Selective catalytic synthesis of amino-silanes at part-per million catalyst loadings

Journal Name COMMUNICATION This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 1 Please do not adjust margins Please do not adjust margins a. Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica, Centro de Innovación en Química Avanzada (ORFEO-CINQA), CSIC / Universidad de Sevilla, Avda. Américo Vespucio 49, 41092 Sevilla (Spain). b. Departamento de Química Física y Analítica, Centro de Innovación en Química Avanzada (ORFEO-CINQA), Universidad de Oviedo, C/ Julián Clavería 8, 33006, Oviedo (Spain). Electronic Supplementary Information (ESI) available: Experimental details for the synthesis and characterization of aminosilanes. NMR spectra of aminosilanes. H 2 evolution graphics. See DOI: 10.1039/x0xx00000x Received 00th January 20xx, Accepted 00th January 20xx DOI: 10.1039/x0xx00000x www.rsc.org/ Selective Catalytic Synthesis of Amino-Silanes at Part-per Million Catalyst Loadings Pablo Ríos,a Marta Roselló-Merino,a Orestes Rivada-Wheelaghan,a Javier Borge,b Joaquín López- Serrano,a Salvador Conejeroa,* Platinum(II) complex [Pt(ItBu’)(ItBu)][BArF] (1a) is a highly active and selective catalyst in the dehydrocoupling of amines and silanes at part-per-million catalyst loadings (up to 10 ppm), achieving the highest TON and TOF numbers reported in the literature (up to 1 mmol scale). NMR studies suggest a process taking place through electrophilic activation of the silane by the platinum species, assisted by an amine. Catalytic dehydrocoupling processes leading to the formation of main-group E-E’ bonds (E = B, Si; E’ = O, N) are gaining importance as a tool for cleaner, atom economic, synthetic methods.1 In this sense, the catalytic synthesis of Si−N bonds from silanes and amines is particularly attractive since the only by-product is dihydrogen. Amino-silanes are useful molecules for the synthesis of silicon-based bases, polymers and ceramics1c that have been traditionally prepared from halosilanes, amines and bases which is less convenient than the direct dehydrocoupling. During the last years some groups have reported efficient catalytic systems based on main group elements (Li, Na, K, Mg, Ca, Sr, Al), lanthanides and actinides (Yb, U) and late transition metals.2-4 These examples required catalyst loadings of 1-5% to achieve good conversions and reaction rates under mild reaction conditions. Most recently, it has been possible to decrease the catalyst loading (up to 0.25 mol%) by using a barium complex.5 Another challenge associated with this type of transformations is to control the selectivity of the reaction. Particularly, primary silanes (RSiH3) can produce monosilazanes (RSiH2(NRR’)) together with di- and trisilazanes (RSiH(NRR’)2 and RSi(NRR’)3). Herein, we report the catalytic synthesis of amino-silanes using an electron-deficient platinum(II) catalyst, stabilized by N-heterocyclic carbenes (Scheme 1), that exhibits high efficiency at part-per million catalyst loadings with an exceptional selectivity toward the formation of monoaminosilanes from primary silanes. Low temperature NMR studies suggest that transfer of a hydride atom from the silane to the platinum center (through σ-SiH complex intermediates) occurs during the catalytic cycle. We have reported that the coordinatively unsaturated complex [Pt(ItBu’)(ItBu)][BArF] ( 1a )6 (where ItBu is 1,3-di-tert- butylimidazolylidene and ItBu’ its cyclometalated form, (Scheme 1) is an efficient catalyst in dehydrocoupling processes leading to amino and diaminoboranes.7 Mechanistic studies support a reaction that proceeds through nucleophilic attack of free dimethylamine to the boron atom in a σ-BH intermediate complex [Pt]  HBH2·NHMe2, exhibiting a η1-BH coordination mode (Shimoi type). The mechanistic similarities of this process and the dehydrocoupling of silanes and alcohols by iridium