Abstract

Open AccessCCS ChemistryCOMMUNICATION7 Nov 2022A Polyimide-Based Photocatalyst for Continuous Hydrogen Peroxide Production Using Air and Water under Solar Light Yun-Xiao Lin, Hua-Yi Kuang, Shi-Nan Zhang, Xiao-Le Zhang, Guang-Yao Zhai, Xiu Lin, Dong Xu, Jinping Jia, Xin-Hao Li and Jie-Sheng Chen Yun-Xiao Lin School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Hua-Yi Kuang School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Shi-Nan Zhang School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Xiao-Le Zhang School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Guang-Yao Zhai School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Xiu Lin School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Dong Xu School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Jinping Jia School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author , Xin-Hao Li School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author and Jie-Sheng Chen *Corresponding author: E-mail Address: [email protected] School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202201777 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Hydrogen peroxide (H2O2) is a clean disinfectant, bleaching agent, and value-added fuel in various systems and is industrially produced through multistep energy-intensive hydrogenation/oxidation using H2 and O2. The exploitation of metal-free catalysts for alternative H2O2 photoconversion is limited by their limited activity and selectivity, especially for the continuous H2O2 production using only air, water, and solar light. Now, a cheap and robust polymer, poly(4,4′,4″-(1,3,5-triazine-2,4,6-triyl)triphenyl pyromellitimide) (PTTPI) was rationally synthesized by thermal structure-defect control under relatively low temperature (80–190 °C), and the resulting metal-free material with a concisely refined band structure functioned as an efficient and ultra-stable photocatalyst for H2O2 production over a wide pH range. The PTTPI photocatalyst with an optimized defective degree (0.51) and moderate bandgap (2.3 eV) exhibited a H2O2 production yield rate of 16.6 μg h−1 mgcat−1 and a high apparent quantum efficiency of 16.2%. By merging the well-designed PTTPI catalyst with the water transport path of natural cotton, a zero-emission and sunlight-only input system is constructed for continuous H2O2 production from air and water in the natural environment. The design of the polyimide-based catalyst reveals a sustainable future toward efficient H2O2 synthesis with sunlight-only energy input, enabling its implementation in other photocatalytic systems and on-site processing communities. Download figure Download PowerPoint Introduction Hydrogen peroxide (H2O2) is a typical green chemical reagent due to its elemental composition and is widely used for pulp and textile bleaching,1,2 medical disinfection,3,4 water treatment,5 and energy storage,6–8 and it only emits water as a residue after consumption. Currently, the industrial production of H2O2 is an indirect process in which H2 and O2 are kept apart using the sequential hydrogenation and oxidation of ethylanthraquinone.9,10 Much effort has been devoted to developing new techniques for alternative H2O2 production routes that can be operated with lower or zero energy input and waste emissions, including electrosynthesis11–16 and photosythesis17–23 ( Supporting Information Figure S1). Highly efficient, inexpensive, and durable photocatalysts are critically important in the H2O2 photosynthesis process. An optimal photocatalyst for H2O2 production should combine the ability to trigger the reaction under mild visible light with a high stability, especially under the highly oxidative H2O2 solution. Furthermore, the photocatalyst should be nontoxic, abundant, and easily processed on substrate for practical uses in continuous flow photoreactors. Despite extensive efforts, developing stable and efficient photocatalysts remains a substantial challenge. To achieve efficient H2O2 photoconversion, a suitable band structure is essential in the construction of photocatalyst. Several metal-based powder24–26 catalysts with designed band structure have been proposed