Palladium-catalysed annulation of β-chloro-α,β-unsaturated esters with internal alkynes leading to 2H-pyran-2-ones

Abstract

Heteroannulation of β-chloro-α,β-unsaturated esters with internal alkynes proceeded in the presence of triethylamine and palladium complexes, bis(triphenylphosphine)palladium species in particular, to afford to 2H-pyran-2-ones. Treatment of methyl (Z)-3-chloro-2-heptenoate with Pd(PPh₃)₄ generates [(Z)-1-butyl-2-methoxycarbonylethenyl]chlorobis(triphenylphosphine)palladium [italic v]ia oxidative addition, which gives the corresponding 2H-pyran-2-one upon addition of 4-octyne. Terminal alkynes also reacted with β-chloro-α,β-unsaturated esters, but the major products were β-alkynylated α,β-unsaturated esters.

---

Introduction

2H-Pyran-2-one derivatives not only are valuable materials for organic synthesis, but also have physiological and biological activities in their own rights. For instance, some trisubstituted 2H-pyran-2-ones are claimed to be potent human immunodeficiency virus protease (HIV PR) inhibitors.¹ Although their synthetic methods have widely been explored,² those based on transition metal complex-catalysed reactions are very limited. Inoue,^2c–e Walther^2f and Tsuda et al.^2g reported nickel-catalysed synthesis starting with alkynes and CO₂. Liebeskind and co-workers^2h,i reported the formation of 2H-pyran-2-ones in rhodium(I)-catalysed carbonylation of cyclopropenyl esters and cyclopropenyl ketones and in palladium-catalysed carbonylative cross-coupling of 4-halogenocyclobutenones with organostannanes.

Recently we have reported the rhodium(I)-catalysed chloroesterification of alkynes with methyl chloroformate.³ This new catalytic reaction provides a simple and efficient method to prepare (Z)-β-chloro-α,β-unsaturated esters (1) of great synthetic potential (Scheme 1). As a part of our studies on the synthetic application of 1, we disclose herein that its palladium-catalysed reactions with internal alkynes successfully proceed in the presence of Et₃N to afford 4,5,6-trisubstituted 2H-pyran-2-ones. After we finished the experiments,⁴ similar reactions starting with β-iodo, β-bromo-, and trifluoromethylsulfonyloxy-α,β-unsaturated esters were reported.⁵

Scheme 1

Results and discussions

We initiated the investigation with the reaction of compound 1a with 4-octyne. In a sealed tube, a toluene (1 mL) solution of 1a, 4-octyne (1.2 equiv.), Et₃N (5 equiv.) and Pd(PPh₃)₂Cl₂ (5 mol% relative to 1a) was heated at 120°C for 20 h. The corresponding heteroannulation product, 4-butyl-5,6-dipropyl-2H-pyran-2-one 2a, was found by GC analysis to be formed in 83% yield, along with the precipitation of insoluble triethylammonium chloride (Scheme 2; Table 1, entry 1). The yield could be increased to 88% when a higher reaction temperature (e.g. 160°C) was used (entry 2).⁶

Scheme 2

Table 1 Reaction of methyl (Z)-3-chloro-2-heptenoate 1a with 4-octyne in the presence of Group 10 metal complex catalysts and Et₃N^a

Entry	Catalyst	Recovery of 1a(%)	2a (%)^b
a All reactions were carried out using 0.2 mmol of 1a, 0.24 mmol of 4-octyne, 1.0 mmol of Et3N, 0.01 mmol of catalyst in 0.4 mL of toluene or toluene-d8 at 120°C for 20 h.b Determined by GC based on the amount of 1a used.c Run at 160°C for 7 h in ethylbenzene.d 0.005 mmol catalyst.e etpo = 4-ethyl-2,4,6-trioxa-1-phosphabicyclo[2.2.2]octane.f dppf = 1,1′-bis(diphenylphosphino)ferrocene.g dppb = 1,4-bis(diphenylphosphino)butane.h Run in a sealed glass tube without using NEt3.i Run under atmospheric pressure of nitrogen in ethylbenzene without using NEt3.
1	PdCl2(PPh3)2	∽0	83
2^c	PdCl2(PPh3)2	∽0	88
3	cis-PdMe2(PPh3)2	∽0	81
4	PdCl2 + 2PPh3	∽0	76
5	[Pd(η^3-C3H5)Cl]2^d + 4PPh3	∽0	78
6	Pd(PPh3)4	84	9
7	PdCl2 + 4PPh3	78	3
8	PdCl2 + 1PPh3	27	59
9	PdCl2 + 2etpo^e	41	44
10	PdCl2(dppf)^f	nd	56
11	PdCl2(dppb)^g	nd	0
12	cis-PdMe2(PMe2Ph)2	75	0
13	PdCl2	>70	0
14	[Pd(η^3-C3H5)Cl]2^d	>70	0
15	Pd(OAc)2	>70	0
16	Ni(cod)2 + 2PPh3	80	12
17	Ni(cod)2 + 1.2dppf^f	71	19
18	Pt(PPh3)4	nd	0
19	Pt(PPh3)2(CH2[double bond, length half m-dash]CH2	nd	0
20^h	cis-PdMe2(PPh3)2	63	14
21^i	cis-PdMe2(PPh3)2	51	27

