Ammonia-borane dehydrogenation catalyzed by Iron pincer complexes: A concerted metal-ligand cooperation mechanism

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

Ammonia-borane dehydrogenation catalyzed by Iron pincer complexes: A concerted metal-ligand cooperation mechanism Yi Zhang, Ye Zhang, Zheng-Hang Qi, Yun Gao, Wei Liu**, Yong Wang* College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, PR China

article info

abstract

Article history:

A quantum-chemical mechanistic investigation using density functional theory (DFT) on

Received 7 April 2016

ammonia-borane dehydrogenation catalyzed by a series of iron bis(phosphinite) pincer

Received in revised form

complexes is reported. A metal-ligand cooperation mechanism has been proposed, in

12 July 2016

which the hydrogen atom of BeH moves to metal Fe and proton of NeH transfers to pincer

Accepted 23 July 2016

ipso carbon simultaneously with the lowest activation barriers. DFT calculations and nat-

Available online 20 August 2016

ural bond orbital (NBO) charge analysis suggest that FeePOCOP complex with an electrondonating MeO group at the para position to the ipso carbon exhibits the highest catalytic

Keywords: Ammonia-borane

activity. A plausible explanation of the observed catalytic activities is also given. © 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Dehydrogenation mechanism Density functional theory Metal-ligand cooperation Ligand-assisted

Introduction The ever-growing concerns on energy crisis have resulted in an urgent quest for the alternative renewable energy. Hydrogen is an ideal candidate as an energy carrier for fossil fuels if effective methods for its storage and release can be solved. Ammonia borane (NH3BH3, AB) is considered to be one of the most promising chemical hydrogen storage materials, which has received considerable attention in recent years due to the high gravimetric storage density (19.6 wt% of H2) and high thermal stability. However, one of the challenges in the development of the hydrogen economy is to enhance the release rate and production of hydrogen released from AB [1e7]. Searching for efficient transition metal catalysts for

reversible H2 release is still critical in dehydrogenation of AB. As a result, a myriad of catalysts for AB dehydrogenation have been investigated not only experimentally but also computationally in recent years [8e41]. Under reasonable conditions, transition metal based catalysts have provided the most potential for controlling both the rate and extent of H2 release yet reported for AB dehydrogenation. Unfortunately, many of these catalysts use either relatively expensive metals (such as Ru [13,17,21,22], Rh [8,23,24], Pd [25,26], Ir [11,27,28] and Os [29], etc.) or suffer from instability under the reaction conditions (Ni [30], Pd [25]). Therefore, developing less expensive and earth abundant metal catalysts (Ti [9,31,32], Fe [20,33e36], Mo [37]) for AB dehydrogenation have attracted growing interest.

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (W. Liu), [email protected] (Y. Wang). http://dx.doi.org/10.1016/j.ijhydene.2016.07.209 0360-3199/© 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

Recently, a number of mechanistic studies have been carried out to explore the transition-metal-catalyzed dehydrogenation reaction of AB. Theoretical investigations by Paul and Musgrave [12] revealed that dehydrogenation of ammonia-borane (AB) using the Ir-pincer catalyst [(POCOP) Ir(H)2] proceeds through a concerted NeH and BeH removal inner-sphere mechanism. Ohno and Luo [10] reported an intramolecular stepwise dehydrogenation process of Me2NH$BH3 catalyzed by Cp2Ti in which NeH activation precedes BeH activation. Weller and coworker [38] suggested that ammonia borane dehydrogenation promoted by a Rh complex occurs via a BeH activation followed by the NeH activation. Oxidative addition of the BeH bond and then NeH b-elimination has also been explored by Douglas [39]. In contrast, Yang and Hall [15] demonstrated that the Ni(NHC)2 catalyzed ammonia-borane dehydrogenation includes an initial NHCligand-assisted NeH proton transfer to ligated carbene carbon step. Fagnou's [13] bifunctional Ru catalysts undergo similar proton transfer to ligand nitrogen step, which they defined as ligand-assisted concerted dehydrogenation mechanism [18]. Meanwhile, our group investigated the catalytic AB dehydrogenation mechanism with Ir pincer complex [Ir(ItBu0 )2]þ [40] and Ni(NHC)2 catalyst [41], respectively, in which a proton transfer mechanism is also favored. A more recent work by Guan's group [36] independently developed a catalytic system using a series of FeePOCOP pincer complexes (Fig. 1) as catalytic precursors for the dehydrogenation of AB, which are best of few examples of first-row transition metal catalysts reported to date. This catalytic reaction can release 2.3e2.5 equiv. of H2 per AB. Among the three Iron catalysts, iron bis(phosphinite) complex M1 exhibits the highest activity in terms of both the rate and the extent of H2 release [36]. As one of our ongoing research interests, we carried out theoretical modeling to investigate the mechanisms of the AB dehydrogenation catalyzed by Guan groups' FeePOCOP pincer complexes (Cat-PMe, Cat-PPh, and M1 shown in Fig. 1). Herein, the most activated complex, a prototype Iron pincer catalyst, [(MeOePOCOPMe)Fe] (M1), has been selected for full theoretical studies of AB dehydrogenation mechanism, in which the iPr2 were replaced by methyl substituents to save computational resources. According to our calculations, the optimized catalytic cycle for catalytic AB dehydrogenation is shown in Fig. 2.