species8 prompted us to study the dehydrocoupling of silanes and amines. Recently, we have reported that complex 1a can interact with silanes to form σ-SiH complexes9 which, according to DFT calculations, exhibit a η1-SiH coordination mode similar to that reported by Brookhart in an iridium complex.10 In this work, we first investigated the dehydrocoupling of phenylsilane (PhSiH3) and tert-butylamine (tBuNH2) (1 to 1 ratio) using a catalyst loading of 0.5% (Scheme 1). Under these conditions a rather violent reaction led to the exclusive formation of PhSiH2(NtBuH) within a few seconds. This selectivity is comparable to that reported using a Mg catalyst,2a for which longer reaction times (24 h) and higher catalyst loadings (5 mol%) were required. Scheme 1. Catalytic dehydrocoupling of primary silanes an amines. Journal Name COMMUNICATION This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 2 Please do not adjust margins Please do not adjust margins Table1. Cross-dehydrocoupling of silanes and amines catalyzed by complex [Pt(ItBu’)(ItBu)][BArF] (1a). Entry Silane Amine (equiv) mol % cata product % yieldb (isolated) TON TOF (h-1) 1 PhSiH3 tBuNH2 (1) 0.005 PhSiH2(tBuNH) 99(90) 2 x 104 9.6 x 104 2 PhSiH3 tBuNH2 (1) 0.003 PhSiH2(tBuNH) 99 3.3 x 104 5.7 x 104 3 PhSiH3 tBuNH2 (2) 0.2 PhSiH(tBuNH)2 92(90) 920 540 4 PhSiH3 NEt2H (1) 0.005 PhSiH2(NEt2) 99(88) 2 x 104 3.3 x 105 5 PhSiH3 NEt2H (1) 0.001 PhSiH2(NEt2) 98 9.8 x 104 2.9 x 104 6 PhSiH3 NEt2H (2) 0.3 PhSiH(NEt2)2 99(86) 666 595 7 PhSiH3 (CH2)4NH (1) 0.01 PhSiH2[N(CH2)4] 99b 1 x 104 6.4 x 104 8 PhSiH3 (CH2)4NH (1) 0.005 PhSiH2[N(CH2)4] 99 2 x 104 3.4 x 104 9 PhSiH3 (CH2)4NH (2) 0.2 PhSiH [N(CH2)4]2 99(96) 1000 8450 10 PhSiH3 iPr2NH (1) 0.3 PhSiH2(N iPr2) 99(93) 333 1600 11 PhSiH3 iPr2NH (1) 0.1 PhSiH2(N iPr2) 99 1000 667 12 PhSiH3 iPr2NH (4) 0.1 PhSiH2(N iPr2) 99 1000 1.7 x 104 13 PhSiH3 MesNH2 (2) 0.1 PhSiH2(NHMes) 95c 950 840 14 PhSiH3 MesNH2 (4) 0.1 PhSiH2(NHMes) 99 1000 2300 15 Ph2SiH2 tBuNH2 (1) 0.1 Ph2SiH(tBuNH) 99(97) 1000 2000 16 Ph2SiH2 NEt2H (1) 0.1 Ph2SiH(NEt2) 99(91) 1000 1300 17 Ph2SiH2 (CH2)4NH (1) 0.1 Ph2SiH[N(CH2)4] 99(92) 1000 1330 18 nBuSiH3 tBuNH2 (1) 0.005 nBuSiH2(tBuNH) 99(90) 2 x 104 1.3 x 105 19 nBuSiH3 tBuNH2 (2) 0.2 nBuSiH(tBuNH) 2 96(91) 1000 1.9 x 104 20 nBuSiH3 NEt2H (1) 0.005 nBuSiH2(NEt2) 99(88) 2 x 104 1.2 x 105 21 nBuSiH3 NEt2H (2) 0.2 nBuSiH(NEt2) 2 99(89) 1000 5.7 x 103 22 nBuSiH3 (CH2)4NH (1) 0.005 nBuSiH2[N(CH2)4] 99(91) c,d 2 x 104 3 x 104 23 nBuSiH3 (CH2)4NH (2) 0.2 nBuSiH[N(CH2)4] 2 99(95) 1000 1.3 x 104 24 nBuSiH3 iPr2NH (1) 0.1 nBuSiH2(N iPr2) 96(86) 1000 3.5 x 104 25 Et2SiH2 tBuNH2 (1) 0.1 Et2SiH(tBuNH) 99(98) 1000 3.3 x 105 26 Et2SiH2 NEt2H (1) 0.1 Et2SiH(NEt2) 99(96) 1000 5.3 x 104 27 Et2SiH2 (CH2)4NH (1) 0.1 Et2SiH[N(CH2)4] 99(98) 1000 9.4 x 103 a With respect to silane. b Determined by NMR. c Not isolated. d ca. 5% of diaminosilane nBuSiH[N(CH2)4] 2 has been detected in this case (see ESI). Other main-group catalysts provide mixtures of mono- and bis(dehydrogenated) phenylsilane with, again, longer reaction times and catalyst loadings of 5 mol%.2b In addition, at the end of the reaction 1a had evolved to its hydrogenated form [PtH(ItBu)2][BArF] ( 1b ).11 The encouraging results obtained with complex 1a prompted us to decrease the catalyst loading. Thus, when the reaction was carried out with 0.005 mol% (50 ppm) of 1a all the silane was consumed in about 12 min (Table 1, entry 1), whereas catalyst loadings of 30 ppm (Table 1, entry 2) required 35 min, which correspond to turnover numbers (TON) and turnover frequencies (TOF) of 2 x 104 and 3.3 x 104, and 9.6 x 104 and 5.7 x 104 h-1, respectively. The reaction with NEt2H (using 50 ppm of 1a ) is faster (Table 1, entry 4) with full conversion of the phenylsilane into the mono(amino)silane PhSiH2(NEt2) in 3.6 min, leading to TONs and TOFs of 2 x 104 and 3.3 x 105 h-1 respectively, the