Dianionic Phase-Transfer Catalyst for Asymmetric Fluoro-cyclization

Communication pubs.acs.org/JACS

Cite This: J. Am. Chem. Soc. 2018, 140, 2785−2788

Dianionic Phase-Transfer Catalyst for Asymmetric Fluoro-cyclization Hiromichi Egami, Tomoki Niwa, Hitomi Sato, Ryo Hotta, Daiki Rouno, Yuji Kawato, and Yoshitaka Hamashima* School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan S Supporting Information *

ABSTRACT: Inspired by the dicationic nature of the electrophilic fluorinating reagent, Selectfluor (1), we rationally designed a series of dicarboxylic acid precatalysts (2), which, when deprotonated, act as anionic phasetransfer catalysts for asymmetric fluorination of alkenes. Among them, 2a having the shortest linker moiety efficiently catalyzed unprecedented 6-endo-fluoro-cyclization of various allylic amides, affording fluorinated dihydrooxazine compounds with high enantioselectivity (up to 99% ee). In addition to cyclic substrates, acyclic trisubstituted alkenes underwent the reaction with good diastereoselectivity, whereas low diastereoselectivity was observed for linear disubstituted alkenes. Results suggest that the reaction proceeds via a fluoro-carbocation intermediate.

F

luorinated compounds have found extensive applications in pharmaceutical, agrochemical, and materials sciences.1 Many methods for the introduction of fluorine atoms into organic frameworks have been reported.2 In particular, asymmetric fluoro-functionalization of alkenes has attracted much attention, because it is a powerful approach to obtain structurally diverse fluorinated compounds.3 However, stereoselective fluorination of alkenes is still challenging, although cinchona alkaloid-derived chiral fluorinating reagents4 and chiral hypervalent iodine reagents5 have been studied. Following Toste’s pioneering work,6,7 we became interested in anionic phase-transfer catalysis for enantioselective fluorination. We recently reported the first successful example of enantioselective fluoro-lactonization of ene-carboxylic acids with a carefully designed hydroxyl-carboxylic acid precatalyst (Figure 1a).8 Upon deprotonation in situ, the carboxylate ion of the active catalyst acts as a phase-transfer unit to form an ion pair with Selectfluor (1), while the hydroxyl group interacts with the anionic substrate to define its position (Figure 1b, left). Mechanistic studies of the fluoro-lactonization8 strongly indicated that a binary complex between the catalyst and the substrate anion is responsible for bringing 1 into the organic phase (Figure 1b, left). This observation prompted us to think that the cooperative action of two carboxylate anions located at an appropriate separation distance would achieve high reaction efficiency.9 Based on this hypothesis, we designed dicarboxylic acid precatalysts 2 (Figure 1c), which would form a series of linked dicarboxylate catalysts (Figure 1b, right). Even though the designed catalysts are conformationally flexible, we envisaged that the two-point ionic pairing of the catalyst with © 2018 American Chemical Society

Figure 1. Background and our design strategy for novel dicarboxylate catalysts.

1 would form a well-defined chiral environment. Based on our halogenation project regarding bromo-cyclization reactions of allylic amides,10 we selected fluoro-cyclization of ene-amides as a model reaction, since the amide has a hydrogen-bonding donor for interaction with a catalyst carboxylate and asymmetric 6-endo-fluoro-cyclization is unprecedented in the literature.11 Herein, we disclose a highly enantioselective fluorocyclization of allylic amides using designed and novel dianionic phase-transfer catalysts (Figure 1c). The precatalyst 2a was readily synthesized in 5 steps from compound 3 (Scheme 1). Thus, monoprotection of 3,3′dibromo-BINOL (3) and etherification with 1,3-propanediolReceived: December 27, 2017 Published: February 9, 2018 2785

