Palladium-catalyzed oxidative C–O cross-coupling of ketene dithioacetals and carboxylic acids

Abstract

Direct oxidative C–O cross-coupling reaction between active alkenes and carboxylic acids is presented. By utilizing Pd(OAc)₂ as a catalyst and PhI(OAc)₂ as an oxidant, ketene dithioacetals were smoothly acyloxylated in carboxylic acid–water solution, affording a variety of vinyl esters in a highly selective way with high efficiency and good functional group tolerance. A plausible mechanism is proposed that features a vinyl iodonium species as key intermediate.

---

Alkene functionalization is an important and long-standing goal in organic chemistry.¹ In this field, oxygenation of olefins represents a key step for the synthesis of numerous structurally interesting compounds.^1,2 However, direct olefinic C–H oxygenation has received less attention likely due to the large driving force for addition reaction or allylation reaction of the double bonds upon coupling reaction.^1–7

Limited methodologies have been developed for the acetoxylation of some active olefins (e.g., enol silyl ethers,³ enamines,⁴ ketene N/S,S-acetals⁵). Pb(OAc)₄ proved to be an efficient acetoxylating reagent for this purpose though the employment of a large amount of highly toxic lead salt is not encouraged.^4d,e,5 Additionally, the acetoxylation was found to accompany with α-iodination and N-arylation of enol silyl ethers³ when treating enol silyl ethers with PhI(OAc)₂. The acetoxylation of acyclic enamines was also observed when performing the homocoupling of them by the action of PhI(OAc)₂ in the presence of BF₃·Et₂O.⁴ However, those reactions involving PhI(OAc)₂ as acetoxylating reagent are usually available to only one or few specific substrates.^{3,4b–d,5b–e,6,7} As a consequence, it is highly desirable to develop an environmentally benign approach toward the goal of direct olefinic C–H acyloxylation with high efficiency and a wide substrate scope.^4a

α-Oxo ketene dithioacetals are kinds of polarized alkenes which have been used as versatile synthons for the synthesis of various carbo- and heterocyclic compounds.⁸ The α-C–H functionalization of them has become attractive transformation as it furnishes more densely functionalized molecules with synthetic potential.⁹ Our continuing interest in the α-functionalization of ketene dithioacetals prompted us to investigate their reactivity in new C–O cross-coupling reactions. Herein, we report a novel Pd(OAc)₂-catalyzed olefinic C–H acyloxylation of ketene dithioacetals by using carboxylic acids as coupling partner and PhI(OAc)₂ as oxidant, furnishing the target vinyl esters.

At the outset of the present investigation, α-oxo ketene dithioacetal 1a was employed as model substrate to screen the reaction conditions. The results were summarized in Table 1. First of all, several oxidants were evaluated to perform the coupling reaction under the catalysis of 10 mol% of Pd(OAc)₂ in acetic acid at 50 °C. It proved that the use of AgOAc, Cu(OAc)₂, Oxone, para-benzoquinone (BQ) and PhI(O₂CCF₃)₂ was unsuccessful (Table 1, entries 1–5). To our delight, upon treatment of 1a with 1.2 equivalents of PhI(OAc)₂ as the oxidant for 1 h, a satisfactory yield (81%) of 2a was achieved (Table 1, entry 6). Next, other solvents instead of acetic acid were tested under otherwise the same conditions. Reactions carried out in CH₂Cl₂, CH₃CN and DMF were found to result in very low yields (Table 1, entries 7–9), even the addition of 10 equivalents of HOAc did not bring forth any improvement (Table 1, entry 10). Interestingly, water proved to benefit the acetoxylation (Table 1, entries 11–14). When the ratio of HOAc to H₂O was 20 : 1 or 10 : 1 (v/v), 2a was separated in nearly quantitative yields (Table 1, entries 12 and 13).‡ Temperature also significantly affected this transformation. When the reaction was run at ambient temperature or 80 °C, a decrease in the yield of 2a was observed (Table 1, entries 15 and 16). Then, other Pd catalysts, such as PdCl₂ and [Pd(PPh₃)₄], were tested in comparison with Pd(OAc)₂ and found to be less efficient (Table 1, entries 17 and 18). It is worthy of notice that a reduced catalyst loading of 5 mol% of Pd(OAc)₂ was sufficient to give a satisfactory yield (84%) within 2 h (Table 1, entry 19). However, in the absence of any Pd catalyst under otherwise optimized conditions, no target product was generated (Table 1, entry 20).

