Mechanistic Insights into Electrochemical Nitrogen Reduction

An ammonia production rate of 1.1 × 10−10 mol s−1 cm−2 can be main- tained for ... In this work, we demonstrate that vanadium nitride (VN) ... ...
0 downloads 0 Views 785KB Size
Download PDF
Subscriber access provided by Duquesne University - Gumberg Library

Article

Mechanistic Insights into Electrochemical Nitrogen Reduction Reaction on Vanadium Nitride Nanoparticles Xuan Yang, Jared Nash, Jacob Anibal, Marco Dunwell, Shyam Kattel, Eli Stavitski, Klaus Attenkofer, Jingguang G Chen, Yushan Yan, and Bingjun Xu J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b08379 • Publication Date (Web): 23 Sep 2018 Downloaded from http://pubs.acs.org on September 23, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Mechanistic Insights into Electrochemical Nitrogen Reduction Reaction on Vanadium Nitride Nanoparticles Xuan Yang,†,‡ Jared Nash,†,‡ Jacob Anibal,† Marco Dunwell,† Shyam Kattel,§,# Eli Stavitski,¶ Klaus Attenkofer,¶ Jingguang G. Chen,§,#,* Yushan Yan,†,* and Bingjun Xu†,* †

Center for Catalytic Science and Technology, Department of Chemical and Biomolecular Engineering, University of Delaware, 150 Academy Street, Newark DE, 19716 § Chemistry Division, Brookhaven National Laboratory, Upton, NY 11973, USA # Department of Chemical Engineering, Columbia University, New York, NY 10027, USA. ¶ National Synchrotron Light Source II, Brookhaven National Laboratory, Upton, New York 11973, USA ‡ These authors contributed equally to this work. *Address correspondence to [email protected], [email protected], and [email protected] KEYWORDS: Electrochemical nitrogen reduction, vanadium nitride, Mar-van Krevelen mechanism, operando XAS, 15N2

ABSTRACT: Renewable production of ammonia, a building block for most fertilizers, via the electrochemical nitrogen reduction reaction (ENRR) is desirable; however, selective electrocatalyst is lacking. Here we show that vanadium nitride (VN) nanoparticles are active, selective, and stable ENRR catalysts with an ENRR rate and Faradaic efficiency (FE) of 3.3 × 10−10 mol s−1 cm−2 and 6.0% at −0.1 V within 1 h, respectively. ENRR with 15N2 as the feed produce both 14NH3 and 15NH3, which indicates that the reaction follows a Mars-van Krevelen mechanism. Ex-situ X-ray photoelectron spectroscopy characterization of fresh and spent catalysts reveals that multiple vanadium oxide, oxynitride and nitride species are present on the surface and identified VN0.7O0.45 as the active phase in ENRR. Operando X-ray absorption spectroscopy and catalyst durability test results corroborate with this hypothesis, and indicate that the conversion of VN0.7O0.45 to the VN phase leads to catalyst deactivation. We hypothesize that only the surface N sites adjacent to a surface O are active in the ENRR. An ammonia production rate of 1.1 × 10−10 mol s−1 cm−2 can be maintained for 116 h, with a steady state turnover number of 431.

INTRODUCTION The ability to fixate nitrogen on an industrial scale marked a turning point in the history of the human society, which supports the livelihood of close to half of world’s population.1 The Haber-Bosch process operates at high temperatures (~700 K) and pressures (~100 atm) on optimized catalysts,2−7 which accounts for ~1% of the global energy consumption.2 Hydrogen consumed in the Haber-Bosch process derives exclusively from the steam reforming of fossil hydrocarbons, making it also a carbon intensive process.5 Further, the centralized ammonia production is a poor fit for the distributed nature of agriculture. Distributed and modular ammonia synthesis via the electrochemical nitrogen reduction reaction (ENRR) at or close to ambient conditions, powered by renewable electricity, emerges as an attractive alternative because it allows for the on-demand, on-site production of ammonia, and in turn Nfertilizers, from ubiquitously available resources, i.e., N2 and water.3 Widespread adoption of ENRR for ammonia production could drastically reduce the carbon footprint of agricultural activities. The ENRR is also compatible with the intermittency of renewable energy sources, e.g., solar and wind, as ammonia and N-fertilizers can be produced and stored when renewable electricity is abundant or even in surplus. Electrochemical fixation of atmospheric nitrogen was first attempted in 1908, before the establishment of the Haber-

