Nickel-Catalyzed β,γ-Dicarbofunctionalization of Alkenyl Carbonyl

Communication pubs.acs.org/JACS

Nickel-Catalyzed β,γ-Dicarbofunctionalization of Alkenyl Carbonyl Compounds via Conjunctive Cross-Coupling Joseph Derosa,‡ Van T. Tran,‡ Mark N. Boulous, Jason S. Chen, and Keary M. Engle* Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, United States S Supporting Information *

Scheme 1. Background and Synopsis of Current Work

ABSTRACT: A nickel-catalyzed conjunctive cross-coupling between non-conjugated alkenes, aryl iodides, and alkylzinc reagents is reported. Excellent regiocontrol is achieved utilizing an 8-aminoquinoline directing group that can be readily cleaved to unmask net β,γ-dicarbofunctionalized carboxylic acid products. Under optimized conditions, both terminal and internal alkene substrates provided the corresponding alkyl/aryl difunctionalized products in moderate to excellent yields. The methodology developed herein represents the first three-component 1,2dicarbofunctionalization of non-conjugated alkenes involving a C(sp3)−C(sp3) reductive elimination step.

S

ince their inception, transition-metal-catalyzed cross-coupling reactions have emerged as invaluable tools for forging C−C bonds in complex molecule synthesis.1 Recent advances have enabled efficient C(sp3)−C(sp2) bond formation, often utilizing privileged ligand scaffolds to facilitate oxidative addition/reductive elimination and prevent otherwise rapid βhydride elimination.2 Alkyl−alkyl C(sp3)−C(sp3) cross-coupling is arguably an even more difficult challenge and has been actively studied during the past several years.3−5 In this context, Fu has elegantly utilized alkyl electrophiles containing chelating groups to promote such transformations.4 While two-component crosscouplings to achieve alkyl−alkyl bond formation have been vigorously pursued, the development of a practically useful threecomponent variant remains an unmet challenge. Specifically, we were attracted to the notion that a non-conjugated alkene containing a coordinating group could be used as a conjunctive reagent3d,6 to facilitate the union of an organometallic nucleophile and an organohalide electrophile. Such a reaction would provide a powerful method for vicinal dicarbofunctionalization that would enable rapid buildup of molecular complexity and would rely on a key C(sp3)−C(sp3) bond-forming step. Previous research on three-component cross-coupling reactions with C−C π bonds has generally employed conjugated alkenes (e.g., styrenes and acrylates),3d,7 allenes,8 1,3-dienes,9 or alkynes.10 In these catalytic processes, regioselectivity is controlled by the innate polarization of the π-system and/or the resulting stability of the organometallic intermediate. Such systems either lack β-hydrogen atoms or proceed via formation of a π-allyl/benzyl intermediate. In the context of alkyne 1,2difunctionalization, Larock and co-workers reported a pioneering example of 1,2-diarylation of internal alkynes with arylboronic acids and aryl halides using catalytic palladium (Scheme 1A).10a In addition to the examples cited above, Sigman has reported 1,2© 2017 American Chemical Society

diarylation of terminal alkenes to introduce the same aryl group at both positions.11 Relevant intramolecular two-component alkene 1,2-dicarbofunctionalization reactions have also been reported, in which the alkyl halide electrophile12 or carbon nucleophile13 is intramolecularly tethered to the alkene. Our group has previously found that, when intramolecularly tethered to non-conjugated alkenes, removable bidentate directing groups can effectively control regiochemical outcomes and prevent β-hydride elimination in alkene hydrofunctionalization and 1,2-dicarbofunctionalization reactions using palladium(II) as catalyst.14,15 Specifically, in this latter category, we recently developed an anti-β,γ-vicinal-dicarbofunctionalization reaction via Pd(II)/Pd(IV) catalysis that facilitates addition of indole, 2-naphthol, and 1,3-dicarbonyl nucleophiles to the γ-position and C(sp2)/C(sp) carbon electrophiles at the β-position.15 In addition to the limited nucleophile scope of this reaction, isomerization of acyclic internal alkenes was observed, leading to mixtures of diastereoisomers, greatly limiting the overall utility of this reaction. We thus questioned whether it would be possible to develop a conjunctive cross-coupling that would proceed with opposite polarity. We envisioned that such a reaction would enable access to syn-difunctionalized products and would potentially allow for C(sp3)−C(sp3) coupling between an intermediate organometallic species and an alkyl organometallic nucleophile. Though such a reaction could potentially be achieved via a Pd(0)/Pd(II) Received: June 24, 2017 Published: July 24, 2017 10657

