High-Performance Photocathodic Bioanalysis Based on Core–Shell Structured Cu ₂ O@TiO ₂ Nanowire Arrays with Air–Liquid–Solid Joint Interfaces

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Mar 2022High-Performance Photocathodic Bioanalysis Based on Core–Shell Structured Cu2[email protected]2 Nanowire Arrays with Air–Liquid–Solid Joint Interfaces Zhaohong Wang†, Liping Chen†, Dandan Wang, Zhenyao Ding, Xiqi Zhang, Xinjian Feng and Lei Jiang Zhaohong Wang† College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123 , Liping Chen† College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123 , Dandan Wang College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123 , Zhenyao Ding College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123 , Xiqi Zhang Key Laboratory of Bio-Inspired Materials and Interfacial Science, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190 , Xinjian Feng *Corresponding author: E-mail Address: [email protected] College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123 and Lei Jiang Key Laboratory of Bio-Inspired Materials and Interfacial Science, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Future Technology, University of Chinese Academy of Sciences, Beijing 101407 https://doi.org/10.31635/ccschem.021.202100842 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Developing photoelectrochemical (PEC) bioassays based on the principle of a photocathodic measurement of enzymatic product H2O2 is highly attractive because it can naturally avoid interfering signals arising from reductive species inherent to biofluids. However, fluctuant oxygen levels in the analyte solution can compromise the accuracy of photocathodic bioanalysis and restrict its application because oxygen reduction potential is similar to H2O2. Herein, we addressed this restriction by constructing a triphase biophotocathode with air–liquid–solid joint interfaces by immobilizing an oxidase enzyme film on the tip part of superhydrophobic p-type semiconductor nanowire arrays. Such a triphase biophotocathode has a reaction zone with steady and air phase-dependent oxygen concentration which stabilizes and increases the oxidase kinetics, and enables the photocathodic measurement principle in reliable PEC bioassay development with high selectivity, good accuracy, and a wide linear detection range. Moreover, the biophotocathode shows good stability during repeated testing under light illumination. This reliable PEC bioassay system has broad potential in the fields of disease diagnosis, medical research, and environmental monitoring. Download figure Download PowerPoint Introduction Photoelectrochemical (PEC) bioanalysis based on the PEC and enzymatic cascade reaction has attracted broad attention and shows great potential in applications including environmental monitoring and biochemical analysis.1–5 In a conventional oxidase-based PEC bioassay system, oxidase catalyzes its substrate (the analyte) in the presence of O2 and generates hydrogen peroxide (H2O2),6–10 which can subsequently react with photogenerated holes on the photoanode, resulting in a significant photocurrent increase.11–14 However, due to the strong oxidation capability of photogenerated holes,15–19 side reactions will inevitably take place when reductive endogenous/exogenous species co-exist in biological solution. This leads to an inherent drawback of photoanodic bioanalysis, that is, poor selectivity. Moreover, photogenerated holes may decompose biomolecules immobilized on the surface of photoanodes, which degrades bioassay performance. Photogenerated electrons are prone to reacting with electron acceptors but inert to those easily oxidizable interferents.20–23 Therefore, developing PEC bioassay systems based on the principle of photocathodic measurement of enzymatic product H2O2 would improve the selectivity. However, practical applications of the photocathodic measurement principle have been restricted by concomitant oxygen reduction reactions at similar potentials.24 Since oxygen levels vary in aqueous solution, photocathodic current changes contributed from oxygen reduction reaction will compromise the measurement accuracy of enzymatic product H2O2. Hence, a biophotocathode with a constant oxygen concentration at the reaction zone is desired to design a reliable photocathodic bioassay system. In this study, we demonstrate a triphase biophotocathode with a constant and air phase-dependent interface oxygen concentration, which