Computational details All calculations were conducted using density functional theory with the Gaussian09 suite of programs [42] at the M06 [43] level of theory. All main group elements were represented by means

17209

of the 6-31G(d) [44,45] basis set along with the quasirelativistic Stuttgart Dresden pseudo potential SDD [46] for iron. For each optimized stationary point vibrational analysis was performed to establish its nature as a minimum (no imaginary frequencies) or saddle point (only one imaginary frequency). In addition, intrinsic reaction coordinate (IRC) [47,48] analysis was carefully carried out to confirm whether it connected the correct configurations of reactant and product on the potential energy surface. The single-point energy calculations with solvent effects (THF) were performed through the SMD continuum model [49] at the M06/6-311 þ G (d,p) [50] level on gas phase-optimized configurations (Fig. S1). Energies reported in the text are based on the gas-phase free energies. Furthermore, for the sake of understanding the differences in catalytic activity of the three precatalysts, the natural bond orbital (NBO) [51] charge analysis was performed at the same level of theory. Optimized structures were visualized by the CYL view program [52].

Results and discussion Our starting point in the dehydrogenation of AB is the dissociation of phosphine ligand trans to the hydride, which provides a vacant coordination site for AB activation and forms AB-bound complex M2 easily (Fig. 2). In the catalytically active intermediate M2, the distance of FeeH bond is 1.76
A, BeH bond is 1.24
A, leading to a h1 sigma complex, involving a 3-center, 2-electron (3c-2e) Fe/H/B bridging bond (Fig. S2). However, the catalytically active species M2 can undergo a geometric isomerization to yield M2′, placing the AB molecule trans to the ipso carbon. DFT calculations show that M2′ is less stable than M2 by 7.0 kcal/mol (Fig. S1), which is in good agreement with the experimental observations. As shown in Fig. 3, the dehydrogenation of AB catalyzed by Iron pincer complex M1 may potentially occur through four possible reaction pathways, including no ligand-assisted (path 1 and path 2) and ligand-assisted (path 3 and path 4) AB dehydrogenation. At this point, the current catalytic reaction system is a good model to investigate competitions among various possible pathways in catalytic dehydrogenation of AB.

No ligand-assisted ammonia-borane dehydrogenation At the beginning of our theoretical investigation, driven by the previous theoretical studies unveiled by Paul and Musgrave, we firstly examined the concerted NeH and BeH transfer inner phase mechanism. As depicted in Fig. 3 (path 1), the NeH and Be H bonds of AB simultaneous undergo oxidative addition to Fe center via a five-membered transition state TS1. In TS1, the distances of FeeH1, BeH1, FeeH2, and NeH2 are 1.67
A, 1.30
A,

Fig. 1 e The increasing catalytic activities of FeePOCOP pincer complexes [36].

17210

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

Fig. 2 e Optimized catalytic cycle for the AB dehydrogenation catalyzed by M1.

