Atomically Dispersed Manganese Lewis Acid Sites Catalyze Electrohydrogenation of Nitrogen to Ammonia

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE6 Jun 2022Atomically Dispersed Manganese Lewis Acid Sites Catalyze Electrohydrogenation of Nitrogen to Ammonia Zhoutai Shang†, Bin Song†, Hongbao Li†, Hong Zhang, Fan Feng, Jacob Kaelin, Wenli Zhang, Beibei Xie, Yingwen Cheng, Ke Lu and Qianwang Chen Zhoutai Shang† Institutes of Physical Science and Information Technology, Key Laboratory of Structure and Functional Regulation of Hybrid Materials of Ministry of Education, Anhui Graphene Engineering Laboratory, Anhui University, Hefei, Anhui 230601 Hefei National Laboratory for Physical Sciences at the Microscale, Hefei, Anhui 230026 , Bin Song† Laboratory of Nanoscale Biochemical Analysis, Institute of Functional Nano and Soft Materials (FUNSOM), Jiangsu Key Laboratory for Carbon-Based Functional Materials and Devices, Soochow University, Suzhou, Jiangsu 215123 , Hongbao Li† Institutes of Physical Science and Information Technology, Key Laboratory of Structure and Functional Regulation of Hybrid Materials of Ministry of Education, Anhui Graphene Engineering Laboratory, Anhui University, Hefei, Anhui 230601 , Hong Zhang School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 , Fan Feng School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240 , Jacob Kaelin Department of Chemistry and Biochemistry, Northern Illinois University, DeKalb, IL 60115 , Wenli Zhang School of Chemical Engineering and Light Industry, Guangdong University of Technology, Guangzhou 510006 , Beibei Xie State Key Laboratory of Quality Research in Chinese Medicine, Institute of Chinese Medical Sciences, University of Macau, Taipa, Macau 999078 , Yingwen Cheng Department of Chemistry and Biochemistry, Northern Illinois University, DeKalb, IL 60115 , Ke Lu *Corresponding author: E-mail Address: [email protected] Institutes of Physical Science and Information Technology, Key Laboratory of Structure and Functional Regulation of Hybrid Materials of Ministry of Education, Anhui Graphene Engineering Laboratory, Anhui University, Hefei, Anhui 230601 Hefei National Laboratory for Physical Sciences at the Microscale, Hefei, Anhui 230026 and Qianwang Chen Institutes of Physical Science and Information Technology, Key Laboratory of Structure and Functional Regulation of Hybrid Materials of Ministry of Education, Anhui Graphene Engineering Laboratory, Anhui University, Hefei, Anhui 230601 Hefei National Laboratory for Physical Sciences at the Microscale, Hefei, Anhui 230026 Department of Materials Science and Engineering, University of Science and Technology of China, Hefei 230026 https://doi.org/10.31635/ccschem.021.202101106 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Ambient electrochemical nitrogen fixation is a promising and environmentally benign route for producing sustainable ammonia, but has been limited by the poor performance of existing catalysts that promote the balanced chemisorption of N2 and subsequent electrochemical activation and hydrogenation. Herein, we describe the highly selective and efficient electrohydrogenation of nitrogen to ammonia using a hollow nanorod-based hierarchically graphitic carbon electrocatalyst with abundant atomically dispersed Mn sites. We discovered that the electron interactions strengthen the interfacial binding between nitrogen and active Mn Lewis acidic hotspots. The Lewis acid–base interactions promote the chemisorption and lock up nitrogen on the active sites and suppress proton adsorption. The proton-coupled electron transfer cleavage of the nitrogen triple bond through an associative mechanism was confirmed under lower overpotential, which delivered high ammonia yield of 67.5 μg h−1 mgcat.