DOI: 10.1002/chem.201405543
Communication
& Density Functional Calculations | Hot Paper |
Theoretical Investigation on the Chemistry of Entrapment of the Elusive Aminoborane (H2N=BH2) Molecule Tanmay Malakar, Sourav Bhunya, and Ankan Paul*[a]
Chem. Eur. J. 2015, 21, 6340 – 6345
6340
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication Abstract: Aminoborane (H2N=BH2) is an elusive entity and is thought to be produced during dehydropolymerization of ammonia borane, a molecule of prime interest in the field of chemical hydrogen storage. The entrapment of H2N=BH2 through hydroboration of exogenous cyclohexene has emerged as a routine technique to infer if free H2N=BH2 is produced or not during metal-catalyzed ammonia borane dehydrogenation reactions. But to date, the underlying mechanism of this trapping reaction remains unexplored. Herein, by using DFT calculations, we have investigated the mechanism of trapping of H2N=BH2 by cyclohexene. Contrary to conventional wisdom, our study revealed that the trapping of H2N=BH2 does not occur through direct hydroboration of H2N=BH2 on the double bond of cyclohexene. We found that autocatalysis by H2N=BH2 is crucial for the entrapment of another H2N=BH2 molecule by cyclohexene. Additionally, nucleophilic assistance from the solvent is also implicated for the entrapment reaction carried out in nucleophilic solvents. In THF, the rate-determining barrier for formation of the trapping product was predicted to be 16.7 kcal mol¢1 at M06 L(CPCM) level of theory.
Aminoborane (H2N=BH2, M), unlike its isoelectronic hydrocarbon analogue ethylene, is an intriguing species due to its fleeting existence.[1] M is highly reactive and oligomerizes much below room temperature. It can only be isolated in argon matrices at ¢156 8C.[1c] The polymerization of M produces B¢N polymers, which can potentially have several applications.[2] Recently, experiments have shown that B¢N oligomers and polymers are produced during the dehydropolymerization of ammonia borane (NH3BH3, AB), a molecule of prime interest in the field of chemical hydrogen storage.[3] The oligopolymerization process of M has elicited much interest both in the experimental and the theoretical community.[4] The rapid polymerization phenomenon has all the more made it difficult to unveil the underlying mechanistic intricacies of the process. Furthermore, experiments have shown that dehydropolymerization of AB by different transition metal (TM) catalysts produce different types of B¢N polymeric end products. High molecular weight and highly dispersed linear polyaminoborane (LPAB) has been obtained by dehydrocoupling of amine boranes by using Brookhart’s [Ir(POCOP)H2] (1) and other TM-containing catalysts.[4b–d] In the cases when the TM catalyst can remove more than two equivalents of H2 from AB, polyborazylene (PBZ) is formed as the by-product.[5] Because the understanding of the oligo-polymerization of M has been scarce, experimentalists have devoted their efforts to[a] T. Malakar, S. Bhunya, Dr. A. Paul Raman Centre for Atomic, Molecular and Optical Sciences Indian Association for the Cultivation of Science Jadavpur, Kolkata-32 (India) E-mail: [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201405543. Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
wards the difficult job of trapping and isolating important early intermediates.[4a–d] One of such elegant strategies uses the hydroborating action of M on an external alkene, typically cyclohexene (C) to trap the in situ generated species M.[4a] Pons et al. suggested that in TM-catalyzed dehydropolymerization reactions, this trapping of the intermediate M can be used to predict whether free M is produced during the TM-catalyzed dehydropolymerization reactions.[4a] Baker and co-workers conducted dehydropolymerization of AB by using [Rh(cod)Cl]2 (cod = 1,5-cyclooctadiene) as the pre-catalyst in presence of large excess of C in diglyme solvent at 25 8C; the expected PBZ or cyclopentaborazane[(NH2BH2])5,CPB] was not observed; rather they identified spectroscopic signature of Cy2B=NH2 (D), which in turn indicates successful entrapment of free M in the form of D through hydroboration of B¢H bond with the external trapping source.[4a] They did not observe the formation of D in the case of dehydropolymerization of AB using Brookhart’s catalyst (1) at room temperature,[4a] which was further reaffirmed by experiments carried out by Manners and coworkers.[4b] However, at elevated temperatures (60 8C), mixtures of products consisting of D were observed both by the groups of Manners and Baker.[4a–b] Similarly, D was not obtained also in the AB dehydropolymerization catalyzed by Ru complexes, as was reported by Schneider and co-workers.[4c] Considering the experimental facts gathered in presence or absence of external trapping agent in the case of 1-catalyzed AB dehydrogenation reactions, Pons et al. concluded that the nascent aminoborane (M) generated immediately after dehydrogenation of AB did not remain free in the reaction mixture, rather they stay bound to the metal center.[4a] In particular, such a scenario has been suggested for other Ir- and Ru-complex-catalyzed amine boranes dehydropolymerization reactions.[4c,d] In contrast, our recent theoretical study on the oligo-polymerization process in the Ir pincer catalyst (1) case has unveiled low-barrier routes, which involve free aminoborane (M) units,[4i] and which is consistent with one of the mechanistic scenarios predicted by Manners and co-workers based on their thorough kinetic studies with the Brookhart’s catalyst (1).[4b] We have suggested that possibly the polymer formation is faster than the hydroboration trapping reaction. To resolve this contradiction, it is imperative to understand the underlying mechanism of hydroboration of C by M. Herein, we show that the hydroboration of C by M does not directly happen by the hydroborating action of M on C, and in some cases, the trapping experiment may simply fail due to its sluggish kinetics compared with that of the fast oligomerization of M. The so-called trapping experiment is essentially based on the fact that the trapping agent can effectively trap an intermediate, if the reaction for trapping has a significantly faster rate compared to any side reaction, which consumes the targeted intermediate from the reaction medium. In our recent report, we have computed the barrier of direct hydroboration reaction between C and M (Scheme 1).[4i] It turned out that the direct hydroboration of C by M leading to formation of D happens through a rate-determining barrier (RDB) of 27 kcal mol¢1.[4i] However, oligomerization of M at the [Ir(POCOP)H2] leading to LPAB formation proceeds at a much
6341
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
Scheme 1. Conventional mechanism of NH2=BH2 trapping by cyclohexene.