enables the development of a reliable photocathodic bioassay system. As illustrated in Figure 1a, a thin oxidase layer is immobilized on the tip part of p-type semiconductor nanowire (NW) arrays with superhydrophobicity. When the biophotocathode is immersed in an analyte solution, the enzyme layer can be wetted, while air pockets will be trapped inside the free space between NWs due to surface hydrophobicity.25–30 This leads to the formation of an air–liquid–solid triphase reaction interface that enables the rapid delivery of oxygen to the reaction zone through the atmosphere-connected air phase.31–35 Compared with conventional biophotocathodes with a liquid–solid diphase cascade reaction interface (see Supporting Information Figure S1), oxygen levels at the triphase reaction zone will be air phase-dependent and constant. Figure 1b shows an illustration of the triphase reaction zone where the oxidase catalytic and photocathodic cascade reaction takes place. In the presence of oxygen, oxidase catalyzes its substrate (analyte) and produces H2O2-the concentration of which is proportional to that of the substrate. Meanwhile, the photogenerated electrons will diffuse outward to the surface of p-type semiconductor NWs and reduce the enzymatic product H2O2, resulting in a proportional photocurrent increase. Due to the fixed oxygen concentration at the triphase reaction zone, the photocathodic current originating from background oxygen reduction reaction will be constant, and consequently a reliable PEC bioassay system dependent on the photocathodic measurement of oxidase catalytic product H2O2 can be developed. Figure 1 | (a) Schematic illustration of the biophotocathode with an air–liquid–solid triphase reaction interfacial microenvironment. A thin oxidase enzyme film is modified on the surface of superhydrophobic p-type semiconductor NW arrays. Such architecture enables the rapid transport of oxygen to the reaction zone through the air phase, leading to a constant and liquid phase-independent oxygen level at the reaction zone. (b) Enlarged view for the triphase reaction zone where oxidase catalytic and photocathodic cascade reaction takes place. In the presence of oxygen, oxidase catalyzes its substrate (analyte) and produces H2O2, which will be subsequently reduced by the photogenerated electrons, leading to a photocathodic current response. Download figure Download PowerPoint Experimental Methods Preparation of Cu2[email protected]2 NWs Before synthesis, 200-mesh copper meshes were cleaned ultrasonically in a mixture of acetone, ethanol, and deionized water. The Cu(OH)2 NWs were prepared by an electrochemical anodization method, which was performed in a 3 M NaOH aqueous solution at a constant current density of 6 mA cm−2 for 14 min. The Cu mesh covered with Cu(OH)2 NWs was washed with deionized water and dried in the air. Then Cu(OH)2 NWs transformed to Cu2O NWs under annealing at 450 °C for 20 min. After that, a conformal TiO2 layer was deposited on Cu2O NWs using an atomic layer deposition (ALD) system (Savannah-100, Cambridge Nanotech Inc., Cambridge, MA) at 180 °C. The precursors used were titanium tetraisopropanolate and water. Each of the 150 cycles of ALD was 30 s and the power was 1000 W. The as-prepared NWs were further annealed in air at 350 °C for 30 min. Fabrication of triphase biophotocathode The Cu2[email protected]2 NWs were submerged in a solution of cyclohexane and trichloro(1H,1H,2H,2H-perfluorooctyl)silane (PFOTS) with a volume ratio of 1000:1 for 90 min, then dried at 120 °C for 90 min to obtain a superhydrophobic surface. To make the tip of the NWs hydrophilic, the PFOTS-modified Cu2[email protected]2 NWs were exposed to an oxygen plasma for 30 s with an area of 0.25 cm2, and then an oxidase solution was coated on the hydrophilic region of the Cu2[email protected]2 NWs. The oxidase solution containing glucose oxidase (GOx; 20 mg mL−1), chitosan (2 mg mL−1 in 1 vol % CH3COOH), H2O, and glutaraldehyde (5 wt %) were thoroughly mixed at a volume ratio of 20:10:9:1. For control experiments, diphase biophotocathodes were synthesized by casting the oxidase solution directly onto the surface of hydrophilic Cu2[email protected]2 NWs. Characterization Morphologies were characterized by scanning electron microscopy (SEM) (S4700; Hitachi, Tokyo, Japan) and transmission electron microscopy (TEM) (Tecnai G2 F20 S-Twin; FEI, Boston, MA). X-ray diffraction (XRD) analysis was performed using an X-ray powder