Ammonia Synthesis from a Pincer Ruthenium Nitride via Metal–Ligand

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Communication pubs.acs.org/JACS

Ammonia Synthesis from a Pincer Ruthenium Nitride via Metal−Ligand Cooperative Proton-Coupled Electron Transfer Brian M. Lindley,† Quinton J. Bruch,† Peter S. White,† Faraj Hasanayn,*,‡ and Alexander J. M. Miller*,† †

Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States Department of Chemistry, American University of Beirut, Beirut 1107 2020, Lebanon

‡

S Supporting Information *

In the facile ammonia synthesis described here, addition of acids to the nitrido complex (PNP)RuN (1; PNP = [N(CH2CH2PtBu2)2]−)25 triggers a new metal−ligand cooperative multiproton/multielectron PCET reaction. After a separate step to protonate the pincer amide, nitride-to-ammine conversion is achieved in as little as 15 min at room temperature upon addition of weak “ligating” acids that feature Lewis basic sites in the conjugate acid and can chelate as the conjugate base. The reaction proceeds without an external reductant, via a distinct mechanism involving 2H+/2e− PCET from the saturated ligand backbone. The reactivity of Ru nitride 1, a complex reported by Schneider in 2011,25 with reductants and acids was assessed initially. Cyclic voltammetry (CV) of 1 in thf showed irreversible reduction features, Ep,c = −3.2 and −3.5 V vs Fc+/0. These ill-defined reductions occur at extremely negative potentials. Treating 1 with 2,6-lutidinium tetrafluoroborate ([HLut][BF4], pKa ∼ 7.2 in thf)28 results in selective protonation at the pincer amide, producing the cationic nitride complex 2 (eq 1). The structure of 2 was characterized by X-ray

ABSTRACT: The conversion of metal nitride complexes to ammonia may be essential to dinitrogen fixation. We report a new reduction pathway that utilizes ligating acids and metal−ligand cooperation to effect this conversion without external reductants. Weak acids such as 4methoxybenzoic acid and 2-pyridone react with nitride complex [(H-PNP)RuN]+ (H-PNP = HN(CH2CH2PtBu2)2) to generate octahedral ammine complexes that are κ2-chelated by the conjugate base. Experimental and computational mechanistic studies reveal the important role of Lewis basic sites proximal to the acidic proton in facilitating protonation of the nitride. The subsequent reduction to ammonia is enabled by intramolecular 2H+/2e− proton-coupled electron transfer from the saturated pincer ligand backbone.

T

he synthesis of ammonia from its elements underpins global agriculture, but carries steep economic and environmental costs.1 Inspired by the mild conditions of natural NH3 synthesis,2 chemists have long sought to develop sustainable systems for N2 reduction under ambient conditions.3 The first molecular catalysts for N2 reduction to NH3 were Mo complexes;4,5 late transition metal (Fe and Co) catalysts have emerged in a surge of recent discoveries.6−10 Nitrides are widely proposed as N2 reduction intermediates, and some nitride complexes are competent molecular catalysts.11,12 The direct reductive cleavage of N2 to generate a metal nitride has also been demonstrated, as in Cummins’ seminal Mo nitride synthesis,13 Holland’s synthesis of a nitridebridged Fe cluster,14,15 and in Schrock’s and Schneider’s respective recent examples of N2 splitting by Mo and Re pincer complexes.16,17 Well-defined examples of efficient ammonia synthesis from nitrides are rare. Most nitride reduction mechanisms involve proton-coupled electron transfer (PCET), wherein proton and electron equivalents transfer in a stepwise or concerted fashion.18,19 Nitride reduction to NH3 has been accomplished through PCET from separate acids and reductants.15,20−22 Hydrogen atom donors such as TEMPOH, which typically provides H+/e− in a concerted fashion (as H·) can also convert nitrides to ammonia.15,23,24 Nitride hydrogenation has also been achieved using reagents such as H2 and silanes.25−27 The discovery of new mechanistic pathways that facilitate nitride-toammonia PCET transformations would bolster efforts in N2 reduction. © 2017 American Chemical Society