Fig. 3 e Gibbs free energy profiles for the catalytic dehydrogenation pathways.
(Fig. 4), respectively, indicating robust 1.68
A, and 1.59 A Fe/H1eB and Fe/H2eN interactions. In the following step, the BeN bond is shortened from 1.62
A (M2) to 1.53
A (TS1), suggesting a double bond is forming. As a result, aminoborane (NH2BH2) is liberated from TS1 giving M7. During no ligandassisted concerted dehydrogenation path, transition state TS1 connects intermediates M2 and M7 with a large Gibbs activation barrier of 40.4 kcal/mol, indicating path 1 is unfavorable. Interestingly, a no ligand-assisted stepwise dehydrogenation pathway is also computationally investigated (Fig. 3). From the sigma complex intermediate M2, a proton detached from N transfers to Fe center through transition state TS2, which contains an NeHeFe three-center interaction (Fig. 4), whose Ne H activation barrier is 31.9 kcal/mol. After that, the formation of

a new FeeH bond (1.52
A) through H elimination of AB leads to the intermediate M3, which is much less stable than M1 by 27.8 kcal/mol (Fig. 3, path 2). Hence, considering the relative higher barrier and instability of intermediate M3, this pathway is also less favorable. For the stepwise dehydrogenation pathways, sequential BeH activation and then NeH activation was also considered. Unfortunately, no evidence of any accessible transition states was found. This was further corroborated by a relaxed potential energy surface scan (Fig. S4).

Ligand-assisted ammonia-borane dehydrogenation Taking into account the higher barriers (TS1 and TS2), we hopefully find other possible pathways to decrease the

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

17211


Fig. 4 e Optimized geometries of transition states, TS1 and TS2. Key bond lengths are given in A. activation barrier. The AB dehydrogenation is initiated from a proton on N of AB transfer to pincer ipso carbon bond to the Fe center. As depicted in Fig. 3, beginning from M2, the NeH proton is drawn towards the carbon atom bounded to the Fe center through transition state TS3 (Fig. 3, path 3), which corresponds to a relatively higher activation barrier of 32.1 kcal/mol. After TS3, M4 is generated via proton transfer with a new CeH bond formation, which is also much less stable than M2 by about 15.0 kcal/mol. Fortunately, a concerted FeePOCPO cooperation pathway was located (Fig. 3, path 4), in which the iron center and ipso carbon of the POCOP ligand cooperatively react with AB. In this case, a proton from N moves to the unsaturated carbon atom while hydride from B moves toward the iron center via six-membered transition state TS4 and forms relatively stable intermediate M5, compared to M3 and M4 (Fig. 3). Not surprisingly, the concerted FeeCipso cooperation pathway is the most kinetically favorable with a lowest activation barrier of 17.6 kcal/mol. As mentioned above, the facile activation of the metal-ipso carbon bond makes the ligand-assisted concerted step kinetically favorable. In TS4 (Fig. 5), the shortened Fe/HeB distance of 1.68
A and C/HeN contact of 1.20
A are found. During the concerted proton transfer step, the N/H distance is greatly elongated to 1.62
A from 1.02
A of M2,

indicating a strong CeHeN three center interaction by the ligand. In intermediate M5, the CipsoeH1 bond length is found to be 1.20
A, which is greater than the standard aromatic CeH bond length of 1.09
A. What's more, the hydrogen of the Cipsoe H1 bond is observed to lie out of the aromatic ring and the Fee H1eCipso angle is observed to be 92.16 , both falling in the range of typical agnostic interactions. The agostic interactions between the ipso carbon and the metal have already been observed and reported in several complexes containing an aromatic ligand [53e56]. After M5, the unsaturated aminoborane NH2BH2 is formed and then released, which leads to the formation of intermediate M6. After that, the hydrogen atom (H1) easily moves to Fe through the transition state TS5 with a negligible lower activation energy of 4.9 kcal/mol. Subsequently, a more stable intermediate M7 is located through TS5, where the H2 molecule coordinates to the Fe center in h2 fashion (Fig. 5). It should be noted that the transformation from M6 to M7 is predicted to be barrierless and exergonic by 7.6 kcal/mol. During the CipsoeH1/Fe actiA vation step, the CipsoeH1 bond is greatly elongated from 1.18
(M6) to 2.29
A (M7), while the H1eH2 bond is shortened from 1.70
A to 0.83
A, suggesting that the CipsoeH1 bond is broken and a new HeH bond is formed. Finally, after M7, the catalytically active species M2 regeneration occurs through H2


Fig. 5 e Optimized geometries of TS4, M5, TS5, and M7. Key bond lengths are given in A.