but were found to generate only small amounts of H2O2 (<0.4 mM). A series of g-C3N4 based27–30 and TiO2 based semiconductors,31–34 as well as other potential photocatalysts,35–37 exhibit high efficiency for H2O2 photoconversion owing to their adjustable and flexible band structure. However, these materials are still hindered by their harsh preparation conditions (e.g. >598 K) and uncertain structure design approach. In contrast, organic polymers,38 with mild preparation conditions and controllable chemical and band structures, are also promising semiconductors for photocatalytic applications. The specific polymer functional groups not only modify the band structure for superb activity but also impart affinity with the stationary substrate during the continuous production. In this case, polymer semiconductors are ideal candidates for exploring best-in-class photocatalysts. We show that precisely tuning the defective degree and band structure makes a polyimide (exemplified here with PTTPI, poly(4,4′,4″-(1,3,5-triazine-2,4,6-triyl)triphenyl pyromellitimide)) act as a metal-free, robust, and highly efficient photocatalyst for H2O2 production over a wide pH range. The PTTPI-120 photocatalyst, with an optimized defective degree (0.51) and moderate bandgap (2.3 eV), provided a H2O2 production yield rate of 16.6 μg h−1 mgcat−1 and a high apparent quantum efficiency of 16.2%. With good durability and processability, we merged the well-designed PTTPI catalyst with the water transport path of natural cotton, realizing a zero-emission and sunlight-only input system for continuous H2O2 production from air and water under sunlight. Results and Discussion Preparation of polyimide photocatalysts We first explored various photoresponsive polymers as possible materials for solar-driven H2O2 production. Polyimides39–41 with very high chemical, mechanical, and thermal stabilities are promising candidates for photocatalytic H2O2 production. Density functional theory (DFT) simulation results (Figures 1a–1c) predict the suitable electronic structure of PTTPI for possible oxygen reduction and water oxidation reactions to directly convert oxygen gas and water molecules. The calculated lowest unoccupied molecular orbital (Figure 1a) and the highest occupied molecular orbital (Figure 1b) of the repeat unit of PTTPI correspond to a conduction and valence band, respectively, with an infinite structure, respectively, based on Kohn–Sham orbitals. The valence band mostly consisted of nitrogen pz orbitals and isolated benzene π orbitals, while the conduction band was mostly distributed on the five-membered ring. Such a significant electron-hole separation leads to a bandgap of 1.75 eV, indicating a potential ability of PTTPI to harvest visible light for H2O2 photoconversion. Inspired by these results, we initially prepared three PTTPI powders via the direct polymerization of 2,4,6-tris(4-aminophenyl)-1,3,5-triazine and pyromellitic dianhydride, which were polymerized at 80 °C (PTPPI-80), 120 °C (PTPPI-120), and 190 °C (PTPPI-190), to tailor the band structure and catalytic activity (Figure 1d, for experimental details, please see the Supporting Information, Materials and Methods). The Fourier transform infrared spectra ( Supporting Information Figure S2) verified the successful polymerization of PTTPI-x. The PTTPI-x powder samples all showed the morphology of aggregated particles by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images (Figure 1e and Supporting Information Figures S3 and S4). The energy-dispersive X-ray (EDX) mapping (Figure 1f) further revealed the even element distribution of C, O, and N. The color change in the PTPPI-x powder from orange to light yellow directly reflected the gradually widened band structures, which were further confirmed by their UV–vis spectra ( Supporting Information Figure S5a). Figure 1 | The calculated band structure and the synthesis of PTTPI. (a) Kohn–Sham orbitals for the conduction band of PTTPI. (b) Kohn–Sham orbitals for the valence band of PTTPI (C, grey; N, blue; O, red; H, white). (c) DFT calculated valence band top and conduction band bottom of PTTPI. The positions of the dashed red and blue lines