Table 1 Optimization of the reaction conditions^a

Entry	Catalyst	Oxidant	Solvent	T/°C	t/h	Yield^b (%)
a Reaction conditions: 1a (1.0 mmol), Pd catalyst (0.1 mmol), oxidant (1.2 mmol), solvent (10 mL).b Isolated yields.c 2.4 equiv. of oxidant was used.d 10 equiv. of HOAc was added.e 5 mol% of Pd(OAc)2 was used.f Only an unidentified great polar product was detected.g 10 mol% of Pd(OAc)2 was added after 1a had reacted with PhI(OAc)2 for 0.5 h, then the mixture was stirred for another 1 h.
1	Pd(OAc)2	AgOAc^c	HOAc	50	12	Trace
2	Pd(OAc)2	Cu(OAc)2^c	HOAc	50	12	Trace
3	Pd(OAc)2	Oxone	HOAc	50	12	Trace
4	Pd(OAc)2	BQ^c	HOAc	50	12	17
5	Pd(OAc)2	PhI(O2CCF3)2	HOAc	50	12	12
6	Pd(OAc)2	PhI(OAc)2	HOAc	50	1	81
7	Pd(OAc)2	PhI(OAc)2	CH2Cl2	50	12	11
8	Pd(OAc)2	PhI(OAc)2	CH3CN	50	12	8
9	Pd(OAc)2	PhI(OAc)2	DMF	50	12	15
10^d	Pd(OAc)2	PhI(OAc)2	DMF	50	12	12
11	Pd(OAc)2	PhI(OAc)2	HOAc/H2O(50 : 1)	50	1	81
12	Pd(OAc)2	PhI(OAc)2	HOAc/H2O(20 : 1)	50	1	95
13	Pd(OAc)2	PhI(OAc)2	HOAc/H2O(10 : 1)	50	1	98
14	Pd(OAc)2	PhI(OAc)2	HOAc/H2O(5 : 1)	50	1	91
15	Pd(OAc)2	PhI(OAc)2	HOAc/H2O(10 : 1)	rt	12	64
16	Pd(OAc)2	PhI(OAc)2	HOAc/H2O(10 : 1)	80	1	87
17	PdCl2	PhI(OAc)2	HOAc/H2O(10 : 1)	50	1	23
18	[Pd(PPh3)4]	PhI(OAc)2	HOAc/H2O(10 : 1)	50	1	60
19	Pd(OAc)2^e	PhI(OAc)2	HOAc/H2O(10 : 1)	50	2	84
20	—	PhI(OAc)2	HOAc/H2O(10 : 1)	50	1	0^f
21	Pd(OAc)2^g	PhI(OAc)2	HOAc/H2O(10 : 1)	50	1	92

---

After achieving the optimized reaction conditions, the scope of ketene dithioacetals 1 was explored by varying the electron-withdrawing groups (EWGs) and SR groups (Table 2). We are pleased to find that the reactions of a series of ketene cyclic dithioacetals with various α-EWGs, including alkanoyl (Table 2, entries 1 and 2), electron-rich/-deficient aroyl (Table 2, entries 3–6) and alkoxycarbonyl groups (Table 2, entry 7), proceeded smoothly to afford the expected acetoxylated products 2a–g in moderate to nearly quantitative yields within 1–6 h. In addition, the conditions were mild enough to be compatible with an aryl C–Br bond, which could be further transformed into different functionalities (Table 2, entry 5). The electron-rich furan ring (Table 2, entry 6) was also tolerant to the reaction, without any side oxidation processes detected. In the case of using acyclic ketene dithioacetal 1h as substrate, low yield of 2h was obtained (Table 2, entry 8). We also attempted to optimize the reaction with substrate 1h by increasing the amounts of Pd(OAc)₂ and PhI(OAc)₂ or changing the ratio of mixed solvent, but the reaction yield could not be improved.