highest reported to the best of our knowledge. In this case the amount of catalyst can be reduced to 10 ppm (Table 1, entry 5) leading to 98% conversion of the silane in 180 min. Interestingly, pyrrolidine, one of the most challenging substrates in terms of selectivity, can be converted with complete selectivity into PhSiH2[N(CH2)4] using 50 ppm of the catalyst (Table 1, entry 8) in 35 min. However, bulkier (HNiPr2) or less nucleophilic amines (MesNH2) required higher catalyst loadings (0.1 mol%, Journal Name COMMUNICATION This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 3 Please do not adjust margins Please do not adjust margins 1000 ppm) to achieve comparable reaction times (45 min) and complete conversions (see Table 1, entries 10-14). Primary alkylsilanes such as nBuSiH3 are also efficiently converted into the corresponding mono(amino)silanes (Table 1, entries 18, 20, 22 and 24) using typically 50 ppm as catalyst loading. The rates of the reactions are comparable to those found for PhSiH3, except when iPr2NH was used, for which an increase of the reaction rate of one order of magnitude was observed (compare table 1, entries 11 and 24). Secondary silanes can also be dehydrocoupled efficiently. Thus, Ph2SiH2 reacts with NEt2NH, tBuNH2 and (CH2)4NH but catalyst loadings of at least 0.1 mol% are necessary to form the desired mono(amino)silanes Ph2SiH(NR2) at good rates (Scheme 2). The bulkier iPr2NH does not undergo reaction at room temperature, while heating at 60 oC for 4 days yields mixtures of products. Significantly, the rates at which Et2SiH2 is dehydrocoupled under the same reaction conditions are very high (Table 1, entries 25-27), achieving TOF numbers up to 3.3 x 105 h-1, that is up to two orders of magnitude higher in comparison to Ph2SiH2 (Table 1, entries 15-17). Finally, the bulkier silane Ph3SiH does not react at all with any of the amines. If an amine:primary-silane molar ratio of 2:1 is used for tBuNH2, NEt2H and (CH2)4NH, the bis(amino)silanes RSiH(NR’R’’)2 are formed but catalyst loadings above 0.2 mol% are required (Table 1, entries 3, 6, 9, 19, 21, and 23). The second dehydrogenation process is considerably faster when nBuSiH3 is used instead of PhSiH3 (compare for example TOF numbers in Table 1, entries 19 and 23 vs 3 and 9, respectively). No bis(amino)silanes have been detected in reactions with iPr2NH or MesNH2 even with high catalyst loadings. At this point it should be noted that when an excess of the amine is used (up to 8 equiv) in reactions with primary silanes, working at low catalyst loadings (below 0.01 mol%),12 the rate of the reaction increases or decreases depending on the nature of the amine (Figures S111-114). The use of an excess of tBuNH2, NEt2H, (CH2)4NH resulted in slower reaction rates, whereas a significant increase of the speed of the process takes place in the case of the bulkier iPr2NH and the less nucleophilic MesNH2. This is related to the coordination ability of the less hindered amines to the platinum atom. The kinetic profile of the reaction of PhSiH3 and the amines tBuNH2, NEt2H and iPr2NH, obtained by measuring the amount of dihydrogen released in the process in a closed system, depends on the bulkiness of the amine (Figure 1). The different kinetic profiles are probably related to a more efficient coordination of the amine to 1a following the sequence tBuNH2 > NEt2H > iPr2NH. In fact, when 1a is mixed with an excess of the less hindered amines a decrease in the PtH coupling constant of the Pt−CH2 Scheme 2 . Dehydrocoupling of secondary silanes. Figure 1. Kinetic profile (H 2 evolution) of the reaction of PhSiH 3 with t BuNH 2 , NEt 2 H and i Pr 2 NH (catalyst loadings of 0.005, 0.005 and 0.3 mol%, respectively). in the 1H NMR spectra (reduction of 15 and 7 Hz in the presence of 5 equiv of tBuNH2 or NEt2H, respectively) serves as an indication of this interaction, whereas