−1 and Faradaic efficiency of 13.7% at −0.25 V versus the reversible hydrogen electrode, along with ∼100% selectivity and significantly enhanced electrochemical stability (about 88.8% current retention over 50 h potentiostatic test) under mild conditions. Our strategy is versatile to tailor the nitrogen fixation performance of single-atom catalysts with atomic accuracy. Download figure Download PowerPoint Introduction Ammonia is an essential platform chemical for the global economy and plays key roles in industry, agriculture, and pharmaceutical chemistry.1–4 It is also a preferred carbon-neutral energy carrier for sustainable energy storage with a 17.6 wt % hydrogen content and a high energy density of 4.3 kW h−1.5–7 Dinitrogen reduction for ammonia is a vital step in the natural nitrogen cycle. Currently, approximately 200 million tons of ammonia is produced annually using the Harber–Bosch process. However, the harsh reaction conditions (typically conducted at 400–500 °C and 100–350 atm) of this incumbent energy-intensive process (N2 + 3H2 → 2NH3, ΔfH0 = −45.9 kJ mol−1) impede its flexible modular production.8–11 Further, other drawbacks, such as process complexity, high energy consumption, and excessive CO2 emission, increase the need for a new sustainable path to produce NH3.12–14 To this end, ambient electrochemical synthesis of ammonia from water and nitrogen under mild conditions is a promising alternative route, especially since it can also deliver about 20% more thermodynamic efficiency compared with the industrial Haber–Bosch process.6,15,16 Since the electrochemical N2-to-NH3 conversion was first reported in 1807,17 many electrocatalysts, including noble metal,18,19 metal oxide,20 metal carbide,9 metal phosphide,21 metal sulfide,22 and metal-free carbon-based materials,23 have been examined to manipulate the absorption and conversion behavior of nitrogen. Unfortunately, the electrochemical nitrogen redox is still plagued by the challenges of low NH3 selectivity and sluggish conversion kinetics.6,18,22,24,25 In addition, this reaction faces strong competition with proton H+ reduction in aqueous electrolytes, which results in very low selectivity and efficiency of ammonia formation.20,26 In this regard, preoccupation of "locked" nitrogen on active sites instead of undesirable hydrogen coverage is expected to strengthen N2 bonding and promote its subsequent electrocatalytic hydrogenation.26 In particular, one of the key approaches to dissociate the strong triple bond (dissociation energy, 942 kJ mol−1) and relieve its electrochemical activation-related kinetics is the use of transition metal-based catalysts with available d-orbital electrons for their favorable geometric and electronic structures.20,23,27,28 Atomically dispersed transition metal active sites with a local coordination unsaturation environment provide an exciting pathway for activating nitrogen.23,29–32 Considering the weak Lewis base character of nitrogen molecules, metallic Lewis acidic active sites anchored on a catalytic framework could enhance the chemisorption between N2 and active binding sites.23,26,33 In addition, the incorporation of a hierarchically porous graphitic carbon matrix could maximize the accessibility between reactant and active hotspots, and simultaneously facilitate efficient mass and electron transfer.2,23,29,34 Accordingly, versatile atomically dispersed metal catalysts are highly desirable to modulate kinetics and boost nitrogen reduction. Compared to bulk and nanoparticle counterparts, single-atom heterogeneous catalysts possess facile reaction thermodynamics and exhibit improved catalytic performance toward hydrogenation catalysis, and the atomically dispersed binding sites could possibly accelerate N2 chemisorption and hydrogenation to produce ammonia. In this contribution, we outline atomically dispersed Lewis acidic Mn sites in a nanorod carbon framework for highly efficient nitrogen conversion. The Lewis acidic Mn–N–C (LA-MnNC) catalysts have abundant and diverse N-coordinated metal hotspots and hierarchically graphitic porous structure. Mesopores especially (total pore volume: 0.47 cm3 g−1, mesopore volume: 0.35 cm3 g−1) have greatly enhanced the chemisorption of N2 through Lewis acid–base interactions. The adsorbed nitrogen in active sites have stronger substrate-reactant binding that favors subsequent N≡N bond activation. 