---

As Table 1 shows, the reaction appears to be best catalysed by bis(triphenylphosphine)palladium species, independent of the structure of the precursor complex. Thus, the catalyst systems generated with cis-PdMe₂(PPh₃)₂, PdCl₂ + 2PPh₃, and [Pd(η³-C₃H₅)Cl]₂ + 4PPh₃ (entries 3–5) all performed nearly equally well to form 2a. On the other hand, the Pd(PPh₃)₄-catalysed reaction gave 2a in only 9% yield under the same reaction conditions (entry 6), suggesting the importance of the vacant coordination site for the oxidative addition of the Cl–C bond ([italic v]ide infra). The species generated from PdCl₂ + 4PPh₃ also resulted in a similar performance (entry 7). The use of phosphine-free palladium compounds such as PdCl₂, [Pd(η³-C₃H₅)Cl]₂ , and Pd(OAc)₂ (entries 13–15) did not form the product at all,⁷ presumably associated with the instability of intermediates and/or the lack of oxidative addition with 1a due to the low electron density at the palladium center as compared with phosphinepalladium species. Accordingly, addition of only equiv. of PPh₃ greatly improved the catalytic performance of the palladium complex (entry 8). Palladium complexes ligated by less donating phosphorus ligand such as etpo (entry 9) and dppf (entry 10) also catalysed the reaction, albeit less actively than the bis(triphenylphosphine) complexes. However, more donating phosphine complexes such as cis-PdMe₂(PMe₂Ph)₂ and PdCl₂(dppb) did not exhibit activity (entries 11 and 12). Nickel complexes also catalysed the reaction, but the activity was much lower than that of the palladium complexes (entries 16 and 17), and platinum complexes did not catalyse the reaction at all (entries 18 and 19). Based on the possible mechanism of the catalysis ([italic v]ide infra), it appeared conceivable that the reaction could hopefully proceed in the absence of triethylamine as chloromethane trapping agent. Indeed, it proceeded even in a sealed glass tube although it was very sluggish (entry 20). Another reaction run in an open system that minimizes accumulation of chloromethane resulted in a better yield (entry 21).

To assess the scope of this annulation, the reactions of compounds 1 with several alkynes have been investigated. The results are summarized in Table 2. It can be seen that the reactions of 1a and 1b with symmetrical dialkyl-substituted alkynes work nicely to give the corresponding trisubstituted 2H-pyran-2-ones in good yields (entries 1, 2, 4). However, unsymmetrical alkynes end up with the formation of regioisomers. At the moment we are unable to control the regiochemistry. For examples the reaction of 1a with 2-pentyne (entry 3) gave an approximately 1:1 mixture of two regioisomers. β-Chloro-α,β-unsaturated esters having cyano (1c) and phenyl (1d) substituents also reacted similarly (entries 5, 6). However, the reaction of 1e having a chloro substituent at the terminus (entry 7) afforded only 9% isolated yield of the corresponding 2H-pyran-2-one (2g; M⁺, m/z = 256 for ³⁵Cl isotope). Another pyranone (2g′), which was isolated in 35% yield as the major product of the reaction, revealed on the basis of mass spectroscopy (M⁺, m/z 220) that the chloro substituent was not retained in this product. ¹H NMR displayed signals in a very high region (δ 0.2–0.5), characteristic of a cyclopropyl group, suggesting that initially formed 2g partially cyclized to form the 4-cyclopropylpyranone (Scheme 3). The cyclization is presumably associated with the acidity of the α-CH₂ in the γ-chloropropyl group bound to the pyranone ring in compound 2g.⁸

Table 2 Pd-catalysed reactions of β-chloroacrylates (1) with alkynes in the presence of Et₃N^a

Scheme 3

More sterically congested alkynes also reacted with compound 1a, albeit very sluggishly (entries 8 and 9). The starting materials were not completely consumed in these reactions even after heating for longer times (70–90 h). The product from 1-phenyl-1-butyne comprised two regioisomers in a ratio of approximately 1.2:1. 2,7-Dimethyloct-1-en-3-yne (90 h) also formed two regioisomers in an approximately 1:1 ratio.