17212

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

release. As a consequence, a ligand-assisted concerted catalytic cycle for AB dehydrogenation by Iron pincer complex is located. In this pathway, the FeeCipso cooperation activation is the rate determining step (RDS) with a relatively lower barrier of 17.6 kcal/mol than those of other possibilities. Thus, the ligand-assisted concerted FeeCipso cooperation dehydrogenation pathway is not only kinetically but also thermodynamically most favorable, which is in good agreement with the experimental observations [36] and in accordance with the mechanism shown in Fig. 2.

Trend in catalytic activity As mentioned before, Guan et al. [36] found marked catalytic activity differences of the three FeePOCOP pincer complexes (Cat-PMe, Cat-PPh, and M1 shown in Fig. 1) in AB dehydrogenation. In order to further verify the catalytic activities of three FeePOCOP pincer complexes, DFT calculations were carefully employed to investigate the dehydrogenation intermediates and transition states of RDS steps (Fig. S4 and S5). From Fig. 3 and S5, the activation barrier of the ratedetermining steps takes the sequence of TS-PMe (24.5 kcal/ mol) > TS-PPh (22.4 kcal/mol) > TS4 (17.6 kcal/mol). In other words, the relative catalytic activity follows the order of CatPMe < Cat-PPh < M1, which is in good agreement with experimental observations [36]. As depicted in Fig. 6, the natural bond orbital (NBO) charge analysis of the three Fee POCOP pincer catalysts and activated complexes has also been carried out. Unexpectedly, the central Fe atom bears a negative charge (M2: 1.989, Int-(PPh-AB): 1.992, Int-(PMeAB): 2.005). Therefore, since the BH hydrogen atom that is transferred carries a partial negative charge, not surprisingly, a shift to Fe center of M2 should be more favorable during the dehydrogenation paths. Simultaneously, in

contrast to the geometries of Cat-PMe, the increasing steric bulk at the phosphorus center of Cat-PPh and M1 could possibly contribute to the rapid dissociation of the phosphine ligand. According to NBO charge analysis, it should be noted that the MeO group in M1 provides a slightly increase of electron density to the Fe center, from 2.479 (Cat-PPh) to 2.493 (M1), to facilitate ligand dissociation. Moreover, from FeePOCOP pincer complexes to activated intermediates, the negative charges on the ipso carbon of the three AB-bound are intermediates (M2, Int-(PPh-AB), Int-(PMe-AB)) increasing, while negative charges on the Fe center are decreasing, illuminating a more nucleophilic character of ipso carbon than that of Fe. Besides, negatively charged ipso carbon of M2 is greater than those of Int-(PMe-AB) and Int(PPh-AB). Since the NeH hydrogen atom carries a positive charge, a shift to the ipso carbon of M2 should be more favorable. The natural bond orbital (NBO) analysis of the transition state TS-PMe, TS-PPh, and TS4 has also been carried out at the M06/6-31G(d)þSDD level to analyze such stabilization effects on the activations. The selected donoreacceptor interactions in Table 1 are the most significant ones with appreciable stabilization energies, which can partly account for the differences in relative stabilities of TS. As expected, the NBO results show the relatively weak orbital interaction of 47.66 kcal/mol in TS-PMe between the lone pair orbital n on the nitrogen atom and the vacant antibonding s*CeH orbital of the new forming CeH bond (as depicted in Table 1) and the orbital interaction of 96.21 kcal/mol between the sBeH orbital of AB and the d orbital of Fe atom. Compared to TS-PMe, similar results were also found in TS-PPh, while in TS4 a strong orbital interaction of 191.97 kcal/mol was found between the sB-H orbital of AB and the d orbital of Fe atom. Therefore, the stabilization of TS have the trend of TS-PMe z TS-PPh < TS4,

Fig. 6 e NBO charge analysis of the three Iron pincer complexes and AB-bound intermediates.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

Table 1 e The stabilization energies (in units of kcal/mol) obtained by the NBO analysis of the transition states (TSPMe, TS-PPh and TS4) of the rate-determining step in the dehydrogenation of AB at the M06/6-31G*þSDD level of theory.

