Cobalt‐Catalyzed Dehydrogenative C−H Silylation of Alkynylsilanes

Abstract Herein, we report that a cobalt catalyst permits the general synthesis of substituted alkynylsilanes through dehydrogenative coupling of alkynylsilanes and hydrosilanes. Several silylated alkynes, including di‐ and trisubstituted ones, were prepared in a one‐step procedure. Thirty‐seven compounds were synthesized for the first time by applying our catalyst system. The alkynylsilanes bearing hydrosilyl moieties provide an opportunity for further functionalization (e. g., hydrosilylation). The use of primary silanes as substrates and precatalyst activators permits the use of inexpensive and easily accessible 3d metal precatalysts, and avoids the presence of additional activators.

features of our strategy are a) sp CÀ H silylation through versatile cobalt catalysis, b) small amount of the precatalyst, c) mild reaction conditions, d) efficient transformation of several functionalized acetylenes, and e) an unprecedented dual role of hydrosilanes -as substrates and activators.
First, we examined the silylation of trimethylsilylacetylene (2 a) with phenylsilane (1 a) in the presence of several previously synthesized Co complexes A-D (Table S1 in the Supporting Information). A variety of readily available bases and commonly used solvents was examined (Tables 1 and S1). Furthermore, we have also recorded some tests in the absence of Co-precatalyst (Table 1, entry 2), and with Co-starting material ( Table 1, entry 3).
Control experiments showed the essential role of the cobalt catalyst ( Table 1, entry 2), whereas simple cobalt chloride was not active in this process (Table 1, entry 3). Moreover, it turned out, that in the presence of 0.5 mol% of A, the observed conversion was slightly lower (Table 1, entries 4). In general, other complexes B-D were significantly less active in the sp CÀ H silylation (Table 1, entry 5-7). Notably, when 1.0 equiv. of 1 a was used, the reaction was less selective (it gives 23 % of trisilyl-bis(acetylene), Table 1, entry 8). The catalytic activity of the most promising cobalt complex A was subsequently checked in the presence of alkali metal-based activators ( Table 1, entries 9-12), whereby better conversion was observed without any additive. This suggested that one of the substrates plays a dual role in the presented catalytic system and further studies have confirmed that hydrosilane was responsible for this effect. Last but not least, THF turned out to be the optimal solvent for further experimentation. At this point, it is also worth to mention, that only silyl-substituted alkynes gave satisfactory results in terms of chemoselectivity. The nonsilylated acetylenes (e. g., phenylacetylene, 4-ethynyltoluene, 4ethynylanisole, 4-ethynylbenzonitrile, 1-chloro-4-ethynylbenzene, 1-ethynyl-4-fluorobenzene, and 4-ethynyl-α,α,α-trifluorotoluene) or silylated unsaturated alcohols (e. g., 3-(trimethylsilyloxy)but-1-yne, 3-(triethylsilyloxy)but-1-yne, and tertbutyldimethyl(2-propynyloxy)-silane) led to the mixture of products and lower conversion rates (even under harsh conditions).
With the optimized conditions in hand, we investigated the scope of the sp CÀ H silylation with 1 a or n-hexylsilane (1 b), as well as commercially inaccessible p-tolylsilane (1 c; Scheme 2, top). Thus, a variety of unsymmetrical bis(silyl)acetylenes was prepared with excellent isolated yields, under mild reaction conditions, and at low loading of the precatalyst A (Scheme 2, 3 aa-ak, 3 ba-bk and 3 cc). We also probed the robustness of this Co-mediated approach by employing more challenging vinyl-substituted silylacetylene 2 j. Thus, the C=C double bond remained untouched (3 aj and 3 bj), showing the high chemoselectivity of this protocol. Additionally, this fact provides the possibility of subsequent functionalization, by using the alkene function. Given the success of the cobalt-catalyzed monodehydrogenative coupling, we wondered whether a second dehydrocoupling would be achieved (keeping in mind our previous observation, Table 1, entry 8). As the result (Table S2), we discovered that only hydrosilane 1 a can selectively lead to products with three silyl substituents. Notably, larger excess of silylacetylene (> 3.5 equiv.) caused inferior results (Table S2). However, when bulkier acetylenes were employed as the coupling partners, the desired bis(silyl)acetylenes were selectively obtained, without any traces of trisubstituted 1 a. All these results are summarized in Scheme 2 (bottom , Table S3). Next, we set out to investigate the scope for secondary hydrosilanes. Due to the beneficial effect of using a greater amount of hydrosilanes, together with elevated temperatures, we wondered whether Co-catalyzed CÀ H silylation would be viable (for detailed information, see Table S4).
To our delight, under forcing conditions (at 100°C, which also forced a change in the type of the solvent), a reaction of 2 c with diphenylsilane (1 d) took place in the presence of 10 mol% of A. Unfortunately, besides the desired product, we have also detected the redistribution of dihydrosilane in moderate quantity. Considering that at a lower temperature (RT-60°C) the redistribution was not detected, we assumed that probably primary silanes (formed at elevated temperature) [17] are still true activators of the cobalt precatalysts.
With this in mind, we examined the use of phenylsilane (1 a) as the precatalyst activator (molar ratio 1 a/A as 2.0 : 1.0) in a dehydrogenative coupling of silylacetylenes with secondary silanes under slightly milder reaction conditions. After considerable experimentation (Table S3), we found that secondary silanes 1 d-g may provide satisfactory results in reaction with 2 c, leading to the corresponding silylated acetylenes in good yields (Scheme 3). Next, we turned our attention to previously obtained alkynes bearing both alkynyl and hydro substituents. Given the success of H 3 SiPh (1 a) as the activator, we wondered whether a dehydrocoupling between sterically hindered bis(silyl) acetylenes with another silylacetylenes would be viable. For this purpose, 2 a and 3 ac were chosen as model substrates (the optimization findings are summarized in Table S5). As we could observe, the temperature value is crucial for this transformation. Thereby, we have established individual reaction conditions for each substrate, and consequently, another six examples of unsymmetrical derivatives were obtained (Scheme 4).
Finally, we have also demonstrated the utility of obtained silylacetylenes with SiH functionalities in the subsequent hydrosilylation process. A classical Karstedt-catalyzed SiÀ H addition to unsaturated systems constitutes a very elegant synthetic pathway in preparation of functionalized organosilicons, [18] and in our case, it enabled the synthesis of novel multifunctional silylacetylenes (Scheme 5).
To gain mechanistic insights into this Co-catalyzed reaction, [19] we conducted preliminary experiments. Firstly, 2.2 equiv. of

