Interfacial Polarization Triggered by Covalent‐Bonded MXene and Black Phosphorus for Enhanced Electrochemical Nitrate to Ammonia Conversion

MXenes have recently emerged as a new class of electrocatalysts for diverse energy conversion reactions, but they are generally considered to be catalytically inert. Herein, a strategy of interfacial polarization is put forward to improve the catalytic activity of MXenes, with the construction of an interfacial structure consisting of black phosphorus (BP) nanoflakes and Nb2C MXene nanosheets for electrochemical nitrate reduction to ammonia as a proof of concept. The experimental and computational results reveal that a strong interfacial polarization is built between BP nanoflakes and Nb2C nanosheets in the prepared BP/Nb2C composite. The polarization of Nb centers by adjacent BP sites gives rise to the formation of positively centered Nb atoms and BP‐polarized Nb atoms, which synergistically catalyze the cleavage of N–O bonds in HNO2* to form NO*. In addition, the stabilization of monatomic *N by the BP/Nb2C composite is also enhanced compared with BP nanoflakes and Nb2C nanosheets. As a result, both the ammonia yield rate and Faraday efficiency of the BP/Nb2C composite are much higher than that of BP nanoflakes and Nb2C nanosheets. Overall, this work shows the great potential for constructing a strong interfacial polarization to improve the catalytic performance of MXenes‐based electrocatalysts.


Introduction
Nitrate is considered one of the most widespread water pollutants in the world and comes mainly from fertilizer runoff, animal DOI: 10.1002/aenm.202301136 farming, and industrial wastewater. [1] The accumulation of nitrate constantly pollutes surface water and underground aquifers and finally poses a serious threat to human health. [2] To remove nitrate from water bodies, some commercial technologies have been developed, including physical adsorption, membrane separation, and biological denitrification. [3] From the sustainability point of view, the development of chemical processes that enable to use nitrate as a raw material to produce valuable chemical feedstocks is highly desirable. Recently, the electrocatalytic nitrate reduction reaction (NO 3 − RR) has gained tremendous interest in converting nitrate pollutants into valueadded ammonia. Since the NO 3 − RR process is extremely complex and involves multiple electron transfers, the selectivity and efficiency of the NO 3 − RR towards ammonia synthesis are often limited by undesired hydrogen evolution reactions (HER) and other side reactions of forming nitrogen gas, nitrite, and hydrazine. Therefore, there is a pressing need to develop a high-performance electrocatalyst with high efficiency www.advancedsciencenews.com www.advenergymat.de and high selectivity for the electrocatalytic reduction of nitrate to ammonia.
To date, transition metal-based electrocatalysts (e.g., copper, cobalt, and nickel) have demonstrated promise for catalyzing nitrate reduction with high activity for ammonia synthesis. [4] However, the d-orbital electrons in transition metals are conducive to the metal-H bond formation, which will intensify competitive HER and limit their NO 3 − RR efficiencies. [5] Compared with transition metals, nonmetallic catalysts hold great potential to provide a more ideal activation center for nitrate reduction due to their weak hydrogen adsorption and abundant valence electrons. [6] As a typical nonmetallic layered material, 2D black phosphorus (BP) has recently gained increasing attention for use in various electrocatalytic reactions, including hydrogen evolution, [7] oxygen evolution, [8] and oxygen reduction, [5a] owing to their abundant accessible active centers and large surface area. However, the electrocatalytic performance of single-component BP is extremely limited by its low conductivity and strong polymerization tendency. Particularly, BP exhibits low stability since the existence of exposed lone pairs makes BP prone to be oxidized. [9] Surface covalent modification has been demonstrated to be an effective strategy to overcome these limitations. [7] Covalent bonding of planar and edge sites of BP with organic groups or molecules will passivate or occupy the exposed lone pairs of BP, thereby improving its durability in catalytic applications. [10] Moreover, due to high carrier mobility and high Fermi level of BP, the surface covalent modification offers the possibilities for inducing effective interfacial charge transfer and thus enhancing the catalytic activity of BP. [11] In addition to BP, a new class of 2D transition metal carbides/nitrides known as MXenes have recently gained considerable interest in electrocatalysis owing to their ultra-thin sheet structure, high conductivity, and chemical diversity. [12] In general, the rich functionality of MXene surfaces facilitates rational interface engineering of electrocatalysts, providing electronic coupling for substantial improvements in electrocatalytic performance. [13] Ti 3 C 2 T x is currently the most reported MXenes applied in electrocatalytic energy conversion reactions, which are composed of three Ti layers and two C layers. However, Ti 3 C 2 T x MXenes with a uniform surface structure are difficult to obtain, and they typically exhibit poor cycling stability during electrocatalytic reactions. [14] As a second representative of the MXene family, Nb 2 C MXene has been extensively studied in recent years. [15] Compared with the most commonly Ti 3 C 2 T x MXenes, Nb 2 C MXene has a thinner atomic structure with more exposed active sites, higher electrical conductivity, and a larger surface area. [16] However, the catalytic activities of pure Nb 2 C MXene are still unsatisfactory in electrochemical applications. [17] In this context, a few strategies have been proposed to improve the electrocatalytic activity of the Nb 2 C MXene, such as hybridizing Nb 2 C MXene with transition metal sulfides, [18] heteroatom doping, [17] and engineering the termination groups of Nb 2 C MXene. [14] Interfacial engineering is another effective way to develop highly active MXene-based catalysts. [19] The large exposed basal plane of MXenes makes it easy to couple with other materials to construct interfacial structures. It has been reported that the covalent bonding of BP nanosheets with MXenes leads to enhanced electrocatalytic activity towards hydrogen evolution [20] and nitrogen reduction [21] arising from efficient electron transfer across the 2D/2D BP/MXene interfaces. Additionally, face-to-face contact is more effective in preventing the active undercoordinated P atoms from being oxidized and reducing the aggregation of BP and MXenes. [22] Until now, the reported BP/MXene composites have primarily been fabricated using Ti 3 C 2 MXene. However, there has been little exploration into constructing BP/Nb 2 C MXene interfacial structures and their potential application in electrochemical nitrate reduction. Additionally, the current understanding of the interfacial interactions between BP nanosheets and Nb 2 C MXene is still limited. It is crucial that we gain a comprehensive understanding of these interactions to fully comprehend the mechanisms involved in nitrate reduction using the BP/Nb 2 C MXene composite.