indicate the reduction level of O2 to H2O2 and the oxidation potential of H2O to O2, respectively. (d) The polymerization of PTTPI at 80, 120, and 190 °C. (e) SEM (left), optical (middle) and TEM (right) images of PTTPI-120. (f) EDX spectroscopy mapping images of the TEM image area. Download figure Download PowerPoint Defective degree and ultrahigh photocatalytic performance of PTTPI It is well-known that the polymerization of polyimide is divided into two parts: (1) the formation of polyamic acid and (2) the further imidization. The reaction temperature is the dividing line for the two steps, especially for this sort of poorly soluble, five-member-ring polyimide. Excluding the size effect from the morphology, the polyamic acid structure is the key factor in the integral band structure of PTTPI. The ratio of polyamic acid in the polymer node is defined as the “defective degree”. Three typical node-to-node structures are proposed with a defective degree of 1, 0.5, and 0, respectively (Figure 2a). As demonstrated by 13C solid-state NMR (Figure 2b and Supporting Information Figure S6), two major peaks around 160–180 ppm were identified and utilized to quantify the defective degree of the PTTPI-x polymer. As shown in Figure 2d, based on the mathematical relationship of Ca:Cb in the 13C NMR spectrum and the proposed structure, the formula is given as follows, x = ( 2 y − 1 ) / ( y + 1 )where x is the defective degree of polymer, and y is the value of Ca:Cb. Figure 2 | The defect effect in the PTTPI polymer chain and corresponding band structure. (a) Proposed typical node-to-node structures of PTTPI-x polymer with different defective degrees. (b) 13C solid-state NMR spectra of PTTPI-80, PTTPI-120, and PTTPI-190 photocatalysts. (c) 1H solid-state NMR spectra of PTTPI-80, PTTPI-120, and PTTPI-190 photocatalysts. (d) Experimental band structure of PTTPI-80, PTTPI-120, and PTTPI-190 powder and corresponding defective degree according to the illustrated curve. (e) The H2O2 and H2 yield rate of PTTPI-80, PTTPI-120, and PTTPI-190 powder photocatalysts in the top-illuminated photoreactor. With a continuous O2 gas flow and vigorous stirring, 200 mg catalyst was dispersed into 60 mL water and photoirradiated at λ > 400 nm for 4 h using a 300 W xenon lamp at 298 K. Download figure Download PowerPoint Thus, the defective degree of PTTPI-x is illustrated on the mentioned curve (Figure 2d), and the defective degree was further verified by 1H solid-state NMR (Figure 2c), where the peak of H1 (representing carboxyl and amino hydrogen) significantly shrank when the polymerization temperature increased. The same conclusion was also drawn based on O 1s X-ray photoelectron spectroscopy (XPS) ( Supporting Information Figure S7). The corresponding experimental band structure of the PTTPI-x was estimated and given (Figure 2d) based on UV–vis absorption spectra ( Supporting Information Figure S5a), Tauc curves ( Supporting Information Figure S5b) and Mott–Schottky plots ( Supporting Information Figure S8). It should be noted that the hole- and electron-concentrated unit spread on both sides of the ideal imide connection in the polymer network. Thus, the polyamic acid (defect) site that exists in the PTTPI structure plays a role as a unique pathway for the electron-hole separation and recombination, with a significant effect on the charge transport driving force. In general, as shown in the experimental band structure and defective degree curve (Figure 2d), once the defective degree reaches 0.51, the PTTPI-120 had a moderate width bandgap (2.3 eV), implying strong visible light absorption and usage. As largely depending on the thermal structure-defect control, the lowest conduction band (0.1 eV) was obtained, which enhanced the reduction selectivity for 2e− oxygen reduction reaction (0.68 V vs reversible hydrogen electrode (RHE)) rather than the harmful 1e− oxygen reduction reaction (O2 + H+ + e− → •OOH (−0.13 V vs RHE)) and hydrogen evolution reactions (0 V vs RHE). We also tested the H2O2 photoconversion performance of PTTPI catalysts with three defective degrees as well as the corresponding H2 generation rate by chromatography (Figure 2e). The PTTPI-120 