DOI: 10.1021/jacs.7b06567 J. Am. Chem. Soc. 2017, 139, 10657−10660

Communication

Journal of the American Chemical Society Table 1. Optimization of β,γ-Dicarbofunctionalizationa

Table 2. Aryl/Vinyl Electrophile and Alkyl/Aryl Nucleophile Scopea

a Reaction conditions: 1a (0.25 mmol), iodoarene/vinyl iodide (0.5 mmol), organozinc reagent (2 equiv). Percentages represent isolated yields. bAlkylzinc halide (3 equiv).

a

Reaction conditions: diethylzinc (1.0 M in NMR analysis using otherwise; n.d. = not M in hexanes).

operations were performed outside of the glovebox (entry 4), providing an alternative procedure that may be more appealing for some end-users. Yields were generally superior with Ni(cod)2, however, so we elected to focus on this precatalyst for the remainder of the investigation (entry 9). As expected, a control experiment showed that the reaction did not occur in the absence of nickel catalyst (entry 10). Though many of the directing groups that were tested proved ineffective throughout the optimization process, both the oxazoline and thioether directing groups showed promising reactivity (entries 16 and 17). Interestingly, using 1 equiv of the dialkylzinc reagent resulted in only 8% yield, suggesting that the N−H bond consumes the first of the 2 equiv needed for the optimized reaction conditions. We first examined the scope of aryl electrophiles and organozinc nucleophiles for this regioselective β,γ-dicarbofunctionalization reaction (Table 2). We were satisfied to find that both electron-rich and electron-deficient arenes gave the desired product in good to excellent yield (2a−2f). Additionally, we were pleased to observe that the reaction appeared to be chemoselective for C(sp2)−I bond cleavage over C(sp2)−Br bond cleavage, allowing for potential downstream diversification (2b). Heterocycles such as thiophene and pyridine were tolerated, though the latter required steric blocking of the coordinating nitrogen atom, providing the corresponding heteroarylation products in modest yields (2g, 2h). Both orthofluoro- and -methoxy-substituted arenes were also found to be competent electrophiles under the optimized reaction conditions (2e, 2f). Although in general alkenyl iodides are not suitable reaction partners (potentially due to secondary reactions that can take place with the corresponding products), we found that cyclohexenyl iodide was a competent electrophile, delivering 2i in 61% yield. We then set out to test the nucleophile scope by substituting diethylzinc with other alkylzinc nucleophiles (2j− 2o). Gratifyingly, we observed the formation of the desired products in high yields using dimethylzinc (2j) and diphenylzinc

alkene (0.1 mmol), iodobenzene (0.2 mmol), hexanes, 2 equiv). bYields determined by 1H CH2Br2 as internal standard unless noted detected. cIsolated yield. d1 equiv ZnEt2 (1.0