Herein, we report a simple route to prepare the BP/MXene composite with layered structure and interlaced structure interface through covalent bonding for electrochemical nitrate reduction. The prepared BP/Nb 2 C composite exhibits a high NO 3 − RR activity with a maximum ammonia production rate of 1967.04 μg·h −1 cm −2 . Besides, the BP/Nb 2 C composite achieves excellent NO 3 − RR selectivity towards ammonia synthesis with the Faraday efficiency of 90.4% at −0.5 V versus RHE in 0.1 M K 2 SO 4 electrolyte with the addition of 0.05 M KNO 3 , comparable to those recently reported NO 3 − RR electrocatalysts. Besides, the ammonia yield rate and Faraday efficiency of the BP/Nb 2 C composite are much higher than that of BP nanoflakes and Nb 2 C nanosheets. Density functional theory (DFT) calculations, together with characterization analysis, reveal that the enhanced NO 3 − RR performance of the BP/Nb 2 C composite is attributed to the presence of polarization effects between BP nanoflakes and Nb 2 C nanosheets, leading to a high electron distribution at the interface of the BP/Nb 2 C composite. The produced BP-polarized Nb and positively centered Nb can synergistically catalyze the cleavage of N-O bonds in HNO 2 * to form NO*. Moreover, the BP/Nb 2 C composite displays high stability during ten recycling tests and a 10 h potentiostatic test with neglectable fluctuation in Faraday efficiency and ammonia yield.

Preparation and Characterizations of the BP/Nb 2 C Composite
The BP/Nb 2 C composite was prepared by a simple self-assembly method (Figure 1a). Black phosphorus (BP) nanoflakes were prepared by stripping bulk BP crystals using liquid nitrogen, and Nb 2 C nanosheets were prepared by etching Nb 2 AlC powder with hydrofluoric acid. [23] Then, the BP/Nb 2 C composite was synthesized by ultrasonic mixing of the obtained BP nanoflakes and Nb 2 C nanosheets. Because of van der Waals interaction, BP nanoflakes and Nb 2 C nanosheets can be easily dispersed evenly. The scanning electron microscopy (SEM) image of the BP/Nb 2 C composite shows a layer-by-layer stacking pattern (Figure 1b). [24] The transmission electron microscopy (TEM) image of the BP/Nb 2 C composite further shows that the BP nanoflakes and Nb 2 C nanosheets were stacked on top of each other. This stack preserves the properties of the material itself, and the crystal lattice of BP nanoflakes can be clearly observed (Figure 1c). The prepared Nb 2 C sheets present a typical accordion-like structure reported in other related literature ( Figure S1, Supporting Information). [25] The prepared BP nanoflakes show an obvious lamellar structure ( Figure S2, Supporting Information). By comparing the SEM images of the BP nanoflakes, Nb 2 C nanosheets, and BP/Nb 2 C composite, it is found that the sizes of BP nanoflakes and Nb 2 C nanosheets are slightly decreased after the formation of the BP/Nb 2 C composite (Figures S1 and S2, Supporting Information). Figure 1d shows an HRTEM image of the BP/Nb 2 C composite. The observed lattice fringes are assigned to the (111) crystal plane of BP nanoflakes and the (100) and (103) crystal planes of Nb 2 C nanosheets. The selected area electron diffraction (SAED) pattern of the BP/Nb 2 C composite shows four diffuse rings corresponding to the (203), (109), and (106) crystal planes of Nb 2 C nanosheets (the inset of Figure 1d). [26] In addition, the (111) and (151) crystal planes of BP nanoflakes are also depicted, [27] which is consistent with the reported structure of the BP nanosheets. [28] The results provide further evidence for the successful synthesis of the BP/Nb 2 C composite. [29] Furthermore, the elemental mapping images of the BP/Nb 2 C composite indicate the presence and uniform distribution of P, C, and Nb elements in the BP/Nb 2 C composite ( Figure 1e). [30] X-ray diffraction (XRD) analysis was performed to study the crystalline structures of the BP/Nb 2 C composite, as shown in Figure 2a. Unlike BP nanoflakes with high crystallinity, the XRD pattern of Nb 2 C nanosheets shows low peak intensity, indicating slightly lower crystallinity and disordered structure. [31] The XRD pattern of the BP/Nb 2 C composite shows the characteristic peaks of BP nanoflakes and Nb 2 C nanosheets, indicating that the crystalline structures of BP nanoflakes and Nb 2 C nanosheets are well retained in the prepared BP/Nb 2 C composite. [32] Besides, those small peaks corresponding to the (002), (100), (101), (106), (110), and (109) crystal planes of Nb 2 C nanosheets are identified in the XRD pattern of the BP/Nb 2 C composite. However, the intensities of these peaks are much lower than that in the XRD pattern of Nb 2 C nanosheets, suggesting a lower crystallinity of Nb 2 C nanosheets in the BP/Nb 2 C composite. [29] Moreover, the (002) peak of Nb 2 C nanosheets and the (020) Figure 2. a) XRD patterns and b) Raman spectra of the BP/Nb 2 C composite, BP nanoflakes, and Nb 2 C nanosheets. c) XPS spectra of Nb 3d obtained from BP/Nb 2 C composite (red solid line) and Nb 2 C nanosheets (black dashed line). d) XPS spectra of P 2p obtained from BP/Nb 2 C composite (red solid line) and BP nanoflakes (blue dashed line) are overlaid for comparison. e) Normalized XANES spectra of Nb 2 C nanosheets and from the BP/Nb 2 C composite samples at Nb K-edge. f) Fourier transform curves at Nb K-edge of Nb foil, Nb 2 C nanosheets, and the BP/Nb 2 C composite.