TS-PMe TS-PPh TS4

Interactions

Stabilization energy

sBeH/dFe nN/s*CeH sBeH/dFe nN/s*CeH sBeH/dFe nN/s*CeH

96.21 47.66 96.32 48.15 191.97 44.43

17213

the benzene group of the ligand, causing steric repulsion and destabilizing M2′, which makes it impossible to protonation of the ipso carbon and prevents the loss of H2. Interestingly, through calculations we located an NeH proton of AB protonation to the Fe center and then promote H2 release. However, according to the unstable intermediate M2′ and relatively higher activation barrier of 17.7 kcal/mol (TS-transAB, path a) than that of path 4 (TS4, 17.6 kcal/mol), the dehydrogenation process starting from M2′ is also impossible. In addition, we investigated the substitution of PMe2Ph in M2′ by a second AB to generate Int-2AB. As shown in Fig. 7, Int2AB is much less stable than M2 by 12.1 kcal/mol. Next, a similar ligand-assisted concerted FeeCipso cooperation pathway through TS-2AB as TS4 is located with the highest activation barrier of 25.0 kcal/mol (path b) than those of path 4 and path a. It shows that the catalyst deactivation from M2 to M2′ and Int-2AB is impossible to promote AB dehydrogenation. Notably, after M2, the BeN bond is possibly elongated from 1.62
A (M2) to 2.71
A (TS-AB-Dis) (Fig. S7) by overcoming a relatively larger barrier of 25.7 kcal/mol (Fig. 7) than that of path 4 (17.6 kcal/mol), and get unstable BH3-coordinated complex FeeBH3 via the loss of NH3, which is an endothermic process with about 18 kcal/mol. Considering the possible pathways leading to the deactivation of catalysts, the dissociation of NH3 moiety from the AB-bound complex may contribute to the decreased catalytic activity at the late stage of the dehydrogenation process.

which is the same as the above-mentioned sequence of the reactivity, Cat-PMe < Cat-PPh < M1.

Conclusions Catalyst deactivation pathways As expected, the three catalysts could potentially lead to a gradual loss of catalytic activity active species during reactions over time. Several pathways leading to the catalyst deactivation have also been carefully studied (Fig. 7). As depicted in Fig. 7, catalytic active species M2 could potentially undergo an endothermic isomerization to M2′ with 7.0 kcal/ mol free energy. In M2′, the NH3 group is positioned trans to

In summary, the AB dehydrogenation mechanisms for the newly developed Iron bis(phosphinite) pincer complexes have been elucidated. We have shown that AB dehydrogenation proceeds through a ligand-assisted concerted FeeCipso cooperation mechanism. The reaction barriers of the ratedetermining steps for the AB dehydrogenation catalyzed by the three FeePOCOP pincer complexes (Cat-PMe, Cat-PPh, and M1) are predicted to be 24.5, 22.4, and 17.6 kcal/mol, respectively. Thus, M1 exhibits the highest catalytic activity among

Fig. 7 e Gibbs free energy profiles for the catalyst deactivation pathways.

17214

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

the three FeePOCOP pincer catalysts, which is in good agreement with experimental observations. In addition, these mechanistic studies are of fundamental importance in light of the development of better-performing catalysts for an efficient AB dehydrogenation. Further investigations on theoretical exploration of AB dehydrogenation are still underway in our group.