lower kinetic barrier (chain initiation barrier: 3.3 kcal mol¢1; chain propagation: 5–7 kcal mol¢1).[4i] Comparison of these barriers would immediately suggest that the trapping agent would certainly fail to capture M as the oligomerization process of M at the metal center is extremely rapid. However, the hydroboration channel might be operative at elevated temperatures. Hence, one could then expect a small amount of hydroboration product (D), as was observed by the groups of Baker and Manners.[4a–b] Confusion arises when we compared the RDB for hydroboration of C (27 kcal mol¢1) to the solvent-assisted M oligo-polymerization barrier (11.9 kcal mol¢1).[4i] We found that the difference in RDBs of the competing processes is too large to obtain a significant hydroboration product (D) in the cases when TM species are not involved in the M oligomerization process.[6] But Manners and co-workers have shown that in situ generated M is trapped as D in the case of AB dehydrogenation by using iPr2N=BH2, a nonmetal bifunctional species.[6a] This led us to investigate further alternative channels for the trapping experiment by using standard DFT.
Computational Details Gas-phase geometry optimizations of all stationary points were done by using the B3LYP hybrid functional using 6–31 + g** basis functions. Following this, single-point solvent-phase computations on the B3LYP optimized geometries were conducted at M06L[7] level of theory by using the CPCM[8] model with THF as solvent employing the same basis functions. Optimizations were also carried out by using wB97XD density functional, which is known to predict reliable barriers and reaction energetics.[9] The ensuing text discusses the stability and free-energy activation barriers in terms of relative solvent-phase Gibbs free-energy difference (DGs), which includes an empirical solvent-phase entropic correction at 298 K and 1 atm pressure, until and unless mentioned otherwise. The respective computed results at wB97XD level of theory are provided in parentheses. All computations were done using Gaussian 09 package.[10] For further computational details together with basis set information, please see the Supporting information.
Interestingly, decades back, the mechanistic intricacies of hydroboration of alkenes by BH3¢THF have generated much debate. Although some experimental and theoretical studies have implicated the role of BH3¢THF in formation of the initial p complex between ethylene,[11] others have suggested that free BH3 is produced, which leads to the p complex formation.[12] Our DFT studies with cyclohexene (C) hydroboration with BH3¢THF are consistent with the former mechanistic scenario. In a typical hydroboration reaction with BH3¢THF, the cyclohexene p cloud attacks the BH3¢THF complex at a barrier of 14.5 kcal mol¢1, and the THF simply acts as a good leaving group leading to the formation of C¢BH3 p complex. However, the Lewis acidity of the boron center in M is significantly quenched compared to that of in BH3 due to strong back donation from the N lone pair. Hence, it does not form a usual p complex, as was observed in the case of BH3¢THF or BH3. This factor probably contributes to a higher free-energy activation barrier for direct hydroboration of C by M. Because the hydroboration product (D) is obtained in copious amount in certain dehydrocoupling cases, it is possible that some intermediate produced by M is responsible for the experimentally observed D. Interestingly, the binding of M with THF is found to be endoergic (Figure S2 in the Supporting Information). Earlier studies from our group have implicated the role of intermediate BH3¢NH2¢BH=NH2 (2) in the oligomerization process.[4e] We found that 2, which is equipped with a vulnerable BH3 moiety, plays a crucial role in the formation of D (Scheme 2). Hydroboration reaction between two entities of M can lead to the formation of 2 via TsM-2. Generation of 2 is favorable by 5.7 kcal mol¢1 (DGwB97XD = ¢5.8 kcal mol¢1), and the associated freeenergy activation barrier is 11.6 kcal mol¢1 (DGwB97XD = 13.0 kcal mol¢1). Nucleophilic attack by the solvent, THF, leads to the formation of BH3·THF (3) and NH2¢BH=NH2 (4) through Ts2–3 with an associated free energy of activation of 15.6 kcal mol¢1 (DGwB97XD = 16.4 kcal mol¢1; Figure 1). Similar role of solvent as a nucleophile has been reported recently.[4e] Moreover, BH(NHMe)2, an analogue of intermediate 4, was recently detected by 11B NMR during a trapping experiment for dehydropolymerization of MeH2NBH3 by Rh catalyst[4h] (Figure S4 in the Supporting Information). Later, a cyclohexene (C) dislodges the
Scheme 2. Proposed trapping mechanism of NH2=BH2 by cyclohexene. Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
6342
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
Figure 1. Gibb’s free-energy profile for trapping of NH2=BH2 by cyclohexene. B3LYP optimized geometries of few important intermediates and transition states. All bond lengths are given in æ; color code: N blue, O red, B yellow, C grey, and H white.