it remain unaltered with iPr2NH.13 Thus, de-coordination of the less-hindered amines is a prerequisite for the reaction to proceed, which causes the induction period observed at early reaction times14 and as the concentration of amine decreases the reaction becomes faster (Figure S104 and comments below). Interestingly, kinetic profiles in all reactions with Ph2SiH2 (and Et2SiH2) exhibit exponential shapes (Figures S107-109). This different behaviour might be related to a more efficient interaction of Ph2SiH2 with the metal compared to PhSiH3 (calculated ∆Eint is 4.8 kcal·mol-1 higher for this silane).15 An analogous result has been previously observed in the interaction of nBuSiH3 and Et2SiH2 with complexes 1a , b .9 Low temperature (-30 ° C) 1H NMR studies of a stoichiometric reaction of Ph2SiH2, NEt2H and 1a conducted in THF-d8 (a solvent in which the reaction takes place at slower reaction rates) indicate the formation of the neutral platinum hydride 3 7a coexisting with 1a (Scheme 3). Very likely, complex 3 is formed through transfer of a hydride atom from Ph2SiH2 to the platinum atom (preceded by formation of a σ-SiH complex),9 assisted by the amine in a similar way to the transfer of a hydride from amine-boranes during dehydrocoupling processes induced by 1a .7 In addition, the 1H NMR spectra revealed the formation of aminosilane Ph2SiH(NEt2) together with another species ( A ) that contains an NEt2 fragment. This species exists in a 1:1 ratio with respect to hydride complex 3 . As the temperature increases the concentration of both 3 and A decreases concurrently, until they disappear completely at r. t. According to 1H,29Si-HMBC NMR experiments, species A has no silicon atom and, furthermore, we have not detected any other siliconcontaining species other than Ph2SiH2 and aminosilane Ph2SiH(NEt2). Additionally, the CH2 groups of the NEt2 fragment in species A correlate in a COSY experiment with another broad signal at ca. 7.6 ppm.16 All this information points at species A being the ammonium cation [NEt2H2][BArF], presumably formed by deprotonation of a transient, unobserved, silylium cation “Ph2SiH(NEt2H)+” by free NEt2H. Moreover, 1H and 15N NMR chemical shifts of A match those of COMMUNICATION Journal Name 4 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Scheme 5 . Low temperature NMR studies. the cation [NEt2H2][BArF] synthesized by reaction of NEt2H with HBArF·Et2O (see ESI). In support to this, when the reaction of 1a with Ph2SiH2 and NEt2H was carried out using 2 eq of NEt2H( 1a : Ph2SiH2 : NEt2H stoichiometry 1 : 1 : 2) under identical reactions conditions, we observed complex 3 together with Ph2SiH(NEt2) and species A as major products but, importantly, no Ph2SiH2 nor 1a are observed (in agreement with the need of two equiv of amine per silane). The final step of the process is the protonation of the platinum hydride 3 by the ammonium salt [NEt2H2][BArF], releasing H2 and complex 1a that is hydrogenated leading to 1b .17 In summary, we have developed a very selective and efficient catalytic system for the generation of amino-silanes (up to 1 mmol scale) using a highly electrophilic, robust, Pt(II) complex that can operate at ppm catalyst loadings. Experimental evidence hints at a process involving the transfer of a hydride from the Si−H bond to platinum without an oxidative addition pathway. Thus, the cationic platinum(II) complex 1a can be viewed as a new example of Lewis acid catalyst in which their electrophilicity is transmitted to a silicon atom upon coordination of the silane leading to a σ-SiH complex.9,18 Current work is geared at determining the mechanism of the dehydrocoupling process. We are grateful for financial support from the Junta de Andalucía (FQM-2126) and Spanish MICINN (CTQ2013-40591- P, CTQ2016-81797-REDC and CTQ2016-76267-P, FEDER support) and Principado de Asturias (FC-15-GRUPIN14-006) is acknowledged. P. R. thanks the Junta de Andalucía for a research grant. The use of Computational facilities of the Supercomputing Center of Galicia (CESGA) is also acknowledged. Conflicts of interest There are no conflicts to declare. Notes and references 1 (a) G. Alcaraz and S. Sabo-Etienne, Angew. Chem., Int. Ed., 2010, 49 , 7170; (b) H. C. Johnson, T. N. Hooper and A. S. Weller, Top. Organomet. Chem., 2012, 49 , 153; (c) E. M. Leitao, T. Jurca and I. Manners, Nat. Chem., 2013, 5 , 817. 