Bosch process, to make nitric acid via electric discharge.8 Further studies of ENRR have generally focused on high temperature proton conductors at 773 K; however, the instability of ammonia at high operating temperature makes it unsuitable for distributed deployment.9 More recently, there has been a surge of interest in developing low temperature ENRR processes. Proton exchange membranes (PEMs) can achieve high proton conductivity (~100 mS⋅cm−1 at 25 °C for Nafion) at ambient or modestly elevated temperatures, which opens the possibility for electrochemical ammonia synthesis at those mild conditions.10 Due to the acidic nature of PEMs, platinum group metal (PGM) catalysts are the most commonly used catalysts because of their excellent stability in acidic environments.11−15 Our recent work showed that PGM catalysts generally have low Faradaic efficiencies (FE 50% (Figure S4-1−S4-5), which are likely reduced by either the negative electrode potential or the produced H2 via HER. The intensity of the peak corresponding to VNxOy also decreases after ENRR, indicating that it is also partially reduced. Interestingly, the ratio between VNxOy and VN, as quantified

by either the V 2p or the N 1s features, decreases from ~0.91 before ENRR to 0.54 after ENRR at potentials ≤ −0.2 V (Figure 3C), while it is 0.77 after 1 h of ENRR at −0.1 V. Control experiments show that V2O3 is not active for ENRR, and V2O5 is unstable at any negative potentials (Table S6-1).36 Therefore, it is reasonable to conclude that the active phase in ENRR is either VN or VNxOy. Since the fraction of V in the VN phase increases after ENRR, enhanced ENRR rates would be expected if VN is the active phase for the reaction. The opposite is observed in the reactivity studies. Thus, we propose that VNxOy is the most likely active phase for ENRR. The constant VNxOy/VN ratio in the spent catalysts after 1 h of ENRR at ≤ −0.2 V correlate well with the reactivity results that they are nearly or completely inactive for ENRR. It is likely that the exposed VNxOy is entirely converted to VN during ENRR, causing the catalyst deactivation. Our hypothesis is also consistent with the observation that only about half of exposed VNxOy was converted during 1 h of ENRR at −0.1 V, leading to its slower deactivation. Based on the ratio of V 2p and N 1s bands (corrected with atomic sensitivity factors) assigned to the VNxOy phase, an x value of 0.7 could be obtained. If a +3 oxidation state of V in VNxOy is assumed, its composition is calculated to be VN0.7O0.45.

Figure 3. XPS spectra of VN catalysts after nitrogen reduction at −0.1 V for 1 h. (A) XPS deconvolutions of the V 2p region. The five peaks in the V 2p region are attributed to VN, VNxOy, VOx, V2O5, and oxygen related species from lower to higher binding energy, respectively. (B) XPS deconvolutions of the N 1s region. The four peaks in the N 1s region are attributed to VNxOy, VN, satellite feature, and ammonia from lower to higher binding energy, respectively. (C) The ratio of VNxOy to VN before and after nitrogen reduction at different potentials for 1 h.

Operando X-ray absorption spectroscopic (XAS) studies further support our hypothesis that VN0.7O0.45 is the active phase for ENRR. X-ray absorption near edge spectroscopic (XANES) results of the vanadium K-edge are collected at different potentials (Figures 4A−C and S5-1). The position and intensity of the white line at all potentials investigated remain constant throughout the experiment, indicating that no detectable level of V leaching occurs during ENRR. The Fourier transform of the extended X-ray absorption fine structure (FT-EXAFS) spectra of the catalysts (Figure S5-2) show two peaks at 1.6 and 2.4 Å, which correspond to the V−N and V−V bonds in cubic VN, respectively. The pronounced amplitude of the V−V shells at higher distances (2−7 Å) confirms the well-defined long-range order of the VN catalysts.37 The XANES and FT-EXAFS results are consistent with the lack of change in the XRD patterns of the fresh and spent catalysts (Figure S1-7). The characteristic pre-edge peak at 5468.4 eV indicates the presence of oxynitride species,37,38 in good agreement with the XPS results. Importantly, the intensity of the pre-edge peak decreases with time during ENRR at all potentials studied, suggesting that the VN0.7O0.45 is gradually