peak of BP nanoflakes are both shifted to a lower angle in the XRD pattern of the BP/Nb 2 C composite ( Figure S3, Supporting Information), indicating that the average layer distances for both BP nanoflakes and Nb 2 C nanosheets are increased after they are composited together. [30] The Raman spectrum of Nb 2 C nanosheets shows a characteristic peak at 263 cm −1 (Figure 2b), attributing to the A1 g vibration outside the plane of symmetry of Nb and C atoms in Nb 2 C nanosheets. [33] The Raman spectrum of BP nanoflakes exhibits three well-defined Raman peaks located at 361, 437, and 465 cm −1 , corresponding to the A1 g, B 2g , and A2 g phonon modes, respectively. [34] These peaks are also identified in the Raman spectrum of the BP/Nb 2 C composite, revealing that the structure of BP nanoflakes is still maintained in the BP/Nb 2 C composite. Compared with the Raman spectrum of BP nanoflakes, the A1 g, B 2g , and A2 g Raman peaks of the BP/Nb 2 C composite show a significant blue shift, indicating the electron transfer between BP nanoflakes and Nb 2 C nanosheets in the BP/Nb 2 C composite. [26] Moreover, the characteristic peaks of BP nanoflakes and Nb 2 C nanosheets are identified in the Raman spectrum of the BP/Nb 2 C composite, demonstrating the successful synthesis of the BP/Nb 2 C composite. [35] The chemical configuration and interfacial interaction between Nb 2 C nanosheets and BP nanoflakes in the BP/Nb 2 C composite were explored by X-ray photoelectron spectroscopy (XPS) analysis. The XPS survey spectrum of the BP/Nb 2 C composite shows the peaks of P 2p, P 2s, C 1s, Nb 3d, Nb 3p, O 1s, and F 1s ( Figure S4, Supporting Information), which is consistent with elemental compositions of the BP/Nb 2 C composite. The identified O 1s signal mainly originates from the intercalated water, sur-face oxidation, and surface functional groups. [23b] The identified F 1s signal is attributed to the etch residue of Nb 2 C nanosheets by hydrofluoric acid. [15] As expected, the signals of Nb 3d, Nb 3p, and F 1s are not detected in the XPS survey spectrum of BP nanoflakes. The high-resolution XPS spectrum of Nb 3d for the BP/Nb 2 C composite is shown in Figure 2c. The main peaks located at 206.9 eV (3d 5/2 ) and 209.6 eV (3d 3/2 ) are attributed to the Nb(4+)-O bonding. [14,36] Compared with Nb 2 C nanosheets, the peak positions of Nb(I), Nb 2 O 5 , and Nb(4+)-O in the Nb 3d XPS spectrum of the BP/Nb 2 C composite are shifted to some extents, while the peak position of Nb-C bonds is almost unchanged. This is probably due to the fact that some Nb atoms in the BP/Nb 2 C composite still maintain their conventional positive centers, while the Nb atoms near the BP layer are shifted due to the polarization of BP. The high-resolution XPS spectrum of P 2p for BP nanoflakes reveals the existence of predominate P-P bonds and a small amount of P-O bonds (Figure 2d). [21] After binding with Nb 2 C nanosheets to form the BP/Nb 2 C composite, the positions of P 2p 1/2 and P 2p 3/2 peaks are shifted from 130.7 and 129.8 to 130.5 and 129.6 eV, respectively. This is attributed to the redistribution of the charge density caused by the interactions between BP nanoflakes and Nb 2 C nanosheets at their interfaces. [10a,37] It is worth noting that the P-O bond in the P 2p XPS spectrum of the BP/Nb 2 C composite is also slightly offset when compared with that of BP nanoflakes, which is due to the formation of P-Nb bonds in the BP/Nb 2 C composite induced by interfacial polarization. Moreover, the peak intensity of the P-Nb bond is enhanced after BP nanoflakes are composited with Nb 2 C nanosheets, indicating that there is a close electron flow and interaction force www.advancedsciencenews.com www.advenergymat.de between BP nanoflakes and Nb 2 C nanosheets in the BP/Nb 2 C composite. [38] This provides further evidence that interfacial polarization exists in the BP/Nb 2 C composite. Figure 2c,d shows that the formation of Nb-P bonds at the interface leads to the increased binding energy of Nb 3d and reduced binding energy of P 2p. This indicates that Nb is partially positively charged and BP is partially negatively charged with the formation of a bonding dipole moment from Nb to BP. The fitting of the deconvoluted O 1s peak of BP nanoflakes shown in Figure S5, Supporting Information reveals the presence of P-O species, including P-O-P (530.6 eV), P-O bonds (531.6 eV), and O-P=O dangling bonding (532.7 eV). [39] These peaks experience a slight blue shift after BP nanoflakes are composited with Nb 2 C nanosheets, indicating the increase of electron concentration and the change of layer spacing for BP nanoflakes in the BP/Nb 2 C composite, [40] which is essential for achieving effective local polarization. [15,41] X-ray absorption near-edge spectroscopy (XANES) and extended X-ray absorption fine structure (EXAFS) were used to further elucidate the oxidation state and structural information of Nb sites in the BP/Nb 2 C composite. The XANES spectrum of the BP/Nb 2 C composite shows higher near-edge absorption energy than that of Nb 2 C nanosheets (Figure 2e), indicating higher Nb chemical states after Nb 2 C nanosheets are composited with BP nanoflakes, which is mainly due to the polarization of Nb sites by BP nanoflakes as demonstrated by the above XPS analyses. The EXAFS spectrum of the BP/Nb 2 C composite exhibits a characteristic peak at around 2.3 Å, which is assigned to the Nb-Nb bonds of the Nb 2 C nanosheets in the BP/Nb 2 C composite. Besides, two characteristic peaks at 0.8 and 1.8 Å are also identified in the EXAFS spectrum of the BP/Nb 2 C composite (Figure 2f), which are assigned to the Nb-O bonds of surface oxidized Nb. [42] In comparison with the EXAFS spectra of Nb 2 C nanosheets and Nb foil, a new peak appearing at 1.7 Å is identified in the EXAFS spectrum of the BP/Nb 2 C composite, which can be assigned to the Nb-P bonds of the BP/Nb 2 C composite, further confirming the presence of the interface polarization between BP nanoflakes and Nb 2 C nanosheets in the BP/Nb 2 C composite.