15N isotopic labeling experiments confirmed that the nitrogen come from the reduction of feeding gas, rather than the electrochemical decomposition of the catalyst. The LA-MnNC catalyst achieved a high ammonia yield rate of 67.5 μg h−1 mgcat.−1 with the Faradaic efficiency of 13.7% at −0.25 V versus the reversible hydrogen electrode (RHE) through the associative electrohydrogenation pathway, and exhibited excellent stability in aqueous electrolytes under ambient conditions, holding high promise for further deployable electrochemical nitrogen reduction. Experimental Methods Synthesis of MnO2 nanowire In a typical procedure, 2.028 g MnSO4 (Sigma-Aldrich, Shanghai, China) was added into 300 mL 0.014 M HCl (Thermo Fisher Scientific, Cleveland, OH, USA) solution.35 1.264 g of KMnO4 (Sigma-Aldrich) was dissolved into the 100 mL deionized (DI) water. The above solutions were mixed together under stirring and stirred for another 2 h. Then the mixture was transferred into a Teflon-lined stainless steel autoclave and heated at 120 °C for 12 h. The resultant powder was washed with DI water and ethanol and then vacuum dried. Synthesis of LA-MnNC catalyst Typically, 400 mg of as-prepared MnO2 powder was added into 100 mL H2O. 3 mmol aniline (Sigma-Aldrich), 1.5 mmol pyrrole (Sigma-Aldrich) monomer, and 4 mL 0.5 M H2SO4 (Sigma-Aldrich) were added into the 100 mL ethanol (Aladdin, Shanghai, China) and water (v∶v = 1∶1) solution. After cooling down to about 4 °C, the aniline–pyrrole solution was added dropwise under stirring and kept at 4 °C for 4 h. The MnO2@copolymer products were washed with a large amount of DI water and vacuum-dried, followed by annealing at 900 °C for 1 h with a heating rate of 7 °C min−1 under Ar atmosphere. Followed by an acid washing step for leaching out the metal oxide/metal clusters, the adsorbed Mn2+ ions were confined within the carbon channel and open porous structures of the Mn–N–C samples. The acid leaching treatment was conducted in 0.5 M H2SO4 solution at 80 °C for 8 h, and about 100 mL acid solution was used for 100 mg powder. The resultant powder was heated to 900 °C under Ar flow for 2 h (heating ramp rate, 5 °C min−1) to repair the carbon structure and increase the Mn–Nx content.3 The final catalysts were achieved by a second pyrolysis and labelled as LA-2MnNC (2 represents the molar ratio between aniline and pyrrole monomers). As control samples, LA-1MnNC, LA-0.5MnNC, LA-aMnNC, and LA-yMnNC samples were prepared through the same procedures by adjusting the aniline and pyrrole molar ratio. LA-1MnNC, 2.25 mmol aniline and 2.25 mmol pyrrole were used; LA-0.5MnNC, 1.5 mmol aniline and 2 mmol pyrrole were used; LA-aMnNC, 4.5 mmol aniline was used; LA-yMnNC, 4.5 mmol pyrrole was used, respectively. N-doped porous carbon (NC) was also prepared using the same experimental conditions, and metal oxide was replaced with ammonium persulfate (APS; Sigma-Aldrich) as the initiator. Materials characterization Powder X-ray diffraction (PXRD) patterns were recorded using a Miniflex 600 rotation anode X-ray diffractometer (Rigaku, Tokyo, Japan). X-ray photoelectron spectroscopy (XPS) measurements were performed on an Escalab 250 X-ray photoelectron spectrometer (Thermo Scientific, New York, USA), and the binding energies were calibrated by assigning the C 1s peak at 284.5 eV. The Raman spectra were collected on LabRAM HR 800 system (Horiba Jobin Yvon, Tokyo, Japan) with a 514 nm laser. Temperature-programmed desorption (TPD) was performed on an Autochem II chemisorption analyser (Micromeritics, Norcross, GA) using N2 as the probe molecule. The N2 sorption experiments were carried out using Micromeritics ASAP 2020 system. Before measurements, samples were degassed at 100 °C for 10 h under vacuum. The specific surface areas and pore size