Table 2 Acetoxylation of ketene dithioacetals 1^a

Entry	1	EWG	R	t/h	2	Yield^b (%)
a Reaction conditions: 1 (1.0 mmol), Pd(OAc)2 (0.1 mmol) and PhI(OAc)2 (1.2 mmol) were stirred in HOAc/H2O (10 : 1, 10 mL) at 50 °C.b Isolated yields.
1	1a	MeCO	–CH2CH2–	1	2a	98
2	1b	t-BuCO	–CH2CH2–	6	2b	70
3	1c	PhCO	–CH2CH2–	3	2c	80
4	1d	4-MePhCO	–CH2CH2–	3	2d	61
5	1e	4-BrPhCO	–CH2CH2–	3	2e	76
6	1f	2-furylCO	–CH2CH2–	3	2f	66
7	1g	CO2Et	–CH2CH2–	3	2g	67
8	1h	MeCO	Et	3	2h	23

---

With the purpose of further probing the reactivity and regioselectivity of this new procedure, a range of reactions was carried out with various α-alkenoyl ketene dithioacetals 1i–p, and the results are summarized in Table 3. In order to improve the reaction efficiency, 20 mol% of Pd(OAc)₂ should be used. It proved that all of the substrates having phenyl (Table 3, entry 1), electron-rich (Table 3, entries 2 and 3), electron-deficient (Table 3, entries 4 and 5) aromatic and hetero aromatic groups (Table 3, entry 6) at the β′-position of the α-alkenoyl moiety can efficiently furnish the corresponding α-C–O coupling products 2i–n in moderate to good yields. Triene 1o gave the desired polyene product 2o in good yield as well (Table 3, entry 7). α-Alkenoyl dithioacetal 1p with aliphatic substitutent at the β′-position was also suitable substrate and provided 2p in 87% yield (Table 3, entry 8). It is noteworthy that in all cases, no α′-acetoxylated products were detected. Thus, we provided here a highly regioselective C–O cross-coupling reaction.

Table 3 Acetoxylation of α-alkenoyl dithioacetals 1i–p^a

Entry	1	R′	2	Yield^b (%)
a Reaction conditions: 1i–p (1.0 mmol), Pd(OAc)2 (0.2 mmol) and PhI(OAc)2 (1.2 mmol) were stirred in HOAc/H2O (10 : 1, 10 mL) at 50 °C for 3 h.b Isolated yields.
1	1i	Ph	2i	55
2	1j	4-MePh	2j	75
3	1k	3,4-O2CH2Ph	2k	68
4	1l	4-CF3Ph	2l	79
5	1m	2-ClPh	2m	66
6	1n	2-Thienyl	2n	53
7	1o	PhCH=CH	2o	70
8	1p	t-Bu	2p	87

---

In order to investigate the reaction mechanism, we next carried out the above reaction by using other carboxylic acids as solvents. As expected, propanoic acid–water (10 : 1) worked well as an alternative solvent under otherwise optimized conditions described in Table 1, entry 13, providing the acyloxylation product 3a in moderate yield (Table 4, entry 1). When dichloroacetic acid–water (10 : 1) was employed, the acyloxylation product 3b was only obtained in 13% yield along with the formation of the hydrolyzed product 4 in 40% yield (Table 4, entry 2). Whereas, the reaction of 1a with trifluoroacetic acid (TFA) afforded 4 as the sole product in the yield of 24% (Table 4, entry 3). These findings suggest that the acyloxy groups incorporated in products 2 and 3 did not ultimately originate from hypervalent iodine reagents but rather from the solvents.

Table 4 Cross-coupling reactions of ketene dithioacetal 1a with carboxylic acids^a

Entry	R′′	3	Yield of 3^b (%)	Yield of 4^b (%)
a Reaction conditions: 1a (1.0 mmol), Pd(OAc)2 (0.1 mmol) and PhI(OAc)2 (1.2 mmol) were stirred in R′′COOH/H2O (10 : 1, 10 mL) at 50 °C for 3 h.b Isolated yields.c Anhydrous TFA was used as solvent, with the addition of 4 Å MS (400 mg).
1	Et	3a	50	0
2	CHCl2	3b	13	40
3^c	CF3	—	—	24

---

In general, it is considered that Pd-catalyzed direct acetoxylation using PhI(OAc)₂ as a terminal oxidant in acetic acid, conditions developed by Crabtree,¹⁰ involves palladation of a C–H bond with Pd(II) species as the first step.¹¹ However, Suna recently revealed that the acetoxylation of electron-rich heterocycles (i.e., indoles, pyrroles) under Crabtree condition occurs via the initial formation of heteroaryl(phenyl)iodonium acetates rather than via the initial carbopalladation by C–H activation. The key iodonium intermediates, which could be isolated, are then smoothly transformed into acetoxylated products in the presence of a Pd catalyst.^12,13

In present investigation, as shown in Table 1, entry 20, ketene dithioacetal 1a disappeared rapidly within 20 minutes in the absence of any Pd catalyst under otherwise optimized conditions, however, no target product was generated even with a somewhat prolonged reaction time (1 h), leading to the formation of a great polar product which was not stable enough to be isolated and characterized, but identified as a vinyl iodonium salt^14,15 by ESI/MS experiment.¹⁶ By contrast, acetylated product 2a could be obtained in excellent yield by a subsequent addition of 10 mol% of Pd(OAc)₂ after the disappearance of starting material 1a (Table 1, entry 21).

