Copper‐Catalyzed Cross‐Coupling Reaction of Organoboron Compounds with Primary Alkyl Halides and Pseudohalides

Non-activated alkyl electrophiles, including alkyl iodides, bromides, tosylates, mesylates, and even chlorides, underwent copper-catalyzed cross-coupling with aryl boron compounds and alkyl 9-BBN reagents (see scheme; 9-BBN=9-borabicyclo[3.3.1]nonane). The reactions proceed with practically useful reactivities and thus complement palladium- and nickel-catalyzed Suzuki–Miyaura coupling reactions of alkyl halides. Cross-coupling reactions of organocopper reagents with alkyl halides are among the most useful CC bond-forming reactions in organic synthesis.1 One drawback of these reactions has been the need to use an excess of the organocopper reagent for high conversion of the starting material.2 This requirement causes serious waste of the precious organometallic species, in particular in the case of cuprate reagents (R2CuLi), which contain two transferable R groups.3 To solve this problem, several research groups have developed copper-catalyzed alkylation reactions of Grignard reagents.3–5 It was found that these catalytic reactions are much easier to carry out and considerably less expensive. Additives, such as 1-methyl-2-pyrrolidinone and 1-phenylpropyne, were originally used to promote the alkylation reactions,4 and the slow addition of the Grignard reagent was found to provide similar results in a recent study.5 Although primary alkyl halides (as well as mesylates and tosylates) are mostly used in these catalytic alkylation reactions, some studies have indicated that secondary and even tertiary alkyl halides may also be possible coupling partners.3, 6 Our interest in copper-catalyzed cross-coupling7 prompted us to consider the possibility of using organoboron compounds in such alkylation reactions. The advantages of replacing Grignard reagents with organoboron reagents are well-known, including the better commercial availability of the reagents and the higher functional-group tolerance.8 However, whereas certain complexes of Pd,9 Ni,10, 11 and Fe12 have been reported to be competent catalysts for the coupling of alkyl halides with organoboron compounds, until now the catalysis of such transformations by Cu complexes has not been reported.13 Therefore, we were surprised to find that in the presence of LiOtBu as a base, CuI could efficiently catalyze the cross-coupling of aryl boronate esters with primary halides and pseudohalides containing a CX bond (X=I, Br, Cl, OTs, and OMs; Ts=p-toluenesulfonyl, Ms=methanesulfonyl). This new transformation not only expands the concept and utility of copper-catalyzed alkylation reactions in a fundamental sense, but also provides practically useful reactivities. It may thus complement palladium- and nickel-catalyzed Suzuki–Miyaura coupling reactions of alkyl halides.9–11 We began our study by examining the coupling of n-dodecyl p-toluenesulfonate with an aryl boronate ester 1. We initially used CuI as the catalyst and tested several different bases (Table 1, entries 1–7). Although the yield of the desired product was very low with most bases, LiOtBu provided a good yield of 83 % at 110 °C (Table 1, entry 7). Interestingly, when the temperature was lowered to 60 °C, the yield increased slightly to 87 % (Table 1, entry 8). The reaction also proceeded at room temperature, but more slowly (Table 1, entry 9). The use of other Li bases, Cu catalysts, and solvents did not improve the reaction (Table 1, entries 10–17). Other organoboron substrates, 2–4, could also be used (Table 1, entries 18–20). Alkyl iodides, bromides, mesylates, and even chlorides are also acceptable coupling partners (Table 1, entries 21–24). The amount of the CuI catalyst could be decreased to 2 mol % (Table 1, entry 25); however, no product was formed with Pd and Ni salts (Table 1, entries 26 and 27). Thus, we could rule out the possible involvement of Pd or Ni contamination in the catalysis. In the absence of a catalyst, the reaction did not occur (Table 1, entry 28). The reaction was also sensitive to water (Table 1, entry 29). Finally, under the present conditions, the reaction of cyclohexyl p-toluenesulfonate with 1 gave only a trace