distributions were calculated by the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods. The acquired Mn K-edge extended X-ray absorption fine structure (EXAFS) data were collected in fluorescence mode and processed according to the standard procedures using the ATHENA module implemented in the IFEFFIT software packages. In situ Fourier transform infrared (FT-IR) spectra of the electrochemical cell were collected on a Nicolet 8700 FT-IR spectrometer (Thermo Fisher) equipped with an EverGlo IR source. The inductively coupled plasma atomic emission spectroscopy (ICP-AES) measurements were conducted on an Optima 7300DV (ICP) spectrometer (Perkin-Elmer, Waltham, MA). Electrochemical studies An electrochemical workstation (Interface 1010E, Gamry, USA) was employed to perform electrochemical measurements in two-compartment H-cell at room temperature, in which a saturated calomel reference electrode (SCE) and a graphite rod were used as the reference and counter electrodes, respectively. Before nitrogen (99.99%) reduction reaction (NRR) tests, the Nafion 211 membrane (DuPont, USA) was pretreated by heating it in H2O2 5% aqueous solution at 80 °C for 1 h, ultrapure water at 80 °C for 1 h, 0.5 M H2SO4 at 80 °C for 3 h, and finally in ultrapure water at 80 °C for another 4 h. In a typical preparation procedure of the working electrode, 10 mg sample and 40 μL Nafion solution (5 wt %) were dispersed in 960 μL water–isopropanol solution with a volume ratio of 1:3 by sonicating it for 30 min to form a homogeneous ink. Then, 30 μL of the ink was loaded onto a carbon paper (1 × 1 cm2) to prepare a working electrode. The mass loading was 0.3 mg cm−2. The obtained data were calibrated with respect to the RHE by E(vs RHE) = E(vs SCE) + 0.242 V + 0.0591 × pH. The potentiostat tests were performed at different potentials including 0, −0.10, −0.20, −0.25, −0.30, −0.40, −0.50 V versus RHE. Ammonia detection The concentrations of NH3 (NH4+) produced in 0.1 M HCl solutions were determined via a widely used indophenol blue method.36 The calibration curve was established using standard NH4Cl stock solution in the concentrations of 0, 1, 5, 10, 30, 50, and 100 μM, respectively. In each test, after the electrochemical reduction reaction test, the following were added together: 3.0 mL of the standard solution or electrolyte, 0.12 mL phenol solution (20.0 g phenol dissolved in 100 mL ethanol), 0.12 mL sodium nitroprusside (C5FeN6Na2O) solution (1.0 g C5FeN6Na2O in 200 mL H2O), and 0.3 mL oxidizing solution. This solution was kept at room temperature for 2 h in the dark for the formation of indophenol blue. The oxidizing mixture solution containing 2.5 mL sodium hyphchlorite (NaClO) solution and 10 mL alkaline reagent (100 g sodium citrate and 5 g sodium hydroxide dissolved in 500 mL H2O). The UV–vis absorption spectrum of each solution was measured, and the absorbance value at the wavelength of 650 nm was obtained. Hydrazine hydrate detection The concentration of the hydrazine present in the electrolyte was determined by the Watt and Chrisp method.37 The calibration curve was established using standard N2H4 stock solution, in the concentrations of 0, 1, 2, 5, 8, and 10 μM. The p-dimethylaminobenzaldehyde (p-C9H11NO, 5.99 g), HCl (30 mL), and ethanol (300 mL) were first mixed as a color reagent. In detail, 1.0 mL of standard solution or electrolyte after NRR was mixed with 1.0 mL of the coloring reagent solution with rapid stirring for 10 min at room temperature. Then, the absorbance of the mixture was measured at a wavelength of 455 nm. 15N isotopic labeling experiment A mixture of 15N2 (99 atom % 15N, Wuhan Newradar Special Gas Co. Ltd., Wuhan, China) and 14N2 mixture at a molar ration of 1∶2 was used as the feeding gas in the labeling experiment. And a low velocity gas flow system was adopted (∼5 mL min−1). After electrolysis at −0.25 V versus RHE for 6 h, the resultant electrolyte (20 mL) was concentrated to 4 mL, and 0.9 mL of the solution was taken