Acknowledgments The authors thank the reviewers for their constructive and pertinent comments. The authors appreciate the financial support from Starting-up Foundation (Q410900111 and Q410900211) and Scientific Research Foundation of Soochow University (SDY2012A07), and Natural Science Foundation of China (21201127). This project was funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ijhydene.2016.07.209.

references

[1] Staubitz A, Robertson APM, Manners I. Ammonia-borane and related compounds as dihydrogen sources. Chem Rev 2010;110:4079e124. [2] Hamilton CW, Baker RT, Staubitz A, Manners I. BeN compounds for chemical hydrogen storage. Chem Soc Rev 2009;38:279e93. [3] Yang J, Sudik A, Wolverton C, Siegel DJ. High capacity hydrogen storage materials: attributes for automotive applications and techniques for materials discovery. Chem Soc Rev 2010;39:656e75. [4] Bockris JO’M. The origin of ideas on a hydrogen economy and its solution to the decay of the environment. Int J Hydrogen Energy 2002;27:731e40. [5] Zhu Q, Xu Q. Liquid organic and inorganic chemical hydrides for high-capacity hydrogen storage. Energy Environ Sci 2015;8:478e512. [6] Dresselhaus MS, Thomas IL. Alternative energy technologies. Nature 2001;414:332e7. [7] Huang Z, Autrey T. Boron-nitrogen-hydrogen (BNH) compounds: recent developments in hydrogen storage, applications in hydrogenation and catalysis, and new syntheses. Energy Environ Sci 2012;5:9257e68. [8] Jaska CA, Temple K, Lough AJ, Manners I. Transition metalcatalyzed formation of boron-nitrogen bonds: catalytic dehydrocoupling of amine-borane adducts to form aminoboranes and borazines. J Am Chem Soc 2003;125:9424e34. [9] Clark TJ, Russell CA, Manners I. Homogeneous, titanocenecatalyzed dehydrocoupling of amine-borane adducts. J Am Chem Soc 2006;128:9582e3. [10] Luo Y, Ohno K. Computational study of titanocene-catalyzed dehydrocoupling of the adduct Me2NH-BH3: an intramolecular, stepwise mechanism. Organometallics 2007;26:3597e600.

[11] Denney MC, Pons V, Hebden TJ, Heinekey DM, Goldberg KI. Efficient catalysis of ammonia borane dehydrogenation. J Am Chem Soc 2006;128:12048e9. [12] Paul A, Musgrave CB. Catalyzed dehydrogenation of ammonia-borane by iridium dihydrogen pincer complex differs from ethane dehydrogenation. Angew Chem Int Ed 2007;46:8153e6. [13] Blaquiere N, Diallo-Garcia S, Gorelsky SI, Black DA, Fagnou K. Ruthenium catalyzed dehydrogenation of ammonia boranes. J Am Chem Soc 2008;130:14034e5. [14] Keaton RJ, Blacquiere JM, Baker RT. Base metal catalyzed dehydrogenation of ammonia-borane for chemical hydrogen storage. J Am Chem Soc 2007;129:1844e5. [15] Yang X, Hall MB. The catalytic dehydrogenation of ammoniaborane involving an unexpected hydrogen transfer to ligated carbene and subsequent carbon-hydrogen activation. J Am Chem Soc 2008;130:1798e9. [16] Alcaraz G, Sabo-Etienne S. Coordination and dehydrogenation of amineeboranes at metal centers. Angew Chem Int Ed 2010;49:7170e9. [17] Marziale AN, Friedrich A, Klopsch I, Drees M, Celinski VR, auf der Gu¨nne JS, et al. The mechanism of boraneamine dehydrocoupling with bifunctional ruthenium catalysts. J Am Chem Soc 2013;135:13342e55. [18] Waterman R. Mechanisms of metal-catalyzed dehydrocoupling reactions. Chem Soc Rev 2013;42:5629e41. [19] Butera V, Russo N, Sicilia E. Hydrogen release from dialkylamineeboranes promoted by Mg and Ca complexes: a DFT analysis of the reaction mechanism. Chem Eur J 2014;20:5967e76. € rster M, Celinski VR, auf der Gu¨nne JS, [20] Glu¨er A, Fo Holthausen MC, Schneider S. Highly active Iron catalyst for ammonia borane dehydrocoupling at room temperature. ACS Catal 2015;5:7214e7. [21] Kab M, Friedrich A, Drees M, Schneider S. Ruthenium complexes with cooperative PNP ligands: bifunctional catalysts for the dehydrogenation of ammoniaeborane. Angew Chem Int Ed 2009;48:905e7. [22] Conley BL, Williams TJ. Dehydrogenation of ammoniaborane by Shvo's catalyst. Chem Commun 2010;46:4815e7. [23] Jaska CA, Temple K, Lough AJ, Manners I. Rhodium-catalyzed formation of boronenitrogen bonds: a mild route to cyclic aminoboranes and borazines. Chem Commun 2001:962e3. [24] Johnson HC, Leitao EM, Whittell GR, Manners I, LloydJones GC, Weller AS. Mechanistic studies of the dehydrocoupling and dehydropolymerization of amineboranes using a [Rh(Xantphos)]þ catalyst. J Am Chem Soc 2014;136:9078e93. [25] Kim SK, Han WS, Kim TJ, Kim TY, Nam SW, Mitoraj M, et al. Palladium catalysts for dehydrogenation of ammonia borane with preferential B-H activation. J Am Chem Soc 2010;132:9954e5. [26] Rossin A, Bottari G, Lozano-Vila AM, Paneque M, Peruzzini M, Rossi A. Catalytic amine-borane dehydrogenation by a PCPpincer palladium complex: a combined experimental and DFT analysis of the reaction mechanism. Dalton Trans 2013;42:3533e41. [27] Dietrich BL, Goldberg KI, Heinekey DM, Autrey T, Linehan JC. Iridium-catalyzed dehydrogenation of substituted amine boranes: kinetics, thermodynamics, and implications for hydrogen storage. Inorg Chem 2008;47:8583e5. [28] Staubitz A, Soto AP, Manners I. Iridium-catalyzed dehydrocoupling of primary amineeborane adducts: a route to high molecular weight polyaminoboranes, boronenitrogen analogues of polyolefins. Angew Chem Int Ed 2008;47:6212e5.  ndez I, Lo  pez AM, Mora M, On ~ ate E. [29] Esteruelas MA, Ferna Osmium-promoted dehydrogenation of amineboranes and