Electrochemical Measurements
Electrocatalytic nitrate reduction was performed using an H-type electrolytic cell ( Figure S6, Supporting Information). The prepared electrocatalysts, Ag/AgCl electrode, and platinum sheet electrode were used as the working electrode, the reference electrode, and the counter electrode, respectively. Figure 3a shows the linear sweep voltammetry (LSV) curves of the BP/Nb 2 C composite in 0.1 M K 2 SO 4 electrolyte with or without 0.05 M NO 3 − . The LSV current density of the BP/Nb 2 C composite in 0.1 M K 2 SO 4 without 0.05 M NO 3 − is smaller than that in the presence of 0.05 M NO 3 − , indicating that the BP/Nb 2 C composite can effectively reduce NO 3 − ions. Besides, the LSV current densities of BP nanoflakes and Nb 2 C nanosheets are also significantly enhanced after the addition of NO 3 − ( Figure S7, Supporting Information), suggesting that both BP nanoflakes and Nb 2 C nanosheets are able to reduce NO 3 − ions. A comparison of LSV curves of the BP/Nb 2 C composite, BP nanoflakes, and Nb 2 C nanosheets with 0.1 M K 2 SO 4 containing 0.05 M KNO 3 as the electrolyte was also performed (the insert of Figure 3a). The BP/Nb 2 C composite ex-hibits a much higher current density than BP nanoflakes and Nb 2 C nanosheets, implying its superior activity towards nitrate reduction. The NH 3 yield rates of the three catalysts are shown in Figure 3b. The amounts of generated NH 3 were quantitatively analyzed by UV-vis spectrophotometry ( Figure S8, Supporting Information). The NH 3 yield rates of these catalysts all increase with the increase of the applied potential. Notably, the NH 3 yield rates of the BP/Nb 2 C composite are much higher than that of BP nanoflakes and Nb 2 C nanosheets at all tested potentials, which agrees well with the results of LSV measurements, further confirming that the prepared BP/Nb 2 C composite exhibits a superior activity for nitrate reduction to ammonia when compared with BP nanoflakes and Nb 2 C nanosheets. The maximum NH 3 yield rate of 1967.0 μg h −1 cm −2 (6491.2 μg h −1 mg cat −1 ) is achieved by the BP/Nb 2 C composite at −0.6 V versus RHE, the maximum NH 3 FE of 90.4% at −0.5 V versus RHE (Figure 3c). The NH 3 FEs of the BP/Nb 2 C composite are also significantly higher than that of BP nanoflakes and Nb 2 C nanosheets at all tested potentials, indicating a higher selectivity of the BP/Nb 2 C composite towards ammonia synthesis. A comparison of the NO 3 − RR activity of the BP/Nb 2 C composite with other reported NO 3 − RR electrocatalysts ( Figure 3d and Table S1, Supporting Information) reveals that the NH 3 FE of the BP/Nb 2 C composite is superior to most reported electrocatalysts, further demonstrating the excellent NO 3 − RR performance of the BP/Nb 2 C composite for ammonia synthesis.
The applicability of the BP/Nb 2 C composite towards NH 3 production from NO 3 − RR was further studied by varying the electrolyte pH, which is an important parameter influencing the NO 3 − RR kinetics. [43] The competition between HER and NO 3 − RR is significantly affected by the electrolyte pH. [44] Figure  S9, Supporting Information shows that the NH 3 FEs of the BP/Nb 2 C composite under neutral conditions are higher than that under acidic and alkaline conditions. The competitive HER is the main reason for the unsatisfactory NO 3 − RR performance of the BP/Nb 2 C composite in a strong acid electrolyte (pH = 1). [45] In a strong alkaline electrolyte (pH = 13), excessive OH − ions may compete with NO 3 − ions to accept electrons, leading to slow hydrogenation kinetics for the synthesis of NH 3 . [46] The maximum NH 3 FEs of the BP/Nb 2 C composite are still over 40% under strongly acidic or alkaline conditions, indicating that the BP/Nb 2 C composite has great application potential in nitratepolluted wastewater with various pH values. [47] To corroborate the source of produced ammonia, the BP/Nb 2 C composite was used as the working electrode in 0.1 M K 2 SO 4 electrolyte without 0.05 M NO 3 − for 1 h of electrolysis at open circuit potential (Figure 3e). It is found that only trace amounts of ammonia are detected at open circuit potential without the addition of nitrate to the electrolyte, which might be due to a small amount of contamination in the experimental setup, such as the electrolytic cell. [48] To unambiguously verify that the produced ammonia originates from nitrate reduction, isotope labeling experiments using K 14 NO 3 or K 15 NO 3 followed by product identification via 1 H-NMR were performed. [49] As shown in Figure 3f, three peaks are observed in the 1 H-NMR spectrum after 10 h of electrolysis when using K 14 NO 3 as the electrolyte, corresponding to 14 NH 4 + . [50] Only double peaks corresponding to 15 NH 4 + are identified in the 1 H-NMR spectrum when using K 15 NO 3 as the electrolyte, no peak is detected when a blank electrolyte is used, suggesting that the produced ammonia totally originates from electrochemical NO 3 − RR. [51] The electrochemical NO 3 − RR process for ammonia synthesis involves the transfer of eight electrons and several possible intermediates, including NO 2 , NO 2 − , NO, N 2 O, N 2 , NH 2 OH, NH 3 , and NH 2 NH 2 . In fact, the NO 3 − RR process can be considered as a successive hydrogenation process, [52] and the step for the hydrogenation of nitrate to produce nitrite is very critical. [53] According to the standard curve at = 540 nm, the NO 2 − FE of the BP/Nb 2 C composite is found to be as low as 7.01% at −0.5 V versus RHE (Figures S10 and S11, Supporting Information), indicating that the NO 3 − RR towards the production of NO 2 − is inhibited. Given that the total FE of NH 3 and NO 2 − is 97.36%, which is close to 100%, it can be inferred that there are almost no other by-products in the NO 3 − RR process. [4a,54] Consecutive electrolysis tests at −0.5 V versus RHE were performed to probe the durability of the BP/Nb 2 C composite. The chronoamperometry test for 10 h of continuous NO 3 − RR electrolysis presents a very steady current density (Figure 3g). The accumulated amount of NH 3 for NO 3 − RR by the BP/Nb 2 C composite at different time intervals also