Ruthenium(II)-Catalyzed C–H Allylation of N,N-Dialkylthiobenzamides with Allyl Methyl Carbonate by Sulfur Coordination

Key words

C–H bond activation - directing group - ruthenium catalysis - allylation - thiobenzamides

The C–H bond is one of the most common and diverse chemical bonds in organic molecules. The C–H bond can be activated to undergo direct arylation, alkylation, allylation, or other functionalization reactions, and this is widely used in the synthesis of complex natural products, organic intermediates, and pharmaceuticals.[1] The method has a good atom economy and reduces chemical waste, and it is an efficient, green, and environmentally friendly method for constructing complex molecular skeletons.[2] Allylic compounds are widely found in natural bioactive molecules and drug molecules, and the allyl group is an important synthetic structural unit in the organic chemical synthesis industry, with a wide range of applications in areas such as flavorings, pharmaceuticals, materials, and supramolecular assemblies.[3]

The introduction of specific directing groups into a substrate can directly and effectively promote C–H bond activation and achieve reaction selectivity.[4] Usually, the directing group forms a metallocyclic intermediate with a metal to facilitate C–H bond activation and subsequent functionalization reactions. However, despite the prevalence of N-directing groups,[5] such as amide-based weak coordination through N-amidate assistance, in the vast majority of C–H functionalization methods, the use of sulfur to direct C–H cleavage is severely underdeveloped.[6]

In general, sulfur directing groups are easier to convert into other functional groups than their nitrogen-based counterparts.[7] Additionally, sulfur-based functional groups are widely recognized as privileged scaffolds in pharmaceuticals, and are commonly found in natural products, agrochemicals, and medicinal chemistry[8] as well as in organic optoelectronic materials, which are the subject of this text. It is estimated that 25% of drugs contain sulfur as a key component, highlighting the significance of developing new strategies for constructing sulfur-containing lead compounds. Thioamides are particularly attractive in this context as amide bond bioisosteres.[9]

In 2015, Kim and co-workers reported a highly selective C7-allylation of indolines with allylic carbonates under rhodium catalysis (Scheme [1]A).[10] Later, the Glorius group reported the first cobalt-catalyzed C–H allylation reaction.[11] This allylation proceeded efficiently at room temperature, which is unprecedented for a Co(III) catalyst (Scheme [1]B). Ruthenium, as a common transition-metal catalyst, has a wide range of applications in C–H bond activation. Kim’s group revealed the Ru(II)-catalyzed oxidative allylation, crotonylation, and prenylation of aromatic and unsaturated carboxamides with allyl carbonates.[12] These transformations were applied to a wide range of substrates (Scheme [1]C). This letter reports the first ruthenium(II)-catalyzed C–H allylation of N,N-dialkylthiobenzamides with allyl methyl carbonate. The method, which employs inexpensive and commercially available reagents and catalysts, is performed in sustainable and ecofriendly MeCN as the solvent (Scheme [1]D).

Initially, 1-(phenylcarbonothioyl)pyrrolidine (1a) and allyl methyl carbonate (2) were chosen as model substrates for condition optimization of the Ru(II)-catalyzed C–H allylation reaction. The solvent was screened first. Several solvents commonly employed in C–H bond activation were tested, some of which successfully propelled the reaction (Table [1], entries 1–5). THF did not perform well in this reaction, although it is often used in oxygen-atom-directed reactions. Ag₂O was the preferred Ru(II) oxidant, with AgSbF₆ and Ag₂CO₃ showing little to no conversation (entries 6 and 7). The effect of various bases on the reaction was next examined. The effects of various acetates, carbonates, and phosphates on the reaction were examined (entries 8–11). Among these, K₂CO₃ and Cs₂CO₃ promoted the reaction, with K₂CO₃ having the superior effect. When Cu(OTf)₂ was used as an additive, the target product was obtained in a yield of 21% (entry 12). Subsequently, lowering the reaction temperature significantly reduced the reaction efficiency (entry 14). When the reaction was not protected by an inert gas, none of the target product was detected, because air caused the sulfur atoms to be replaced by oxygen atoms under heating conditions (entry 15). Finally, shortening the reaction time to 12 hours led to a yield of about 50%. (entry 16). Therefore, the optimal conditions involve the use of [RuCl₂(p-cymene)]₂ (10 mol%), Cu(OAc)₂ (2 equiv), Ag₂O (0.5 equiv), and K₂CO₃ (2 equiv) in MeCN at 100 °C for 20 hours.