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 1 7 2 0 8 e1 7 2 1 5

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

BH bond activation of the resulting aminoboranes. Organometallics 2014;33:1104e7. Zimmerman PM, Paul A, Zhang Z, Musgrave CB. The role of free N-heterocyclic carbene (NHC) in the catalytic dehydrogenation of ammoniaeborane in the nickel NHC system. Angew Chem Int Ed 2009;48:2201e5. Sloan ME, Staubitz A, Clark TJ, Russell CA, Lloyd-Jones GC, Manners I. Homogeneous catalytic dehydrocoupling/ dehydrogenation of amine-borane adducts by early transition metal, group 4 metallocene complexes. J Am Chem Soc 2010;132:3831e41. Helten H, Dutta B, Vance JR, Sloan ME, Haddow MF, Sproules S, et al. Paramagnetic titanium(III) and zirconium(III) metallocene complexes as precatalysts for the dehydrocoupling/dehydrogenation of amineeboranes. Angew Chem Int Ed 2013;52:437e40. Vance JR, Robertson APM, Lee K, Manners I. Photoactivated, iron-catalyzed dehydrocoupling of amineeborane adducts: formation of boronenitrogen oligomers and polymers. Chem Eur J 2011;17:4099e103. Baker RT, Gordon JC, Hamilton CW, Henson NJ, Lin PH, Maguire S, et al. Iron complex-catalyzed ammoniaborane dehydrogenation. A potential route toward BN-containing polymer motifs using earth-abundant metal catalysts. J Am Chem Soc 2012;134:5598e609. € fer A, Robertson APM, Lee K, Turner J, Vance JR, Scha Whittell GR, et al. Iron-catalyzed dehydrocoupling/ dehydrogenation of amineboranes. J Am Chem Soc 2014;136:3048e64. Bhattacharya P, Krause JA, Guan H. Mechanistic studies of ammonia borane dehydrogenation catalyzed by iron pincer complexes. J Am Chem Soc 2014;136:11153e61. Buss JA, Edouard GA, Cheng C, Shi J, Agapie T. Molybdenum catalyzed ammonia borane dehydrogenation: oxidation state specific mechanisms. J Am Chem Soc 2014;136:11272e5. Chaplin AB, Weller AS. B-H activation at a rhodium (I) center: isolation of a bimetallic complex relevant to the transitionmetal-catalyzed dehydrocoupling of amineeboranes. Angew Chem Int Ed 2010;49:581e4. Douglas TM, Chaplin AB, Weller AS, Yang X, Hall MB. Monomeric and oligomeric amine-borane s-complexes of rhodium. Intermediates in the catalytic dehydrogenation of amine-boranes. J Am Chem Soc 2009;131:15440e56. Ai DX, Qi ZH, Ruan GY, Zhang Y, Liu W, Wang Y. DFT studies of dehydrogenation of ammoniaeborane catalyzed by [Ir(ItBu0 )2]þ: a proton transfer mechanism. Comput Theor Chem 2014;1048:1e6. Ai DX, Guo Y, Liu W, Wong Y. DFT studies on catalytic dehydrogenation of ammonia borane by Ni(NHC)2. J Phys Org Chem 2014;27:597e603. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al. Gaussian 09, revision C.01. Wallingford, CT: Gaussian Inc.; 2010.