Chemistry-A European Journal
Research Article doi.org/10.1002/chem.202103629 1 a were added to 1 equiv. of precatalyst B in [D 8 ]THF and stirred at 50°C for 1 h. The reaction was tracked on 1 H NMR, initially giving the mixture of [CoÀ H] species, as well as PhSiH 2 Cl and dihydrogen (but without any phenylsilane residues; Supplement 2 in the Supporting Information). Next, we increased the amount of 1 a to 10 equiv., which exclusively resulted in the formation of one [CoÀ H] entity (Supplement 3 in the Supporting Information), indicating the generation of (PN5P) Co III H 2 (SiH 2 Ph) (Scheme 6). [19c] Notably, such Co I /Co III mechanism was also suggested in other TM-catalyzed CÀ H activation processes. [18b,d] Furthermore, a dehydrogenative coupling between 1 a and 2 a was performed in the presence of TEMPO (1 equiv.), and led to the expected product (90 %), thereby implying that radical pathway is likely, not operative.
A plausible catalytic cycle based on previous literature and our experimental results is presented in Scheme 7.
The cobalt complex (PN5P)Co III H 2 (SiH 2 Ph) undergoes ligand replacement with the silylacetylene molecule, with simultaneous liberation of a dihydrogen molecule. Consequently, the alkynyl cobalt complex is generated, in the form of two possible isomers (Supplement 4 in the Supporting Information). [20] Finally, the reaction proceeds by reductive elimination to afford the final product and regenerates the active cobalt(I) catalyst.
In conclusion, we have reported the selective sp CÀ H silylation of silylacetylenes with primary and secondary hydrosilanes by using cobalt catalysis. Under environmentally benign reaction conditions, a series of symmetrical and unsymmetrical silylacetylenes (44 compounds, including products of hydrosilylation) were synthesized in good to excellent yields (up to 99 %). Considering the combination of desirable features, such as high chemoselectivity, high atom economy, benign reaction conditions, and the use of a 3d-metal catalyst, this reaction system is expected to provide a promising alternative to existing methodologies and an attractive approach for the synthesis of complex organosilicon compounds. The use of primary silanes as substrates and precatalyst activators is also beneficial as additional activators can be avoided. Mechanistic studies provided strong support for the involvement of Co I /Co III pathway.