Carbon Dioxide-Mediated C(sp3)–H Arylation of Amine Substrates

Communication Cite This: J. Am. Chem. Soc. 2018, 140, 6818−6822

pubs.acs.org/JACS

Carbon Dioxide-Mediated C(sp3)−H Arylation of Amine Substrates Mohit Kapoor, Daniel Liu, and Michael C. Young* Department of Chemistry and Biochemistry, School of Green Chemistry and Engineering, University of Toledo, Toledo, Ohio 43606, United States S Supporting Information *

featured either activation of more reactive C(sp2)−H bonds9 or intramolecular cyclization.10 We present herein an alternative strategy for overcoming these barriers using carbon dioxide to mediate C−H activation. Recent work has suggested that in situ protection of the amine as an ammonium ion can be used to protect it from oxidation, thereby facilitating remote C(sp3)−H oxidation reactions by deactivation,11 while a recent report even suggests 1° amine- or ammonium-directed C(sp3)−H oxidation.12 However, 1° and 2° amines generally require the presence of a directing group (DG) to facilitate C−C bond-forming C(sp3)−H activation.13 The only exception to this is when 2° amine substrates bearing fully substituted α,α′-centers are used.14 The strategy can also be applied to primary β-amino alcohols by first converting them into highly substituted aminoketals, although neither strategy enjoys a particularly broad substrate scope.15 The traditional method to facilitate DG-mediated C(sp3)−H arylation of amines has been to use static functional groups such as amides with or without a second chelating moiety (Scheme 1b),16 while a recent trend has focused on transient

ABSTRACT: Elaborating amines via C−H functionalization has been an important area of research over the past decade but has generally relied on an added directing group or sterically hindered amine approach. Since freeamine-directed C(sp3)−H activation is still primarily limited to cyclization reactions and to improve the sustainability and reaction scope of amine-based C−H activation, we present a strategy using CO2 in the form of dry ice that facilitates intermolecular C−H arylation. This methodology has been used to enable an operationally simple procedure whereby 1° and 2° aliphatic amines can be arylated selectively at their γ-C−H positions. In addition to potentially serving as a directing group, CO2 has also been demonstrated to curtail the oxidation of sensitive amine substrates.

A

mines are a key functional group in pharmaceuticals1 and agrochemicals2 (Figure 1) as well as materials3 and have

Scheme 1. Strategies for Amine-Directed C(sp3)−H Arylation

Figure 1. Biologically active compounds bearing a γ-arylamine motif.

DGs (Scheme 1c).17 Though each can be used to facilitate C(sp3)−H activation reactions, both possess disadvantages that might be alleviated. Static DGs require additional stoichiometric reagents and are both atom- and step-uneconomical. Meanwhile, transient DGs, which are generally aldehyde-based, suffer from the presence of oxidation-sensitive imines as intermediates. Furthermore, static DGs are rarely used for C−H activation of 2° amine substrates, while transient DG methods are nonexistent. To circumvent the challenges of these two strategies, we sought a hybrid DG strategy (Scheme 1d). Carbon dioxide was

been the subject of numerous synthetic approaches.4 Although transition-metal-catalyzed C−H activation has revolutionized the installation of functional groups at otherwise inert C−H bonds,5 amines are still a complicated substrate class for this strategy because of their reactivity. Primary (1°) and secondary (2°) amines are especially sensitive to oxidation, making them a challenge for organometallic reactions:6 palladium, for example, is well-known for its ability to oxidize 1° and 2° amines.7 Furthermore, amines can react with organometallics to produce substituted amines, such as in the Buchwald−Hartwig coupling.8 Despite these challenges, there are examples of palladium-catalyzed C−H activation of both 1° and 2° freeamine substrates in the literature, yet these have generally © 2018 American Chemical Society