diffractometer (X'Pert Pro MPD; Holland Panalytical, Almelo, Holland). X-ray photoelectron spectroscopy (XPS) spectra were obtained with an ESCLAB 250 XI from Thermo Fisher Scientific ((Waltham MA). The water contact angle (CA) was measured by a CA goniometer (JC2000D6; Powereach, Shanghai, China). The depth of enzyme infiltration was examined by laser scanning confocal microscopy (LSCM; LSM 800 with Airyscan, Aalen, Germany). For LSCM characterizing, the prepared enzyme photoelectrodes were immersed in the residual rhodamine B (RhB) solution (25 mg L−1) for 15 min; the residual RhB molecules were removed before ambient drying. The fluorescent signal of the absorbed RhB was recorded by LSCM with a 48-fold oil lens. PEC bioassay All the PEC bioassays were carried out with an electrochemical workstation (660e; CH Instruments, Inc., Shanghai, China) at room temperature. The Cu2[email protected]2 NW-based biophotocathode served as the working electrode, Pt wire electrode served as counter electrode, and Ag/AgCl electrode served as the reference electrode. A Tris–HCl buffer solution (0.01 M, pH 7.4) was used as the electrolyte. A visible light with wavelength <420 nm was illuminated on Cu2[email protected]2 NWs (0.25 cm2) for PEC bioassay. A stirring rate of 600 rpm was used in all measurements. The linear sweep voltammetric (LSV) experiments were performed from +0.1 to −0.1 V, and amperometric (i-t) tests were performed at 0 V with a scan rate of 50 mV s−1. The oxygen concentration fluctuation experiment was performed in a nitrogen-bubbled Tris–HCl solution. Oxygen was allowed to diffuse from the air into the solution via continuous stirring. The dissolved oxygen meter (MP516; Sanxin, China) was used to measure the dissolved oxygen concentration in the electrolyte. The selectivity tests were conducted by i-t with various interferents (0.05 mM of methanol, xylose, urea, sucrose, galactose, acetaminophen, mannose, ethanol, dopamine, and ascorbic acid) added to the electrolyte after the introduction of 0.5 mM glucose. The kinetics of the oxidase catalytic reaction were measured in a 0.01 M Tris–HCl solution (pH 7.4) in the absence of light illumination (without photocathodic reaction). The initial H2O2 formation rate (v0) was calculated by measuring the amount of H2O2 produced at 5 min after the addition of glucose. The concentrations of H2O2 were determined by iodometry using a UV–vis spectrophotometer in all experiments.36 In detail, 840 μL Tris–HCl solution containing enzymatic product H2O2 was mixed with 375 μL 0.1 M potassium phthalate and 375 μL iodide reagent (0.18 M NaOH, 1.2 M potassium iodide, 0.3 mM ammonium molybdate mixed at a volume ratio 1:1:1), then the absorbance value of the mixture at 350 nm after stirring for 5 min was recorded. Results and Discussion In a typical experiment, core–shell structured Cu2[email protected]2 NW arrays were chosen as a model photocathode. Cu2O is a p-type semiconductor with a narrow band gap of ∼2.0 eV that guarantees efficient visible light absorption.37–39 The NW arrays architecture facilitates fast transport of photogenerated holes to the substrate (along the NWs) and electrons to the surface of the photoelectrode (across the NWs). To improve its photocorrosion resistance during light illumination, a thin TiO2 layer was further conformally coated onto the surface. Figure 2a shows the fabrication process of the Cu2[email protected]2 NW arrays on a copper mesh substrate. The photocathode was initially fabricated by growing Cu(OH)2 NWs via an electrochemical anodizing method, and then the Cu2O NWs were formed in situ by treating Cu(OH)2 NWs in an annealing procedure. Finally, the surfaces of the Cu2O NWs were coated with a thin layer of TiO2 via the ALD process. Figure 2 | Preparation and morphology characterizations of the Cu2[email protected]2 core–shell structured NW arrays. (a) Schematic illustration of the preparation procedure for Cu2[email protected]2 NWs on a Cu mesh; (b) SEM top view of the NWs grown on Cu mesh. (c and d) SEM side views of Cu2[email protected]2 NWs at low and high magnification, respectively. (e) TEM image of the NWs. (f and g) High-resolution TEM images of a single NW at different magnifications. (h and i) High-angle annular dark-field scanning TEM image and elemental mapping distribution, respectively. Download figure Download PowerPoint Figure 2b and Supporting Information Figure S2 show SEM images of the NW arrays grown on the copper mesh