weakly ligated THF moiety from the in situ generated species 3 and produces the complex 5. Generation of 5 is endergonic by 8.5 kcal mol¢1 and (DGwB97XD = 8.5 kcal mol¢1) the associated free-energy hurdle due to Ts3–5 is predicted to be 16.7 kcal mol¢1 (DGwB97XD = 17.5 kcal mol¢1). This is followed by a hydroboration reaction through Ts5–6 from 5 to form CyBH2 (6) at a barrier of 10.5 kcal mol¢1 (DGwB97XD = 12.3 kcal mol¢1). Subsequently, a THF molecule gets ligated to 6 and forms a Lewis acid/base complex CyBH2·THF (7). Later, a C displaces the weakly ligated THF moiety from the in situ generated species 7 and produces the p complex 8.Transformation of 7 into 8 happens through Ts7–8 with a free-energy hurdle of 13.9 kcal mol¢1 (DGwB97XD = 13.5 kcal mol¢1). The intermediate 8 undergoes the hydroboration reaction and leads to generation of the species Cy2BH (9). Formation of 9 from 8 is thermodynamically downhill by 18.6 kcal mol¢1 (DGwB97XD = ¢22.3 kcal mol¢1). Association of THF with species 9 forms a Lewis acid/base complex Cy2BH·THF (10), which remain in equilibrium with intermediate 9. Interestingly, intermediate 9 possesses a B¢H bond, which cannot facilitate any further hydroboration reaction between 9 and C. Avoiding the huge steric crowding in accommodating three cyclohexyl moieties in the concerned transition state might be the reason behind not getting any Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
further hydroboration between 9 and C. However, B¢N bondforming reaction between 9 and a smaller molecule, such as 4, is in principle possible. We have been able to locate the transition state of the reaction in between 9 and 4 at B3LYP level of theory. Thus, the species 11 is formed through Ts9–11 with a Gibbs free-energy activation barrier of 6.1 kcal mol¢1. Unfortunately, despite several attempts, Ts9–11 could not be located at wB97XD level of theory. Formation of 11 is exergonic by 3.4 kcal mol¢1 (DGwB97XD = ¢3.4 kcal mol¢1). Following this, an intramolecular hydride shift within 11 occurs through Ts11–12 and leads to the formation of an intermediate 12 having a 3 c– 2 e B¢H¢B bond. The associated Gibb’s free-energy activation barrier due to Ts11–12 is essentially zero. Subsequently, a reverse hydroboration from 12 leads to the formation of Cy2B=NH2 (D) and regenerates a H2N=BH2 (M) entity. The associated free-energy cost for this last step was estimated to be 9.2 kcal mol¢1 (DGwB97XD = 10.1 kcal mol¢1). Hence, the consumed M in the first step of the trapping reaction is again regenerated in the last step (Scheme 2), and therefore, M is considered to be playing an autocatalytic role in its own trapping by external agent. Thus, trapping of M by C to produce D happens at an overall RDB of only 16.7 kcal mol¢1 (17.5 kcal mol¢1 at wB97XD). Interestingly, trapping of M have also been carried
6343
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication out in non-nucleophilic solvents (C6H5F, C6H5Me). In that case, the direct transformation of intermediate 2 into 5 could be effectuated by the nucleophilic action of the p cloud of C on the electrophilic boron center in 2 (Figure S3 in the Supporting Information). The associated Gibbs free-energy activation barrier for this particular conversion was estimated to be 19.9 kcal mol¢1 (in C6H5F solvent; 17.4 kcal mol¢1 at wB97XD). Thus, THF is simply playing a role of a nucleophilic catalyst in bringing down the free-energy activation barrier for conversion of 2 into 5 (shown in Scheme 2). In other words, trapping of M in non-nucleophilic solvent (C6H5F) is expected to be slow compared to that of in ethereal solvent (THF or any ether). The molecular details emerging from the aforementioned mechanistic pathway delineated by us consists of the following important features: 1) dimerization of H2N=BH2 (M) to form BH3¢NH2¢BH=NH2 (2) is crucial for the trapping of M by cyclohexene (C); and 2) solvent molecule (THF) has a pivotal role as a nucleophilic catalyst to produce the reactive entity BH3·THF (3). Incidentally, we have shown earlier that oligo-polymerization of M in absence of TM catalysts initiates also through the formation of species 2.[4e] Our predicted RDB (16.7 kcal mol¢1) for entrapment of M to produce hydroboration product (D) via the intermediate 2 and its solvent catalyzed oligomerization to form cyclotriborazane (CTB) through 2 is kinetically competitive in solvent medium (TM-free environment), because the oligomerization happens at an overall RDB of 11.9 kcal mol¢1.[4e] It must be noted that the overall rate of oligomerization is further controlled by the rate of monomer formation. Moreover, in the trapping experiments, it is a usual practice to use high excess of C.[4a] These factors are likely to assist the trapping channel, although it has a slightly higher barrier compared to solvent-catalyzed oligomerization and make it competitive. Hence, one can expect that these factors would result into significant amount of D formation due to trapping of M by C in absence of any TM catalyst. Earlier, it was suggested that the trapping of M is not observed for the Ir pincer (1) catalyzed dehydrocoupling of AB, because no free units of M are released from the catalytic metal center.[4a] We have recently shown that in this particular Ir catalyst (1), the oligo-polymerization of M initiates at much smaller free-energy activation barrier of 3.3 kcal mol¢1 and leads to the formation of highly thermodynamically stable LPAB.[4i] Consequently, the two separate events, namely, polymerization and trapping of M, no longer remain kinetically competitive. The TM-facilitated M oligomerization is so rapid that the trapping product (D) is not obtained at room temperature. However, at elevated temperatures, the reaction channel associated with entrapment of M by exogenous C becomes accessible and results into successful entrapment of M in the form of D.[4a–b] Our predicted result is thus in sound agreement with the experimental data obtained both in presence or absence of TM-catalyzed dehydropolymerization of AB. Furthermore, we strongly believe that failure of entrapment of M in presence of Brookhart’s Ir catalyst (1) at room temperature is not necessarily due to the absence of free M in the reaction medium; rather we feel that the low-barrier faster polymerization channel shuts out the trapping of M. The same reasoning can be extrapolated to Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
other TM-catalyzed amine borane dehydrocoupling reactions, as was recently observed by Kumar et al.[4d] and Schneider and co-workers[4c] (Figure S3 in the Supporting Information). Hence, our findings are in agreement with the general experimental observations. In summary, our careful computational effort provides a new mechanistic paradigm of H2N=BH2 trapping, which is in reasonable agreement with available experimental results. Autocatalytic function of H2N=BH2 and solvent’s nucleophilic assistance plays a vital role in the trapping of this reactive entity by external reagent.