Received: May 14, 2018 Published: May 22, 2018 6818

DOI: 10.1021/jacs.8b05061 J. Am. Chem. Soc. 2018, 140, 6818−6822

Communication

Journal of the American Chemical Society

CO2 (see the Supporting Information (SI)), we determined that the reactions could be easily screened in standard reaction vials by adding CO2 in the form of dry ice. Adding approximately 1−2 equiv of dry ice on a 0.3 mmol scale and immediately sealing the vial could achieve a satisfactory pressure to promote the desired C−H activation without concomitant failure of the vial (see the SI for further details and notes on safety). Interestingly, no exogenous ligand was required to facilitate this transformation. With the optimized conditions in hand, we first explored the range of aryl iodides that could participate in the reaction. In general, electron-poor aryl iodides containing fluorides, trifluoromethyls, and esters were all effective in the reaction (Table 1, 1b−l). While ortho substituents are often challenging in C−H activation reactions, the presence of an ortho ester was well-tolerated, likely because of a chelate effect that promotes rather than inhibits oxidative addition to the C−I bond.20 Halophenyl iodides were also tolerated (1m−o), with no sign of activation of either aryl chlorides or bromides. Even a diiodide gave good selectivity for monofunctionalization (1o), while both halides could be functionalized in the presence of excess amine (see the SI). More complex heterocycles can also participate in the reaction (1p and 1q), although with somewhat limited success due we suspect to partial decomposition of the heterocycles during the reaction. Iodides with moderately and strongly electron-donating substituents (1r−w) as well as iodides with extended conjugation (1x and 1y) were also tolerated in the reaction. While use of ortho functional groups other than a chelating ester were not tolerated under the standard conditions, switching the silver additive to AgOTf allowed substituents with o-fluoro (1z) and o-hydroxyl (1aa) groups to participate in the reaction. The reason for this modulation of reactivity is not clear at the present time, although the reaction using AgOTf is more acidic than the AgTFA reaction by ∼1 pH unit after completion. Considering the amine scope, substrates containing saturated (2a−d) and unsaturated (2e) carbocycles all participated in the reaction with relatively good yields (Table 2). The precursor for 2d was an inseparable mixture of cis (major) and trans (minor) isomers, yet we were able to cleanly isolate the cis isomer of 2d after the reaction (the minor trans isomer also gave product, although it could not be cleanly isolated). While 2e has more reactive γ-C(sp2)−H bonds, good selectivity was observed for arylation of only the C(sp3)−H bond. This selectivity can be rationalized on the basis of the inflexibility of the fused rings, which likely prevents the more reactive C−H bonds from accessing the coordinated palladium catalyst. As long as a single type of γ-C−H bond is present on the substrate, good selectivity could be achieved despite the length of other aliphatic chains (2f and 2g). When there were symmetrical positions with γ-C−H bonds, however, a mixture of mono- and diarylation was observed (2h and 2i), albeit with some selectivity for mono- and diarylation onlytriarylation was not observed when three separate ethyl groups were attached (2i). We found that the 1° amine substrates did not require an αtertiary center, and simply lowering the temperature enabled less substituted substrates to participate in the reaction (2j−m). Despite symmetrical γ-C−H bonds in both 2k and 2m, only monoarylation was observed. The enhanced selectivity for monoarylation (compared with 2h and 2i) is likely due to the decreased reaction temperature. Despite the presence of a chelating ester moiety, the ethyl ester of L-valine could be

subsequently identified as a viable candidate: it contains a carbonyl that can reversibly react with amines, while the carbamate products can readily coordinate to palladium but are more chemically robust than imines.18 CO2 has previously been used as a traceless DG for meta-C−H activation of phenols (although it requires separate and harsh installation and removal steps).19 We reasoned that CO2 could therefore serve as a transient DG in the presence of a suitable nucleophile such as an amineserving as a protecting group, preventing substrate oxidation, and directing site-selective C−H bond scission by Pd. The use of a transient carbamate moiety as a DG would have the potential to make C−H activation of 1° amines feasible under more oxidizing conditions where simple ammonium formation is inadequate to prevent amine oxidation while simultaneously allowing expansion of a transient DG approach to 2° amines. We began our studies using the model substrate tertamylamine. Previous work17 had shown that in the absence of a DG, only trace C−H arylation of this substrate is observed. Gratifyingly, using conditions similar to Ge’s,17a we were able to selectively γ-arylate tert-amylamine using phenyl iodide in acetic acid with silver trifluoroacetate as an additive simply by performing the reaction under CO2 pressure (Table 1, 1a). As expected, the absence of any of these components leads to only trace or no product. While the reaction can occur under 1 atm Table 1. Aryl Iodide Substrate Scope for γ-C(sp3)−H Arylation of 1° Amines Directed by CO2

a

AgOTf was used in place of AgTFA. 6819

DOI: 10.1021/jacs.8b05061 J. Am. Chem. Soc. 2018, 140, 6818−6822

Communication

Journal of the American Chemical Society Table 2. 1° Amine Substrate Scope for γ-C(sp3)−H Arylation Directed by CO2