2 (a) J. F. Dunne, S. R. Neal, J. Engelkemier, A. Ellern and A. D. Sadow, J. Am. Chem. Soc., 2011, 133 , 16782; (b) M. S. Hill, D. J. Liptrot, D. J. MacDougall, M. F. Mahon and T. P. Robinson, Chem. Sci., 2013, 4 , 4212; (c) L. K. Allen, R. García-Rodríguez and D. S. Wright, RSC Advances, 2015, 44 , 12112. 3 (a) J. X. Wang, A. K. Dash, J. C. Berthet, M. Ephritikhine, M. S. Eisen, J. Organomet. Chem., 2000, 610 , 49; (b) W. Xie, H. Hu and C. Cui, Angew. Chem., Int. Ed., 2012, 51 , 11141. 4 (a) W.-D. Wand and R. Eisenberg, Organometallics, 1991, 10 , 2222; (b) J. He, H. Q. Liu, J. F. Harrod and R. Hynes, Organometallics, 1994, 13 , 336; (c) S. Itagaki, K. Kamata, K. Yamaguchi and N. Mizuno, Chem. Commun., 2012, 48 , 9269; (d) C. D. Königs, M. F. Müller, N. Aiguabella, H. F. T. Klare and M. Oestreich, Chem. Commun., 2013, 49 , 1506; (e) A. E. Nako, W. Chen, A. J. P. White and M. R. Crimmin, Organometallics, 2015, 34 , 4369. 5 (a) C. Bellini, J.-F. Carpentier, S. Tobisch and Y. Sarazin, Angew. Chem., Int. Ed., 2015, 54 , 7679; (b) C. Bellini, V. Dorcet, J.-F. Carpentier, S. Tobisch, Y. Sarazin, Chem.—Eur. J., 2016, 22 , 4564; (c) C. Bellini, T. Roisnel, J.-F. Carpentier, S. Tobisch and Y. Sarazin, Chem.—Eur. J., 2016, 22 , 15733. 6 O. Rivada-Wheelaghan, B. Donnadieu, C. Maya and S. Conejero, Chem.—Eur. J., 2010, 16 , 10323. 7 (a) M. Roselló-Merino, J. López-Serrano and S. Conejero, J. Am. Chem. Soc., 2013, 135 , 10910; (b) M. Roselló-Merino, R. J. Rama, J. Díez and S. Conejero, Chem. Commun., 2016, 52 , 8389. 8 (a) X.-L. Luo and R. H. Crabtree, J. Am. Chem. Soc., 1989, 111 , 2527; (b) G. J. Kubas, Adv. Inorg. Chem., 2004, 56 , 127. 9 P. Ríos, J. Díez, J. López-Serrano, A. Rodríguez and S. Conejero, Chem.—Eur., J. 2016, 22 , 16791. 10 J. Y. Yang, P. S. White, C. K. Schauer and M. Brookhart, Angew. Chem., Int. Ed., 2008, 47 , 4141. 11 O. Rivada-Wheelaghan, M. Roselló-Merino, M. A. Ortuño, P. Vidossich, E. Gutiérrez-Puebla, A. Lledós and S. Conejero, Inorg. Chem., 2014, 53 , 4257. 12 At catalyst loadings below 0.01% no bis(amino)silanes are produced in the presence of an excess of amines. We are currently investigating the reasons for this effect. 13 M. A. Ortuño, S. Conejero and A. Lledós, Beilstein J. Org. Chem., 2013, 9 , 1352. 14 Similar induction periods, associated to the need to generate a vacant site, have been observed during dehydrocoupling process of amino-boranes by Ruthenium complexes: A. E. W. Ledger, C. E. Ellul, M. F. Mahon, J. M. J. Williams and M. K. Whittlesey, Chem.—Eur. J., 2011, 17 , 8704. 15 Calculations have been performed with the Gaussian 09 program: Gaussian 09, Revision E. 01, Gaussian, Inc. Wallingford CT, 2013. 16 Overlapping of this signal with those of the phenyl groups in the Ph 2 SiH 2 and Ph 2 SiH(NEt 2 ) precluded integration of this signal. However, when a similar experiment was carried out in the reaction of Et 2 SiH 2 and NEt 2 H the relative integral of this signal is 2 with respect to the Et fragments. Significantly, the chemical shifts of species A in both experiments are identical, with independence of the silane used, hinting at A being the same in nature. 17 We have previously proved that ammonium salts can protonate complex 3 producing H 2 and 1a (see ref. 8a). 18 (a) M. C. Lipke, A. L. Liberman-Martin, T. D. Tilley, Angew. Chem., Int. Ed., 2017, 56 , 2260; (b) J. Y. Corey, Chem. Rev. 2016, 116 , 11291.