Acknowledgements A.P. would like to thank DST, India for providing financial support through “Fast track” project (No. SR/FT/CS-118/2011). S.B would like to thank CSIR, India, for providing a research fellowship. T.M. would like to thank Mr. Sabyasachi Sutradhar for preparing the frontispiece graphic. Keywords: aminoborane · autocatalysis · density functional calculations · nucleophilic assistance · trapping [1] a) K. W. Bçddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. Soc. 1966, 88, 4396 – 4401; b) C. T. Kwon Jr., H. A. McGee, Inorg. Chem. 1970, 9, 2458 – 2461; c) J. D. Carpenter, B. S. Ault, J. Phys. Chem. 1991, 95, 3502 – 3506; d) M. C. L. Gerry, W. Lewis-Bevan, A. J. Merer, N. P. C. Westwood, J. Mol. Spectrosc. 1985, 110, 153 – 163. [2] A. Staubitz, A. P. M. Robertson, M. E. Sloan, I. Manners, Chem. Rev. 2010, 110, 4023 – 4078. [3] a) T. B. Marder, Angew. Chem. Int. Ed. 2007, 46, 8116 – 8118; Angew. Chem. 2007, 119, 8262 – 8264; b) A. Staubitz, A. P. M. Robertson, I. Manners, Chem. Rev. 2010, 110, 4079 – 4124. [4] a) V. Pons, R. T. Baker, N. K. Szymczak, D. J. Heldebrant, J. C. Linehan, M. H. Matus, D. J. Grant, D. A. Dixon, Chem. Commun. 2008, 6597 – 6599; b) A. Staubitz, M. E. Sloan, A. P. M. Robertson, A. Friedrich, S. Schneider, P. J. Gates, J. Schmedt auf der Gìnne, I. Manners, J. Am. Chem. Soc. 2010, 132, 13332 – 13345; c) A. N. Marziale, A. Friedrich, I. Klopsch, M. Drees, V. R. Celinski, J. Schmedt auf der Gìnne, S. Schneider, J. Am. Chem. Soc. 2013, 135, 13342 – 13355; d) A. Kumar, H. C. Johnson, T. N. Hooper, A. S. Weller, A. G. Algarra, S. A. Macgregor, Chem. Sci. 2014, 5, 2546 – 2553; e) T. Malakar, L. Roy, A. Paul, Chem. Eur. J. 2013, 19, 5812 – 5817; f) P. M. Zimmerman, A. Paul, Z. Zhang, C. B. Musgrave, Inorg. Chem. 2009, 48, 1069 – 1081; g) H. C. Johnson, A. Robertson, A. B. Chaplin, L. J. Sewell, A. L. Thompson, M. F. Haddow, I. Manners, A. S. Weller, J. Am. Chem. Soc. 2011, 133, 11076 – 11079; h) H. C. Johnson, E. M. Leitao, G. R. Whittell, I. Manners, G. C. Lloyd-Jones, A. S. Weller, J. Am. Chem. Soc. 2014, 136, 9078 – 9093; i) S. Bhunya, T. Malakar, A. Paul, Chem. Commun. 2014, 50, 5919 – 5922. [5] a) B. L. Conley, T. J. Williams, Chem. Commun. 2010, 46, 4815 – 4817; b) B. L. Conley, D. Guess, T. J. Williams, J. Am. Chem. Soc. 2011, 133, 14212 – 14215. [6] a) A. P. M. Robertson, E. M. Leitao, I. Manners, J. Am. Chem. Soc. 2011, 133, 19322 – 19335; b) X. Yang, T. Fox, H. Berke, Org. Biomol. Chem. 2012, 10, 852 – 860. [7] Y. Zhao, D. G. Truhlar, J. Chem. Phys. 2006, 125, 1 – 18. [8] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24, 669 – 681. [9] a) J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10, 6615 – 6620; b) S. Bhunya, A. Paul, Chem. Eur. J. 2013, 19, 11541 – 11546. [10] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
6344
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ©. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
[11] a) D. J. Pasto, B. Lepeska, T. C. Cheng, J. Am. Chem. Soc. 1972, 94, 6083 – 6090; b) T. Clark, P. v. R. Schleyer, J. Chem. Soc. Chem. Commun. 1983, 606 – 608. [12] H. C. Brown, J. Chandrasekharan, J. Am. Chem. Soc. 1984, 106, 1863 – 1865.