Scheme 1 Transition-metal-catalyzed C–H allylation reactions

Table 1 Optimization of the Reaction Conditions^a

Entry	Oxidant	Additive	Base	Solvent	Yield (%)
1	Ag2O	Cu(OAc)2	K2CO3	THF	51
2	Ag2O	Cu(OAc)2	K2CO3	2-MeTHFb	43
3	Ag2O	Cu(OAc)2	K2CO3	DMF	–
4	Ag2O	Cu(OAc)2	K2CO3	DCE	22
5	Ag2O	Cu(OAc)2	K2CO3	toluene	25
6	AgSbF6	Cu(OAc)2	K2CO3	MeCN	44
7	Ag2CO3	Cu(OAc)2	K2CO3	MeCN	0
8	Ag2O	Cu(OAc)2	K2CO3	MeCN	21
9	Ag2O	Cu(OAc)2	Cs2CO3	MeCN	35
10	Ag2O	Cu(OAc)2	K3PO4	MeCN	19
11	Ag2O	Cu(OAc)2	NaOAc	MeCN	–
12	Ag2O	Cu(OTf)2	K2CO3	MeCN	21
13	Ag2O	Cu(OAc)2	K2CO3	MeCN	76
14c	Ag2O	Cu(OAc)2	K2CO3	MeCN	40
15d	Ag2O	Cu(OAc)2	K2CO3	MeCN	<2
16e	Ag2O	Cu(OAc)2	K2CO3	MeCN	46

^a Reaction conditions: 1a (0.2 mmol), 2 (0.4 mmol), [RuCl₂(p-cymene)]₂ (10 mol%), oxidant (0.1 mmol), additive (0.4 mmol), base (0.4 mmol), solvent (1.5 mL), 100 °C, under Ar, 20 h.

^b At 80 °C.

^c Under air.

^d Reaction time: 12 h.

With the optimized reaction conditions in hand, we investigated the scope and limitations of this transformation (Scheme [2]). The coupling of para-substituted thiobenzamides with allyl methyl carbonate was favored, affording the desired products with high levels of regioselectivity and good yields. Substrates with an electron-withdrawing group (3b) or an electron-donating group (3c) at the para-position of the benzene ring exhibited favorable reactions, with yields of 72 and 59%, respectively. However, when the para-substituent was a strongly electron-donating methoxy group, only a trace of the corresponding product was detected. It is noteworthy that the reaction tolerates readily modifiable halide substituents such as chloro (3e), bromo (3f), or iodo (3g). In addition, when we attempted to link substrates containing a dimethylamino or methoxycarbonyl group, the corresponding product 3h was obtained in a 16% yield from the methoxycarbonyl derivative, whereas the dimethylamino-substituted substrate did not react. Substrates with an electron-deficient group at the ortho- or meta-position also reacted to give the corresponding allylation products 3i–l. Moreover, the reaction could be extended to polyaromatic substrates, such as a naphthyl derivative (3m). We also tested various heterocyclic substrates containing a thiophene or furan group, but unfortunately these two types of substrate did not give the expected products. Next, we investigated the effect of N-substitution on the thiobenzamide scaffold on our C–H functionalization. The reaction proceeded smoothly, and the yield of 3o from the N,N-diethyl-substituted reactant was only slightly lower than that of 3n from the N,N-dimethyl substrate. A morpholine derivative, on the other hand, did not react. Significantly, only mono-C–H allylation products were observed in all cases.

Scheme 2 Substrate scope of the C–H allylation. Reaction conditions: 1 (0.2 mmol), 2 (2.0 equiv), [RuCl₂(p-cymene)]₂ (10 mol%), Ag₂O (0.5 equiv), Cu(OAc)₂ (2.0 equiv), K₂CO₃ (2.0 equiv), MeCN (1.5 mL), 100 °C, under Ar, 20 h. N.D. = not detected.

Based on the previous work by the Maiti group,[13] we used allyl acetate as an alternative reagent for a similar allylation reaction under the same experimental conditions and we were pleased to find that the reaction proceeded with a 33% yield (Scheme [3]).

To elucidate the mechanism of this allylation reaction, several intermolecular competition experiments were performed. First, we conducted two intermolecular competition experiments with various substituted thioamides (Schemes 4A and 4B). The electron-rich thioamide 1b reacted more readily with the allylating reagent than did the electron-deficient thioamide 1b, and the pyrrole-substituted substrate 1a was more reactive than the diethylamine-substituted substrate 1m (Scheme [4]B). As we can see, the reactivity of the directing group is consistent with the steric effect of S-coordination. Deuterium doping revealed a reversible C–H cycloruthenation (Scheme [4]C).

