Synthesis of 1-Tetrasubstituted 2,2,2-Trifluoroethylamine Derivatives via Palladium-Catalyzed Allylation of sp³ C–H Bonds

Abstract

This note describes the construction of tetrasubstituted carbon stereocenters via palladium-catalyzed allylation of sp³ C–H bonds of 2,2,2-trifluoroethylamine derivatives. The presence of 2-pyridyl group of the imines derived from 1-substituted-2,2,2-trifluoroethylamine was key to promoting the reaction efficiently, allowing an access to a variety of 1-allylated 2,2,2-trifluoroethylamine derivatives with tetrasubstituted carbon stereocenters.

2,2,2-Trifluoroethylamines are important structures in organic molecules¹⁾ and pharmaceuticals because they improve bioactivities compared with the corresponding ethylamine derivatives.^2,3) Thus, the development of synthetic methods for 1-tetrasubstituted-2,2,2-trifluoroethylamines that cannot be accessed by hydrogenation of imines and reductive amination of ketones is an important challenge.^4,5) Catalytic nucleophilic addition to trifluoromethyl ketimines to gain access to these structures has been extensively studied over the last decade (Chart 1, Eq. a).^4,5)

Chart 1. Approaches to 1-Substituted 2,2,2-Trifluoroethylamines

On the other hand, the reaction of 2,2,2-trifluoroethylamine derivatives with electrophiles is a complementary approach for the synthesis of 1-substituted-2,2,2-trifluoroethylamines (Chart 1, Eq. b).^6–12) This reaction comprises one of these steps toward functionalization of 2,2,2-trifluoroethylamines when combining condensation with carbonyl compounds and subsequent hydrolysis. Very recently, several groups reported ketimines derived from 2,2,2-trifluoroethylamine as a pro-nucleophile. In these reactions, azomethine ylides are formed as nucleophiles through deprotonation of the sp³ C–H bond of 2,2,2-trifluoroethylamines. The scope of these reactions is mostly limited to unsubstituted 2,2,2-trifluoroethylamines (R¹=H), however, and reported reactions with 1-substituted-2,2,2-trifluoroethylamines mostly have an electron-withdrawing group as the substituent.^13–15) In addition, most of the reported reactions are limited to 1,3-dipolar cycloaddition with electron-deficient olefins,^6–8,10,11) and these cyclized products are not readily transformed into N-unprotected 2,2,2-trifluoroethylamines.¹⁶⁾

To overcome these limitations, herein we report our preliminary studies on Pd-catalyzed allylation of sp³ C–H bonds of 1-mono-substituted 2,2,2-trifluoroethylamines, providing a variety of 1-allylated 2,2,2-trifluoroethylamines with tetrasubstituted carbon stereocenters. Our results demonstrated that the cooperative activation of 2,2,2-trifluoroethylamines and allyl carbonate using Pd catalysts is important for promoting the reaction. In addition, the product was readily transformed into N-unprotected 2,2,2-trifluoroethylamine.

The design of our reaction is depicted in Chart 2. We selected Pd-catalyzed allylation of the sp³ C–H bond of N-(1-substituted-2,2,2-trifluoroethyl)aldimine (1) with allyl tert-butyl carbonate (2) for the following reasons: 1) the tert-butoxide anion generated from 2 and Pd(0) would work as a base for the deprotonation of 1; 2) the coordination of 1 to the palladium catalyst would allow for facile deprotonation of the sp³ C–H bond to generate the azomethine ylide intermediate; and 3) soft electrophilic (π-allyl)Pd would promote the reaction with the generated azomethine ylide without concomitant 1,3-cycloaddition reactions.^17,18)

Chart 2. Design of Our Reaction

To evaluate whether the hypothetical catalytic cycle described above would work as expected, we performed the reaction with imine 1a and 2 in the presence of 2.5 mol% of the allylpalladium(II) chloride dimer (5 mol% for Pd) with 10 mol% of PPh₃, and the desired product 3a was obtained in 25% yield in 2 h at room temperature (Table 1, entry 1). When the reaction time was prolonged to 20 h, most of substrate 1a was consumed and the yield was increased to 82% (entry 2). Reactions without PPh₃ or with 5 mol% of PPh₃ did not afford 3a, suggesting the importance of PPh₃ for promoting the reaction (entries 3, 4). The use of 5 mol% of tetrakis(triphenylphosphine)palladium(0) as the catalyst improved the reactivity (entry 5), producing 88% yield of 3a when the reaction was performed over 24 h (entry 6), and this reaction condition was selected as optimal. An elevated reaction temperature did not improve the yield of 3a (entry 7). Addition of base did not change the reactivity (entry 8), whereas performing the reaction without the Pd catalyst did not afford the product at all (entry 9). These results suggest that deprotonation of 1a was likely promoted by tert-butoxide anions generated from 2 through the formation of the (π-allyl)Pd(II) species.