17215

[43] Zhao Y, Truhlar DG. The m06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four m06-class functionals and 12 other functionals. Theor Chem Acc 2008;120:215e41. [44] Stephens PJ, Devlin FJ, Chabalowski CF, Frisch MJ. Ab initio calculation of vibrational absorption and circular dichroism spectra using density functional force fields. J Phys Chem 1994;98:11623e7. [45] Rassolov VA, Pople JA, Ratner MA, Windus TL. 6-31G* basis set for atoms K through Zn. J Chem Phys 1998;109:1223e9. [46] Bergner A, Dolg M, Ku¨chle W, Stoll H, Preuß H. Ab initio energy-adjusted pseudopotentials for elements of groups 1317. Mol Phys 1993;80:1431e41. [47] Fukui K. Formulation of the reaction coordinate. J Phys Chem 1970;74:4161e3. [48] Fukui K. The path of chemical reactions e the IRC approach. Acc Chem Res 1981;14:363e8. [49] Marenich AV, Cramer CJ, Truhlar DG. Universal solvation model based on solute electron density and on a continuum model of the solvent defined by the bulk dielectric constant and atomic surface lensions. J Phys Chem B 2009;113:6378e96. [50] Zhao Y, Truhlar DG. A new local density functional for maingroup thermochemistry, transition metal bonding, thermochemical kinetics, and noncovalent interactions. J Chem Phys 2006;125:194101e18. [51] Reed AE, Curtiss LA, Weinhold F. Intermolecular interactions from a natural bond orbital, donor-acceptor viewpoint. Chem Rev 1988;88:899e926.  de [52] Legault CY. CYLview, 1.0 b. Que-bec(Canada): Universite Sherbrooke; 2009. http://www.cylview.org. [53] Montag M, Schwartsburd L, Cohen R, Leitus G, Ben-David Y, Martin JML, et al. The unexpected role of CO in C-H oxidative addition by a cationic rhodium (I) complex. Angew Chem Int Ed 2007;46:1901e4. [54] Lewis JC, Wu J, Bergman RG, Ellman JA. Preagostic RhH interactions and CH bond functionalization: a combined experimental and theoretical investigation of Rhodium (I) phosphinite complexes. Organometallics 2005;24:5737e46. [55] Koridze AA, Polezhaev AV, Safronov SV, Sheloumov AM, Dolgushin FM, Ezernitskaya MG, et al. Cationic ruthenium hydridocarbonyls derived from metallocene-based pincers: unusual rearrangements and H2 evolution with formation of cationic Ruthenium metallocenylidenes. Organometallics 2010;29:4360e8. [56] Ghatak k, Mane M, Vanka K. Metal or nonmetal cooperation with a phenyl group: route to catalysis? A computational investigation. ACS Catal 2013;3:920e7.