substrate. The NW has a length of about 6 μm (Figures 2c and 2d). According to XRD analysis shown in Supporting Information Figure S3, peaks at 29.55°, 36.41°, 52.45°, 61.34°, and 73.52° can be indexed to the (110), (111), (200), (220), and (311) crystal planes of Cu2O (JCPDS no. 05­0667), respectively, suggesting good crystallinity of the Cu2O NWs. No peaks originating from titanium oxide (TiO2) were observed, which can be attributed to the amorphous feature of the surface coating. Supporting Information Figure S4 shows XPS of O 1s and Ti 2p for the Cu2[email protected]2 NWs. In Supporting Information Figure S4a, peaks at 529.6 and 530.2 eV can be ascribed to the O2− state in TiO2 and Cu2O, respectively, while the peak at 531.5 eV can be attributed to the OH− in Cu(OH)2.40 In Supporting Information Figure S4b, peaks at 458.5 and 464.4 eV can be attributed to Ti 2p3/2 and Ti 2p1/2, respectively.41 This confirms the presence of TiO2. The TEM image shown in Figure 2e indicates that the NWs have a rough surface and a diameter of about 200 nm. A TiO2 layer with a thickness of 10 nm was uniformly coated on the Cu2O surface (Figure 2f). Figure 2g shows a fringe spacing of 0.246 nm assigned to the {111} facet of Cu2O, while the irregular shape of the outside coating layer represents the amorphous structure of TiO2; these observations are consistent with the XRD analysis discussed above. High-angle annular dark-field scanning TEM image (Figure 2h), elemental mapping distribution (Figure 2i), and energy-dispersive X-ray spectroscopy (EDXS) ( Supporting Information Figure S5) further demonstrate that the NW has a Cu2[email protected]2 core–shell structure. To fabricate the triphase biophotocathode, the surfaces of the Cu2[email protected]2 NW arrays were further chemically functionalized with PFOTS. The wettability of the NW arrays changed from hydrophilic into hydrophobic with a water CA measurement of 147 ± 2° ( Supporting Information Figure S6a) after surface modification. This can be attributed to the cooperative effect between the low surface energy and the micro-nanocomposite structure. One side of the NWs covered copper substrate was then rinsed by a short-term oxygen plasma, and a GOx/chitosan composite layer was drop-cast onto the surface with an area of 0.25 cm2. This led to the formation of an air–liquid–solid triphase biophotocathode, and the corresponding SEM image is presented in Figure 3a. The surface of the photocathode exhibits hydrophilicity with a water CA of 53 ± 2° ( Supporting Information Figure S6b), and the GOx/chitosan composite can only be observed on the tip region of NWs according to the cross-section SEM image shown in Figure 3b. Figure 3 | (a and b) Top- and side-field emission scanning microscopic images of the Cu2[email protected]2 NW arrays that are covered with a thin GOx/chitosan layer at the tip part, respectively. (c) Top views of the dye-sensitized triphase biophotoelectrode as illustrated in panel (d) obtained by LSCM. Download figure Download PowerPoint To prove the existence of the air–liquid–solid triphase interface, the oxidase-modified Cu2[email protected]2 NWs biophotocathode was soaked in RhB dye aqueous solution which could adsorb in the region where the solution and biophotocathode contacted. The dye-absorbed enzyme photoelectrode was then characterized with LSCM. The scanning area is marked in Figure 3a. Figure 3c shows optical images of the dye-sensitized area at different depths of NWs obtained by continuous layer scanning from top to bottom as illustrated in Figure 3d. The fluorescence images shown in Figure 3c demonstrate that the RhB molecules adsorb only on the tip region of NWs with a length of about 1.5 μm. These results suggest that the aqueous solution could only moisten the enzyme and was forbidden from penetrating into the free space between hydrophobic NWs. This confirms that an air–liquid–solid triphase reaction zone will be formed when such a biophotoelectrode is immersed in an aqueous analyte solution. In contrast, for hydrophilic NWs-based biophotocathodes ( Supporting Information Figure S7a), the distribution of the fluorescence signal in each scanning layer ( Supporting Information Figure S7b) indicates that aqueous solution can wet the oxidase layer along the entire NWs, and that a liquid–solid diphase cascade reaction zone will be formed when immersed in an aqueous analyte solution. The performance of the triphase biophotocathode was then evaluated