with the lowest conduction band has a negligible H2 yield rate and the best H2O2 production yield rate of 16.6 μg h−1 mgcat−1 in pure water with visible light (λ > 400 nm) in a top-illuminated photoreactor ( Supporting Information Figure S9) according to the standard quantitative methods ( Supporting Information Figure S25). In contrast, the 0.51 defective degree photocatalyst also has the lowest valence band (2.4 eV), implying a boosted oxidation property for both 2e− (1.23 V vs. RHE) and 4e− water oxidation reaction (1.76 V vs RHE), which is not easy to achieve due to the uphill thermodynamics. Once the in situ-generated O2 from 2e− water oxidation reaction forms rapidly for O2 reduction consumption, it will kinetically facilitate the H2O2 generation.35 Isotopic photoreaction experiments also confirmed that the mechanism of H2O2 photoreaction was the combination of both water oxidation and O2 reduction31 ( Supporting Information Figure S19). With an optimized electron reduction selectivity and the enhanced hole oxidation property, the thermal structure-defect control strategy successfully modifies the charge separation inside the semiconductor and boosts the H2O2 photoconversion activity of the PTTPI photocatalyst. As the benchmark H2O2 photocatalysts in this work, the PTTPI-120 was used as the main sample without further notice for the following measurements. Indeed, the PTTPI catalyst effectively accumulated the H2O2 concentration in the solution with high production efficiency and low degradation rate. The weight percentages of H2O2 in the water dispersion of PTTPI powder increased linearly within 18 h and then reached a plateau of 0.09 wt % at 24 h due to the upper limit of the reaction balance (Figure 3a). The apparent quantum efficiency of PTTPI catalyst for H2O2 generation at λ = 400 nm was recorded to be 16.2% ( Supporting Information Figure S10), far surpassing the performance of state-of-the-art photocatalysts in the literature (Figure 3b and Supporting Information Table S1). The reverse decomposition of the as-formed H2O2, confronted by most noble metal-based photocatalysts,42 is negligible on our PTTPI catalyst, with 93.3% of the H2O2 content remaining after 100 h, comparable to the decay rate of pure H2O2 solution with 95.5% left under fixed conditions (Figure 3c). The XPS results ( Supporting Information Figure S11) further reveal the metal-free nature of the PTTPI powder containing only C, N, and O. Furthermore, the high chemical stability of PTTPI also afforded high durability, maintaining 98.6% of its H2O2 production activity after nine cycles of 4 h reactions (Figure 3d). In addition, the PTTPI photocatalyst generally performed well as photocatalyst in both acidic and neutral electrolytes with a high yield rate and H2O2 concentration upper limit (Figure 3e) due to the pH tolerance of polyimide. Figure 3 | H2O2 production performance of best-in-class PTTPI photocatalyst. (a) H2O2 weight percentage during the reaction in the O2 bubbling top-illuminated photoreactor produced by PTTPI (200 mg catalyst in 60 mL water, λ > 400 nm). Inset: the top view of the top-illuminated photoreactor. (b) The H2O2 generation yield rate and apparent quantum efficiency of PTTPI during 4 h irradiation and other benchmarked H2O2 photocatalysts in the literature (circle, PTTPI; top triangle, reduced g-C3N4; left triangle, Au-BiVO4; bottom triangle, g-C3N4/mellitic triimide (MTI); square, g-C3N4/pyromellitic diimide (PDI); pentagon, rGO/Cd3(TMT)2; hexagon, RTF523; details listed in Supporting Information Table S1). (c) H2O2 decomposition curves in the dark with (red) and without (green) 60 mg of PTTPI in 200 mL of 20% H2O2 solution. (d) H2O2 production cycles of PTTPI (200 mg catalyst in 60 mL water) under visible light irradiation (λ > 400 nm), with water replaced every 4 h during photoirradiation. (e) H2O2 yield rate (4 h) and H2O2 concentration upper limit (24 h) of PTTPI within a pH range of 1–11, which was adjusted by phosphoric acid and potassium hydroxide. Download figure Download PowerPoint Water and H2O2 transport path of microreactors With a powerful photocatalyst in hand, we then integrated the PTTPI polymer with