cycle, we were attracted to the use of nickel as the catalyst, given its efficacy in promoting C(sp3)−C(sp3) couplings. We set out to reduce our hypothesis to practice by optimizing reaction conditions and the directing group structure in parallel. To begin, we selected a collection of representative mono- and bidentate directing groups. Notably, Chatani has previously found that N,N-bidentate directing groups enable C(sp2)−H and C(sp3)−H activation under nickel catalysis.16,17 In order to achieve the desired carbonickelation/C(sp3)−C(sp3) crosscoupling sequence, we elected to screen using diethylzinc as the nucleophile and iodobenzene as the electrophile (Table 1). At the outset, we anticipated that Negishi coupling between diethylzinc and iodobenzene could potentially be competitive with the desired three-component reaction, but in practice this could be avoided through choice of appropriate reaction conditions. In a series of initial experiments, we were pleased to observe that the desired product was formed in 49% yield with only 10 mol% Ni(cod)2 at room temperature using Daugulis’s 8aminoquinoline (AQ) directing group (entry 1). Interestingly, the reaction also resulted in the formation of a 1,2-diarylation product in modest yield (30% by 1H NMR), suggesting a potential competitive pathway from the putative nickelacycle intermediate (see Supporting Information (SI)). This byproduct was effectively suppressed by using dioxane as solvent and increasing the temperature to 50 °C, yielding the desired product in 88% isolated yield (entry 7). We were also pleased to find that NiCl2·glyme provided a moderate yield in DMF, even when all 10658

DOI: 10.1021/jacs.7b06567 J. Am. Chem. Soc. 2017, 139, 10657−10660

Communication

Journal of the American Chemical Society Table 3. Alkene Substrate Scope

Scheme 2. Multi-Gram-Scale Reaction and AQ Removal

Scheme 3. Proposed Catalytic Cycle

a Reaction conditions: 1b−1k (0.25 mmol), iodobenzene (0.5 mmol), dialkylzinc solution (2 equiv). Percentages represent isolated yields. Stereochemical outcomes were assigned by analogy to 3c and 3i, which were determined by X-ray crystallography.

access to a cyclopentane stereotriad that would otherwise be difficult to prepare. Given our success with β,γ-dicarbofunctionalization, we then sought to access γ,δ-dicarbofunctionalized products by subjecting a pentenoic acid-derived substrate, which contains one additional methylene spacer, to the reaction conditions; however, in this case we were only able to detect isomerization and a mixture of diarylation products. This result speaks to the importance of the 5-membered nickelacycle intermediate in productive catalysis. To highlight the practicality of this nickel-catalyzed β,γdicarbofunctionalization reaction, the procedure was carried out on multi-gram scale to yield 2.1 g of 2h (82%), which was readily deprotected in nearly quantitative yield to provide 1.2 g of 4 (98%) (Scheme 2).19 A plausible mechanism for this nickel-catalyzed β,γ-dicarbofunctionalization reaction is proposed below (Scheme 3). Initially, the active nickel(0) species, which could potentially already be bound to the directing group, performs oxidative addition on the aryl iodide. The resulting nickel(II) oxidative addition complex coordinates the alkene moiety and undergoes a regioselective 1,2-insertion to yield 5-membered, AQ-chelated intermediate A. At this stage, an organozinc reagent can intercept A in a transmetalation event to yield intermediate B, which can then undergo C(sp3)−C(sp3) reductive elimination to provide the desired product and regenerate nickel(0) to close the cycle. To gain a better understanding of relative rates of the desired three-component coupling and the potential alternative Negishi coupling reaction, we performed the reaction of 4-iodoanisole and diethylzinc in the absence of alkene (see SI). The reaction yielded 10% of the Negishi coupling product, 70% iodoarene starting material, and 20% reduced arene. When a surrogate AQbased ligand was added, formation of the reduction byproduct was suppressed, but the Negishi byproduct was still only formed in 30% yield. These data suggest that, under the optimized conditions, Negishi cross-coupling is substantially slower than the three-component reaction. The origins of this phenomenon are currently being probed in more detail. In conclusion, we have developed a highly regioselective, three-component conjunctive cross-coupling between organozinc nucleophiles, aryl electrophiles, and non-conjugated terminal and internal alkenes by employing a removable