The present procedure was not well applicable to terminal alkynes. However, a Hagihara–Heck type coupling product was formed readily (Scheme 4). Thus the reaction between compound 1a and 1-hexyne run for 5 h under the standard conditions using Pd Cl₂(PPh₃)₂ as the catalyst gave methyl 3-butyl-(Z)-2-nonen-4-ynoate 3 in 54% yield.⁹ GC-MS showed formation of the corresponding 2H-pyran-2-one derivatives only in a small quantity (<5%).

Scheme 4

Under the standard reaction conditions, other functionalized alkynes, such as dimethyl acetylenedicarboxylate, methyl 2-octynoate, 3-hexyn-2-one and 2-butyn-1-yl acetate, did not form 2H-pyran-2-one derivatives in appreciable quantities. Oligomerization of these alkynes appeared to precede the annulation reaction.

On the basis of the results and the known palladium chemistry, a possible mechanistic explanation for the formation of compound 2 can be summarized as shown in Scheme 5. This proposal is closely related to one for isocoumarin formation.^5b It involves oxidative addition of 1 to bis(phosphine)palladium(0) species, generating the (β-methoxycarbonylvinyl)palladium intermediate 4, insertion of alkyne into the Pd–C bond and nucleophilic attack of the car bonyl oxygen on the palladium center to result in the palladacycle 5. Finally, elimination of triethylmethylammonium chloride and reductive elimination furnish 2H-pyran-2-one and regenerate Pd⁰.¹⁰ Among the elemental processes involved in the catalytic cycle, those relevant to the oxidative addition and the reactivity of the resulting complex are substantiated by the following experiments. Thus, treatment of Pd(PPh₃)₄ with 1a at 90°C for 3 h in toluene caused the solution to change from yellow to pale yellow and very cleanly generated Z-4, which was isolated in 83% yield after recrystallization. Complex Z-4, when used as the catalyst (5 mol%) for the reaction of 1a and 4-octyne under the standard conditions, afforded 2a in 87% yield. In addition, when Z-4 was treated in toluene with 4-octyne (5 equiv.) in the presence of Et₃N (5 equiv.) at 120°C for 10 h, 2a was obtained in 65% yield, supporting that the generation of Z-4 triggered the catalysis. We have been unsuccessful in isolating or spectroscopically characterizing the intermediates that follow Z-4. However, according to the mechanistic proposal, it was envisaged that an external nucleophile added to the reaction between complex 4 and alkyne could retard the conversion of 4 into 2 by suppressing nucleophilic attack of the carbonyl oxygen on the palladium(II) center at the intermediate stage like 5. Indeed, when the foregoing reaction of Z-4 with 4-octyne (5 equiv.) in the presence of Et₃N (5 equiv.) was repeated in the presence of 1 equiv. of triphenylphosphine the yield of 2a decreased to 23% ([italic v]s. 65% observed in the absence of the extra PPh₃). Another similar reaction in the presence of tetrahydrofuran (50 equiv. relative to Z-4) also resulted in a lower yield of 2a (44%).¹¹ The PdCl₂(PPh₃)₂-catalysed reaction of 1a with 4-octyne (entry 1, Table 1) was also retarded when run in the presence of THF (5 equiv. relative to 1a) to give only 54% GC yield of 2a. The low catalytic activity of Pd(PPh₃)₄ as compared with PdCl₂(PPh₃)₂ is presumably associated with the oxidative addition step. However, if we consider that the oxidative addition of 1a with Pd(PPh₃)₄ could proceed even at 90°C ([italic v]ide supra), we cannot rule out the possibility that the low activity of Pd(PPh₃)₄ is due at least partially to the free PPh₃ generated from it under the catalytic conditions, which causes a similar negative effect on the nucleophilic attack of the carbonyl oxygen.

Scheme 5

In summary, we have developed a new palladium-catalysed annulation reaction to furnish 4,5,6-trisubstituted 2H-pyran-2-ones, starting with internal alkynes and β-chloro-α,β-unsaturated esters 1, the latter of which can readily be obtained by rhodium-catalysed addition of chloroformates to terminal alkynes. Further study of the synthetic elaboration of 1 is in progress.

Experimental

General

All ¹H, ³¹P-{H} and ¹³C NMR spectra were recorded on a Bruker ARX-300 spectrometer at 300, 121.5 and 75.5 MHz, and referenced to SiMe₄ (¹H), appropriate solvent resonances (¹³C) and H₃PO₄ (³¹P), respectively. GC-MS was performed on a Shimadzu GC-17A/QP-5000 mass spectrometer by using EI (70 eV) with an OV-1701 (25 m) column. Infrared spectra were obtained on a JASCO FT/IR-5000 spectrometer.