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

converted during ENRR, and that the rate of conversion increases as the potential becomes more negative (Figure 4D). In order to analyze and interpret the pre-edge features, the background of the edge under the pre-edge features needs to be removed (Figure 4B, left panel), which is achieved by fitting a polynomial that smoothly reproduces the slope of the data prior to the pre-edge and immediately following the pre-edge peaks.38 35.7% of VN0.7O0.45 is converted at −0.1 V within 2 h, while 57.8% conversion is achieved in 1 h at −0.2 V (Figure 4D). The fractions of VN0.7O0.45 that survives after 1 h of ENRR at ≤ −0.2 V are similar, suggesting that all accessible VN0.7O0.45 (~58%) has been converted to VN. This is consistent with our DFT calculations showing that oxygen vacancies on the VN0.75O0.25 (111) surface bind with N much more strongly (by more than 2 eV) than nitrogen vacancies (Table S7-1). The XAS results corroborate with the XPS and the reactivity results, e.g., the slow conversion of VN0.7O0.45 and slow deactivation at −0.1 V. Based on the reactivity and characterization results, the turnover number (TON) per active N site, i.e., N sites on the accessible VN0.7O0.45 (57.8% of total), could be estimated. The amount of VN0.7O0.45 in the catalyst is determined to be 3.5 µmol per MEA (with a total catalyst loading of 2.5 mg) based on the elemental analysis. Together with the total amount of produced ammonia via ENRR (253.1 µmol −0.1 V), an average TON of ~179 could be obtained at −0.1 V within 120 h. Procedures for the TON calculation are detailed in the Supplementary Information (Section 8). Because there is substantial catalyst deactivation within the first 4 h, the TON for active sites at steady state (5−120 h) is estimated to be 431. It is worth noting the total amount of nitrogen in the ammonia produced within 120 h is 9.1 times that in the VN catalyst loaded on the MEA. Furthermore, the total amount of produced ammonia in this work is more than one order of magnitude higher than those in previous reports (Table S6-2). Combined with the result from the 15N2 experiment (Figure 2D), we conclusively show that the produced ammonia originates from N2 activation, rather than N leaching from VN.

show the pre-edge and white line peaks at −0.1 V as a function of time, respectively. The color scheme used in (A) applies to (B). Inset in (C) is the corresponding pre-edge peak. (D) Timedependent pre-edge area at different potentials.

Our results suggest that ENRR on VN proceeds via a Marsvan Krevelen mechanism (Figure 5, right circle): an ammonia molecule is formed by extracting a surface N atom with adsorbed hydrogen atoms and leaves behind a N vacancy, and the catalyst is regenerated by activating N2 and thus healing the vacancy. This mechanistic hypothesis is supported by the observation of a mixture of 14NH4+ and 15NH4+ with using 15N2 as the feed in ENRR (Figure 2D), which indicates that surface N is involved in the activation of N2, as suggested by a few computational studies.23−26 We hypothesize that only surface N that are adjacent to a surface oxygen in the VN0.7O0.45 phase are active in facilitating the catalytic turnover, as catalyst deactivation is accompanied by the accumulation of the VN phase. The surface oxygen in the VN0.7O0.45 phase is unstable at the reducing environment for ENRR, with more negative potentials leading to faster removal of surface oxygen atoms and generation of surface vacancies. Since the calculated binding energy of N on a vacancy is much higher than that of O, vacancies will be preferentially filled with N, leading to the inactive VN phase (Figure 5, left circle). Thus, increasing the density and stability of the surface oxygen of the VN0.7O0.45 phase under ENRR conditions, e.g., via introducing dopants, could be effective strategies to improve the activity and durability of the VN catalyst.

Figure 5. Proposed reaction pathway for nitrogen reduction on the surface of VN0.7O0.45 via a Mars-van Krevelen mechanism and the catalyst deactivation mechanism.