Table 1. Optimization of Reaction Conditions

Entry	Catalyst	Temp. (°C)	Time (h)	Yield (%)a)
1	2.5 mol% of [Pd(allyl)Cl]2+10 mol% of PPh3	r.t.	2	25
2	2.5 mol% of [Pd(allyl)Cl]2+10 mol% of PPh3	r.t.	20	82
3	2.5 mol% of [Pd(allyl)Cl]2	60	2	<5
4	2.5 mol% of [Pd(allyl)Cl]2+5 mol% of PPh3	60	2	<5
5	5 mol% of Pd(PPh3)4	r.t.	2	58
6	5 mol% of Pd(PPh3)4	r.t.	24	88
7	5 mol% of Pd(PPh3)4	60	2	85
8	5 mol% of Pd(PPh3)4+20 mol% of KOtBu	60	2	86
9	20 mol% of KOtBu	60	2	<5

a) Determined by ¹⁹F-NMR analysis of the crude mixture. r.t., room temperature.

With the optimized conditions in hand, we investigated the substrate scope of imine 1 (Table 2). The desired product 3a was isolated in 66% yield after silica gel column chromatography without significant decomposition. The reduced isolated yield was due to the difficulties in separating the product 3a from byproducts.¹⁹⁾ Imines with an electron-donating and -withdrawing group on the arene ring were also applicable to the reaction, affording the product 3b–d in moderate to good isolated yields. meta-Substituted imine 1e also gave the products 3e in 68% isolated yield. 1-Alkyl-substituted imine 1f, a more challenging substrate due to possible isomerization, was applicable to the reaction but afforded 3f in a reduced yield, while more sterically demanding 1-cyclohexyl-substituted imine 1g did not provide the product 3g at all, and 1g remained unchanged even at higher temperatures. The reaction with 1a and 2 and subsequent hydrolysis of 3a yielded N-unprotected 1-allylated 2,2,2-trifluoroethylamine 4a in 63% yield in two steps (Chart 3).

Table 2. Scope and Limitations of Substrates^a)

a) Isolated yield was reported unless otherwise noted. Yield in parenthesis was determined by ¹⁹F-NMR analysis of the crude mixture.

Chart 3. Access to N-Unprotected 1-Tetrasubstituted 2,2,2-Trifluoroethylamine

To gain mechanistic insight into the catalytic system, we performed several control experiments. Imines 1h and 1i derived from benzaldehyde and benzophenone did not provide the allylated products (Chart 4, Eqs. a, b), and both imines 1h and 1i remained unchanged. Replacement of the 2-pyridyl moiety with 3-pyridyl group reduced the reactivity (Eq. c).²⁰⁾ These results indicated that both the electron-withdrawing nature and the coordinating ability of the imine moiety are important to promote the reaction efficiently, although bidentate chelation may not be indispensable. In addition, treatment of 1a in the absence of 2 resulted in the recovery of 1a with no formation of isomerized 1a′ (Eq. d), supporting the assumption that tert-butoxide anion generated from 2 promotes the deprotonation of 1a.

Chart 4. Control Experiments

In conclusion, we developed a palladium-catalyzed allylation of the sp³ C–H bonds of N-(1-substituted-2,2,2-trifluoroethyl)aldimines, affording 1-allylated 2,2,2-trifluoroethylamine derivatives with tetrasubstituted carbon stereocenters. In situ generation of tert-butoxide anion and coordination of the imines was important for promoting the reaction. Further studies to improve the scope of the reaction and application to one-pot sp³ C–H bond functionalization of N-unprotected amines in the presence of catalytic amounts of coordinating aldehydes are ongoing in our laboratory.