Synthesis and product characterization

General procedure for palladium-catalysed annulation of compounds 1 with alkynes. In a thick-walled Pyrex tube was placed a mixture of compound 1 (0.5 mmol), alkyne (0.6 mmol), Et₃N (2.5 mmol), palladium catalyst (0.025 mmol) and toluene (1.0 mL) under nitrogen. The tube was then flame-sealed and heated at 120°C. The progress of the reaction could be followed visually by the precipitation of insoluble ammonium chloride. After cooling, removal of precipitate by filtration and evaporation of volatiles in [italic v]acuo, the residue was extracted with hexane (5.0 mL) (except for 2e, which was extracted with diethyl ether). The extract, after concentration (to ca. 1 mL), was chromatographed on alumina with an appropriate eluent to afford pure product 2.

The experiments summarized in Table 1 were conducted on an 0.2 mmol scale (with respect to compound 1a). After removal of precipitate by filtration, the filtrate was diluted with toluene to 2.0 mL and analysed by gas chromatography after addition of an appropriate amount of ferrocene as internal standard.

4-Butyl-5,6-dipropyl-2H-pyran-2-one 2a. Chromatography on alumina with hexane as eluent gave a pale yellow oil (87.0 mg, 0.37 mmol, 74%), which was distilled on Kugelrohr to furnish an analytically pure colorless oil; bath temperature 120°C/0.9 Torr. ¹H NMR (C₆D₆): δ 5.92 (s, 1H, olefinic CH), 2.13 (t, 2H, J = 7.4 Hz), 1.95–1.90 (m, 4H), 1.52 (m, 2H), 1.12 (m, 6H) and 0.77–0.72 (m, 9H). ¹³C NMR (C₆D₆): δ 161.6, 160.9, 159.4, 114.7, 111.4, 32.8, 32.0, 30.7, 28.0, 24.1, 22.6, 21.1, 14.1, 13.9 and 13.8. IR (neat): 2964, 2936, 2876, 1729, 1632 and 1545 cm⁻¹. GC-MS: m/z (% relative intensity) 236 (M⁺, 21), 207 (11), 194 (24), 179 (40), 166 (100), 151 (56), 137 (41) and 71 (72). Calc. for C₅H₈O₂: C, 76.27; H, 10.17. Found: C, 76.30; H, 10.31%. HRMS: calc. for C₁₅H₂₄O₆m/z 236.1775, found 236.1795.

4-Butyl-5,6-diethyl-2H-pyran-2-one 2b. Chromatography on alumina with hexane as eluent gave a pale yellow oil (75.0 mg, 0.36 mmol, 72%), which was distilled from Kugelrohr to furnish an analytically pure colorless oil; bath temperature 110°C/1.2 Torr. ¹H NMR (C₆D₆): δ 5.90 (s, 1H, olefinic CH), 2.05–1.79 (m, 6H) and 1.10–0.67 (m, 13H). ¹³C NMR (C₆D₆): δ 161.8, 161.7, 159.3, 115.4, 111.5, 31.8, 32.6, 24.1, 22.6, 19.1, 15.0, 13.9 and 12.1. IR (neat): 2962, 2936, 2876, 1723, 1634 and 1547 cm⁻¹. GC-MS: m/z (% relative intensity) 208 (M⁺, 13), 166 (23), 138 (100), 123 (40), 109 (51) and 57 (84). Calc. for C₁₃H₂₀O₂: C, 75.00; H, 9.62. Found: C, 75.10; H, 9.74%.

Reaction of compound 1a with 2-pentyne. Chromatography on alumina with hexane as eluent gave a pale yellow oil (65.0 mg, 0.34 mmol, 67%), which was a 1:1 regioisomeric mixture 2c. Kugelrohr distillation furnished an analytically pure colorless oil; bath temperature 105–110°C/1.5 Torr. ¹H NMR (C₆D₆): δ 5.88 (s, 1H, olefinic CH) and 2.05–0.48 (m, 17H). ¹³C NMR (C₆D₆): δ 161.7, 161.6, 161.3, 159.7, 159.1, 157.4, 116.0, 111.4, 111.0, 109.1, 32.7, 31.8, 30.6, 30.0, 24.6, 22.6, 22.5, 19.3, 16.7, 14.1, 13.9, 11.5 and 11.1. IR (neat): 2962, 2936, 2876, 1717, 1634 and 1549 cm⁻¹. GC-MS: m/z (% relative intensity) 194 (M⁺, 10), 152 (11), 124 (100), 109 (40), 95 (25), 79 (16), 67 (24) and 57 (34). Calc. for C₆H₉O: C, 74.23; H, 9.28. Found: C, 73.95; H, 9.44%.