Received: October 7, 2014 Published online on March 18, 2015
6345
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
& Density Functional Calculations | Hot Paper |
Theoretical Investigation on the Chemistry of Entrapment of the Elusive Aminoborane (H2N=BH2) Molecule Tanmay Malakar, Sourav Bhunya, and Ankan Paul*[a]
Chem. Eur. J. 2015, 21, 6340 – 6345
6340
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication Abstract: Aminoborane (H2N=BH2) is an elusive entity and is thought to be produced during dehydropolymerization of ammonia borane, a molecule of prime interest in the field of chemical hydrogen storage. The entrapment of H2N=BH2 through hydroboration of exogenous cyclohexene has emerged as a routine technique to infer if free H2N=BH2 is produced or not during metal-catalyzed ammonia borane dehydrogenation reactions. But to date, the underlying mechanism of this trapping reaction remains unexplored. Herein, by using DFT calculations, we have investigated the mechanism of trapping of H2N=BH2 by cyclohexene. Contrary to conventional wisdom, our study revealed that the trapping of H2N=BH2 does not occur through direct hydroboration of H2N=BH2 on the double bond of cyclohexene. We found that autocatalysis by H2N=BH2 is crucial for the entrapment of another H2N=BH2 molecule by cyclohexene. Additionally, nucleophilic assistance from the solvent is also implicated for the entrapment reaction carried out in nucleophilic solvents. In THF, the rate-determining barrier for formation of the trapping product was predicted to be 16.7 kcal mol¢1 at M06 L(CPCM) level of theory.
Aminoborane (H2N=BH2, M), unlike its isoelectronic hydrocarbon analogue ethylene, is an intriguing species due to its fleeting existence.[1] M is highly reactive and oligomerizes much below room temperature. It can only be isolated in argon matrices at ¢156 8C.[1c] The polymerization of M produces B¢N polymers, which can potentially have several applications.[2] Recently, experiments have shown that B¢N oligomers and polymers are produced during the dehydropolymerization of ammonia borane (NH3BH3, AB), a molecule of prime interest in the field of chemical hydrogen storage.[3] The oligopolymerization process of M has elicited much interest both in the experimental and the theoretical community.[4] The rapid polymerization phenomenon has all the more made it difficult to unveil the underlying mechanistic intricacies of the process. Furthermore, experiments have shown that dehydropolymerization of AB by different transition metal (TM) catalysts produce different types of B¢N polymeric end products. High molecular weight and highly dispersed linear polyaminoborane (LPAB) has been obtained by dehydrocoupling of amine boranes by using Brookhart’s [Ir(POCOP)H2] (1) and other TM-containing catalysts.[4b–d] In the cases when the TM catalyst can remove more than two equivalents of H2 from AB, polyborazylene (PBZ) is formed as the by-product.[5] Because the understanding of the oligo-polymerization of M has been scarce, experimentalists have devoted their efforts to[a] T. Malakar, S. Bhunya, Dr. A. Paul Raman Centre for Atomic, Molecular and Optical Sciences Indian Association for the Cultivation of Science Jadavpur, Kolkata-32 (India) E-mail: [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201405543. Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
wards the difficult job of trapping and isolating important early intermediates.[4a–d] One of such elegant strategies uses the hydroborating action of M on an external alkene, typically cyclohexene (C) to trap the in situ generated species M.[4a] Pons et al. suggested that in TM-catalyzed dehydropolymerization reactions, this trapping of the intermediate M can be used to predict whether free M is produced during the TM-catalyzed dehydropolymerization reactions.[4a] Baker and co-workers conducted dehydropolymerization of AB by using [Rh(cod)Cl]2 (cod = 1,5-cyclooctadiene) as the pre-catalyst in presence of large excess of C in diglyme solvent at 25 8C; the expected PBZ or cyclopentaborazane[(NH2BH2])5,CPB] was not observed; rather they identified spectroscopic signature of Cy2B=NH2 (D), which in turn indicates successful entrapment of free M in the form of D through hydroboration of B¢H bond with the external trapping source.[4a] They did not observe the formation of D in the case of dehydropolymerization of AB using Brookhart’s catalyst (1) at room temperature,[4a] which was further reaffirmed by experiments carried out by Manners and coworkers.[4b] However, at elevated temperatures (60 8C), mixtures of products consisting of D were observed both by the groups of Manners and Baker.[4a–b] Similarly, D was not obtained also in the AB dehydropolymerization catalyzed by Ru complexes, as was reported by Schneider and co-workers.[4c] Considering the experimental facts gathered in presence or absence of external trapping agent in the case of 1-catalyzed AB dehydrogenation reactions, Pons et al. concluded that the nascent aminoborane (M) generated immediately after dehydrogenation of AB did not remain free in the reaction mixture, rather they stay bound to the metal center.[4a] In particular, such a scenario has been suggested for other Ir- and Ru-complex-catalyzed amine boranes dehydropolymerization reactions.[4c,d] In contrast, our recent theoretical study on the oligo-polymerization process in the Ir pincer catalyst (1) case has unveiled low-barrier routes, which involve free aminoborane (M) units,[4i] and which is consistent with one of the mechanistic scenarios predicted by Manners and co-workers based on their thorough kinetic studies with the Brookhart’s catalyst (1).[4b] We have suggested that possibly the polymer formation is faster than the hydroboration trapping reaction. To resolve this contradiction, it is imperative to understand the underlying mechanism of hydroboration of C by M. Herein, we show that the hydroboration of C by M does not directly happen by the hydroborating action of M on C, and in some cases, the trapping experiment may simply fail due to its sluggish kinetics compared with that of the fast oligomerization of M. The so-called trapping experiment is essentially based on the fact that the trapping agent can effectively trap an intermediate, if the reaction for trapping has a significantly faster rate compared to any side reaction, which consumes the targeted intermediate from the reaction medium. In our recent report, we have computed the barrier of direct hydroboration reaction between C and M (Scheme 1).[4i] It turned out that the direct hydroboration of C by M leading to formation of D happens through a rate-determining barrier (RDB) of 27 kcal mol¢1.[4i] However, oligomerization of M at the [Ir(POCOP)H2] leading to LPAB formation proceeds at a much
6341
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
Scheme 1. Conventional mechanism of NH2=BH2 trapping by cyclohexene.