(2m). Additionally, the reaction is also compatible with monoalkylzinc halides (2k, 2l, 2n, and 2o), which are milder, more stable, and easier to access. We were delighted to obtain cyclopropylated product 2n, establishing the viability of secondary−secondary C(sp3)−C(sp3) coupling; however, more hindered secondary alkyl nucleophiles were not tolerated in the reaction. For example, when diisopropylzinc was subjected to the reaction conditions, we observed only a mixture of monoand diarylation byproducts. This could be due to the increased number of accessible β-hydrogen atoms on the transmetalating agent, promoting undesired β-hydride elimination. Alternatively, the increased steric demands could suppress the rate of transmetalation onto the relatively hindered nickelacycle (vide inf ra). Another limitation of this method at present is incompatibility with coordinating heteroatoms, such as free alcohols or amines, which we hypothesize can bind the nickel species, preventing successive turnovers. Next, we examined the scope of non-conjugated α-substituted terminal alkenes and internal alkenes (1b−1k) (Table 3). To probe the effect of alkene structure on reactivity, we used iodobenzene as the electrophile and either dimethyl- or diethylzinc as the nucleophile. To our delight, α-substituted substrates were reactive, providing the desired products in moderate yields (3a−3d). Notably, with mono-α-substituted substrates, increasing the steric bulk of the α-substituent led to significantly increased diastereomeric ratios, though this came at the expense of yield.18 The relative stereochemistry of the major diastereomer of 3c was established by X-ray crystallography. The trans orientation of the α and β substituents presumably arises from formation of the more stable trans-nickelacycle upon carbonickelation.15 As for internal substrates, we were pleased to observe high yields of only the syn-diastereomer under the optimal reaction conditions (3e−3j). The syn selectivity of the reaction was established by analysis of the X-ray crystal structure of 3i. Notably, both diastereomers could be accessed based on the stereochemistry of the alkene substrate (i.e., E to 3g, Z to 3h), suggesting that this nickel-catalyzed process does not isomerize the starting material. A pendant phthalimide-protected amine was also tolerated in the reaction (3j). Interestingly, a cyclopentenyl substrate was viable in the reaction (3k), providing 10659