out, followed by adding 0.1 mL of D2O as an internal standard. The produced 15NH4+ was identified using 1H nuclear magnetic resonance measurements (Bruker DRX600, Karlsruhe, Germany). Faraday efficiency calculation The Faraday efficiency (FE) and mass-normalized yield rate of NH3 were calculated as below: FE ( NH 3 ) = [ 3 F × c ( NH 3 ) × V ] / Q Yield rate mass ( NH 3 ) = [ 17 × c ( NH 3 ) × V ] / ( t × m ) where F is the Faraday constant (96,485 C mol−1), t is the electrolysis time, m is the loading mass of the catalyst (0.3 mg), Q is the total charge passed through the electrode, V is the volume of the electrolyte, and c(NH3) is the measured ammonia concentration. Computational methods To identify the catalysis site and explore the NRR reaction mechanism, first-principles calculations were performed by using the Gaussian 16 software package. A nanotube characterized by (n, m) (n = 3 and m = 2) was built, including 76 carbon atoms and a 3.5 Å diameter. The gas-phase conformers were optimized at the M062X/6-311+G (d, p) level to better describe the weak interactions. Since no clear structural differences were observed between nanotubes with two ends hydrogenated or not, only the hydrogenated nanotube was explored as a model to reduce the computational burden. Four carbon atoms in a middle benzene ring were replaced by nitrogen, on which one manganese (Mn) atom was absorbed with them. The chemical potential of proton and electron pairs (H+ + e−) was equal to half of that of the gaseous H2 molecule.38,39 Results and Discussion Structure characterization of single-atom dispersed Mn catalyst Figure 1a illustrates the synthesis of atomically dispersed and N-coordinated Mn sites in a porous graphitic carbon framework. The general procedure includes four steps: (1) in situ copolymerization of aniline and pyrrole; (2) Mn-assisted thermal pyrolytic carbonization; (3) acid etching; and (4) second thermal treatment. The α-MnO2 nanowires were synthesized by a facile hydrothermal method.35 The synthesized nanowires were highly crystalline and had well-defined nanowire morphology, with the diameter of ∼50 nm and the length of several micrometers ( Supporting Information Figure S1). The MnO2 nanowires were used as sacrificial template, and their gradual dissolution initiated the simultaneous interfacial polymerization of pyrrole and aniline, generating coaxial crosslinked [email protected]2 ([email protected]2) nanostructure with uniform polymer wrapping shells (see Figure 1b and Supporting Information Figure S2).40–42 The redox reaction between oxidant and monomer can be expressed by the following: MnO 2 + 4 H + + 2 e − → Mn 2 + + 2 H 2 O ( 1.23 V vs RHE ) (1) n C 4 H 4 NH or n C 6 H 5 NH 2 → ( C 4 H 2 NH ) n or ( C 6 H 4 NH ) n + 2 n H + + 2 n e − ( ∼ 0.7 V vs RHE ) (2) Figure 1 | and characterization of single-atom dispersed LA-MnNC catalyst. of the process of atomically dispersed and Mn sites in porous graphitic carbon framework and its atomic structure of the MnO2@copolymer and and the hollow nanorod of catalyst and the of and atomically dispersed Mn atoms in porous carbon The of the Mn Download figure Download PowerPoint The redox had redox potential than the polymerization potential of the interfacial The Mn2+ ions with the in polymer producing the 2 to the molar The coaxial polymer was then at 900 °C, and electron confirmed that the MnO2 was ( Supporting Information and The of Mn2+ could the as in the where metal in carbon could facilitate the catalytic The reaction the formation of and with the simultaneous incorporation of and Lewis acid of and Mn into the carbon The was washed with acid to Mn including and Mn clusters, and a second thermal activation process was employed to repair the structure of into atomically dispersed single-atom catalysts and increase the producing the LA-2MnNC hollow nanorod catalysts 2 to the molar ratio between aniline and pyrrole The produced LA-2MnNC catalyst abundant and with of nm The electron diffraction