Table 3. 2° Amine Substrate Scope for γ-C(sp3)−H Arylation Directed by CO2

(3e−k). These are interesting substrates from the perspectives of potentially competitive C(sp2)−H bonds and sensitivity to oxidation. Again, the putative more favored conformation during the reaction facilitates complete selectivity for the less reactive C−H bonds, even though the increased flexibility was initially predicted to lead to γ-C(sp2)−H functionalization (2f). It is noteworthy that negligible oxidation of the benzylic amines to the corresponding imines was observed in these reactions, with significant mass balance being the unreacted amine. Regrettably, increasing the catalyst loading or adding the catalyst portionwise failed to drive these reactions to completion, a result that demands further scrutiny to understand. Less substituted 2° amine substrates bearing a more oxidatively sensitive α-2° center could also be utilized (3l and 3m). Finally, we wanted to explore the mechanism and role of CO2 in the reaction. Salt 4 was prepared and then subjected to the reaction conditions without additional CO2 and gave greater than stoichiometric conversion with respect to CO2, suggesting that CO2 acts transiently (Figure 2A). When the reaction was performed in AcOD, no deuteration was observed, suggesting that the concerted metalation−deprotonation step may be irreversible (Figure 2B).22 This was further corroborated by kinetic isotope effect (KIE) experiments, which showed a significantly faster reaction for the proteo substrate. Although Pd−carbamato complexes are known,23 it is possible that CO2 could actually have an off-cycle role in this reaction. To investigate whether CO2 actually serves to disperse catalytically inactive Pd−diamine complexes, we prepared complex 5 (Figure 2C).24 Dissolution of 5 in AcOH-d4 without CO2 was satisfactory to partially dissociate the complex to give a mixture of trinuclear complex 6 and free ammonium, suggesting that CO2 is not necessary under the reaction conditions to disrupt the formation of complex 5. Meanwhile, introduction of CO2 into CDCl3 solutions of this complex gave no change extra base is required to facilitate proton transfer to lead to stable Pd−carbamate complexes.23 To simulate the reaction conditions, PdCl2 and amine were mixed in 1% AcOH in DMSO-d6, followed by bubbling CO2 through the solution for 12 h. This gave a complex spectrum containing a carbamate signal as well as new resonances between 121 and 124 ppm in the 13C NMR spectrum from different CO2 species. We believe that this supports an on-cycle role for CO2 as a transient DG acting through a rare seven-membered palladacycle,25 though further studies are needed.

a 90 °C. bFrom isolated Bz products. cExtra AcOH molecule removed for clarity.

selectively arylated once to give the unnatural amino acid product (2n). This compound was functionalized with retention of the S stereocenter of L-valine and with moderate diastereoselectivity (see the SI). We were delighted to find that when a rigid bornylamine was used, in which there is a 1° γ-C− H bond that cannot be reached by the catalyst, selective transannular arylation at a methylene γ-C−H bond position was achieved instead of at the methyl group (2n).21 When a larger biphenyl iodide was used, a similar product was obtained (2m), which facilitated X-ray analysis to confirm location of the aryl group. We next turned our attention to the challenge of whether CO2 could also serve to promote the γ-C(sp3)−H arylation of 2° amines, a transformation that had required highly substituted substrates up to this point, generally gave β- rather than γfunctionalization, and was inaccessible via imine-type DGs. Gratifyingly, the reaction could be performed by modifying our conditions, most notably by increasing the CO2 loading. The excess CO2 was criticalat lower concentrations not only was the product yield decreased, but significant oxidation of the starting material was observed, giving a mixture of imine, amine, and aldehyde. The reaction tolerates 2° amines with different-length alkyl substituents on the nitrogen (Table 3, 3a−c). Although 3a possesses two distinct terminal γ-C−H bonds, the reaction occurs selectively on the more substituted side, presumably because of a more favorable comformation during the C−H activation step.10a A homobenzylic 2° amine substrate was also tolerated without concomitant oxidation (3d), as were a variety of benzylic amines, all with complete selectivity observed for the γ-C(sp3)−H bond rather than functionalization on the arene 6820

DOI: 10.1021/jacs.8b05061 J. Am. Chem. Soc. 2018, 140, 6818−6822

Communication

Journal of the American Chemical Society

Foundation in partial support of this work. Ms. T. Perera and Ms. K. Rajanayake are acknowledged for collecting highresolution ESI-MS data at The University of Toledo, as is Dr. K. Suhr at The University of Texas at Austin MS Core Facility.