4-Hexyl-5,6-dipropyl-2H-pyran-2-one 2d. Chromatography on alumina first with hexane and then with a 5:95 (v/v) mixture of ethyl acetate and hexane gave a pale yellow oil (81.0 mg, 0.31 mmol, 62%), which was distilled over Kugelrohr to furnish an analytically pure colorless oil; bath temperature 130°C/0.8 Torr. ¹H NMR (C₆D₆): δ 5.95 (s, 1H, olefinic CH), 2.12 (t, 2H, J = 7.6), 1.95 (m, 4H), 1.50 (m, 2H), 1.22–1.08 (m, 10H), 0.87 (t, 3H, J = 7.0), 0.75 (t, 3H, J = 7.3) and 0.74 (t, 3H, J = 7.4 Hz). ¹³C NMR (C₆D₆): δ 161.6, 160.9, 159.3, 114.6, 111.5, 32.8, 32.3, 31.8, 29.3, 28.6, 28.1, 24.1, 22.8, 21.1, 14.2, 14.1 and 13.8. IR (neat): 2962, 2934, 2874, 1729, 1632 and 1547 cm⁻¹. GC-MS: m/z (% relative intensity) 264 (M⁺, 16), 235 (4), 207 (18), 194 (52), 179 (23), 166 (100), 151 (75), 137 (31), 123 (31) and 71 (74). Calc. for C₁₇H₂₈O₂: C, 77.27; H, 10.61. Found: C, 77.00; H, 10.85%.

4-(3-Cyanopropyl)-5,6-dipropyl-2H-pyran-2-one 2e. Chromatography on alumina first with hexane and then with a 15:85 (v/v) mixture of ethyl acetate and hexane gave an orange oil (69.2 mg, 0.28 mmol, 56%), which was distilled over Kugelrohr to furnish an analytically pure colorless oil; bath temperature 125–130°C/1.0 Torr. ¹H NMR (C₆D₆): δ 5.64 (s, 1H, olefinic CH), 2.08 (t, 2H, J = 7.5), 1.79 (m, 4H), 1.49 (m, 2H), 1.25 (t, 2H, J = 6.8), 1.11 (m, 2H), 0.89 (m, 2H), 0.77 (t, 3H, J = 7.3) and 0.73 (t, 3H, J = 7.4 Hz). ¹³C NMR (C₆D₆): δ 161.4, 161.2, 156.9, 118.6, 114.3, 111.6, 32.8, 30.4, 27.9, 24.1, 23.8, 21.1, 16.1, 14.0 and 13.8. IR (neat): 2960, 2932, 2876, 2248, 1721, 1630 and 1545 cm⁻¹. GC-MS: m/z (% relative intensity) 247 (M⁺, 20), 218 (30), 207 (17), 190 (100), 166 (22), 148 (19), 91 (22), 77 (31), 71 (60) and 55 (20). Calc. for C₁₅H₂₁NO₂: C, 72.87; H, 8.50; N, 5.67. Found: C, 72.39; H, 8.66; N, 5.72%.

4-Phenyl-5,6-dipropyl-2H-pyran-2-one 2f. Chromatography on alumina with hexane followed by crystallization from pentane (−40°C) gave colorless crystals (72.3 mg, 0.28 mmol, 57%); mp 96.0–97.5°C. ¹H NMR (CDCl₃): δ 7.42–7.22 (m, 5H), 6.03 (s, 1H, olefinic CH), 2.54 (t, 2H, J = 7.6), 2.22 (t, 2H, J = 7.8), 1.74 (m, 2H), 1.16 (m, 2H), 1.01 (t, 3H, J = 7.3) and 0.69 (t, 3H, J = 7.2 Hz). ¹³C NMR (CDCl₃): δ 162.6, 162.1, 160.3, 137.6, 128.6, 128.4, 127.4, 115.2, 112.9, 32.1, 28.6, 23.5, 21.2, 13.9 and 13.8. IR (KBr): 2968, 2936, 2876, 1711, 1628, 1539, 1390, 940, 899, 768 and 706 cm⁻¹. GC-MS: m/z (% relative intensity) 256 (M⁺, 23), 228 (32), 199 (100), 157 (28), 128 (20) and 71 (50). Calc. for C₁₇H₂₀O₂: C, 79.69; H, 7.81. Found: C, 79.57; H, 7.89%.