crystallography (Figure 1A). The distorted square planar geometry (τ4 = 0.26)29 features a relatively long Ru−N2 bond (2.302(4) Å) implicating a strong trans influence of the nitride ligand. Protonation at a site other than the nitride is common.16,17,30,31 In pincer examples, a Ru(IV) nitride undergoes selective protonation at the pincer amide,31 a Mo(IV) nitride is protonated at a phosphine donor of the pincer,16 and the Os analogue of 1 is protonated at the Os center to form a hydride.26 Coercing nitrides to react with acids to generate NH3 has been identified as a key challenge.3 We turned our attention to the reactivity of protonated nitride 2 with reductants and acids. CV of 2 in thf shows an irreversible reduction (Ep,c = −2.0 vs Fc+/0) that is anodically shifted by more than 1 V relative to 1 (Figure S13). We chose to explore the reactivity of 2 with benzoic acid derivatives (pKa Received: February 7, 2017 Published: April 6, 2017 5305

DOI: 10.1021/jacs.7b01323 J. Am. Chem. Soc. 2017, 139, 5305−5308

Communication

Journal of the American Chemical Society

cooperation most commonly involves either proton transfer or redox reactions.35−37 The present example of multiproton/ multielectron metal−ligand cooperation38 would offer a pathway that is complementary to other nitride reductions. The choice of acid is also surprisingly important: 2 is impervious to further reaction with the much stronger acid [HLut]+. To understand better the pathway from nitride to NH3, DFT and experimental mechanistic studies were conducted.39 A viable pathway for the transformation was calculated on the closed shell singlet spin-state potential energy surface (PES), starting with 2 and benzoic acid. Following formation of a loose adduct (2·BzOH, Figure 2), the entry step is addition of

Figure 1. Molecular structures of 2 (A) and 3 (B), with ellipsoids drawn at the 50% probability level. Counter ions, solvent molecules and most H atoms are omitted for clarity. Figure 2. Computed addition of benzoic acid to 2, with HOMOs depicted below. ΔG (kcal·mol−1) in brackets given relative to the separated reactants. Distances in Å. L = PtBu2.

∼ 25 in thf32), as these weak acids are not expected to undergo competing proton reduction to H2 at the potentials required to reduce 2.33 4-Methoxybenzoic acid (4-OMe-BzOH) was selected for initial reactions on the basis of NMR features that are easy to follow and the expectation of good solubility. Addition of 1 equiv 4-OMe-BzOH to a thf solution of 2, in the absence of an external reductant, led to a new product isolated in 97% yield after 15 h (eq 2). Using 50 equiv acid, the

H−OBz across the Ru nitride to form a κ1-benzoate imido intermediate (i1). This step is mediated by a pericyclic transition state (TS1) featuring concomitant benzoate coordination to Ru and proton delivery to the nitride. A significant electronic reorganization occurs moving from 2 to i1. Nitride π-donation into Ru dxy and dxz orbitals imposes a (dyz)2(dz2)2 configuration on the d4-Ru(IV) center in 2 and 2· BzOH with predominantly dz2 character in the HOMO (Figure 2). The electronic structure changes dramatically along the intrinsic reaction coordinate of TS1 such that the HOMO of i1 is largely d xz in character and an empty d z 2 orbital accommodates benzoate coordination. Analysis of the PES defined by the N−Ru−N angle (Figure S27) reveals a low barrier pathway for isomerization of i1 into i2, which has a 78° angle between the imido nitrogen and pincer amino nitrogen and a κ2-bound benzoate (Figure 3A). The amino proton in i2 is in ideal position to transfer to the imide. Nevertheless, the computed barrier for this step (via TS2) is substantial (17.2 kcal·mol−1 above i2). TS2 initially affords high-energy intermediate i3a, which features an NH2 ligand aligned with the xz plane. Effectively barrierless 90° rotation of the NH2 group of i3a leads to another intermediate (i3b in Figure 3B) that is a full 24 kcal·mol−1 more stable than i3a. This dramatic relaxation energy can be attributed to (i) removal of an antibonding interaction between the filled pz orbital on the pincer nitrogen and the filled Ru dxz orbital in i3a, and (ii) electronic

reaction was complete within 15 min at 25 °C. Formation of an ammine complex is evidenced by a 1H−15N HSQC NMR experiment featuring a one-bond N−H correlation between a broad singlet 1H resonance (δ1H = 1.57, 3H) and 15N resonance (δ15N = −64.23). The ammine product contains an asymmetric pincer backbone with a diagnostic downfield doublet (δ1H = 7.98, 3JHP = 23 Hz), characteristic of the imine form of the pincer ligand.34 X-ray crystallography verified the formation of [(PNCP)Ru(OBz-4-OMe)(NH3)]+ (3), with an NH3 ligand and a κ2-benzoate completing a distorted octahedral geometry (Figure 1B). The overall transformation of eq 2 is a 3H+/2e− reduction of the Ru(IV) nitride fragment, with the saturated H-PNP pincer ligand providing two protons and two electrons. Metal−ligand 5306