Figure 4. Operando XAS results of VN catalysts: K-edge XANES spectra of VN catalysts at different potentials (A) −0.1 V and (C) −0.2 V as a function of time. (B) Left and Right panels

CONCLUSIONS To summarize, we demonstrate that VN nanoparticles are active and selective ENRR catalysts with an ammonia production rate and FE of 3.3 × 10−10 mol s−1 cm−2 and 6.0% at −0.1 V within 1 h, respectively. A steady state ammonia production rate of 1.1 × 10−10 mol s−1 cm−2, with an FE of 1.6%, can be maintained for 116 h. The ENRR with 15N2 as the feed produces a mixture of 14NH3 and 15NH3, which proves the involvement of surface N in the catalytic turnovers. This provides concrete evidence that the reaction proceeds via the Mars-van Krevelen mechanism. Combined ex-situ and operando characterization results indicate that the surface VN0.7O0.45 species is the active phase in ENRR, and the conversion of this phase to VN is proposed as the deactivation pathway. We hypothesize that only the surface N sites adjacent to a surface O are active in the ENRR, which leads to a steady state turnover number of 431.

ACS Paragon Plus Environment

Page 4 of 6

Page 5 of 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

ASSOCIATED CONTENT Supporting Information. Experiment methods, quantification of ammonia, Nitrogen reduction in batch mode under 15N2, deconvolution of XPS spectra, operando XAS characterizations of the catalysts, production rate and FE of other ENRR catalysts, computational methods, and procedure for turnover number calculations are included in the supporting information. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Authors * Address correspondence to [email protected], [email protected], and [email protected]

Author Contributions ‡

These authors contributed equally.

ACKNOWLEDGMENT This work is supported by the US Department of Energy under the grant number of DE-SC0016537. The DFT work was supported by the US Department of Energy under the grant number of DEFG02-13ER16381.

REFERENCES 1.

Erisman, J. W.; Sutton, M. A.; Galloway, J.; Klimont, Z.; Winiwarter, W. Nat. Geosci. 2008, 1, 636−639. 2. Appl, M. Ullmann’s Encyclopedia of Industiral Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, 2000. 3. McEnaney, J. M.; Singh, A. R.; Schwalbe, J. A.; Kibsgaard, J.; Lin, J. C.; Cargnello, M.; Jaramillo, T. F.; Nøskov, J. K. Energy Environ. Sci. 2017, 10, 1621−1630. 4. Industrial Efficiency Technology Database, http://ietd.iipnetwork.org/content/ammonia, accessed September 2017. 5. Nøskov, J.; Chen, J. Sustainable Ammonia Synthesis, DOE Roundtable Report, U.S. Department of Energy, 2016. 6. Schlögl, R. Ammonia synthesis, Handbook of Heterogeneous Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, 2008. 7. Chen, J. G.; Crooks, R. M.; Seefeldt, L. C.; Bren, K. L.; Bullock, R. M.; Darensbourg, M. Y.; Holland, P. L.; Hoffman, B.; Janik, M .J.; Jones, A. K.; Kanatzidis, M. G.; King, P.; Lancaster, K. M.; Lymar, S. V.; Pfromm, P.; Schneider, W. F.; Schrock, R. R. Science 2018, 360, eaar6611. 8. Cramp, W.; Hoyle, B. Journal of the Institution of Electrical Engineers 1908, 42, 297−338. 9. Marnellos, G.; Stoukides, M. Science 1998, 282, 98−100. 10. Amar, I. A.; Lan, R.; Petit, C. T. G.; Tao, S. J. Solid State Electrochem. 2011, 15, 1845−1860. 11. Kordali, V.; Kyriacou, G.; Lambrou, C. Chem. Commun. 2000, 1673−1674. 12. Lan, R.; Irvine, J. T. S.; Tao, S. Sci. Rep. 2013, 3, 1145.