shows a good linear upward trend (the inset of Figure 3g), suggesting that the BP/Nb 2 C composite has a stable ammonia yield for nitrate reduction. The neglectable fluctuation in FEs and NH 3 yields during the tests for a total of 15 cycles further confirms the excellent stability of the BP/Nb 2 C composite for NO 3 − RR (Figure 3i). [55] Besides, no obvious change is observed on the LSV curves of the BP/Nb 2 C composite after 500 cycles, also corroborating the high stability of the BP/Nb 2 C composite for NO 3 − RR (Figure 3h). The characterizations of the BP/Nb 2 C composite before and after cycling tests by SEM and TEM revealed that the microscopic morphology of the BP/Nb 2 C composite still maintains a layered structure without obvious changes after cycling experiments (Figures S12 and S13, Supporting Information), indicating the good durability of BP/Nb 2 C composite. Moreover, the Nb 3d, P 2p, and O 1s XPS spectra of the BP/Nb 2 C composite after cycling experiments are almost the same as the pristine BP/Nb 2 C composite, and the peaks of Nb-C, P x O y , and Nb-O are still evident ( Figure S14, Supporting Information). All these results fully demonstrate that the BP/Nb 2 C composite exhibits high durability for NO 3 − RR towards ammonia synthesis.

Elucidation of the NO 3 − RR Mechanism
To understand the origin of high NO 3 − RR activity of the BP/Nb 2 C composite, the electrochemically active surface areas (ECSA) of the BP/Nb 2 C composite, BP nanoflakes, and Nb 2 C nanosheets were determined via the electrochemical double-layer capacitance (C dl ) measurements (Figure 4a). [56] Cdl values were obtained by cyclic voltammetry (CV) measurements at different scan rates (20-150 mV −1 ) (Figures S15-S17, Supporting Information). The Cdl value of the BP/Nb 2 C composite is 0.791 mF cm −2 , much higher than that of BP nanoflakes (0.462 mF cm −2 ) and Nb 2 C nanosheets (0.618 mF cm −2 ), indicating that the higher ECSA of the BP/Nb 2 C composite contributes to its supe-rior NO 3 − RR performance when compared with BP nanoflakes and Nb 2 C nanosheets. Subsequently, electrochemical impedance spectroscopy (EIS) measurements were performed to gain further insights into the interfacial electron-transfer kinetics of the BP/Nb 2 C composite-based NO 3 − RR, as shown in Figure 4b. Compared with BP nanoflakes and Nb 2 C nanosheets, the diameter of the arcs representing the charge transfer resistance of the BP/Nb 2 C composite is far smaller, indicating that the BP/Nb 2 C composite has a better electrocatalytic NO 3 − RR kinetic performance than BP nanoflakes and Nb 2 C nanosheets. [37,57] Electrokinetic analysis was conducted to determine the rate-determining steps (RDS) of the nitrate reduction catalyzed by the three catalysts. Figure 4c shows that the Tafel slope BP/Nb 2 C composite is 62.78 dec −1 , much lower than 120 mV dec −1 , suggesting that the RDS for the BP/Nb 2 C composite-based NO 3 − RR is the first oneelectron transfer in the step for the NO 3 − to NO 2 − conversion. [58] The much higher Tafel slope of BP nanoflakes (173.74 mV dec −1 ) indicates that the NO 3 − RR over BP nanoflakes is limited by the initial adsorption and activation of NO 3 − . [59] Subsequently, the www.advancedsciencenews.com www.advenergymat.de HER kinetics of the BP/Nb 2 C composite was probed by acquiring linear fitting of the Tafel curve based on the Tafel equation, as shown in Figure S18, Supporting Information. The BP/Nb 2 C composite exhibits a larger Tafel slope (544.2 mV dec −1 ) than that of BP nanoflakes (321.5 mV dec −1 ) and Nb 2 C nanosheets (471.19 mV dec −1 ), suggesting that the BP/Nb 2 C composite has a lower HER activity than BP nanoflakes and Nb 2 C nanosheets, which may be a critical factor for the BP/Nb 2 C composite to afford such a high NH 3 FE. [60] During electrolysis, the reduction reaction at the cathode is crucial (the inset of Figure 4d). As a specific quenching agent for hydrogen radicals, tert-butanol (TBA) was used to identify the role of active hydrogen. [61] As shown in Figure 4d, with the increase of TBA concentration (0, 10, 25, and 50 mM), the amount of produced ammonia in the electrolyte is gradually decreased. This result indicates that the active hydrogen plays a key role in triggering the conversion of nitrate, and the more accessible active hydrogen from the BP/Nb 2 C composite also supports its high catalytic activity. [62] In addition, the conversion rates of the BP/Nb 2 C composite for the reduction of NO 3 − to NO 2 − and the reduction of NO 2 − to NH 3 at −0.5 V versus RHE were measured, as shown in Figure S19 and − to NO 2 − is slightly faster than that from NO 2 − to NH 3 due to the difference in the number of electron transfers, which agrees well with our experimental results. [50b] Based on these results, it can be concluded that the reduction of NO 2 − to NH 3 is much easier than NO 3 − reduction. [63] To further understand the mechanism underlying the enhanced NO 3 − RR performance of the BP/Nb 2 C composite, the charge distribution on the surfaces of Nb 2 C nanosheets and the BP/Nb 2 C composite was studied by kelvin probe force microscopy (KPFM) analysis. Figure 4e shows the accordion shape of the edge of Nb 2 C nanosheets, which agrees well with the surface morphology of Nb 2 C nanosheets captured by SEM ( Figure S1, Supporting Information). After compositing with the BP nanoflakes, this accordion-like appearance disappears (Figure 4f). The contact potential difference (CPD) image of Nb 2 C nanosheets shows a uniform hue and the CPD image of the BP/Nb 2 C composite shows different shades, indicating the changes in the surface potential and charge distribution of Nb 2 C nanosheets after compositing with BP nanoflakes. This is likely due to the formation of covalent bonds between BP nanoflakes and Nb 2 C nanosheets, as well as the surface roughness variation after they are composited together ( Figure S20, Supporting Information). [64] The KPFM results further reveal that the polarization of Nb atoms is present in the BP/Nb 2 C composite, which is in good agreement with the above characterization results shown in Figure 2.