DOI: 10.1021/jacs.7b01323 J. Am. Chem. Soc. 2017, 139, 5305−5308

Communication

Journal of the American Chemical Society

The reactivity of 2 with several other acids was examined. Like benzoic acid, the substitution pattern of 2-pyridone (2hydroxypyridine, Figure 4) would be appropriate for pericyclic

Figure 4. Structures of pyridone isomers (A) and calculated transition state of 2-pyridone addition to 2 (B).

proton delivery to the nitride of 2 and subsequent κ2-chelation. The reaction of 2 with 3 equiv 2-pyridone in thf-d8 at 25 °C reached 85% conversion after 15 h (50% yield of an ammine complex by NMR spectroscopy). A preparative scale reaction stirred for 1 week produced the ammine complex [(PNCP)Ru(2-pyridonate)(NH3)]+ (4) in 65% isolated yield. The reaction of 2 with 3 equiv of isomer 4-pyridone in thf-d8 at 25 °C reached only 18% conversion after 15 h, forming an unidentified product along with intractable decomposition. Although 4-pyridone is a stronger acid than 2-pyridone (pKa ∼ 27 vs 29 in thf),43,44 it lacks a donor atom proximate to the acidic proton. The computed energy profile of the reaction of 2 with 2-pyridone is comparable to that of benzoic acid discussed above, while the reaction with 4-pyridone encounters barriers that are ∼10 kcal·mol−1 higher in energy (see SI). Treating 2 with 10 equiv [HLut][BF4] led to essentially no observed reaction over 24 h at 25 °C. When 10 equiv of the less sterically bulky acid [HPy][BF4] was utilized, 7% yield of NH4+ was observed by 1H NMR spectroscopy after 30 h at 25 °C. The reaction of 2 with 10 equiv [HPy][OTf], which features a more coordinating counterion, gave higher yields of NH4+ (61% after 16 h at 25 °C), albeit along with a multitude of unknown Ru-containing products. Taken together, these experiments suggest that ligating acids better support clean conversion of nitrides to ammine complexes. In brief, a new metal−ligand cooperative PCET pathway is reported, wherein the saturated pincer ligand provides 2H+/2e− to the Ru−N fragment. The mechanistic themes that emerge are of interest from fundamental perspectives, and could enable new strategies in N2 fixation and other applications in electrocatalysis.

Figure 3. Computed proton transfer (A) and PCET (B) processes and simplified MOs for i3a to i3b isomerization (C). ΔG (kcal·mol−1) given relative to the separated benzoic acid and 2. Distances in Å. L = PtBu2.

reorganization that empties the dxz orbital and allows delocalized bonding with two coplanar filled nitrogen p orbitals in i3b (Figure 3C).40,41 The transformation of i3b into the final ammine product 3 is computed to proceed via TS3 (Figure 3B). Characterized by a puckered five-membered ring that enables transfer of a pincer CH proton to RuNH2, TS3 also involves formation of a CN π-bond in the pincer backbone. This final step is a concerted 1H+/2e− metal−ligand cooperative PCET process that results in formal 2e− reduction of the Ru(IV) amide by the ligand to form the Ru(II) ammine 3. Equation 2 is computed to be highly exergonic (ΔG° = −29.3 kcal·mol−1) and free of any prohibitively high barriers. The facile reaction is attributed to the favorable geometry of benzoic acid for protonation of the nitride and κ2-binding to Ru, the ability of the d4 metal to modulate its electronic structure, and the role of the pincer ligand as both two electron reductant and proton shuttle in the cooperative PCET process. The present mechanism provides an interesting contrast to the hydrogenation of neutral Ru nitride 1 with H2 at 50 °C, which forms a Ru polyhydride and free NH3 over 48 h.25 Metal− ligand cooperation is also proposed to play a role in hydrogenation, with the pincer ligand acting as a base in heterolytic H2 cleavage. Intramolecular proton transfer from the amine of the H-PNP ligand to a Ru(IV) imido group is calculated to be the ratedetermining step (TS2 of 25.9 kcal·mol−1 relative to the separated reactants), much higher in energy than TS1 (18.2 kcal·mol−1) or TS3 (3.9 kcal·mol−1). If intramolecular proton transfer via TS2 is the rate-determining process, a primary kinetic isotope effect (KIE) would be expected upon deuteration of the pincer amine. In parallel experiments, the reaction of 2 with 4-OMe-BzOH and the reaction of [(DPNP)RuN]+ with 4-OMe-BzOD were monitored by NMR spectroscopy. Analysis of the initial rates provided a KIE of 4.6 (see SI), consistent with our predictions. The corresponding calculated KIE (composite for TS1 and TS2) is 8.2 (using unscaled harmonic frequencies).42