On the basis of the above observations and seminal works,^12,13 a plausible mechanism for this transformation is outlined in Scheme 1. To start with, ketene dithioacetals 1 react with PhI(O₂CR′′)₂, generated in situ by ligand metathesis of PhI(OAc)₂ with carboxylic acids, to form the intermediate vinyl iodonium salts A, which are subsequently coordinated to Pd-catalyst via the sulfur atom, furnishing intermediates B. Then, Pd-catalyst coordinates both the sulfur atom and the double bond, affording the η³-coordination complexes C, which ensure the high regioselectivity of the ketene dithioacetal moiety transfer to the Pd center. Subsequent oxidative addition would generate transient Pd(IV)¹⁷ or dinuclear Pd(III) complexes¹⁸ D. Finally, reductive elimination of D delivers acyloxylated products 2 or 3, as well as regenerates Pd(II) salts to finish the catalytic cycle.

Scheme 1 Proposed mechanism for acyloxylation of ketene dithioacetals 1.

Conclusions

In conclusion, a novel palladium-catalyzed oxidative acyloxylation reaction of ketene dithioacetals with carboxylic acids has been successfully developed to synthesize functionalized vinyl esters in a highly regioselective way with high efficiency and tolerance of diverse functional groups. This protocol could complement the powerful direct cross-coupling strategy and make it more universal. Detailed mechanistic investigations are currently underway in our group.

Acknowledgements

Financial support of this research by the NCET-11-0613, NNSFC (21172031/21272034/21372040) is gratefully acknowledged.