DFT Calculations and Practical Applications
DFT calculations were performed to further study the catalytic mechanism of the BP/Nb 2 C composite-based NO 3 − RR. Figure 5a shows the intermediates and corresponding free energy changes that may be involved in the process of the BP/Nb 2 C composite-based NO 3 − RR. As discussed above, the first step of NO 3 − RR is the adsorption of NO 3 − onto the surface of the BP/Nb 2 C composite. Next, NO 3 * is converted into HNO 3 * by the hydrogenation reaction. The subsequent hydrogenation steps result in the formation of *NOH, *N, *NH, *NH 2 , and *NH 3 . Finally, ammonia is desorbed from the surface of the BP/Nb 2 C composite. The process of the BP/Nb 2 C composite-based NO 3 − RR at 0 V versus RHE (red line) is limited by the following three steps with high barrier potentials: *NO to *NOH (0.36 eV), *NH to *NH 2 (0.42 eV), and *NH 2 to *NH 3 (0.55 eV), which are also the potential speedlimiting steps for the BP/Nb 2 C composite-based NO 3 − RR at 0 V versus RHE (yellow box area). [65] According to the molecular configuration of each intermediate and the corresponding calculation results, the reduction of *NO to *NOH requires the cleavage of stable Nb-O bonds, and the cleavage of Nb-N bonds is necessary for the reduction steps of *NH to *NH 2 and *NH 2 to *NH 3 . [66] Besides, all reduction steps in the BP/Nb 2 C compositebased NO 3 − RR process become spontaneous when the applied potential is −0.55 V versus RHE. It is worth noting that there are two dehydration steps (*HNO 3 to *NO 2 and *NOH to *N) in the whole NO 3 − RR process. Since the dehydration steps do not involve electron transfer, the free energy changes of the dehydration steps at 0 or −0.55 V versus RHE are the same. Our experimental results have revealed that the highest FE for the production of NO 2 − is achieved at −0.3 V versus RHE ( Figure S21, Supporting Information). This is because the thermodynamic barrier for the conversion of NO* to HNO* (0.36 eV) is difficult to overcome when the applied potential is −0.3 V versus RHE, leading to the low NH 3 FE and high NO 2 − FE of the BP/Nb 2 C compositebased NO 3 − RR at this potential. [67] When the applied potential is increased to −0.5 V versus RHE, the NH 3 FE of the BP/Nb 2 C composite-based NO 3 − RR increases sharply, while its NO 2 − FE decreases significantly. This is highly consistent with our calculation results that the applied potential of −0.5 V versus RHE can fully overcome the thermodynamic barriers for the conversions of NO* to HNO* and *NH to *NH 2 . Moreover, the applied potential of −0.5 V versus RHE is close to the thermodynamic barrier for the conversions of *NH 2 to *NH 3 , leading to the high NH 3 FE of the BP/Nb 2 C composite-based NO 3 − RR at this potential. The subsequently calculated charge difference reveals the electronic state and charge transfer of the BP/Nb 2 C composite interface. The top and side views of the charge difference clearly show that there is a higher electron distribution at the interface of the composite (Figure 5b,c). The heat map of the charge difference along the (110) lattice surface of this region shows that the inter-layer charge density has increased significantly after Nb 2 C nanosheets and BP nanoflakes are composited together, which is due to the polarization effects between BP nanoflakes and Nb 2 C nanosheets, leading to the increased electron transfer at the interface of the BP/Nb 2 C composite. [68] The calculated charge difference of the BP/Nb 2 C composite along the z-axis reveals that the inter-layer charge moves significantly in the positive direction. This is strong evidence for an increase in charge density at the interface of the BP/Nb 2 C composite (Figure 5d,e). The change of charge on the surface of Nb 2 C nanosheets after compositing with BP nanoflakes is not uniform because the diffusion of electrons around P atoms of BP nanoflakes causes more charges to be concentrated on Nb atoms of Nb 2 C nanosheets. [22b] The calculated charge difference of the surface Nb layer shows that Nb atoms mainly exist in two forms after Nb 2 C nanosheets are composited with BP nanoflakes: type-I and type-II. Type-I is a typical positively centered Nb due to its intrinsic metallic properties. [69] Type-II is BP-polarized Nb. Based on the molecular structure of each intermediate shown in Figure 5a, type-I Nb and type-II Nb in the elliptic region of Figure 5f tend to bind with both sides of N-O bonds in HNO 2 *, respectively, thus synergistically promoting the cleavage of N-O bonds to form NO*. Moreover, negative charges are concentrated in the triangular region of Figure 5f, which favors the stabilization of monatomic *N and thus promotes the reduction of nitrate to ammonia. In contrast, only positively centered Nb atoms and no BP-polarized Nb atoms existed in Nb 2 C nanosheets, so the synergistic effects between positively centered Nb atoms and BP-polarized Nb atoms existed in the BP/Nb 2 C composite for nitrate reduction are absent in Nb 2 C nanosheets. This can be used to explain why the BP/Nb 2 C composite exhibits a much higher ammonia yield and FE of NO 3 − RR than Nb 2 C nanosheets (Figure 3b,c). Based on our experimen-tal and calculation results, it is clear that the enhanced NO 3 − RR performance of the prepared BP/Nb 2 C composite is attributed to the interfacial polarization between BP nanoflakes and Nb 2 C nanosheets.