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b01323. Experimental and computational details (PDF) Computed geometries (TXT) Crystallographic geometries (CIF)

■

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *[email protected] ORCID

Alexander J. M. Miller: 0000-0001-9390-3951 Notes

The authors declare no competing financial interest. 5307

DOI: 10.1021/jacs.7b01323 J. Am. Chem. Soc. 2017, 139, 5305−5308

Communication

Journal of the American Chemical Society

■

(27) Yi, X.-Y.; Ng, H.-Y.; Williams, I. D.; Leung, W.-H. Inorg. Chem. 2011, 50, 1161−1163. (28) Rodima, T.; Kaljurand, I.; Pihl, A.; Mäemets, V.; Leito, I.; Koppel, I. A. J. Org. Chem. 2002, 67, 1873−1881. (29) Yang, L.; Powell, D. R.; Houser, R. P. Dalton Trans. 2007, 955− 964. (30) Odom, A. L.; Cummins, C. C. Polyhedron 1998, 17, 675−688. (31) Walstrom, A.; Fan, H.; Pink, M.; Caulton, K. G. Inorg. Chim. Acta 2010, 363, 633−636. (32) Barbosa, J.; Barrón, D.; Bosch, E.; Rosés, M. Anal. Chim. Acta 1992, 265, 157−165. (33) McCarthy, B. D.; Martin, D. J.; Rountree, E. S.; Ullman, A. C.; Dempsey, J. L. Inorg. Chem. 2014, 53, 8350−8361. (34) Friedrich, A.; Drees, M.; Käss, M.; Herdtweck, E.; Schneider, S. Inorg. Chem. 2010, 49, 5482−5494. (35) Khusnutdinova, J. R.; Milstein, D. Angew. Chem., Int. Ed. 2015, 54, 12236−12273. (36) DuBois, D. L. Inorg. Chem. 2014, 53, 3935−3960. (37) Chirik, P. J.; Wieghardt, K. Science 2010, 327, 794. (38) Henthorn, J. T.; Lin, S.; Agapie, T. J. Am. Chem. Soc. 2015, 137, 1458−1464. (39) The computed results in the figures were obtained at the M06L//cc-pVTZ basis set in a THF continuum. See SI for details and additional computational results. (40) Li, L.; White, P. S.; Brookhart, M.; Templeton, J. L. J. Am. Chem. Soc. 1990, 112, 8190−8192. (41) Powell, K. R.; Perez, P. J.; Luan, L.; Feng, S. G.; White, P. S.; Brookhart, M.; Templeton, J. L. Organometallics 1994, 13, 1851−1864. (42) The individual OH/OD and pincer NH/ND KIEs could not be probed experimentally because of rapid H/D scrambling between the acid and the pincer amine of 2. (43) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456−463. (44) Ding, F.; Smith, J. M.; Wang, H. J. Org. Chem. 2009, 74, 2679− 2691.

ACKNOWLEDGMENTS B.M.L. and A.J.M.M. gratefully acknowledge support from the NSF Center for Enabling New Technologies through Catalysis (CENTC), CHE-1205189. Q.J.B. acknowledges support from the University of North Carolina at Chapel Hill (Morrison Research Fellowship) and NSF Graduate Research Fellowship Program (DGE-1650116). F.H. thanks AUB for supporting a research leave at Rutgers University. A.J.M.M. is an Alfred P. Sloan Fellow. The authors thank Alan Goldman, Patrick Holland, James Mayer, Karsten Krogh-Jespersen, and Sven Schneider for helpful discussions.