DOI: 10.1021/jacs.7b06567 J. Am. Chem. Soc. 2017, 139, 10657−10660

Communication

Journal of the American Chemical Society

S.; Maxwell, B. D.; Eastgate, M. D.; Baran, P. S. Science 2016, 352, 801− 805. (4) For examples of C(sp3)−C(sp3) cross-coupling using chelating electrophiles, see: (a) Owston, N. A.; Fu, G. C. J. Am. Chem. Soc. 2010, 132, 11908−11909. (b) Lu, Z.; Wilsily, A.; Fu, G. C. J. Am. Chem. Soc. 2011, 133, 8154−8157. (c) Zultanski, S. L.; Fu, G. C. J. Am. Chem. Soc. 2011, 133, 15362−15364. (d) Wilsily, A.; Tramutola, F.; Owston, N. A.; Fu, G. C. J. Am. Chem. Soc. 2012, 134, 5794−5797. (5) For reviews of C(sp3)−C(sp3) cross-coupling, see: (a) Netherton, M. R.; Fu, G. C. Adv. Synth. Catal. 2004, 346, 1525−1532. (b) Frisch, A. C.; Beller, M. Angew. Chem., Int. Ed. 2005, 44, 674−688. (c) Rudolph, A.; Lautens, M. Angew. Chem., Int. Ed. 2009, 48, 2656−2670. (d) Jana, R.; Pathak, T. P.; Sigman, M. S. Chem. Rev. 2011, 111, 1417−1492. (6) (a) Zhang, L.; Lovinger, G. J.; Edelstein, E. K.; Szymaniak, A. A.; Chierchia, M. P.; Morken, J. P. Science 2016, 351, 70−74. During preparation of this manuscript, an example of reductive 1,2dicarbofunctionalization of alkenes was reported: (b) GarcíaDomínguez, A.; Li, Z.; Nevado, C. J. Am. Chem. Soc. 2017, 139, 6835−6838. (7) For syn- or anti-selective carboboration of styrenyl alkenes, see: Logan, K. M.; Smith, K. B.; Brown, M. K. Angew. Chem., Int. Ed. 2015, 54, 5228−5231. (8) Huang, T.-H.; Chang, H.-M.; Wu, M.-Y.; Cheng, C.-H. J. Org. Chem. 2002, 67, 99−105. (b) Aftab, T.; Grigg, R.; Ladlow, M.; Sridharan, V.; Thornton-Pett, M. Chem. Commun. 2002, 1754−1755. (c) Shu, W.; Jia, G.; Ma, S. Angew. Chem., Int. Ed. 2009, 48, 2788−2791. (9) Liao, L.; Jana, R.; Urkalan, K. B.; Sigman, M. S. J. Am. Chem. Soc. 2011, 133, 5784−5787. (10) (a) Zhou, C.; Emrich, D. E.; Larock, R. C. Org. Lett. 2003, 5, 1579−1582. (b) Xue, F.; Zhao, J.; Hor, T. S. A.; Hayashi, T. J. Am. Chem. Soc. 2015, 137, 3189−3192. (c) Li, Z.; Garcı ́a-Domı ́nguez, A.; Nevado, C. Angew. Chem., Int. Ed. 2016, 55, 6938−6941. (d) Yada, A.; Yukawa, T.; Nakao, Y.; Hiyama, T. Chem. Commun. 2009, 3931−3933. (11) Urkalan, K. B.; Sigman, M. S. Angew. Chem., Int. Ed. 2009, 48, 3146−3149. (12) (a) Wakabayashi, K.; Yorimitsu, H.; Oshima, K. J. Am. Chem. Soc. 2001, 123, 5374−5375. (b) Powell, D. A.; Maki, T.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 510−511. (c) Phapale, V. B.; Buñuel, E.; GarcíaIglesias, M.; Cárdenas, D. J. Angew. Chem., Int. Ed. 2007, 46, 8790−8795. (13) (a) Cong, H.; Fu, G. C. J. Am. Chem. Soc. 2014, 136, 3788−3791. (b) You, W.; Brown, M. K. J. Am. Chem. Soc. 2014, 136, 14730−14733. (c) You, W.; Brown, M. K. J. Am. Chem. Soc. 2015, 137, 14578−14581. (14) (a) Zaitsev, V. G.; Shabashov, D.; Daugulis, O. J. Am. Chem. Soc. 2005, 127, 13154−13155. (b) Shabashov, D.; Daugulis, O. J. Am. Chem. Soc. 2010, 132, 3965−3972. (15) Liu, Z.; Zeng, T.; Yang, K. S.; Engle, K. M. J. Am. Chem. Soc. 2016, 138, 15122−15125. (16) (a) Shiota, H.; Ano, Y.; Aihara, Y.; Fukumoto, Y.; Chatani, N. J. Am. Chem. Soc. 2011, 133, 14952−14955. (b) Yokota, A.; Aihara, Y.; Chatani, N. J. Org. Chem. 2014, 79, 11922−11932. (c) Aihara, Y.; Chatani, N. J. Am. Chem. Soc. 2014, 136, 898−901. (d) Aihara, Y.; Wuelbern, J.; Chatani, N. Bull. Chem. Soc. Jpn. 2015, 88, 438−446. (e) Misal-Castro, L. C.; Obata, A.; Aihara, Y.; Chatani, N. Chem. - Eur. J. 2016, 22, 1362−1367. (f) Aihara, Y.; Chatani, N. ACS Catal. 2016, 6, 4323−4329. (17) For reviews on bidentate directing groups in C−H activation, see: (a) Rouquet, G.; Chatani, N. Angew. Chem., Int. Ed. 2013, 52, 11726− 11743. (b) Daugulis, O.; Roane, J.; Tran, L. D. Acc. Chem. Res. 2015, 48, 1053−1064. (18) With an enantiopure sample of 3d, no evidence of epimerization was found during the reaction (see SI). (19) With more hindered substrates, such as 3c, this hydrolysis procedure led to erosion of d.r. and low conversion to product. An alternative protocol using Cp2ZrHCl to reduce the AQ amide to the corresponding aldehyde was found to be effective, though partial epimerization was observed (see SI). For previous use of this method, see: Chapman, L. M.; Beck, J. C.; Wu, L.; Reisman, S. E. J. Am. Chem. Soc. 2016, 138, 9803−9806.