■

(1) (a) Allred, T. K.; Manoni, F.; Harran, P. G. Chem. Rev. 2017, 117, 11994−12051. (b) Genovino, J.; Lütz, S.; Sames, D.; Touré, B. B. J. Am. Chem. Soc. 2013, 135, 12346−12352. (2) Beatty, J. W.; Stephenson, C. R. J. Acc. Chem. Res. 2015, 48, 1474−1484. (3) (a) Froidevaux, V.; Negrell, C.; Caillol, S.; Pascault, J.-P.; Boutevin, B. Chem. Rev. 2016, 116, 14181−14224. (b) Dutcher, B.; Fan, M.; Russell, A. G. ACS Appl. Mater. Interfaces 2015, 7, 2137− 2148. (4) (a) Xu, H.-C.; Chowdhury, S.; Ellman, J. A. Nat. Protoc. 2013, 8, 2271−2280. (b) Pirnot, M. T.; Wang, Y.-M.; Buchwald, S. L. Angew. Chem., Int. Ed. 2016, 55, 48−57. (c) Tafazolian, H.; Samblanet, D. C.; Schmidt, J. A. R. Organometallics 2015, 34, 1809−1817. (5) (a) He, J.; Wasa, M.; Chan, K. S. L.; Shao, Q.; Yu, J.-Q. Chem. Rev. 2017, 117, 8754−8786. (b) Petrone, D. A.; Ye, J.; Lautens, M. Chem. Rev. 2016, 116, 8003−8104. (c) Bedell, T. A.; Hone, G. A. B.; Valette, D.; Yu, J.-Q.; Davies, H. M. L.; Sorensen, E. J. Angew. Chem., Int. Ed. 2016, 55, 8270−8274. (6) (a) Mack, J. B. C.; Gipson, J. D.; Du Bois, J.; Sigman, M. S. J. Am. Chem. Soc. 2017, 139, 9503−9506. (b) Xu, Y.; Dong, G. Chem. Sci. 2018, 9, 1424−1432. (c) Zultanski, S. L.; Zhao, J.; Stahl, S. S. J. Am. Chem. Soc. 2016, 138, 6416−6419. (7) (a) Sheng, J.; Guo, Y.; Wu, J. Tetrahedron 2013, 69, 6495−6499. (b) Wang, J.-R.; Fu, Y.; Zhang, B.-B.; Cui, X.; Liu, L.; Guo, Q.-X. Tetrahedron Lett. 2006, 47, 8293−8297. (8) (a) Ruiz-Castillo, P.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc. 2015, 137, 3085−3092. (b) Vo, G. D.; Hartwig, J. F. J. Am. Chem. Soc. 2009, 131, 11049−11061. (9) (a) Font, H.; Font-Bardia, M.; Gómez, K.; González, G.; Granell, J.; Macho, I.; Martínez, M. Dalton Trans. 