Subsequently, the practical application of the proposed NO 3 − RR towards ammonia synthesis was studied, and the application potential of the electrolyte after NO 3 − RR as a plant nitrogen fertilizer was evaluated. Figure 6a,b shows the results of pot experiments for tomato seedlings fertilized by the electrolytes before (control group) and after (experiment group) NO 3 − RR. The basic physical and chemical properties of used soil for pot experiments are shown in Table S3, Supporting Information. In the first week, there is no significant difference in plant height for tomato seedlings fertilized by the electrolytes before and after NO 3 − RR. However, the growth of tomato seedlings fertilized by the electrolyte after NO 3 − RR is much better than that fertilized by the electrolyte before NO 3 − RR, as shown in Figure 6c. The average plant height of tomato seedlings for the experiment Figure 6. a) The plant heights and the final root lengths of tomato seedlings fertilized by the electrolytes before (control group) and after (experimental group) NO 3 − RR. b) The dry weight, wet weight, plant height, root length, total nitrogen, and total carbon content of tomato seedlings in the experiment and control groups. c) Control group and experimental group tomato plants over time.
group is around 34.3 cm, which is much higher than that of the control group, with a value of 19.2 cm. The final root length of tomato seedlings for the experiment group reaches 8.6 cm, while it is only 5.9 cm for the control group. Besides, yellowing leaves are observed in the control group due to the lack of nitrogen fertilizer. [70] Figure 6a,b shows that the dry weight, wet weight, plant height, root length, total nitrogen, and total carbon content of tomato seedlings in the experiment group are far higher than that in the control group, indicating that the electrolyte after NO 3 − RR enables to provide tomato seedlings with more adequate and easy-to-use nitrogen fertilizer. This is because the BP/Nb 2 C composite can effectively convert nitrate-N into ammonia-N and thus promote the growth of tomato seedlings in this study.

Conclusion
In summary, we have experimentally and theoretically demonstrated that the presence of interfacial polarization between BP nanoflakes and Nb 2 C nanosheets is vital to the enhanced NO 3 − RR performance of the BP/Nb 2 C composite towards ammonia synthesis. The resulting BP-polarized Nb and positively centered Nb tend to bind with both sides of N-O bonds in HNO 2 *, respectively, synergistically promoting the cleavage of N-O bonds to form NO*. We have reported here that the BP/Nb 2 C composite exhibits excellent NO 3 − RR performance with the Faraday efficiency of 90.4% at −0.5 V versus RHE and high stability during ten recycling tests and a 10 h potentiostatic test. Admittedly, the BP/Nb 2 C composite developed here does not demonstrate any significant advantages in terms of the rate of ammonia yield. However, this study suggests that constructing interfacial polarization could be a promising approach to adjusting the reactivity and selectivity of MXene-based electrocatalysts. These findings may inspire the design of more efficient electrocatalysts for energy-related reactions in the future.

Experimental Section
Materials: Salicylic acid (C 7  Synthesis of BP Nanoflakes: A simple green liquid nitrogen-assisted exfoliation method was used to synthesize fewer-layer BP nanoflakes. The synthetic procedures were as follows: Firstly, 500 mg of BP powder was soaked in liquid nitrogen for 1 h. Then, the frozen pretreated BP was transferred into a 50 mL plastic tube. After the residual liquid nitrogen in the plastic tube was volatilized, 40 mL of dimethyl sulfoxide solution (DMSO: H 2 O = 1:1) was added to the plastic tube. After ultrasonication for 2 h, the obtained dispersion was centrifugated at 5000 rpm for 3 min. The supernatant was freeze-dried for 36 h to obtain a gray-brown powder. The resulting powder was then vacuum-dried overnight at 70°C to further remove water and isopropanol. The gray-brown powder was identified as successfully exfoliated few-layer BP nanosheets. [23] The as-prepared few-layer BP nanoflakes were mixed with ethanol. Subsequently, the nafion perfluorinated resin solution was added to the mixture in a ratio of 1:0.5 (molar ratio). After ultrasonication for 2 h, the resulting dispersion was dropped on carbon cloth to obtain the BP working electrode.
Synthesis of Nb 2 C-MXene: Nb 2 AlC powder was soaked in a 50% aqueous hydrofluoric acid solution at 55°C for 48 h. The resulting suspension was then repeatedly washed with deionized water and centrifuged at 3500 rpm until the pH of the solution was around 6. The last portion of the suspension was filtered through a polyester membrane and washed with 50 mL of deionized water. [24] The obtained Nb 2 C-MXene suspension was added to an equal proportion of naphthol resin solution. After ultrasonication, the mixture was dropped on carbon cloth to obtain the Nb 2 C-MXene working electrode.