bidentate AQ directing group. By utilizing the chelation control imparted by the AQ auxiliary, the putative nickel(II) intermediate was stabilized to enable C(sp3)−C(sp3) crosscoupling without significant β-hydride elimination, providing a powerful strategy for β,γ-dicarbofunctionalization. The reaction was found to proceed with a broad range of aryl electrophiles and tolerated both alkyl and aryl nucleophiles. After the reaction, the AQ group could be easily removed via hydrolysis, illustrating the potential utility of this method for practitioners. Work seeking to elucidate the reaction mechanism and to expand the nucleophile and electrophile scope is currently underway and will be reported in due course.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b06567. Experimental details, graphical guide, analytical data for new compounds, and 1H and 13C NMR spectra (PDF) X-ray crystallographic data for 3c (CIF) X-ray crystallographic data for 3i(CIF) NMR spectra in MNova format (ZIP)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Keary M. Engle: 0000-0003-2767-6556 Author Contributions ‡

J.D. and V.T.T. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by TSRI, Pfizer, Inc., and the Donald E. and Delia B. Baxter Foundation (Young Faculty Award to K.M.E.). We thank Prof. Arnold L. Rheingold and Dr. Milan Gembicky (UCSD) for X-ray crystallographic analysis and Dr. Dee-Hua Huang and Dr. Laura Pasternack for assistance with NMR spectroscopy. We further thank Dr. Josep Cornella (MaxPlanck-Institut für Kohlenforschung), Jacob T. Edwards (Baran lab, TSRI), and Tyler G. Saint-Denis (Yu lab, TSRI) for support and encouragement.

■

REFERENCES

(1) For reviews on transition-metal-catalyzed cross-coupling reactions, see: (a) Johansson-Seechurn, C. C. C.; Kitching, M. O.; Colacot, T. J.; Snieckus, V. Angew. Chem., Int. Ed. 2012, 51, 5062−5085. (b) Tasker, S. Z.; Standley, E. A.; Jamison, T. F. Nature 2014, 509, 299−309. (c) Cahiez, G.; Moyeux, A. Chem. Rev. 2010, 110, 1435−1462. (2) For representative examples of palladium- and nickel-catalyzed C(sp3)−C(sp2) cross-coupling reactions, see: (a) Dreher, S. D.; Lim, S. E.; Sandrock, D. L.; Molander, G. A. J. Org. Chem. 2009, 74, 3626−3631. (b) Joshi-Pangu, A.; Wang, C.-Y.; Biscoe, M. R. J. Am. Chem. Soc. 2011, 133, 8478−8481. (c) Mlynarski, S. N.; Schuster, C. H.; Morken, J. P. Nature 2014, 505, 386−390. (3) For representative examples of C(sp3)−C(sp3) cross-coupling, see: (a) Ishiyama, T.; Abe, S.; Miyaura, N.; Suzuki, A. Chem. Lett. 1992, 21, 691−694. (b) Devasagayaraj, A.; Stüdemann, T.; Knochel, P. Angew. Chem., Int. Ed. Engl. 1996, 34, 2723−2725. (c) Netherton, M. R.; Dai, C.; Neuschutz, K.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 10099−10100. (d) Qin, T.; Cornella, J.; Li, C.; Malins, L. R.; Edwards, J. T.; Kawamura, 10660

DOI: 10.1021/jacs.7b06567 J. Am. Chem. Soc. 2017, 139, 10657−10660