lower kinetic barrier (chain initiation barrier: 3.3 kcal mol¢1; chain propagation: 5–7 kcal mol¢1).[4i] Comparison of these barriers would immediately suggest that the trapping agent would certainly fail to capture M as the oligomerization process of M at the metal center is extremely rapid. However, the hydroboration channel might be operative at elevated temperatures. Hence, one could then expect a small amount of hydroboration product (D), as was observed by the groups of Baker and Manners.[4a–b] Confusion arises when we compared the RDB for hydroboration of C (27 kcal mol¢1) to the solvent-assisted M oligo-polymerization barrier (11.9 kcal mol¢1).[4i] We found that the difference in RDBs of the competing processes is too large to obtain a significant hydroboration product (D) in the cases when TM species are not involved in the M oligomerization process.[6] But Manners and co-workers have shown that in situ generated M is trapped as D in the case of AB dehydrogenation by using iPr2N=BH2, a nonmetal bifunctional species.[6a] This led us to investigate further alternative channels for the trapping experiment by using standard DFT.
Computational Details Gas-phase geometry optimizations of all stationary points were done by using the B3LYP hybrid functional using 6–31 + g** basis functions. Following this, single-point solvent-phase computations on the B3LYP optimized geometries were conducted at M06L[7] level of theory by using the CPCM[8] model with THF as solvent employing the same basis functions. Optimizations were also carried out by using wB97XD density functional, which is known to predict reliable barriers and reaction energetics.[9] The ensuing text discusses the stability and free-energy activation barriers in terms of relative solvent-phase Gibbs free-energy difference (DGs), which includes an empirical solvent-phase entropic correction at 298 K and 1 atm pressure, until and unless mentioned otherwise. The respective computed results at wB97XD level of theory are provided in parentheses. All computations were done using Gaussian 09 package.[10] For further computational details together with basis set information, please see the Supporting information.
Interestingly, decades back, the mechanistic intricacies of hydroboration of alkenes by BH3¢THF have generated much debate. Although some experimental and theoretical studies have implicated the role of BH3¢THF in formation of the initial p complex between ethylene,[11] others have suggested that free BH3 is produced, which leads to the p complex formation.[12] Our DFT studies with cyclohexene (C) hydroboration with BH3¢THF are consistent with the former mechanistic scenario. In a typical hydroboration reaction with BH3¢THF, the cyclohexene p cloud attacks the BH3¢THF complex at a barrier of 14.5 kcal mol¢1, and the THF simply acts as a good leaving group leading to the formation of C¢BH3 p complex. However, the Lewis acidity of the boron center in M is significantly quenched compared to that of in BH3 due to strong back donation from the N lone pair. Hence, it does not form a usual p complex, as was observed in the case of BH3¢THF or BH3. This factor probably contributes to a higher free-energy activation barrier for direct hydroboration of C by M. Because the hydroboration product (D) is obtained in copious amount in certain dehydrocoupling cases, it is possible that some intermediate produced by M is responsible for the experimentally observed D. Interestingly, the binding of M with THF is found to be endoergic (Figure S2 in the Supporting Information). Earlier studies from our group have implicated the role of intermediate BH3¢NH2¢BH=NH2 (2) in the oligomerization process.[4e] We found that 2, which is equipped with a vulnerable BH3 moiety, plays a crucial role in the formation of D (Scheme 2). Hydroboration reaction between two entities of M can lead to the formation of 2 via TsM-2. Generation of 2 is favorable by 5.7 kcal mol¢1 (DGwB97XD = ¢5.8 kcal mol¢1), and the associated freeenergy activation barrier is 11.6 kcal mol¢1 (DGwB97XD = 13.0 kcal mol¢1). Nucleophilic attack by the solvent, THF, leads to the formation of BH3·THF (3) and NH2¢BH=NH2 (4) through Ts2–3 with an associated free energy of activation of 15.6 kcal mol¢1 (DGwB97XD = 16.4 kcal mol¢1; Figure 1). Similar role of solvent as a nucleophile has been reported recently.[4e] Moreover, BH(NHMe)2, an analogue of intermediate 4, was recently detected by 11B NMR during a trapping experiment for dehydropolymerization of MeH2NBH3 by Rh catalyst[4h] (Figure S4 in the Supporting Information). Later, a cyclohexene (C) dislodges the
Scheme 2. Proposed trapping mechanism of NH2=BH2 by cyclohexene. Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
6342
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
Figure 1. Gibb’s free-energy profile for trapping of NH2=BH2 by cyclohexene. B3LYP optimized geometries of few important intermediates and transition states. All bond lengths are given in æ; color code: N blue, O red, B yellow, C grey, and H white.