The reaction of methyl 3,6-dichloro-2-hexenoate with 4-octyne. The reaction formed two products, 2g and 2g′, approximately in a 1:2.7 ratio. Chromatography on alumina first with hexane and then with a 5:95 (v/v) mixture of ethyl acetate and hexane gave analytically pure samples of 2g (11.0 mg, 0.043 mmol, 9%) and 2g′ (38.1 mg, 17.3 mmol, 35%) both as a pale yellow oil and a mixture of these.
4-(3-Chloropropyl)-5,6-dipropyl-2H-pyran-2-one 2g. ¹H NMR (C₆D₆): δ 5.77 (s, 1H, olefinic CH), 2.91 (t, 2H, J = 6.1), 2.08 (t, 2H, J = 7.3), 1.96 (t, 2H, J = 7.7), 1.86 (t, 2H, J = 8.1), 1.54–1.06 (m, 6H), 0.74 (t, 3H, J = 7.3) and 0.73 (t, 3H, J = 7.4 Hz). ¹³C NMR (C₆D₆): δ 161.4, 161.3, 157.7, 114.4, 111.7, 44.0, 32.8, 31.0, 29.1, 27.9, 24.1, 21.1, 14.0 and 13.8. IR (neat): 2966, 2936, 2876, 1725, 1632 and 1545 cm⁻¹. GC-MS: m/z (% relative intensity) 256 (M⁺ for ³⁵Cl isotope, 18), 228 (15), 199 (89), 166 (100), 151 (26), 123 (20), 107 (13), 91 (37), 71 (82) and 55 (57). Calc. for C₁₄H₂₁ClO₂: C, 65.50; H, 8.19. Found: C, 65.03; H, 8.35%.
4-(Cyclopropyl)-5,6-dipropyl-2H-pyran-2-one 2g′. ¹H NMR (C₆D₆): δ 5.61 (s, 1H, olefinic CH), 2.15–2.04 (m, 4H), 1.50 (m, 2H), 1.29–1.07 (m, 3H), 0.77 (t, 3H, J = 7.4), 0.74 (t, 3H, J = 7.4 Hz), 0.42 (m, 2H) and 0.25 (m, 2H). ¹³C NMR (C₆D₆): δ 161.9, 160.8, 160.4, 115.2, 106.9, 33.8, 28.4, 23.7, 21.1, 14.1, 13.8, 12.7 and 9.1 (2C). IR (neat): 2964, 2936, 2876, 1721, 1632 and 1545 cm⁻¹. GC-MS m/z (% relative intensity) 220 (M⁺ for ³⁵Cl isotope, 35), 192 (28), 163 (100), 149 (24), 121 (30), 91 (48), 71 (79) and 55 (23).

Reaction of compound 1a with 1-phenyl-1-butyne forming isomeric 2h. The reaction formed two regioisomers 2h-A and 2h-B in a 1.2:1 ratio. Chromatography on alumina first with hexane and then with a 5: 95 (v/v) mixture of ethyl acetate and hexane allowed isolation of analytically pure 2h-A (15.4 mg, 0.06 mmol, 12%) and 2h-B (20.5 mg, 0.08 mmol, 16%) as a pale yellow oil in addition to mixture fractions.
Isomer 2h-A. ¹H NMR (CDCl₃): δ 7.44–7.37 (m, 3H), 7.15–7.12 (m, 2H), 6.06 (s, 1H, olefinic CH), 2.27–2.07 (m, 4H), 1.34–1.08 (m, 7H) and 0.74 (t, 3H, J = 7.2 Hz). ¹³C NMR (CDCl₃): δ 163.1, 163.0, 160.8, 134.4, 130.1, 128.7, 128.0, 118.8, 110.1, 33.2, 30.1, 25.3, 22.1, 13.8 and 12.2. IR (neat): 2962, 2934, 2874, 1727, 1632, 1545, 768 and 704 cm⁻¹. GC-MS: m/z (% relative intensity) 256 (M⁺, 13), 214 (30), 186 (100), 171 (23), 128 (24) and 57 (68). Calc. for C₁₇H₂₀O₂: C, 79.69; H, 7.81. Found: C, 79.55; H, 7.90%.
Isomer 2h-B. ¹H NMR (CDCl₃): δ 7.49–7.40 (m, 5H), 6.15 (s, 1H, olefinic CH), 2.49 (t, 2H, J = 7.2), 2.40 (t, 2H, J = 7.4), 1.64–1.37 (m, 4H), 1.07 (t, 3H, J = 7.4) and 0.96 (t, 3H, J = 7.2 Hz). ¹³C NMR (CDCl₃): δ 162.4, 161.1, 157.7, 133.1, 129.6, 128.7, 128.3, 118.2, 112.3, 31.8, 30.8, 22.5, 19.8, 15.3 and 13.8. IR (neat): 2962, 2934, 2876, 1727, 1630, 1543, 1062 and 698 cm⁻¹. GC-MS: m/z (% relative intensity) 256 (M⁺, 24), 214 (49), 185 (91), 171 (23), 129 (12), 105 (84) and 77 (100).