Scheme 3 Allylation reaction with allyl acetate. Reaction conditions: 1a (0.2 mmol), AllOAc (2.0 equiv), [RuCl₂(p-cymene)]₂ (10 mol%), Ag₂O (0.5 equiv), Cu(OAc)₂ (2.0 equiv), K₂CO₃ (2.0 equiv), MeCN (1.5 mL), 100 °C, under Ar, 20 h.

Scheme 4 Mechanistic studies

According to our experimental results and precedents in the literature,[14] a possible reaction pathway for the reaction of an N,N-disubstituted thiobenzamide 1 with allyl methyl carbonate (2) catalyzed by a Ru(II) complex is proposed (Scheme [5]). First, the thioamide derivative is activated by C–H bonding and coordination with Ru(II) to form the stable five-membered cyclometallic intermediate A. Cyclic intermediate A is then transmetalated with allyl methyl carbonate in the presence of a base to form intermediate B, which then undergoes an elimination reaction to form the allylated product 3, and the catalytic cycle proceeds through participation by Ru(II).

In summary, we have developed a novel approach involving a Ru(II)-catalyzed C–H allylation reaction of N,N-dialkylthiobenzamides with allyl methyl carbonate in an adjacent position.[15] This reaction has a broad substrate tolerance, mild reaction conditions, and inexpensive and readily obtainable reaction materials. The method exploits the weaker sulfur ligand for the challenging C–H arylation reaction, using thiobenzamide as a versatile sulfur-containing director group. The choice of Ru(II) complexes as catalysts for this reaction system further enriches the pool of transition-metal-catalyzed C–H bond-activation reactions for constructing functionalized thioamide derivatives, and lays a foundation for the use of ruthenium catalysts.