DOI: 10.1021/jacs.7b13690 J. Am. Chem. Soc. 2018, 140, 2785−2788

Communication

Journal of the American Chemical Society Table 1. Catalyst Screeninga

Scheme 1. Preparation of 2a

ditosylate, followed by acidic hydrolysis, provided linkedBINOL derivative 4 in 64% yield over 2 steps. After the introduction of phosphate groups, all bromine atoms were replaced with phenyl groups. Finally, reductive carboxylation with lithium-naphthalenide12 furnished the desired diacid 2a in 69% yield. In order to identify the optimum catalyst, fluorination reactions of chromene derivative 5a were carried out in toluene using Na3PO4 as an insoluble base (Table 1). To our delight, 2a promoted the desired reaction smoothly to give 6a in 87% yield with 93% ee as a single diastereomer (entry 1). The length of the methylene linker affected the reaction efficiency, revealing that the C3 linker gave the best results (entries 2 and 3). Both the chemical yield and the enantioselectivity decreased significantly when 2d and 2e were used, revealing that full substitution at the 3,3′-positions of the catalyst is essential for this reaction (entries 4 and 5). Interestingly, previously known precatalysts7,8 were less effective. Binaphthyl-dicarboxylic acid 9 provided the product with only 17% ee (entry 6). Hydroxylcarboxylic acid 10 and phosphoric acid 11 were also examined, and again, lower enantioselectivities were observed (entries 7 and 8). These control experiments supported our working hypothesis regarding the optimum ionic valency within the catalyst as well as the distance between the ionic parts. Further optimization13 revealed that the optimum base depends on the nature of the substrate, since Na2HPO4 gave slightly better results for 5a (entry 9). Notably, reactions of less reactive acyclic substrate 7a clearly demonstrated that 2a is superior to other known catalysts (entries 10−13). For example, phosphoric acid 11 was totally ineffective (entry 13). As found in our previous study,8 the addition of Na 2SO4 was effective for accelerating the fluorination of 7a (entry 14). Although the diastereomer ratio was almost 1:1, the enantioselectivity was excellent for both diastereomers (vide inf ra). Having optimized the reaction conditions, we examined the generality for cyclic substrates (Table 2). Although minor tuning of the reaction conditions (conditions A or B) was needed depending on the substrate, the desired reaction occurred to give the corresponding fluorinated tricyclic compound as a single diastereomer. In these reactions, various functional groups including ester, ether, benzylic methyl group, and halogens were tolerated. Substituents on the chromene framework did not impact largely on the enantioselectivity (6b−6d). The dimethyl groups of the chromene core were not essential as a stereocontrolling element. Thus, sterically less hindered chromene and dihydronaphthalene substrates also underwent the fluorination reaction with excellent enantioselectivity (6e−6i). In addition to six-membered ring substrates, a five-membered indene derivative was also a good substrate, affording 6j in 63% yield with 93% ee. Furthermore,

a

The reactions were carried out with precatalyst, 1, and Na3PO4 in toluene (1 mL) on a 0.1 mmol scale, unless otherwise mentioned. b Yields were determined by 1H NMR analysis using 1,1,2,2tetrabromoethane as an internal standard. cEe values were determined by chiral HPLC analysis. dRun with Na2HPO4 instead of Na3PO4. e Diastereomer ratio was 1:1. fEe values of each diastereomers. g Diastereomer ratio was 2:1. hRun with Na2SO4.

substituted aryl amide groups were also applicable, and high asymmetric induction was observed (6k−6n). It should be noted that reactions generally gave complex mixtures when carried out without catalyst and in MeCN. This observation emphasizes the use of phase-transfer conditions as crucial for this type of fluorination reaction. Next, we turned our attention to acyclic alkene substrates. First, disubstituted allylic amides were subjected to the reaction conditions (Table 3a). While the electronic nature of the substituents did not significantly affect the diastereoselectivity, the ee values of these reactions were generally excellent (8a− 8e).14 To probe the stereochemistry of the products, Mosher’s ester method was applied after hydrolysis of the dihydrooxazine ring of 8a.13 This revealed that the diastereomers are derived 2786

DOI: 10.1021/jacs.7b13690 J. Am. Chem. Soc. 2018, 140, 2785−2788

Communication

Journal of the American Chemical Society Table 2. Fluorination of Cyclic Allylic Amidesa

Table 3. Fluorination of Acyclic Substratesa

a

The conditions were the same as in Table 2. bDetermined by NMR analysis of the crude mixture. cRun at 15 °C. dRun for 4 d. eRun for 3 d. fRun for 2 d. a

Conditions A: The reaction was carried out with 2a, 1, and Na2HPO4 at 15 °C in toluene for 72 h. Conditions B: The reaction was carried out with 2a, 1, Na3PO4, and Na2SO4 at 25 °C in toluene for 24 h. b Run with 20 mol% of 2a. cRun for 5 d. dRun with 3 equiv of Na2HPO4.