2014, 43, 13525−13536. (b) Albert, J.; Ariza, X.; Calvet, T.; Font-Bardia, M.; Garcia, J.; Granell, J.; Lamela, A.; López, B.; Martinez, M.; Ortega, L.; Rodriguez, A.; Santos, D. Organometallics 2013, 32, 649−659. (c) Lazareva, A.; Daugulis, O. Org. Lett. 2006, 8, 5211−5213. (10) (a) Hogg, K. F.; Trowbridge, A.; Alvarez-Pérez, A.; Gaunt, M. J. Chem. Sci. 2017, 8, 8198−8203. (b) Cabrera-Pardo, J. R.; Trowbridge, A.; Nappi, M.; Ozaki, K.; Gaunt, M. J. Angew. Chem., Int. Ed. 2017, 56, 11958−11962. (c) He, C.; Gaunt, M. J. Chem. Sci. 2017, 8, 3586− 3592. (d) Willcox, D.; Chappell, B. G. N.; Hogg, K. F.; Calleja, J.; Smalley, A. P.; Gaunt, M. J. Science 2016, 354, 851−857. (e) McNally, A.; Haffemayer, B.; Collins, B. S. L.; Gaunt, M. J. Nature 2014, 510, 129−133. (f) Haffemayer, B.; Gulias, M.; Gaunt, M. J. Chem. Sci. 2011, 2, 312−315. (11) (a) Mbofana, C. T.; Chong, E.; Lawniczak, J.; Sanford, M. S. Org. Lett. 2016, 18, 4258−4261. (b) Lee, M.; Sanford, M. S. J. Am. Chem. Soc. 2015, 137, 12796−12799. (c) Lee, M.; Sanford, M. S. Org. Lett. 2017, 19, 572−575. (12) Chen, K.; Wang, D.; Li, Z.-W.; Liu, Z.; Pan, F.; Zhang, Y.-F.; Shi, Z.-J. Org. Chem. Front. 2017, 4, 2097−2101. (13) (a) Han, J.; Zheng, Y.; Wang, C.; Zhu, Y.; Shi, D.-Q.; Zeng, R.; Huang, Z.-B.; Zhao, Y. J. Org. Chem. 2015, 80, 9297−9306. (b) Zaitsev, V. G.; Shabashov, D.; Daugulis, O. J. Am. Chem. Soc. 2005, 127, 13154−13155. (c) Shao, Q.; he, J.; Wu, Q.-F.; Yu, J.-Q. ACS Catal. 2017, 7, 7777−7782. (14) He, C.; Gaunt, M. J. Angew. Chem., Int. Ed. 2015, 54, 15840− 15844. (15) Calleja, J.; Pla, D.; Gorman, T. W.; Domingo, V.; Haffemayer, B.; Gaunt, M. J. Nat. Chem. 2015, 7, 1009−1016. (16) (a) Chan, K. S. L.; Wasa, M.; Chu, L.; Laforteza, B. N.; Miura, M.; Yu, J.-Q. Nat. Chem. 2014, 6, 146−150. (b) Rodríguez, N.; Romero-Revilla, J. A.; Fernández-Ibáñez, M. Á .; Carretero, J. C. Chem. Sci. 2013, 4, 175−179. (c) Zhang, Y.-F.; Zhao, H.-W.; Wang, H.; Wei,