weakly ligated THF moiety from the in situ generated species 3 and produces the complex 5. Generation of 5 is endergonic by 8.5 kcal mol¢1 and (DGwB97XD = 8.5 kcal mol¢1) the associated free-energy hurdle due to Ts3–5 is predicted to be 16.7 kcal mol¢1 (DGwB97XD = 17.5 kcal mol¢1). This is followed by a hydroboration reaction through Ts5–6 from 5 to form CyBH2 (6) at a barrier of 10.5 kcal mol¢1 (DGwB97XD = 12.3 kcal mol¢1). Subsequently, a THF molecule gets ligated to 6 and forms a Lewis acid/base complex CyBH2·THF (7). Later, a C displaces the weakly ligated THF moiety from the in situ generated species 7 and produces the p complex 8.Transformation of 7 into 8 happens through Ts7–8 with a free-energy hurdle of 13.9 kcal mol¢1 (DGwB97XD = 13.5 kcal mol¢1). The intermediate 8 undergoes the hydroboration reaction and leads to generation of the species Cy2BH (9). Formation of 9 from 8 is thermodynamically downhill by 18.6 kcal mol¢1 (DGwB97XD = ¢22.3 kcal mol¢1). Association of THF with species 9 forms a Lewis acid/base complex Cy2BH·THF (10), which remain in equilibrium with intermediate 9. Interestingly, intermediate 9 possesses a B¢H bond, which cannot facilitate any further hydroboration reaction between 9 and C. Avoiding the huge steric crowding in accommodating three cyclohexyl moieties in the concerned transition state might be the reason behind not getting any Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
further hydroboration between 9 and C. However, B¢N bondforming reaction between 9 and a smaller molecule, such as 4, is in principle possible. We have been able to locate the transition state of the reaction in between 9 and 4 at B3LYP level of theory. Thus, the species 11 is formed through Ts9–11 with a Gibbs free-energy activation barrier of 6.1 kcal mol¢1. Unfortunately, despite several attempts, Ts9–11 could not be located at wB97XD level of theory. Formation of 11 is exergonic by 3.4 kcal mol¢1 (DGwB97XD = ¢3.4 kcal mol¢1). Following this, an intramolecular hydride shift within 11 occurs through Ts11–12 and leads to the formation of an intermediate 12 having a 3 c– 2 e B¢H¢B bond. The associated Gibb’s free-energy activation barrier due to Ts11–12 is essentially zero. Subsequently, a reverse hydroboration from 12 leads to the formation of Cy2B=NH2 (D) and regenerates a H2N=BH2 (M) entity. The associated free-energy cost for this last step was estimated to be 9.2 kcal mol¢1 (DGwB97XD = 10.1 kcal mol¢1). Hence, the consumed M in the first step of the trapping reaction is again regenerated in the last step (Scheme 2), and therefore, M is considered to be playing an autocatalytic role in its own trapping by external agent. Thus, trapping of M by C to produce D happens at an overall RDB of only 16.7 kcal mol¢1 (17.5 kcal mol¢1 at wB97XD). Interestingly, trapping of M have also been carried
6343
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication out in non-nucleophilic solvents (C6H5F, C6H5Me). In that case, the direct transformation of intermediate 2 into 5 could be effectuated by the nucleophilic action of the p cloud of C on the electrophilic boron center in 2 (Figure S3 in the Supporting Information). The associated Gibbs free-energy activation barrier for this particular conversion was estimated to be 19.9 kcal mol¢1 (in C6H5F solvent; 17.4 kcal mol¢1 at wB97XD). Thus, THF is simply playing a role of a nucleophilic catalyst in bringing down the free-energy activation barrier for conversion of 2 into 5 (shown in Scheme 2). In other words, trapping of M in non-nucleophilic solvent (C6H5F) is expected to be slow compared to that of in ethereal solvent (THF or any ether). The molecular details emerging from the aforementioned mechanistic pathway delineated by us consists of the following important features: 1) dimerization of H2N=BH2 (M) to form BH3¢NH2¢BH=NH2 (2) is crucial for the trapping of M by cyclohexene (C); and 2) solvent molecule (THF) has a pivotal role as a nucleophilic catalyst to produce the reactive entity BH3·THF (3). Incidentally, we have shown earlier that oligo-polymerization of M in absence of TM catalysts initiates also through the formation of species 2.[4e] Our predicted RDB (16.7 kcal mol¢1) for entrapment of M to produce hydroboration product (D) via the intermediate 2 and its solvent catalyzed oligomerization to form cyclotriborazane (CTB) through 2 is kinetically competitive in solvent medium (TM-free environment), because the oligomerization happens at an overall RDB of 11.9 kcal mol¢1.[4e] It must be noted that the overall rate of oligomerization is further controlled by the rate of monomer formation. Moreover, in the trapping experiments, it is a usual practice to use high excess of C.[4a] These factors are likely to assist the trapping channel, although it has a slightly higher barrier compared to solvent-catalyzed oligomerization and make it competitive. Hence, one can expect that these factors would result into significant amount of D formation due to trapping of M by C in absence of any TM catalyst. Earlier, it was suggested that the trapping of M is not observed for the Ir pincer (1) catalyzed dehydrocoupling of AB, because no free units of M are released from the catalytic metal center.[4a] We have recently shown that in this particular Ir catalyst (1), the oligo-polymerization of M initiates at much smaller free-energy activation barrier of 3.3 kcal mol¢1 and leads to the formation of highly thermodynamically stable LPAB.[4i] Consequently, the two separate events, namely, polymerization and trapping of M, no longer remain kinetically competitive. The TM-facilitated M oligomerization is so rapid that the trapping product (D) is not obtained at room temperature. However, at elevated temperatures, the reaction channel associated with entrapment of M by exogenous C becomes accessible and results into successful entrapment of M in the form of D.[4a–b] Our predicted result is thus in sound agreement with the experimental data obtained both in presence or absence of TM-catalyzed dehydropolymerization of AB. Furthermore, we strongly believe that failure of entrapment of M in presence of Brookhart’s Ir catalyst (1) at room temperature is not necessarily due to the absence of free M in the reaction medium; rather we feel that the low-barrier faster polymerization channel shuts out the trapping of M. The same reasoning can be extrapolated to Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
other TM-catalyzed amine borane dehydrocoupling reactions, as was recently observed by Kumar et al.[4d] and Schneider and co-workers[4c] (Figure S3 in the Supporting Information). Hence, our findings are in agreement with the general experimental observations. In summary, our careful computational effort provides a new mechanistic paradigm of H2N=BH2 trapping, which is in reasonable agreement with available experimental results. Autocatalytic function of H2N=BH2 and solvent’s nucleophilic assistance plays a vital role in the trapping of this reactive entity by external reagent.