in a three-electrode PEC system ( Supporting Information Figure S8). The excitation light source was a xenon lamp equipped with a 420 nm cut-off filter (adjusted to an intensity of 20 mW cm−2). The LSV (Figure 4a) shows that the photocathodic current (blue curve) negatively increases when the applied potential changes from +0.1 to −0.1 V in the absence of glucose (analyte). The system presents a typical p-type semiconductor conductivity. After increasing the glucose concentration to 10 mM, the photocurrent increases accordingly due to the reduction reactions between enzymatically generated H2O2 and photogenerated electrons (red curve). The absolute photocurrent response (the gap between the red and blue curves) barely increases when the applied potential negatively shifted after 0 V; thus, the potentiometric test at 0 V was performed for the assessment of the PEC bioassay system. Figure 4 | (a) LSV characteristic curves of the biophotoelectrode in test solution without (blue curve) and with glucose (red curve) under light illumination. (b) Photocathodic currents of the diphase and triphase bio-PEC system in blank test electrolyte with changing oxygen concentrations from 0.09 to 0.32 mM at 0 V versus a Ag/AgCl electrode. (c) Histograms for interference effects on the triphase bio-PEC system while measuring 0.5 mM glucose at 0 V under visible light irradiation. Interferents include methanol, xylose, urea, d(+)-sucrose, galactose, acetaminophen, mannose, ethanol, dopamine, and ascorbic acid at 0.05 mM. (d) Photocurrent signals for the triphase and normal PEC bioassay systems to 3 mM glucose with different oxygen concentrations in the electrolyte. Download figure Download PowerPoint Oxygen level fluctuations in the electrolyte are a key factor affecting the stability of the photocathodic current. This limits the practical application of the photocathodic reaction of H2O2 in a reliable PEC bioassay development. Figure 4b (blue shows that in the diphase system the photocathodic current increases with oxygen concentration in electrolyte (without This can be ascribed to the reduction of dissolved oxygen by the photogenerated electrons at the Such photocurrent the accuracy of the photocathodic detection of enzymatic product H2O2 and practical applications of photocathodic detection principle in reliable bio-PEC system development. In contrast, a triphase system has oxygen level fluctuations in analyte solution that the photocurrent response (Figure This can be ascribed to the rapid transport of oxygen in the reaction interface from the atmosphere-connected air Due to the high value of oxygen in air versus that in aqueous solution the interfacial oxygen concentration of triphase biophotocathode will be by the air and constant, a for detection of H2O2 using photocathodic measurement The of photocathodic detection principle in PEC bioassay system development can naturally avoid from those oxidizable endogenous/exogenous species in leading to a bioassay selectivity. Figure shows the photocurrent response the addition of different of in the presence of 0.5 mM glucose. The introduction of methanol, xylose, urea, sucrose, galactose, acetaminophen, mannose, ethanol, dopamine, and ascorbic acid with a concentration of 0.05 changes in selectivity of the triphase The constant interface oxygen level at the triphase biophotocathode can the oxidase kinetics and formation rate of H2O2, leading to a bioassay Figure (red curve) shows that for a fixed glucose level the photocurrent response of the triphase PEC bioassay system constant as the oxygen level changes from to in the solution. In contrast, the photocurrent response of a diphase system in Figure for the glucose concentration increases when the oxygen level changes from to 0.32 mM. These results that the triphase PEC bioassay system is to the oxygen level fluctuation in solution and exhibits a The air–liquid–solid triphase biophotoelectrode architecture design oxygen to transport from the air to the reaction zone and then into the thin oxidase which oxygen layer thickness in electrolyte and enables a high oxidase The kinetics of oxidase catalytic reaction at diphase and triphase biophotocathode were by measuring the formation rate of enzymatic product H2O2 using an iodometry (without the photocathodic reaction). Figure presents the UV–vis spectra of the changes after 5 min at various glucose concentrations using the triphase The corresponding initial formation (v0) of H2O2 were