■

REFERENCES

(1) Appl, M. In Ullmann’s Encyclopedia of Industrial Chemistry; WileyVCH: Weinheim, Germany, 2006; pp 1−155. (2) Hoffman, B. M.; Lukoyanov, D.; Yang, Z.-Y.; Dean, D. R.; Seefeldt, L. C. Chem. Rev. 2014, 114, 4041−4062. (3) Shaver, M. P.; Fryzuk, M. D. Adv. Synth. Catal. 2003, 345, 1061− 1076. (4) Yandulov, D. V.; Schrock, R. R. Science 2003, 301, 76−78. (5) Tanaka, H.; Nishibayashi, Y.; Yoshizawa, K. Acc. Chem. Res. 2016, 49, 987−995. (6) Anderson, J. S.; Rittle, J.; Peters, J. C. Nature 2013, 501, 84−87. (7) Rittle, J.; Peters, J. C. J. Am. Chem. Soc. 2016, 138, 4243−4248. (8) Del Castillo, T. J.; Thompson, N. B.; Peters, J. C. J. Am. Chem. Soc. 2016, 138, 5341−5350. (9) Kuriyama, S.; Arashiba, K.; Tanaka, H.; Matsuo, Y.; Nakajima, K.; Yoshizawa, K.; Nishibayashi, Y. Angew. Chem., Int. Ed. 2016, 55, 14291−14295. (10) Kuriyama, S.; Arashiba, K.; Nakajima, K.; Matsuo, Y.; Tanaka, H.; Ishii, K.; Yoshizawa, K.; Nishibayashi, Y. Nat. Commun. 2016, 7, 12181. (11) Ritleng, V.; Yandulov, D. V.; Weare, W. W.; Schrock, R. R.; Hock, A. S.; Davis, W. M. J. Am. Chem. Soc. 2004, 126, 6150−6163. (12) Arashiba, K.; Kinoshita, E.; Kuriyama, S.; Eizawa, A.; Nakajima, K.; Tanaka, H.; Yoshizawa, K.; Nishibayashi, Y. J. Am. Chem. Soc. 2015, 137, 5666−5669. (13) Laplaza, C. E.; Cummins, C. C. Science 1995, 268, 861−863. (14) Rodriguez, M. M.; Bill, E.; Brennessel, W. W.; Holland, P. L. Science 2011, 334, 780−783. (15) MacLeod, K. C.; McWilliams, S. F.; Mercado, B. Q.; Holland, P. L. Chem. Sci. 2016, 7, 5736−5746. (16) Hebden, T. J.; Schrock, R. R.; Takase, M. K.; Muller, P. Chem. Commun. 2012, 48, 1851−1853. (17) Klopsch, I.; Finger, M.; Würtele, C.; Milde, B.; Werz, D. B.; Schneider, S. J. Am. Chem. Soc. 2014, 136, 6881−6883. (18) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.; Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.; Meyer, T. J. Chem. Rev. 2012, 112, 4016−4093. (19) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110, 6961−7001. (20) Pipes, D. W.; Bakir, M.; Vitols, S. E.; Hodgson, D. J.; Meyer, T. J. J. Am. Chem. Soc. 1990, 112, 5507−5514. (21) Yandulov, D. V.; Schrock, R. R. J. Am. Chem. Soc. 2002, 124, 6252−6253. (22) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252− 6254. (23) Scepaniak, J. J.; Young, J. A.; Bontchev, R. P.; Smith, J. M. Angew. Chem., Int. Ed. 2009, 48, 3158−3160. (24) Scheibel, M. G.; Abbenseth, J.; Kinauer, M.; Heinemann, F. W.; Würtele, C.; de Bruin, B.; Schneider, S. Inorg. Chem. 2015, 54, 9290− 9302. (25) Askevold, B.; Nieto, J. T.; Tussupbayev, S.; Diefenbach, M.; Herdtweck, E.; Holthausen, M. C.; Schneider, S. Nat. Chem. 2011, 3, 532−537. (26) Schendzielorz, F. S.; Finger, M.; Volkmann, C.; Würtele, C.; Schneider, S. Angew. Chem., Int. Ed. 2016, 55, 11417−11420. 5308

DOI: 10.1021/jacs.7b01323 J. Am. Chem. Soc. 2017, 139, 5305−5308