Notes and references

1. Recent reviews on alkenyl C–H functionalization, see: (a) M. S. Sigman and E. W. Werner, Acc. Chem. Res., 2012, 45, 874–884; (b) R. I. McDonald, G. Liu and S. S. Stahl, Chem. Rev., 2011, 111, 2981–3019; (c) E. M. Beccalli, G. Broggini, M. Martinelli and S. Sottocornola, Chem. Rev., 2007, 107, 5318–5365.
2. For selected examples, see: (a) S. R. Chidipudi, M. D. Wieczysty, I. Khan and H. W. Lam, Org. Lett., 2013, 15, 570–573; (b) M. Yamamoto, S. Nakaoka, Y. Ura and Y. Kataoka, Chem. Commun., 2012, 48, 1165–1167; (c) W. Liu, Y. Li, K. Liu and Z. Li, J. Am. Chem. Soc., 2011, 133, 10756–10759; (d) T. P. Pathak, K. M. Gligorich, B. E. Welm and M. S. Sigman, J. Am. Chem. Soc., 2010, 132, 7870–7871.
3. I. I. Brunovlenskaya, K. M. Kusainova and A. N. Kashin, Zh. Org. Khim., 1988, 24, 358–363.
4. (a) X. Liu, R. Cheng, F. Zhao, D. Zhang-Negrerie, Y. Du and K. Zhao, Org. Lett., 2012, 14, 5480–5483; (b) Y. Chen, T. Ju, J. Wang, W. Yu, Y. Du and K. Zhao, Synlett, 2010, 231–234; (c) J. Wang, S. Liu and W. Yu, Synlett, 2009, 2529–2533; (d) G. R. Lenz and C. Costanza, J. Org. Chem., 1988, 53, 1176–1183; (e) C. R. Dorn, F. J. Koszyk and G. R. Lenz, J. Org. Chem., 1984, 49, 2642–2644.
5. (a) L. Bhat, A. Thomas, H. Ila and H. Junjappa, Tetrahedron, 1991, 47, 305–312; (b) A. Thomas, J. N. Vishwakarma, S. Apparao, H. Ila and H. Junjappa, Tetrahedron, 1988, 44, 1667–1672; (c) K. Hiroi, S. Sato and K. Matsuo, Chem. Pharm. Bull., 1980, 28, 558–566; (d) B. M. Trost, K. Hiroi and L. N. Jungheim, J. Org. Chem., 1980, 45, 1839–1847; (e) K. Hiroi and S. Sato, Chem. Lett., 1979, 923–926.
6. (a) W. H. Henderson, C. T. Check, N. Proust and J. P. Stambuli, Org. Lett., 2010, 12, 824–827; (b) A. M. Balija, K. J. Stowers, M. J. Schultz and M. S. Sigman, Org. Lett., 2006, 8, 1121–1124; (c) S. Fukuzawa, K. Takahashi, H. Kato and H. Yamazaki, J. Org. Chem., 1997, 62, 7711–7716.
7. S. E. Mann, A. E. Aliev, G. J. Tizzard and T. D. Sheppard, Organometallics, 2011, 30, 1772–1775.
8. For a recent review on ketene dithioacetal chemistry, see: L. Pan, X. Bi and Q. Liu, Chem. Soc. Rev., 2013, 42, 1251–1286.
9. Recent examples for α-functionalization of ketene dithioacetals, see: (a) Y. Liu, J. Liu, M. Wang, J. Liu and Q. Liu, Adv. Synth. Catal., 2012, 354, 2678–2682; (b) H. Yuan, M. Wang, Y. Liu, L. Wang, J. Liu and Q. Liu, Chem. – Eur. J., 2010, 16, 13450–13457; (c) D. Liang, M. Wang, B. Bekturhun, B. Xiong and Q. Liu, Adv. Synth. Catal., 2010, 352, 1593–1599; (d) Y. Liu, M. Wang, H. Yuan and Q. Liu, Adv. Synth. Catal., 2010, 352, 884–892.
10. T. Yoneyama and R. H. Crabtree, J. Mol. Catal. A: Chem., 1996, 108, 35–40.
11. It has been shown that C–H activation by a Pd catalyst is the rate-determining step: (a) M. H. Emmert, A. K. Cook, Y. J. Xie and M. S. Sanford, Angew. Chem., Int. Ed., 2011, 50, 9409–9412; (b) L. V. Desai, K. J. Stowers and M. S. Sanford, J. Am. Chem. Soc., 2008, 130, 13285–13293.
12. E. Suna, Chem. Heterocycl. Compd., 2012, 48, 44–48.
13. (a) D. Lubriks, I. Sokolovs and E. Suna, J. Am. Chem. Soc., 2012, 134, 15436–15442; (b) E. Suna, I. Sokolovs and D. Lubriks, Org. Lett., 2011, 13, 4324–4327.
14. M. S. Yusubov, A. V. Maskaev and V. V. Zhdankin, ARKIVOC, 2011, i, 370–409.
15. K. Miyamoto, T. Okubo, M. Hirobe, M. Kunishima and M. Ochiai, Tetrahedron, 2010, 66, 5819–5826.
16. A vinyl iodonium salt E was recognized under the conditions described in Table 1, entry 20 by ESI/MS experiment. For details of ESI/MS spectra, see the ESI.†
    .
17. L. Xu, B. Li, Z. Yang and Z. Shi, Chem. Soc. Rev., 2010, 39, 712–733.
18. (a) D. Powers and T. Ritter, Top. Organomet. Chem., 2011, 35, 129–156; (b) D. C. Powers and T. Ritter, Nat. Chem., 2009, 1, 302–309.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental details, characterization data, copies of the ¹H and ¹³C NMR spectra of all final products and ESI/MS spectra of E. See DOI: 10.1039/c3ra47282e

‡ General procedure for cross-coupling reactions of 1 with carboxylic acids (1a as example): a 25 mL flask, equipped with a magnetic-stirring bar, was charged with ketene dithioacetal 1a (160 mg, 1.0 mmol), PhI(OAc)₂ (387 mg, 1.2 mmol), and Pd(OAc)₂ (23 mg, 0.1 mmol), followed by addition of 9.1 mL acetic acid and 0.91 mL water. The reaction mixture was stirred at 50 °C for 1 h. Then it was cooled to room temperature and poured into 50 mL ice-water under stirring. After neutralized by saturated aqueous K₂CO₃ solution, the resulting mixture was extracted with CH₂Cl₂ three times. The extract was dried over anhydrous MgSO₄. After removal of solvents, the residue was purified by column chromatography on silica gel (petroleum ether : diethyl ether = 7 : 1, V/V) to afford the product 2a as a white solid (213 mg, 98% yield).

---