Preparation of BP/Nb 2 C Composite: The BP supernatant and the Nb 2 C suspension were mixed in a mass ratio of 1:1. After the two were well mixed through ultrasonication, the BP/Nb 2 C composite colloidal suspension was obtained. After ultrasonication, the mixture was dropped on carbon cloth to obtain the BP/Nb 2 C composite working electrode. The loading amount for each catalyst on carbon cloth was 0.3-0.5 mg.
Characterizations: The crystal information was examined by the X-ray diffraction (XRD, Rigaku Ultima IV, Japan) technology using Cu-K radiation ( = 1.5418 Å) at a rate of 2°min −1 . The morphology was surveyed by scanning electron microscopy (SEM, Hitachi MC1000, Japan). TEM and high-resolution TEM images (TEM/HRTEM, JEOL JEMF200, Japan) were operated at 200 kV. KPFM images were obtained by measuring the surface potential of the material using Kelvin probe force microscopy (KPFM, Bruker Dimension Icon). The chemical composition was tested via X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, USA) with an Al K X-ray light source. Raman spectroscopy was performed on a Raman microscope (Horiba LabRAM HR Evolution, Japan) with a wavelength range of ≈50-4000 cm −1 ( = 532 nm). X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) data were collected at Singapore Synchrotron Light Source (SSLS) and Shanghai Synchrotron Radiation Facility (SSRF).
Electrochemical Nitrate Reduction: All electrochemical measurements were performed at room temperature in custom-made H-type glass cells separated by Nafion 117 membranes (DuPont) using an electrochemical analysis (CHI 760E, Chenhua Shanghai, China). All electrochemical tests were performed in triplicate. The as-prepared catalysts, Ag/AgCl electrode, and platinum sheet electrode were used as working, reference, and counter electrodes, respectively. Unless otherwise noted, all potentials reported in this study were calibrated to reversible hydrogen electrodes (RHE) based on Equation (1).
The carbon cloth was cut into a "convex" shape with a size of 1 × 2 cm. K 2 SO 4 solution with a concentration of 0.1 M was used as electrolyte, and KNO 3 (0.05 M NO 3 -N) was added to the cathode chamber as a reactant. A magnetic stirrer bar was placed in the cathode chamber to ensure that the solution was well mixed during the reaction. The electrochemical performance of the catalysts was investigated with LSV at a scan rate of 5 mV s −1 . Electrochemical impedance spectroscopy (EIS) was recorded in a frequency range of 100 kHz to 1 Hz with an ac signal of 5 mV amplitude. The electrochemically active areas (EASA) of the as-prepared catalysts were determined by scanning CV at rates ranging from 20 to 150 mV −1 . Stability and durability tests were performed through chronopotentiometry and chronoamperometry measurements.
Pretreatment Procedures of Carbon Cloth: The carbon cloth was soaked in concentrated HCl for 24 h, then rinsed repeatedly with deionized water, and dried overnight in a 60°C oven. The procedures for the treatment of proton exchange membranes were as follows: proton exchange membranes were firstly boiled with 3% H 2 O 2 for 1 h, washed with deionized water several times, and then boiled with deionized water for 2 h. Subsequently, proton exchange membranes were boiled with 0.5 M H 2 SO 4 for 1 h and finally rinsed with deionized water several times before use.
Calculation of FE and NH 3 Yield Rates: The ammonia-evolving rate was calculated using Equation (2).
The FE was calculated using Equation (3).
The FE of nitrite was calculated using Equation (4).
where C NH 3 is the concentration of NH 3 , C NO − 2 is the concentration of nitrite, V is the volume of the electrolyte, t is the electrolysis time, m is the mass of the catalyst, M is the molar mass of the related substance, Q is the total charge passing the electrode, and F is the Faradaic constant (96 485 C mol −1 ). [25] Determination of Produced Ammonia: Concentration-absorbance curves were prepared using a series of NH 4 Cl solutions at 655 nm. The resulting NH 3 concentration was measured by a UV spectrophotometer (UV2600, SDPTOP, Shanghai, China). First, a certain amount of electrolyte was taken out and diluted to the concentration within the detection range. Then, 2 mL of diluted electrolyte was added into 2 mL of 1 M sodium hydroxide solution containing 5 wt.% salicylic acid and 5 wt.% sodium citrate. Subsequently, 1 mL of 0.05 M sodium hypochlorite solution and 0.2 mL of 1 wt.% sodium nitroprusside solution were added. The obtained solution was shaken to ensure that it was evenly mixed. The resulting solution was allowed to stand in the dark for 1 h, then the absorbance of the resulting solution was measured at a wavelength of 655 nm. [26] Determination of Produced Nitrite: A standard concentrationabsorbance curve was obtained by plotting the absorbance data at a wavelength of 540 nm versus a series of concentrations of KNO 2 solutions. First, 0.2 g C 12 H 14 N 2 ·2HCl, 4 g C 6 H 8 N 2 O 2 S, and 10 mL phosphoric acid solution with the concentration of 85 wt.% were added into 50 mL deionized water to prepare color developer. Then, a certain amount of electrolyte was taken out and diluted to the concentration within the detection range. Subsequently, 5 mL of diluted electrolyte was mixed with 0.1 mL of color developer, and the resulting solution was stood for 20 min at room temperature. Finally, the absorbance of the resulting solution was measured at a wavelength of 540 nm. Pot Experiment: At first, tomato seedlings were cultivated in a seedling tray and then similar seedlings were selected to be transplanted into cultivation pots after germination. Two control groups were set up in the experiment, the electrolyte group after the reaction (group A) and the electrolyte group before the reaction (group B). Each group was set up with five parallel. The cultivated potted soil was mixed with nutrient-free peat soil and vermiculite in a ratio of 3:1. The diluted electrolytes were applied to provide nutrients for the growth of tomato seedlings. All cultivation pots were exposed to sunlight for 40 days after transplanting.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.