Reaction of compound 1a with 2,7-dimethyl-1-octen-3-yne forming 2i. The reaction formed two regioisomers 2i-A and 2i-B in a 1:1 ratio. Chromatography on alumina first with hexane and then with a 2: 98 (v/v) mixture of ethyl acetate and hexane allowed isolation of analytically pure 2i-A (12.0 mg, 0.046 mmol, 9%) and 2i-B (22.2 mg, 0.085 mmol, 17%) as a pale yellow oil in addition to mixture fractions.
Isomer 2i-A. ¹H NMR (C₆D₆): δ 5.95 (s, 1H, olefinic CH), 4.96 (q, 1H, J = 1.6, olefinic CH), 4.60 (q, 1H, J = 0.9 Hz, olefinic CH), 2.23–1.86 (m, 4H), 1.56 (m, 3H), 1.41–1.31 (m, 7H) and 0.78–0.74 (m, 9H). ¹³C NMR (C₆D₆): δ 161.5, 160.7, 158.4, 139.5, 119.1, 118.6, 110.9, 37.2, 32.1, 30.8, 30.2, 29.8, 28.1, 24.4, 22.6, 22.3 and 13.9. IR (neat): 2962, 2936, 2876, 1734 and 1456 cm⁻¹. GC-MS: m/z (% relative intensity) 262 (M⁺, 16), 205 (4), 191 (73), 177 (18), 163 (25), 149 (28), 105 (27), 91 (83), 77 (75), 69 (22) and 55 (100).
Isomer 2i-B. ¹H NMR (C₆D₆): δ 5.98 (s, 1H, olefinic CH), 4.98–4.95 (m, 2H, olefinic CH₂), 2.15–1.95 (m, 4H), 1.79 (s, 3H), 1.16–1.04 (m, 7H) and 0.81–0.75 (m, 9H). ¹³C NMR (C₆D₆): δ 161.0, 159.6, 159.5, 138.1, 118.6, 115.1, 112.9, 40.5, 31.9, 30.9, 28.6, 24.8, 22.7, 22.4, 21.5 and 13.9. IR (neat): 2960, 2932, 2874, 1734, 1543 and 1075 cm⁻¹. GC-MS: m/z (% relative intensity) 262 (M⁺, 10), 205 (14), 191 (24), 177 (27), 163 (34), 150 (28), 107 (24), 91 (59), 79 (47), 69 (100) and 55 (69). HRMS: calc. for C₁₇H₂₆O₂ m/z 262.1931, found 262.1925.

Palladium-catalysed cross-coupling of compound 1a with 1-hexyne affording methyl 3-(n-butyl)-(Z)-2-nonen-4-ynoate 3. The reaction was carried out under conditions similar to those described for compound 2. After 5 h of heating, the reaction mixture was similarly worked up. Chromatography on alumina with hexane gave an oil (60.0 mg, 0.27 mmol, 54%), which was distilled over Kugelrohr to furnish a pale yellow oil; bath temperature 110°C/1.2 Torr. ¹H NMR (C₆D₆): δ 5.98 (s, 1H, C[double bond, length half m-dash]CH), 3.44 (s, 3H), 2.24 (t, 2H, J = 6.5), 2.06 (t, 2H, J = 7.5 Hz) and 1.54–0.76 (m, 14H). ¹³C NMR (C₆D₆): δ 165.1, 140.6, 123.2, 103.3, 79.8, 50.6, 39.2, 30.6, 30.4, 22.2, 22.1, 19.9, 13.9 and 13.7. IR (neat): 2962, 2934, 2866, 2220, 1734, 1615, 1218, 1193 and 1154 cm⁻¹. GC-MS: m/z (% relative intensity) 222 (M⁺, 1), 193 (38), 180 (26), 165 (100), 151 (46), 138 (20), 105 (24), 91 (35), 77 (28) and 55 (28). HRMS: calc. for C₁₄H₂₂O₂m/z 222.1619, found 222.1640.

Complex 4. A solution of compound 1a (35.2 mg, 0.2 mmol) and Pd(PPh₃)₄ (115.4 mg, 0.1 mmol) in toluene (2.0 mL) was stirred at 90°C for 3 h. Removal of volatiles in [italic v]acuo followed by recrystallization from a toluene–hexane mixture gave analytically pure complex 4 (67.0 mg, 0.083 mmol, 83.0%) as a white solid, mp 172.5–174.0°C (decomp.). ¹H NMR (CDCl₃): δ 7.78–7.72 (m, 12H), 7.40–7.33 (m, 18H), 4.96 (s, 1H, olefinic CH), 3.43 (s, 3H), 2.00 (t, 2H, J = 7.3 Hz), 0.77–0.73 (m, 2H) and 0.55–0.47 (m, 5H). ³¹P NMR (CDCl₃): δ 23.1. Calc. for C₄₄H₄₃ClO₂P₂Pd: C, 65.51; H, 5.34; Cl, 4.34. Found: C, 65.41; H, 5.17; Cl, 4.11%.