Acknowledgments

This work was financially supported by Grant-in-Aid for Scientific Research (B) (Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number JP24390004 and JP17H03972) and (C) (JSPS KAKENHI Grant Number JP15K07860) and Grant-in-Aid for Scientific Research on Innovative Areas (JSPS KAKENHI Grant Number JP15H05846 in Middle Molecular Strategy) from JSPS, Platform for Drug Discovery, Informatics, and Structural Life Science from AMED, Uehara Memorial Foundation and Takeda Science Foundation. K.M. and M.S. thank JSPS for Research Fellowships for Young Scientists.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

References and Notes

• 1) Zimmer L. E., Sparr C., Gilmour R., Angew. Chem. Int. Ed., 50, 11860–11871 (2011).
• 2) Meanwell N. A., J. Med. Chem., 54, 2529–2591 (2011).
• 3) Zhou Y., Wang J., Gu Z., Wang S., Zhu W., Aceña J. L., Soloshonok V. A., Izawa K., Liu H., Chem. Rev., 116, 422–518 (2016).
• 4) Nie J., Guo H.-C., Cahard D., Ma J.-A., Chem. Rev., 111, 455–529 (2011).
• 5) Kumagai N., Shibasaki M., Bull. Chem. Soc. Jpn., 88, 503–517 (2015).
• 6) Ma M., Zhu Y., Sun Q., Li X., Su J., Zhao L., Zhao Y., Qiu S., Yan W., Wang K., Wang R., Chem. Commun., 51, 8789–8792 (2015).
• 7) Sun Q., Li X., Su J., Zhao L., Ma M., Zhu Y., Zhao Y., Zhu R., Yan W., Wang K., Wang R., Adv. Synth. Catal., 357, 3187–3196 (2015).
• 8) Ponce A., Alonso I., Adrio J., Carretero J. C., Chem. Eur. J., 22, 4952–4959 (2016).
• 9) Li X., Su J., Liu Z., Zhu Y., Dong Z., Qiu S., Wang J., Lin L., Shen Z., Yan W., Wang K., Wang R., Org. Lett., 18, 956–959 (2016).
• 10) Wang Z.-H., Wu Z.-J., Yue D.-F., Hu W.-F., Zhang X.-M., Xu X.-Y., Yuan W.-C., Chem. Commun., 52, 11708–11711 (2016).
• 11) Dong Z., Zhu Y., Li B., Wang C., Yan W., Wang K., Wang R., J. Org. Chem., 82, 3482–3490 (2017).
• 12) Gao Y.-N., Shi M., Eur. J. Org. Chem., 2017, 1552–1560 (2017).
• 13) For example, see: Konno T., Kanda M., Ishihara T., Yamanaka H., J. Fluor. Chem., 126, 1517–1523 (2005).
• 14) For addition reactions of 2,2,2-trifluoromethylated isocyanates, see: Zhang X., Wang X., Gao Y., Xu X., Chem. Commun., 53, 2427–2430 (2017).
• 15) For a recent example on synthesis of 2,2,2-trifluoroethylamine with tetrasubstituted carbon via umpolung addition of trifluoromethyl ketimines, see: Chen P., Yue Z., Zhang J., Lv X., Wang L., Zhang J., Angew. Chem. Int. Ed., 55, 13316–13320 (2016).
• 16) For our recent contribution regarding facile synthesis of N-unprotected 2,2,2-trifluoroethylamine with tetrasubstituted carbon center, see: Morisaki K., Morimoto H., Ohshima T., Chem. Commun., 53, 6319–6322 (2017).
• 17) For a selected example of transition metal-catalyzed allylation with azomethine ylide, see: Liu J., Cao C.-G., Sun H.-B., Zhang X., Niu D., J. Am. Chem. Soc., 138, 13103–13106 (2016).
• 18) For a selected example of transition metal-catalyzed allylation with azomethine ylide, see: Su Y.-L., Li Y.-H., Chen Y.-G., Han Z.-Y., Chem. Commun., 53, 1985–1988 (2017).
• 19) Although the identification of the byproducts was not possible, we assume that imidazolidine derivatives resulted from dimerization of the imine 1a via 1,3-cycloaddition are the possible byproducts. Such dimerization of the imines are known in the literature: Grigg R., Donegan G., Gunaratne H. Q. N., Kennedy D. A., Malone J. F., Sridharan V., Thianpatanagul S., Tetrahedron, 45, 1723–1746 (1989).
• 20) We also tried the reaction of 1k with 4-pyridyl group instead of 2-pyridyl moiety, and the desired product 3k was observed in 35% yield based on ¹⁹F-NMR analysis of the crude mixture. Due to the difficulty of separating the byproduct, however, we were unable to isolate 3k in pure form.