natural cotton fibers (CFs) to introduce a possible mass diffusion path, avoiding the use of electricity-driven stirring and pressured gas that is widely used in standard reactors.43 CFs ( Supporting Information Figure S12) were chosen as an example for fabricating microreactors due to their good stability and natural capability of water transport. The PTTPI coated cotton fiber (PTTPI/CF) microreactors were prepared via the direct polymerization of the two monomers in the presence of CF, as depicted in Figure 4a. 13C solid-state NMR spectra ( Supporting Information Figure S13) and attenuated total reflection-infrared spectra ( Supporting Information Figure S14) revealed the successful combination of PTTPI and CFs. The as-formed PTTPI nanoparticles in the yellow PTTPI/CF hybrid with a mean size of approximately 200 nm were dispersed randomly on the surface of each cotton fiber according to the SEM (Figure 4b and Supporting Information Figure S15) and high-angle annular dark-field (HAADF) observations (Figure 4b inset). Figure 4 | Fabrication of PTTPI coated cotton fiber (PTTPI/CF) and H2O2 production performance of the zero-energy-input system under sunlight. (a) The in situ PTTPI polymerization process on the CFs. (b) TEM image of PTTPI/CF. Insert; HAADF images of PTTPI nanoparticles on PTTPI/CF. (c) Time-dependent anti-gravity water transport height of the CFs (grey) and PTTPI/CF (red). Inset: schematic diagram of water transport in the PTTPI/CF microreactors. (d) Image of the quartz dish container with an unsealed quartz cover filled with PTTPI/CF microreactor. (e) The integral continuous production system. (f) The H2O2 concentration curve under 0.6 sunlight irradiation (1.94 g PTTPI in 620 mL water) before (sampling from the container, red hollow circles) and after (sampling from the collector, red filled circles) water dropping. (g) The H2O2 concentration curve under 1.8 sunlight irradiation (1.94 g PTTPI in 620 mL water) before (sampling from the container, orange hollow circles) and after (sampling from the collector, orange filled circles) water dropping. (h) Image of the integral continuous production system with a Fresnel lens concentrator. Download figure Download PowerPoint By merging the natural CFs with the new polymer photocatalysts, the role of CFs as the “self-driven pump” for antigravity water transport completed the function of mass transport in the microreactor.44 After the introduction of the PTTPI species, the wettability velocity of the bare CFs was well maintained with a similar time of approximately 0.2 s for the spray of a water droplet (2 μL) on the sample surface ( Supporting Information Figure S16). To further investigate the antigravity water transport capability and velocity of the PTTPI/CF microreactor, the PTTPI/CF microreactors and bare CFs (average width 2 mm, length 12 cm) were kept in deionized water mixed with Rhodamine B. Observation of the in situ water transport ( Supporting Information Figure S17) and transport height (Figure 4c) showed that the PTTPI/CF microreactors maintained 64–83% transport efficiency of that of bare CFs for transporting water uphill. Thus, PTTPI/CF microreactors can work sufficiently as an efficient water and H2O2 transport path during the reaction, making the reaction of oxygen gas and water proceed smoothly at the three-phase interface of the solar-driven microreactor, as well as inheriting the photocatalytic performance of the PTTPI powder. Under fixed reaction conditions, the PTTPI/CF hybrid provided an average H2O2 yield rate of 14.2 μg h−1 mgcat−1 under 4 h irradiation even without stirring ( Supporting Information Table S1) and was comparable to that of PTTPI powder (16.6 μg h−1 mgcat−1), indicating the inert nature of the CFs in the catalytic system with the multidimensional network structure for better mass diffusion. The chemical structure of the PTTPI components remained unchanged after long periods of light irradiation, as verified by the infrared spectrum analysis results ( Supporting Information Figure S18). Zero-emission and sunlight-only input system for H2O2 production and water treatment The solar-driven reaction process and the self-driven