highly functionalized, this method would be useful in synthesizing a variety of fluorinated compounds. Further investigations to expand the scope of the reaction and to elucidate the reaction mechanism are underway in our laboratory.

from the stereoisomer at the benzylic position, while the facial selectivity of the fluorination step is well regulated. Coupled with findings in the literature,15 these results suggest that the fluoro-cyclization proceeds via the formation of a benzylic carbocation intermediate, which undergoes less stereoselective intramolecular C−O bond formation. Based on the above consideration, we expected the stereoselectivity of the cyclization step to be improved if the conformation of the carbocation intermediate was restricted sterically by additional substituents.16 Thus, we examined the fluoro-cyclization of acyclic trisubstituted allylic amides (Table 3b). Gratifyingly, the reactions proceeded with high diastereoselectivity, affording products 8f−8h with a fluorinated quaternary carbon center in up to 99% ee. The reaction was applicable to heteroaromatic compounds (8i−8k). Reduced diastereoselectivity observed for indole and benzofuran derivatives 8i and 8j is attributed to higher stability of the carbocation intermediate. In contrast, thiophene derivative was converted to the cyclized compound 8k with high diastereoand almost perfect enantioselectivity (dr = 1:8.4, 99% ee). In summary, we have developed a new dianionic phasetransfer catalyst for highly enantioselective 6-endo fluorocyclization of allylic amides with Selectfluor. The catalyst was able to control the fluorine delivery step with high enantioselectivity in all the examples reported herein, indicating that the designed catalyst would be applicable to other types of fluoro-functionalization. Since the obtained compounds are

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b13690. Experimental procedures and spectroscopic data (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Yoshitaka Hamashima: 0000-0002-6509-8956 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by a Grant-in-Aid for Scientific Research (B) (No. 16H05077) from JSPS, Basis for Supporting Innovative Drug Discovering and Life Science Research (BINDS) from AMED, the Naito Foundation (Japan), The Research Foundation for Pharmaceutical Sciences, and The FUGAKU Trust for Medicinal Research. We thank Prof. Kenji Watanabe and Dr. Yuta Tsunematsu of University of Shizuoka for their kind help with mass spectrometry analysis. 2787

