Communication pubs.acs.org/JACS
Cu-Catalyzed Hydrocarbonylative C−C Coupling of Terminal Alkynes with Alkyl Iodides Li-Jie Cheng and Neal P. Mankad* Department of Chemistry, University of Illinois at Chicago, 845 W. Taylor Street, Chicago, Illinois 60607, United States S Supporting Information *
Scheme 1. Previous and Current Work
ABSTRACT: A Cu-catalyzed hydrocarbonylative C−C coupling of terminal alkynes with unactivated alkyl iodides has been developed, enabling highly chemo- and regioselective synthesis of unsymmetrical dialkyl ketones. A variety of functional groups are tolerated, and both primary and secondary alkyl iodides react well. An autotandem sequence of two Cu-catalyzed processes is proposed: first hydrocarbonylative coupling of the alkyne and the alkyl iodide, followed by reduction of the intermediate unsaturated ketone to the saturated product. Mechanistic experiments indicate that an alkenylcopper intermediate activates the alkyl iodide by single electron transfer to enable a radical carbonylation pathway.
K
etones are ubiquitous in organic chemistry and play important roles in organic synthesis due to their versatile reactivity. The development of efficient methods for the synthesis of ketones has been a persistent area of interest for decades.1 Compared to catalytic reactions employing carboxylic acid derivatives2 or aldehydes3 in which the carbonyl group is preincorporated, transition-metal-catalyzed carbonylative C−C coupling of organic electrophiles with organometallic reagents has attracted much interest due to its modularity and its use of cheap and abundant CO as a C1 feedstock.4 However, although the carbonylative coupling reaction has been long established, several challenges still remain. For example, precious metals, especially palladium, are invariably employed in the catalysts. In addition, the stoichiometric organometallic reagents such as organoboron, organotin, or organozinc reagents often require synthesis and purification prior to cross-coupling reactions.5 Finally, and most importantly, the electrophiles are mainly limited to aryl, vinyl, allyl, and benzyl halides. The use of unactivated alkyl electrophiles is rarely reported (Scheme 1a),6,7 possibly due to their slow rate of oxidative addition as well as competitive β-hydride elimination under carbonylative conditions. The very few examples of dialkyl ketone formation by Pd-catalyzed carbonylative coupling of unactivated alkyl electrophiles either suffer from limited substrate scope6b,f or require photoirradiation.6a Thus, developing new strategies in carbonylative C−C coupling using earth-abundant catalysts and readily available starting materials to synthesize ketones, especially dialkyl ketones, is highly desirable. Alkynes are readily available compounds, and the potential involvement of their two π-bonds in multistep reaction sequences increases their flexibility as starting materials. Recently, intense efforts have been directed toward the Cu© 2017 American Chemical Society
catalyzed hydrofunctionalization of alkynes.8 The hydrocarboxylation,9 hydroboration,10 hydrobromination,11 hydroalkylation12 and hydroallylation13 of alkynes have been well developed, mainly by the Tsuji and Lalic groups. The key to these successful transformations is the regioselective antiMarkovnikov addition of a catalytic Cu−H intermediate to an alkyne, giving the corresponding alkenylcopper species that can react with various electrophiles to generate the functionalized alkenes (Scheme 1b). These alkene products are commonly inert to the Cu−H intermediates, and thus further reduction to the alkane oxidation state is typically not observed. Nevertheless, the Buchwald group did recently report the reductive amination of alkynes to synthesize alkyl amines by Cu−H catalysis.14 We hypothesized that a tandem sequence of Cu−H catalysis could serve as a platform for the synthesis of dialkyl ketones (Scheme 1c). Specifically, we envisioned that the alkenylcopper intermediate could undergo carbonylative coupling with an alkyl halide under CO atmosphere, affording an α,βunsaturated ketone that would subsequently be reduced to the saturated ketone by the same Cu catalyst.15 Although the carbonylation of alkynes has been established,16 including the Pd-catalyzed carbonylative Sonogashira coupling reaction,4,17 to the best of our knowledge the hydrocarbonylative C−C Received: May 19, 2017 Published: July 12, 2017 10200
DOI: 10.1021/jacs.7b05205 J. Am. Chem. Soc. 2017, 139, 10200−10203
Communication
Journal of the American Chemical Society coupling of alkynes with electrophiles has not been reported.18 Challenges to this approach include potentially competitive reduction of the alkyl halide19 and/or hydroalkylation of the alkyne,12 both known to be mediated by Cu−H catalysis. Nonetheless, we felt that if suitable conditions could be identified, this strategy would constitute a practical and modular way to synthesize dialkyl ketones using readily available alkynes and alkyl halides as starting materials. Herein, we report a highly chemo- and regioselective synthesis of unsymmetrical dialkyl ketones from alkynes and alkyl halides by this approach. We began our work by studying the reaction between 1decyne and 1.5 equiv of 1-iodooctane under 6 atm CO pressure. After intensive investigation,20 we found that the desired dialkyl ketone 3a could be obtained in 85% yield when IPrCuCl was used as catalyst in the presence of KOMe as the base and polymethylhydrosiloxane (PMHS) as the reducing reagent (Table 1, entry 1). A control experiment demonstrated
entry 12). As PMHS is a cheap, easy-to-handle, and environmentally friendly reducing agent, we decided to continue using 6.0 equiv of PMHS. Changing the solvent to toluene resulted in no product formation (Table 1, entry 13). It is noteworthy that, although we typically performed reactions at 60 °C and 6 atm CO pressure, conducting the reaction either at room temperature under 6 atm CO or at 60 °C under 3 atm CO could also afford the ketone product in 71% and 65% yield, respectively (Table 1, entry 14−15). Lowering the catalyst loading to 5 mol % still efficiently led to the product in 80% yield (Table 1, entry 16). With the optimal conditions in hand, we next investigated the substrate scope (Table 2). The mild reaction conditions Table 2. Substrate Scopea
Table 1. Optimizing Hydrocarbonylative Couplinga
entry
variations from optimal conditions
yield (%)b
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
None No IPrCuCl SIPrCuCl instead of IPrCuCl IMesCuCl instead of IPrCuCl Octyl-Br instead of Octyl-I NaOMe instead of KOMe LiOMe instead of KOMe KOtBu instead of KOMe (EtO)3SiH instead of PMHS (Me2HSi)2O instead of PMHS (EtO)2MeSiH instead of PMHS 3.0 equiv PMHS instead of 6.0 Toluene instead of THF r.t. instead of 60 °C 3 atm CO instead of 6 atm 5 mol % IPrCuCl instead of 10 mol %
85 0 29