mass transfer process make it possible to produce H2O2 solution outdoors using only water and air under sunlight. The assistance of any electricity-driven facilitator is not essential, and thus, the energy consumption, except for solar rays, is nearly zero. As a proof-of-concept application of the PTTPI/CF microreactors in real-life devices, a quartz container covered by quartz cover (diameter: 0.285 m) with a gap of 3 mm for air exchange (Figure 4d) and connected with a valve-controlled water tank for water addition was directly used to host the PTTPI/CF microreactors for the continuous production of pure H2O2 solution under ambient sunlight (0.6 sun) (Figure 4e). Such a simple container containing 1.94 g of PTTPI active materials and 620 mL of water produced H2O2 solution with a concentration between 99–104 mg L−1 at a water flow rate of 120 mL h−1 for 24 h (Figure 4f). The same batch of PTTPI/CF microreactors was used for more than 100 h without obvious deactivation. Such a H2O2 production technique can also be extended naturally to the purification of a water system simply by replacing pure water with river water from the Huangpu River ( Supporting Information Figure S20). After 1 h of irradiation, the total organic content (TOC) of the river water decreased to 4.97 mg L−1, which is already lower than the standards (5 mg L−1) for drinking water quality.45 The TOC of river water was further reduced to 1.83 mg L−1 within 3 h to meet stricter environmental quality standards (<2 mg L−1).46 Then, the system can continuously purify the river water at a flow rate of 250 mL h−1 at the same TOC level. To further increase the generation rate of H2O2 under sunlight, a Fresnel lens concentrator was applied to focus the sunlight beam with light intensity up to 1.8 sun (Figure 4h). Thus, the concentration of the H2O2 solution rapidly increased to 225 mg L−1 after 100 min and continued to produce a 225–230 mg L−1 H2O2 solution at a flow rate of 500 mL h−1 (Figure 4g). By increasing or decreasing the water flow rate, we could tune the product concentration of H2O2 or the water treatment efficiency over a wide range for different application scenarios with nearly no feedstock cost ( Supporting Information Figures S21–S23). This continuous H2O2 photoconversion strategy is suitable for scale-up practical use, which enlarges the H2O2 laboratory yield rate (λ > 400 nm) by 3.35 times via a 100-fold volume reactor under sunlight, achieving an ultrahigh H2O2 yield rate >59.17 μg h−1 mgcat−1 ( Supporting Information Figure S24). Conclusion We successfully designed a polyimide photocatalyst PTTPI for H2O2 production from air and water under sunlight. By tuning the defective degree on the polymer chain, we effectively improved the band structure of PTTPI for the designated H2O2 production reaction, reaching an ultrahigh H2O2 generation rate of 16.6 μg h−1 mgcat−1. Taking inspiration from nature, we successfully merged the water transport path of natural CFs together with best-in-class photocatalysts PTTPI-120 as microreactors to achieve the function of plant leaves for the direct production of useful H2O2 using solar light, air, and water. Notably, the PTTPI-based microreactors are highly efficient for outdoor H2O2 production, which continuously produce high concentrations of H2O2 (225–230 mg L−1) for either direct use or further industrial application. The present result enables new directions in the search for robust polymer-based photocatalyst and demonstrates the feasibility of continuous H2O2 production in the natural environment. Supporting Information Supporting Information is available and includes experimental procedures, catalyst preparation, isotopic photoreaction, UV–vis, IR, XPS, 13C solid-state NMR, and other details. Conflict of Interest There is no conflict of interest to report. Funding Information This research was made possible by a generous grant from the National Natural Science Foundation of China (grant nos. 21931005, 21720102002, 21737002, and 22071146), Shanghai Science and Technology Committee (grant nos. 19JC1412600 and 20520711600) and the SJTU-MPI partner group.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call