DOI: 10.1021/jacs.7b13690 J. Am. Chem. Soc. 2018, 140, 2785−2788

Communication

Journal of the American Chemical Society

■

REFERENCES

(1) (a) Kirsch, P. Modern Fluoroorganic Chemistry: Synthesis, Reactivity, Applications, 2nd ed.; Wiley-VCH: Weinheim, 2013. (b) Ojima, I. Fluorine in Medicinal Chemistry and Chemical Biology; Wiley-Blackwell: Oxford, 2009. (c) Gouverneur, V.; Müller, K. Fluorine in Pharmaceutical and Medicinal Chemistry: From Biophysical Aspects to Clinical Applications; Imperial College Press: London, 2012. (d) Müller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881−1886. (2) (a) Liang, T.; Neumann, C. N.; Ritter, T. Angew. Chem., Int. Ed. 2013, 52, 8214−8264. (b) Yang, X.; Wu, T.; Phipps, R. J.; Toste, F. D. Chem. Rev. 2015, 115, 826−870. (3) Wolstenhulme, J. R.; Gouverneur, V. Acc. Chem. Res. 2014, 47, 3560−3570. (4) (a) Shibata, N.; Suzuki, E.; Asahi, T.; Shiro, M. J. Am. Chem. Soc. 2001, 123, 7001−7009. (b) Wang, M.; Wang, B. M.; Shi, L.; Tu, Y. Q.; Fan, C.-A.; Wang, S. H.; Hu, X. D.; Zhang, S. Y. Chem. Commun. 2005, 41, 5580−5582. (c) Ishimaru, T.; Shibata, N.; Horikawa, T.; Yasuda, N.; Nakamura, S.; Toru, T.; Shiro, M. Angew. Chem., Int. Ed. 2008, 47, 4157−4161. (d) Wilkinson, S. C.; Lozano, O.; Schuler, M.; Pacheco, M. C.; Salmon, R.; Gouverneur, V. Angew. Chem., Int. Ed. 2009, 48, 7083−7086. (e) Wolstenhulme, J. R.; Rosenqvist, J.; Lozano, O.; Ilupeju, J.; Wurz, N.; Engle, K. M.; Pidgeon, G. W.; Moore, P. R.; Sandford, G.; Gouverneur, V. Angew. Chem., Int. Ed. 2013, 52, 9796− 9800. (5) (a) Kong, K.; Feige, P.; de Haro, T.; Nevado, C. Angew. Chem., Int. Ed. 2013, 52, 2469−2473. (b) Molnár, I. G.; Gilmour, R. J. Am. Chem. Soc. 2016, 138, 5004−5007. (c) Banik, S. M.; Medley, J. W.; Jacobsen, E. N. J. Am. Chem. Soc. 2016, 138, 5000−5003. (d) Woerly, E. M.; Banik, S. M.; Jacobsen, E. N. J. Am. Chem. Soc. 2016, 138, 13858−13861. (6) Rauniyar, V.; Lackner, A. D.; Hamilton, G. L.; Toste, F. D. Science 2011, 334, 1681−1684. (7) (a) Phipps, R. J.; Hiramatsu, K.; Toste, F. D. J. Am. Chem. Soc. 2012, 134, 8376−8379. (b) Phipps, R. J.; Toste, F. D. J. Am. Chem. Soc. 2013, 135, 1268−1271. (c) Wu, J.; Wang, Y.-M.; Drljevic, A.; Rauniyar, V.; Phipps, R. J.; Toste, F. D. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 13729−13733. (d) Zi, W.; Wang, Y.-M.; Toste, F. D. J. Am. Chem. Soc. 2014, 136, 12864−12867. (e) Romanov-Michailidis, F.; Guénée, L.; Alexakis, A. Angew. Chem., Int. Ed. 2013, 52, 9266−9270. (f) Liang, X.-W.; Liu, C.; Zhang, W.; You, S.-L. Chem. Commun. 2017, 53, 5531−5534. (8) Egami, H.; Asada, J.; Sato, K.; Hashizume, D.; Kawato, Y.; Hamashima, Y. J. Am. Chem. Soc. 2015, 137, 10132−10135. (9) Toste proposed a 1:2 assembly between Selectfluor and chiral phosphate catalysts. See refs 6 and 7. (10) (a) Kawato, Y.; Kubota, A.; Ono, H.; Egami, H.; Hamashima, Y. Org. Lett. 2015, 17, 1244−1247. (b) Kawato, Y.; Ono, H.; Kubota, A.; Nagao, Y.; Morita, N.; Egami, H.; Hamashima, Y. Chem. - Eur. J. 2016, 22, 2127−2133. (c) Nagao, Y.; Hisanaga, T.; Egami, H.; Kawato, Y.; Hamashima, Y. Chem. - Eur. J. 2017, 23, 16758−16762. (11) Shunatona, H. P.; Früh, N.; Wang, Y.-M.; Rauniyar, V.; Toste, F. D. Angew. Chem., Int. Ed. 2013, 52, 7724−7727. (12) Egami, H.; Sato, K.; Asada, J.; Kawato, Y.; Hamashima, Y. Tetrahedron 2015, 71, 6384−6388. (13) See Supporting Information. (14) The ee values between the diastereomers are generally different, probably because a kinetic resolution of the fluorocarbenium ion intermediate occurs during the cyclization step. (15) (a) Stavber, S.; Sotler-Pecan, T.; Zupan, M. Bull. Chem. Soc. Jpn. 1996, 69, 169−175. (b) Stavber, S.; Sotler-Pecan, T.; Zupan, M. Tetrahedron 2000, 56, 1929−1936. (c) Zupan, M.; Skulj, P.; Stavber, S. Tetrahedron 2001, 57, 10027−10031. (d) Olah, G. A.; Prakash, G. K. S.; Rasul, G. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 8427−8430. (16) (a) Mühlthau, F.; Schuster, O.; Bach, T. J. Am. Chem. Soc. 2005, 127, 9348−9349. (b) Sparr, C.; Gilmour, R. Angew. Chem., Int. Ed. 2010, 49, 6520−6523.

2788

DOI: 10.1021/jacs.7b13690 J. Am. Chem. Soc. 2018, 140, 2785−2788