Acknowledgements A.P. would like to thank DST, India for providing financial support through “Fast track” project (No. SR/FT/CS-118/2011). S.B would like to thank CSIR, India, for providing a research fellowship. T.M. would like to thank Mr. Sabyasachi Sutradhar for preparing the frontispiece graphic. Keywords: aminoborane · autocatalysis · density functional calculations · nucleophilic assistance · trapping [1] a) K. W. Bçddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. Soc. 1966, 88, 4396 – 4401; b) C. T. Kwon Jr., H. A. McGee, Inorg. Chem. 1970, 9, 2458 – 2461; c) J. D. Carpenter, B. S. Ault, J. Phys. Chem. 1991, 95, 3502 – 3506; d) M. C. L. Gerry, W. Lewis-Bevan, A. J. Merer, N. P. C. Westwood, J. Mol. Spectrosc. 1985, 110, 153 – 163. [2] A. Staubitz, A. P. M. Robertson, M. E. Sloan, I. Manners, Chem. Rev. 2010, 110, 4023 – 4078. [3] a) T. B. Marder, Angew. Chem. Int. Ed. 2007, 46, 8116 – 8118; Angew. Chem. 2007, 119, 8262 – 8264; b) A. Staubitz, A. P. M. Robertson, I. Manners, Chem. Rev. 2010, 110, 4079 – 4124. [4] a) V. Pons, R. T. Baker, N. K. Szymczak, D. J. Heldebrant, J. C. Linehan, M. H. Matus, D. J. Grant, D. A. Dixon, Chem. Commun. 2008, 6597 – 6599; b) A. Staubitz, M. E. Sloan, A. P. M. Robertson, A. Friedrich, S. Schneider, P. J. Gates, J. Schmedt auf der Gìnne, I. Manners, J. Am. Chem. Soc. 2010, 132, 13332 – 13345; c) A. N. Marziale, A. Friedrich, I. Klopsch, M. Drees, V. R. Celinski, J. Schmedt auf der Gìnne, S. Schneider, J. Am. Chem. Soc. 2013, 135, 13342 – 13355; d) A. Kumar, H. C. Johnson, T. N. Hooper, A. S. Weller, A. G. Algarra, S. A. Macgregor, Chem. Sci. 2014, 5, 2546 – 2553; e) T. Malakar, L. Roy, A. Paul, Chem. Eur. J. 2013, 19, 5812 – 5817; f) P. M. Zimmerman, A. Paul, Z. Zhang, C. B. Musgrave, Inorg. Chem. 2009, 48, 1069 – 1081; g) H. C. Johnson, A. Robertson, A. B. Chaplin, L. J. Sewell, A. L. Thompson, M. F. Haddow, I. Manners, A. S. Weller, J. Am. Chem. Soc. 2011, 133, 11076 – 11079; h) H. C. Johnson, E. M. Leitao, G. R. Whittell, I. Manners, G. C. Lloyd-Jones, A. S. Weller, J. Am. Chem. Soc. 2014, 136, 9078 – 9093; i) S. Bhunya, T. Malakar, A. Paul, Chem. Commun. 2014, 50, 5919 – 5922. [5] a) B. L. Conley, T. J. Williams, Chem. Commun. 2010, 46, 4815 – 4817; b) B. L. Conley, D. Guess, T. J. Williams, J. Am. Chem. Soc. 2011, 133, 14212 – 14215. [6] a) A. P. M. Robertson, E. M. Leitao, I. Manners, J. Am. Chem. Soc. 2011, 133, 19322 – 19335; b) X. Yang, T. Fox, H. Berke, Org. Biomol. Chem. 2012, 10, 852 – 860. [7] Y. Zhao, D. G. Truhlar, J. Chem. Phys. 2006, 125, 1 – 18. [8] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24, 669 – 681. [9] a) J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10, 6615 – 6620; b) S. Bhunya, A. Paul, Chem. Eur. J. 2013, 19, 11541 – 11546. [10] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
6344
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ©. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
Chem. Eur. J. 2015, 21, 6340 – 6345
www.chemeurj.org
[11] a) D. J. Pasto, B. Lepeska, T. C. Cheng, J. Am. Chem. Soc. 1972, 94, 6083 – 6090; b) T. Clark, P. v. R. Schleyer, J. Chem. Soc. Chem. Commun. 1983, 606 – 608. [12] H. C. Brown, J. Chandrasekharan, J. Am. Chem. Soc. 1984, 106, 1863 – 1865.
Received: October 7, 2014 Published online on March 18, 2015
6345
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