the of LA-2MnNC with the of crystalline in the catalyst ( Supporting Information Figure which with dark and of Mn and that the were over the carbon framework The atomic Mn sites can be identified in the with The diameter of the was nm the atomically dispersed of Mn sites. The of Mn was further from X-ray absorption spectroscopy The Mn K-edge X-ray absorption structure spectrum of the LA-2MnNC catalyst exhibited a energy between and and was very to that of the well-defined structure in which is in with results and the strong with Mn to the Figure the Fourier transform spectra of the LA-2MnNC catalyst with standard The and a are at 1.5 and which are of and bonding in the the of a peak in the spectra further the of Mn and the atomic of Mn metal The was as which is very to the Å of in well-defined chemical characterized by a the and of a hollow nanorod catalyst can be by the molar between monomer in the interfacial copolymerization The of 2, 1, and 0.5 to the molar of aniline to pyrrole the interfacial polymerization And a and aniline and pyrrole were added the synthesis of MnO2@copolymer Figure 2 | characterization of atomically dispersed sites in hierarchically porous carbon framework. Mn K-edge spectra and spectra from the LA-2MnNC catalyst and standard and MnO2 reference samples. and Raman spectra and N2 for different samples as The optimized of nitrogen adsorbed on the surface of the (1) (2) N-doped and (3) respectively. million charge of Mn and in (1) and (2) after nitrogen adsorption. Download figure Download PowerPoint and present the spectra of 1s and Mn The peak at from the 1s spectra can be to the in the and its content as high as in the LA-2MnNC catalyst ( Supporting Information S1). In addition, no metallic Mn peak was observed for catalysts with the above that atomic of Mn sites. the with Mn content be to charge transfer between Mn sites and electronic was performed to better Mn concentration as a of the polymer and the results are in Supporting Information The of pyrrole enhanced Mn to the formation of more polymer and the Mn content was identified as wt % atom %) in the LA-2MnNC catalyst. The specific surface and of the catalysts with A very high specific surface of with abundant was measured ( Supporting Information Figure the same time, limited surface and were with the metal-free N-doped carbon A increase of surface and favorable especially in was observed with the of aniline and pyrrole This is of the level of aniline and pyrrole the formation of crosslinked and open which the of the The hierarchically porous hollow structural were favorable for nitrogen and Raman spectra of catalysts with different of are recorded and compared in Figure Compared LA-2MnNC catalysts with its counterparts, the lower value and of peak confirmed the of and lower ( Supporting Information Figure which are for nitrogen of nitrogen was performed to N2 behavior A between chemisorption and sites a content was and samples, the surface areas are and g−1, respectively. However, the catalyst exhibited chemisorption to its lower content in the carbon framework vs atom the as the active to dissolved nitrogen for subsequent Computational results in Figure the of the nitrogen of catalyst. As the key step in the electrochemical nitrogen reduction the bond length and lower energy the strong between nitrogen and hotspots. The favorable Lewis acid–base charge in the charge adsorbed reactant and nitrogen and Supporting Information Figure The and chemical nitrogen in the porous catalyst be more for subsequent the energy of carbon nanotube was that of strong of N2 with the carbon the LA-2MnNC was identified as the promising for nitrogen activation and conversion. It was that N2 chemisorption could be enhanced to abundant Lewis acid sites which is the step for electrochemical Further, hierarchically porous graphitic structure is for efficient reactant Electrochemical nitrogen reduction of catalysts NRR was performed in the M HCl solution using a typical