13. Nash, J.; Yang, X.; Anibal, J.; Wang, J.; Yan, Y.; Xu, B. J. Electrochem. Soc. 2017, 164, F1712−F1716. 14. Bao, D.; Zhang, Q.; Meng, F.-L.; Zhong, H.-X.; Shi, M.-M.; Zhang, Y.; Yan, J.-M.; Jiang, Q.; Zhang, X.-B. Adv. Mater. 2017, 29, 1604799. 15. Shi, M.-M.; Bao, D.; Wulan, B.-R.; Li, Y.-H.; Zhang, Y.-F.; Yan, J.-M.; Jiang, Q. Adv. Mater. 2017, 29, 1606550. 16. Rod, T. H.; Logadottir, A.; Nørskov, J. K. J. Chem. Phys. 2000, 112, 5343−5347. 17. Skúlason, E.; Bligaard, T.; Gudmundsdóttir, S.; Studt, F.; Rossmeisl, J.; Abild-Pedersen, F.; Vegge, T.; Jónsson, H.; Nørskov, J. K. Phys. Chem. Chem. Phys. 2012, 14, 1235−1245. 18. Montoya, J. H.; Tsai, C.; Vojvodic, A.; Nørskov, J. K. ChemSusChem 2015, 8, 2180−2186. 19. Wang, J.; Yu, L.; Hu, L.; Chen, G.; Xin, H.; Feng, X. Nat. Commun. 2018, 9, 1795. 20. Lee, H. K.; Koh, C. S. L.; Lee, Y. H.; Liu, C.; Phang, I. Y.; Han, X.; Tsung, C.-K.; Ling, X. Y. Sci. Adv. 2018, 4, eaar3208. 21. Nazemi, M.; Panikkanvalappil, S. R.; El-Sayed, M. A. Nano Energy 2018, 49, 316−323. 22. Yang, D.; Chen, T.; Wang, Z. J. Mater. Chem. A 2017, 5, 18967−18971. 23. Abghoui, Y.; Skúlasson, E. Procedia Comput. Sci. 2015, 51, 1897−1906. 24. Abghoui, Y.; Garden, A. L.; Howalt, J. G.; Vegge, T.; Skúlason, E. ACS Catal. 2016, 6, 635−646. 25. Abghoui, Y.; Skúlason, E. Catal. Today 2017, 286, 78−84. 26. Abghoui, Y.; Skúlason, E. J. Phys. Chem. C 2017, 121, 6141−6151. 27. Zhang, X.; Kong, R.-M.; Du, H.; Xia, L.; Qu, F. Chem. Commun. 2018, 54, 5323−5325. 28. Giordano, C.; Erpen, C.; Yao, W.; Milke. B.; Antonietti, M. Chem. Mater. 2009, 21, 5136−5144. 29. Zhao, D.; Cui, Z.; Wang, S.; Qin, J.; Cao, M. J. Mater. Chem. A 2016, 4, 7914−7923. 30. Silversmit, G.; Depla, D.; Poelman, H.; Marin, G. B.; De Gryse, R. J. Electron. Spectrosc. Relat. Phenom. 2004, 135, 167−175. 31. Glaser, A.; Surnev, S.; Netzer, F. P.; Fateh, N.; Fontalvo, G. A.; Mitterer, C. Surf. Sci. 2007, 601, 1153−1159. 32. Saha, N. C.; Tompkins, H. G. M. J. Appl. Phys. 1992, 72, 3072−3079 33. Wagner C. D.; Naumkin, A. V.; Kraut-Vass, A.; Allison, J. W.; Powell, C. J.; Rumble, J. R. Jr. NIST Standard Reference Database 20, Version 3.4 (web version) 2003. 34. Gondek, K.; Mierzwa-Hersztek, M. Soil Use Manage. 2016, 32, 357−367. 35. Gondek, K.; Mierzwa-Hersztek, M.; Kopeć, M. J. Environ. Manage. 2018, 210, 87−95. 36. Vanysek, P. Electrochemical Series, CRC Handbook of Chemistry and Physics, CRC Press, Boca Raton, 2002. 37. Nakhal, S.; Hermes, W.; Ressler, T.; Pöttgen, R.; Lerch, M. Z. Anorg. Allg. Chem. 2009, 635, 2016−2020. 38. Henderson, G. S.; de Groot, F. M. F.; Moulton, B. J. A. Rev. Mineral. Geochem. 2014, 78, 75−138.

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 6

TOC

ACS Paragon Plus Environment

6