drying. The dried white product (4.6% yield with respect to urea precursor) was further ground to a fine powder and stored as such for photochemical performance evaluation.
4.2. Synthesis of Alkali Metal-Halides (MX) Modulated PHI (MX → PHI). The MX → PHI was synthesized by controlled ionothermal polymerization processes. The distinctly structured MX → PHI was obtained by mixing a fixed ratio of urea to KCl−LiCl eutectic mixture (5:3) to carry out the polymerization in a tube furnace under a continuous flow of Ar gas at 550°C at a ramp rate of 3°C min −1 for 5 h. The synthesis is sensitive to atmospheric conditions, therefore, Ar gas was continuously purged into the reaction mixture to minimize the O 2 and water content. The greenishyellow colored product, obtained from the cooled polymerized sample, was washed with DI water and collected by filtration followed by vacuum drying. The dried product was ground into a fine powder with an agate mortar and stored as such in an amber vial for photochemical studies and characterization. The final yield of MX → PHI (7.5% with respect to urea precursor) was higher than BCN. Ionothermal polymerization facilitates more uniform doping in the basic framework of the PHI molecular photocatalyst and simultaneously might introduce surface functionality and performanceenhancing structural defects. Moreover, ordering and stabilization of the intermediates result in the synthesis of a highly efficient MX → PHI molecular photocatalyst.

Hydrogen Peroxide Detection.
The H 2 O 2 concentration in the aliquot collected at different time intervals from the reaction suspension during light irradiation was measured by the DPD colorimetric method using a UV−visible spectrophotometer (UV-1800, Shimadzu). 18 Depending on the H 2 O 2 concentration, the collected samples were diluted multiple times (10−6000) before its estimation so that the photogenerated H 2 O 2 concentration lies in the calibrated range. To perform the colorimetric estimation of H 2 O 2 in an aqueous solution, 0.4 mL of 0.1 M sodium phosphate buffer (pH 6) was mixed with 1.12 mL of DIW followed by the addition of 1 mL of sample. To the buffered solution, 0.05 mL of N,N,-Diethyl-pphenylene-diamine sulfate (DPD) solution followed by 0.05 mL of peroxidase (POD) was mixed to catalyze the oxidation of DPD in the presence of H 2 O 2 to generate a pink color due to radical cations as shown in Figure S25. The resultant colored solution was used for spectrophotometric measurement of H 2 O 2 concentration at λ max 551 nm using an external standard curve (R 2 > 0.998). Moreover, a zero/ blank reading for reaction suspension was conducted with the aliquot collected before irradiation for accurate quantification of photogenerated H 2 O 2 in each experiment.
The apparent quantum yield (AQY) for solar H 2 O 2 production was calculated using the following equation: The solar-to-chemical conversion (SCC) efficiency 50  The optical irradiance was 175 mW cm −2 and the irradiated area was 5.2 cm 2 . During the PCORR, the clear liquid samples were collected at fixed time intervals by using a 1 mL of syringe followed by syringe filtration (0.20 μM pore, 15 mm Minisart RC, syringe filter). The H 2 O 2 amount in the solution was quantified by DPD colorimetric method using a UV−visible spectrophotometer. Here j indicates current density (mA cm −2 ), j k kinetic current density (mA cm −2 ), n electron transfer number (n), F Faradaic constant (96485 C mol −1 ), D o diffusion coefficient of dissolved oxygen in the 0.1 M KOH at 298 K (1.9 × 10 −5 cm s −1 ), ν kinematic viscosity of the 0.1 M KOH (0.01 cm 2 s −1 ), C o saturation concentration of dissolved oxygen in the 0.1 M KOH (1.2 × 10 −3 mol L −1 ), and ω angular velocity of the disk electrode (rad s −1 ). The slope (B −1 ) of the plot j −1 as a function of ω −1/2 is used to calculate the electron transfer number (n).

* sı Supporting Information
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.chemmater.2c00528. Materials details; characterization information; Supporting Note 1, ATR-FTIR and Raman spectroscopy; Supporting Note S2, N 2 adsorption−desorption; Supporting Note S3, effect of different aliphatic alcohols having fixed water content (50 vol %) on solar H 2 O 2 production; additional data on the characterization of assynthesized samples SEM micrographs, XEDS overlay images, TEM/HRTEM images, FFTs patterns, XPS survey scan/core level spectra, XRD patterns, FTIR spectra, Raman spectra, N 2 adsorption−desorption isotherm, UPS spectra, energy diagram, and TGA thermograms; photochemical H 2 O 2 production plots; absorbance spectra and calibration curve for spectrophotometric quantification of photogenerated H 2 O 2 content in the reaction solution, elemental composition table, and comparison table of photocatalysts for