calculated through the Figure shows that the changes with the glucose concentration 15 mM after which the This is consistent with the Figure shows that the reaction rate of the triphase and diphase can be obtained as and respectively. the oxidase catalytic constant can be calculated by the value into The of the triphase enzyme electrode is s−1. This value is about that obtained using a conventional diphase enzyme electrode as shown in the in Figure A 0 V ( V Figure 5 | (a) UV–vis spectra of enzymatic product H2O2 with increasing glucose levels based on the triphase system. (b) The formation rate of H2O2 based on triphase and diphase systems versus glucose indicates the oxidase catalytic constant of normal and triphase (red bioassay (c) i-t curves of the triphase PEC bioassay system with different glucose concentration from 0 to 50 mM at 0 V versus (d) for the triphase PEC bioassay system. The linear detection is about 20 conventional diphase systems in panel Download figure Download PowerPoint The oxidase kinetics leads to an linear range. Figure shows the photocurrent of the triphase PEC bioassay system the addition of different of glucose. The photocurrent intensity increases with the of glucose levels from 0.1 to 20 mM (Figure which is that of normal liquid–solid diphase systems in Figure Moreover, under the of glucose to the triphase system, a detection of 20 is ( Supporting Information Figure These results demonstrate that the performance of the bio-PEC system was based on the fabricated biophotocathode where photocathodic and enzymatic cascade reaction at an air–liquid–solid triphase The stability of the triphase biophotocathode was The background amperometric curves of the photoelectrode were recorded with the light illumination As shown in Supporting Information Figure the triphase biophotoelectrode presents a photocathodic current under light illumination. After 1 mM glucose (Figure a photocurrent response of the triphase biophotoelectrode was under light illumination 2 with This the good stability of the Cu2[email protected]2 NW-based triphase PEC bioassay system. Compared with the Cu2O NW-based biophotoelectrode ( Supporting Information Figure the stability can be attributed to the conformal ALD coating of the TiO2 layer which the contact between the Cu2O and the Figure 6 | i-t curves of the photocathodic current response to 1 mM glucose for light illumination Download figure Download PowerPoint fabricated a biophotocathode with a triphase enzymatic and PEC cascade reaction zone based on the of superhydrophobic Cu2[email protected]2 core–shell structured NW arrays. The triphase biophotocathode constant and air phase-dependent interface oxygen concentration, which stabilizes and kinetics of oxidase catalytic and enables the application of photocathodic measurement principle in the development of reliable PEC bioassay system with high selectivity, good accuracy, and wide Moreover, the triphase biophotocathode shows good stability during repeated testing under light illumination. results its to design and the reaction interface in the development of for reliable bio-PEC Supporting Information Supporting Information is and the SEM LSCM XRD XPS test and of is of to This was by the Key of and of and W. The of the for of Photoelectrochemical Zhang Zhang Photoelectrochemical Based on Nanowire into and for in Wang and in Zhang Ding Wang Feng Zhang Jiang Feng from by a Interfacial Wang Based Wang Jiang Feng Based on TiO2 Nanowire Wang Ding Feng at a TiO2 Nanowire Wang by TiO2 Nanowire Photoelectrochemical Wang Zhang TiO2 Arrays as for to in TiO2 of and Photoelectrochemical of to TiO2 with W. TiO2 Nanowire Arrays with on for Wang Feng for Photoelectrochemical Wang Zhang to the of for Photoelectrochemical as a Zhang W. Bioanalysis for Based on in of with Zhang of Photocathodic Based on by W. Photoelectrochemical and of at an with a and of the from in Feng Zhang Jiang to Wang Zhang for and Jiang of for Jiang A by Feng Jiang and of Jiang along the of Feng Jiang at Joint Ding Wang Zhang Feng Jiang the of via Feng Feng Jiang Interfacial Chemical W. on Based on of and in of and Wang Jiang W. Zhang Wang Zhang the of Cu2O in Photoelectrochemical on Cu2O Nanowire Arrays under Zhang Photocathodic with for of H2O2 from W. protected] and of Photocathodic and of to from in Cambridge University Zhang Zhang Wang Feng of and for Photoelectrochemical Information Chinese Chemical was by the Key of and of and

Topics & Concepts