Scheme 5 A proposed reaction mechanism

• References and Notes

• 1a Lakshman MK, Deb AC, Chamala RR, Pradhan P, Pratap R. Angew. Chem. Int. Ed. 2011; 50: 11400
• 1b Li Z.-K, Jia X.-S, Yin L. Synthesis 2018; 50: 4165
• 1c Rossi R, Lessi M, Manzini C, Bellina F. Adv. Synth. Catal. 2015; 357: 3777
• 2a Cho SH, Kim JY, Kwak J, Chang S. Chem. Soc. Rev. 2011; 40: 5068
• 2b Wencel-Delord J, Dröge T, Liu F, Glorius F. Chem. Soc. Rev. 2011; 40: 4740
• 2c Sambiagio C, Schönbauer D, Blieck R, Dao-Huy T, Pototschnig G, Schaaf P, Wiesinger T, Zia MF, Wencel-Delord J, Besset T, Maes BU. W, Schnürch M. Chem. Soc. Rev. 2018; 47: 6603
• 2d Sedelmeier G, Sedelmeier J. Chimia 2017; 71: 730
• 3a Wang P.-S, Gong L.-Z. Acc. Chem. Res. 2020; 53: 2841
• 3b Nobile E, Castanheiro T, Besset T. Angew. Chem. Int. Ed. 2021; 60: 12170
• 3c Tsuji J. Pure Appl. Chem. 1999; 71: 1539
• 3d Han JF, Guo P, Zhang X.-G, Liao J.-B, Ye K.-Y. Org. Biomol. Chem. 2020; 18: 7740
• 4a Chen Z, Wang B, Zhang J, Yu W, Liu Z, Zhang Y. Org. Chem. Front. 2015; 2: 1107
• 4b Zhu R.-Y, Farmer ME, Chen Y.-Q, Yu J.-Q. Angew. Chem. Int. Ed. 2016; 55: 10578
• 5a Engle KM, Mei T.-S, Wasa M, Yu J.-Q. Acc. Chem. Res. 2012; 45: 788
• 5b Park H, Li Y, Yu J.-Q. Angew. Chem. Int. Ed. 2019; 58: 11424
• 5c He J, Wasa M, Chan KS. L, Shao O, Yu J.-Q. Chem. Rev. 2017; 117: 8754
• 5d Zhang X.-G, Dai H.-X, Wasa M, Yu J.-Q. J. Am. Chem. Soc. 2012; 134: 11948
• 5e Li X, Zhang F, Wu D, Liu Y, Xu G, Peng Y, Liu Z, Huang Y. Tetrahedron 2018; 74: 7364
• 6a Li H.-L, Kanai M, Kuninobu Y. Org. Lett. 2017; 19: 5944
• 6b Rej S, Ano Y, Chatani N. Chem. Rev. 2020; 120: 1788
• 7 Kausar A, Zulfiqar S, Sarwar MI. Polym. Rev. 2014; 54: 185
• 8a Ilardi EA, Vitaku E, Njardarson JT. J. Med. Chem. 2014; 57: 2832
• 8b Beno BR, Yeung KS, Bartberger MD, Pennington LD, Meanwell NA. J. Med. Chem. 2015; 58: 4383
• 8c Liao Y, Wang M, Jiang X. Curr. Opin. Chem. Biol. 2023; 75: 102336
• 9a Saito M, Murakami S, Nanjo T, Kobayashi Y, Takemoto Y. J. Am. Chem. Soc. 2020; 142: 8130
• 9b Verma H, Khatri B, Chakraborti S, Chatterjee J. Chem. Sci. 2018; 9: 2443
• 9c Newberry RW, VanVeller B, Guzei IA, Raines RT. J. Am. Chem. Soc. 2013; 135: 7843
• 9d Arockiam PB, Bruneau C, Dixneuf PH. Chem. Rev. 2012; 112: 5879
• 10 Park J, Mishra NK, Sharma S, Han S, Shin Y, Jeong T, Oh JS, Kwak JH, Jung YH, Kim IS. J. Org. Chem. 2015; 80: 1818
• 11 Yu D.-G, Gensch T, de Azambuja F, Vásquez-Céspedes S, Glorius F. J. Am. Chem. Soc. 2014; 136: 17722
• 12 Kim M, Sharma S, Mishra NK, Han S, Park J, Kim M, Shin Y, Kwak JH, Han SH, Kim IS. Chem. Commun. 2014; 50: 11303
• 13 Dutta S, Bhattacharya T, Werz DB, Maiti D. Chem 2021; 7: 555
• 14a Duan Y.-T, Wang Z.-X. J. Org. Chem. 2023; 88: 16076
• 14b Binnani C, Arora S, Priya B, Gupta P, Singh SK. Chem-Asian J. 2023; 18: 11
• 14c Zhang Y.-B, Li B.-S, Xu G.-J, Sun W, Sun M. Org. Lett. 2023; 25: 3922
• 14d Mondal S, Bera R, Chowdhury D, Dana S, Baidya M. Org. Lett. 2023; 25: 70
• 15 Allylation of N,N-Dialkylthiobenzamides; General Procedure A pressure tube was charged with [RuCl₂(p-cymene)]₂ (10 mol%), Cu(OAc)₂ (2.0 equiv), Ag₂O (0.5 equiv), K₂CO₃ (2.0 equiv), the appropriate N,N-dialkylthiobenzamide 1 (0.20 mmol, 1.0 equiv), and allyl methyl carbonate (2; 2.0 equiv). MeCN (1.5 mL) was then added with vigorous stirring at r.t. The tube was placed in an oil bath, and the mixture was stirred for the appropriate time at 100 °C. The mixture was then diluted with EtOAc (20 mL), filtered, and concentrated. The crude product was purified by column chromatography [silica gel, hexanes–EtOAc (10:1)]. 1-[(2-Allylphenyl)carbonothioyl]pyrrolidine (3a) Yellow oil; yield: 77%. ¹H NMR (500 MHz, CDCl₃): δ = 7.23 (d, J = 11.4 Hz, 3 H), 7.09 (d, J = 7.1 Hz, 1 H), 5.92 (ddd, J = 24.1, 9.0, 7.0 Hz, 1 H), 5.07 (dd, J = 24.7, 13.5 Hz, 2 H), 4.03 (dt, J = 13.8, 6.8 Hz, 1 H), 3.91 (dt, J = 14.2, 7.3 Hz, 1 H), 3.51 (dd, J = 15.2, 8.1 Hz, 1 H), 3.38 (dd, J = 15.2, 5.4 Hz, 1 H), 3.34–3.24 (m, 1 H), 3.26–3.11 (m, 1 H), 2.10–2.01 (m, 2 H), 1.93 (dp, J = 19.8, 6.5 Hz, 2 H). ¹³C NMR (126 MHz, CDCl₃): δ = 196.58, 136.53, 133.84, 129.96, 128.81, 126.65, 124.86, 116.19, 53.03, 52.41, 37.18, 26.19, 24.60. HRMS (ESI): m/z [M + H]⁺ calcd for C₁₄H₁₈NS: 232.35; found: 232.19.