Figure 2. Mechanistic investigations of CO2-mediated γ-C(sp3)−H activation of aliphatic amines.

In conclusion, we have described the first example of CO2mediated amine C−H activation. The ability of CO2 to transiently form carbamates may be useful not only for C−H activation but also for other directing-group-mediated reactions. Furthermore, we anticipate that the use of CO2 rather than a traditional protecting group may be a viable strategy for improving the sustainability of organic synthesis. Work is underway in our lab to better understand the mechanism and intermediates at play in these reactions and to further develop the scope of these transformations.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.8b05061. Experimental procedures, characterization data, and NMR spectra (PDF) Crystallographic data for (2p − H)+(OAc)−·HOAc (CIF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Michael C. Young: 0000-0002-3256-5562 Notes

The authors declare the following competing financial interest(s): The authors have a provisional patent submitted related to this chemistry (United States Provisional Patent #62/ 608,074).

■

ACKNOWLEDGMENTS The authors acknowledge startup funding from the University of Toledo as well as a grant from the ACS Herman Frasch 6821

DOI: 10.1021/jacs.8b05061 J. Am. Chem. Soc. 2018, 140, 6818−6822

Communication

Journal of the American Chemical Society J.-B.; Shi, Z.-J. Angew. Chem., Int. Ed. 2015, 54, 13686−13690. (d) Reddy, M. D.; Watkins, E. B. J. Org. Chem. 2015, 80, 11447− 11459. (e) He, G.; Chen, G. Angew. Chem., Int. Ed. 2011, 50, 5192− 5196. (f) Nack, W. A.; Wang, X.; Wang, B.; He, G.; Chen, G. Beilstein J. Org. Chem. 2016, 12, 1243−1249. (17) (a) Liu, Y.; Ge, H. Nat. Chem. 2017, 9, 26−32. (b) Wu, Y.; Chen, Y.-Q.; Liu, T.; Eastgate, M. D.; Yu, J.-Q. J. Am. Chem. Soc. 2016, 138, 14554−14557. (c) Xu, Y.; Young, M. C.; Wang, D.; Magness, D. M.; Dong, G. Angew. Chem., Int. Ed. 2016, 55, 9084−9087. (d) Yada, A.; Liao, W.; Sato, Y.; Murakami, M. Angew. Chem., Int. Ed. 2017, 56, 1073−1076. (18) (a) Anillo, A.; Dell’Amico, D. B.; Calderazzo, F.; Nardelli, M.; Pelizzi, G.; Rocchi, L. J. Chem. Soc., Dalton Trans. 1991, 2845−2849. (b) Srivastava, R. S.; Singh, G.; Nakano, M.; Osakada, K.; Ozawa, F.; Yamamoto, A. J. Organomet. Chem. 1993, 451, 221−229. (c) Ozawa, F.; Ito, T.; Yamamoto, A. Chem. Lett. 1979, 8, 735−738. (19) (a) Luo, J.; Preciado, S.; Larrosa, I. J. Am. Chem. Soc. 2014, 136, 4109−4112. (b) Luo, J.; Preciado, S.; Araromi, S. O.; Larrosa, I. Chem. - Asian J. 2016, 11, 347−350. (20) Mu, D.; Gao, F.; Chen, G.; He, G. ACS Catal. 2017, 7, 1880− 1885. (21) (a) Topczewski, J. J.; Cabrera, P. J.; Saper, N. I.; Sanford, M. S. Nature 2016, 531, 220−224. (b) Coomber, C. E.; Benhamou, L.; Bučar, D.-K.; Smith, P. D.; Porter, M. J.; Sheppard, T. D. J. Org. Chem. 2018, 83, 2495−2503. (c) Cabrera, P. J.; Lee, M.; Sanford, M. S. J. Am. Chem. Soc. 2018, 140, 5599−5606. (22) (a) Simmons, E. M.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51, 3066−3072. (b) Carr, K. J. T.; Davies, D. L.; Macgregor, S. A.; Singh, K.; Villa-Marcos, B. Chem. Sci. 2014, 5, 2340−2346. (c) Stephens, D. E.; Larionov, O. V. Tetrahedron 2015, 71, 8683− 8716. (d) Valpuesta, J. E. V.; Á lvarez, E.; López-Serrano, J.; Maya, C.; Carmona, E. Chem. - Eur. J. 2012, 18, 13149−13159. (23) (a) Srivastava, R. S.; Singh, G.; Nakano, M.; Osakada, K.; Ozawa, F.; Yamamoto, A. J. Organomet. Chem. 1993, 451, 221−229. (b) Ozawa, F.; Ito, T.; Yamamoto, A. Chem. Lett. 1979, 8, 735−738. (c) Anillo, A.; Dell’Amico, D. B.; Calderazzo, F.; Nardelli, M.; Pelizzi, G.; Rocchi, L. J. Chem. Soc., Dalton Trans. 1991, 2845. (24) Smalley, A. P.; Gaunt, M. J. J. Am. Chem. Soc. 2015, 137, 10632− 10641. (25) (a) Frutos-Pedreño, R.; García-Sánchez, E.; Oliva-Madrid, M. J.; Bautista, D.; Martínez-Viviente, E.; Saura-Llamas, I.; Vicente, J. Inorg. Chem. 2016, 55, 5520−5533. (b) Piou, T.; Bunescu, A.; Wang, Q.; Neuville, L.; Zhu, J. Angew. Chem., Int. Ed. 2013, 52, 12385−12389. (c) Nicasio-Collazo, J.; Á lvarez, E.; Alvarado-Monzón, J. C.; Andreude-Riquer, G.; Jimenez-Halla, J. O. C.; De León-Rodríguez, L. M.; Merino, G.; Morales, U.; Serrano, O.; López, J. A. Dalton Trans. 2011, 40, 12450−12453.

6822

DOI: 10.1021/jacs.8b05061 J. Am. Chem. Soc. 2018, 140, 6818−6822