amount of the desired product. Thus, with this catalyst system, secondary alkyl electrophiles do not undergo the coupling reaction. Entry X Catalyst (10 mol %) Base Solvent T [°C] t [h] Yield [%][a] 1 OTs CuI Cs2CO3 DMF 110 12 6 2 OTs CuI K3PO4 DMF 110 12 7 3 OTs CuI CsOAc DMF 110 12 trace 4 OTs CuI NaN(SiMe3)2 DMF 110 12 10 5 OTs CuI KOtBu DMF 110 12 33 6 OTs CuI NaOtBu DMF 110 12 35 7 OTs CuI LiOtBu DMF 110 12 83 8 OTs CuI LiOtBu DMF 60 12 87 9 OTs CuI LiOtBu DMF RT 24 38 (76[b]) 10 OTs CuI LiOMe DMF 60 12 54 11 OTs CuI LiOEt DMF 60 12 47 12 OTs CuI Li2CO3 DMF 60 12 8 13 OTs CuBr LiOtBu DMF 60 12 54 14 OTs CuCl LiOtBu DMF 60 12 20 15 OTs Cu(OTf)2 LiOtBu DMF 60 12 4 16 OTs CuI LiOtBu DMSO 60 12 67 17 OTs CuI LiOtBu PhMe 60 12 18 18[c] OTs CuI LiOtBu DMF 60 12 76 19[d] OTs CuI LiOtBu DMF 60 12 70 20[e] OTs CuI LiOtBu DMF 60 12 65 21 I CuI LiOtBu DMF RT 12 78 22 Br CuI LiOtBu DMF 80 12 69 23 Cl CuI LiOtBu DMF 110 12 56 24 OMs CuI LiOtBu DMF 110 12 71 25 OTs CuI[f] LiOtBu DMF 60 24 83 26 OTs Pd(OAc)2[f] LiOtBu DMF 60 12 0 27 OTs NiI2[f,g] LiOtBu DMF 60 12 0 28 OTs – LiOtBu DMF 60 12 0 29 OTs CuI[h] LiOtBu DMF 60 12 14 The above results indicate that CuI is a competent catalyst for the cross-coupling of organoboron reagents with primary halides and pseudohalides. To test the scope of this method, we examined a variety of substrates (Table 2). We found that alkyl halide and pseudohalide reagents with different chain lengths and branching could participate in the reaction (to give 5 aa–5 af). A variety of substituents were well-tolerated on the alkyl reagent, including an olefin (product 5 ag), an ether (product 5 ah), an aryl group (product 5 aj), an ester (product 5 ak), a cyano group (product 5 al), and an amide (product 5 am). This feature compares favorably with previous examples of the copper-catalyzed coupling of primary alkyl halides with Grignard reagents, in which nearly all reported alkyl groups were hydrocarbons.3–5 Furthermore, for the synthesis of 5 ab, we changed the catalyst from CuI to CuOtBu generated in situ through the equimolar reaction of anhydrous CuI with LiOtBu in THF at room temperature under nitrogen. The yield with CuOtBu was 90 %, which is slightly higher than that observed with CuI (86 %). Finally, moderate yields were observed when alkyl chlorides were used as substrates (for the synthesis of 5 ab, 5 bf, 5 bs, and 5 bt). The reaction also tolerates an array of substituents on the organoboron substrate (products 5 ba–bl). Importantly, arylX (X=F, Cl, Br, I) bonds (in products 5 be–bg, 5 ad, and 5 af) do not interfere with the transformation, so that additional cross-coupling reactions at the halogenated positions are possible. Reactive groups that were previously not compatible with Grignard reagents (i.e. ester (5 bh, 5 bj), cyano (5 bi), amide (5 bl), and even nitro groups (5 bk)) can be present in this new reaction. Furthermore, steric hindrance at the ortho position is tolerated to some extent (products 5 bm, 5 bn). Naphthyl, vinyl, and heteroaryl boronate esters can also participate in the reaction (products 5 bo–bs). Under the present conditions, alkyl boronate esters are not viable substrates. However, when alkyl 9-BBN reagents (9-BBN=9-borabicyclo[3.3.1]nonane) were used as nucleophiles, the copper-catalyzed coupling reaction also proceeded to form a sp3sp3 bond (Table 3). Both alkyl tosylates and alkyl bromides could be used in the reaction. The products were formed in moderate yields. In Table 1 it is shown that for different leaving groups, the required reaction temperature is not the same. To further compare the reactivity trend, we conducted the competitive experiments described in Table 4. According to the results, alkyl iodides are the most reactive electrophiles (Table 4, entry 1), and alkyl bromides are slightly more reactive than alkyl tosylates (Table 4, entry 2). On the other hand, alkyl tosylates (and to a lesser extent, alkyl mesylates) are much more reactive than alkyl chlorides (Table 4, entries 3 and 4). These results are consistent with previous results for the copper-catalyzed coupling of Grignard reagents with alkyl electrophiles and confirm that the reactivity increases in the order: chloride