Acknowledgements

We thank the Japan Science and Technology Corporation (JST) for financial support through the CREST (Core Research for Evolutional Science and Technology) program and for a postdoctoral fellowship to R.H.

References and notes

1. J. V. N. Vara Prasad, K. S. Para, E. A. Lunney, D. F. Ortwine, J. B. Dunbar, Jr., D. Ferguson, P. J. Tummino, D. Hupe, B. D. Tait, J. M. Domagala, C. Humblet, T. N. Bhat, B. Liu, D. M. A. Guerin, E. T. Baldwin, J. W. Erickson and T. K. Sawyer, J. Am. Chem. Soc., 1994, 116, 6989.
2. (a) For synthesis and chemistry of 2H-pyran-2-ones, see: L. Staunton, in Comprehensi[italic v]e Organic Chemistry, ed. P. G. Sammes, Pergamon Press, Oxford, 1979, vol. 4, p. 629; (b) G. P. Ellis and J. D. Hepworth, in Comprehensi[italic v]e Heterocyclic Chemistry, ed. A. J. Boulton and A. Mckillop, Pergamon Press, Oxford, 1984, vol. 3, p. 675.; (c) Other publications on the synthesis:Y. Inoue, Y. Itoh and H. Hashimoto, Chem. Lett., 1977, 855; (d) Y. Inoue, Y. Itoh and H. Hashimoto, Chem. Lett., 1978, 633; (e) Y. Inoue, Y. Itoh, H. Kazama and H. Hashimoto, Bull. Chem. Soc. Jpn., 1980, 53, 3329; (f) D. Walther, H. Schonberg and E. Dinjus, J. Organomet. Chem., 1987, 334, 377; (g) T. Tsuda, T. Kiyoi, N. Hasegawa and T. Saegusa, Yuki Gosei Kagaku Kyokaishi, 1990, 48, 362 and references therein; (h) L. S. Liebeskind and J. Wang, Tetrahedron, 1993, 49, 5461; (i) S. H. Cho and L. S. Liebeskind, J. Org. Chem., 1987, 52, 2631; (j) R. K. Dieter and J. R. Fishpaugh, J. Org. Chem., 1988, 53, 2031; (k) W. Brady and M. O. Agho, J. Org. Chem., 1983, 48, 5337; (l) V. Kvita and H. Sauter, Hel[italic v]. Chim. Acta, 1990, 73, 883; (m) H. Stetter and H.-J. Kogelnik, Synthesis, 1986, 140; (n) X. Shi, W. S. Leal, Z. Liu, E. Schrader and J. Meinwald, Tetrahedron Lett., 1995, 36, 71; (o) L. Viallon, O. Reinaud, P. Capdevielle and M. Maumy, Tetrahedron Lett., 1995, 36, 6669; (p) Recent publications on the reactions of 2H-pyran-2-ones: M. E. Jung, M. Node, R. W. Pfluger, M. A. Lyster and J. A. Lowe, J. Org. Chem., 1982, 47, 1150; (q) F. Effenberger and T. Ziegler, Chem. Ber., 1987, 120, 1339; (r) K. Afarinkia and G. H. Posner, Tetrahedron Lett., 1992, 33, 7839; (s) K. Afarinkia, V. Vinader, T. D. Nelson and G. H. Posner, Tetrahedron, 1992, 48, 9111; (t) G. H. Posner, H. Dai, D. S. Bull, J.-K. Lee, F. Eydoux, Y. Ishihara, W. Welsh, N. Pryor and S. Petr, J. Org. Chem., 1996, 61, 671 and references therein.
3. R. Hua, S. Shimada and M. Tanaka, J. Am. Chem. Soc., 1998, 120, 12365.
4. M. Tanaka and R. Hua, Jpn. Appl., 1999/061234, March 9, 1999..
5. (a) R. C. Larock, M. J. Doty and X. Han, J. Org. Chem., 1999, 64, 8770; (b) Palladium-catalysed annulation of internal alkynes with methyl o-iodobenzoate was reported to give isocoumarins: W. Tao, L. J. Silverberg, A. L. Rheingold and R. F. Heck, Organometallics, 1989, 8, 2550; (c) R. C. Larock, E. K. Yum, M. J. Doty and K. C. Sham, J. Org. Chem., 1995, 60, 3270. However, attempted reactions of methyl o-chlorobenzoate with 4-octyne under our reaction conditions did not proceed..
6. The reaction run at higher temperatures (160°C), using 2 mol% of the catalyst, required a longer time (24 h). This procedure formed a by-product (13%), which was found to be methyl 3-diethylamino-2-heptenoate. For N–C bond cleavage in palladium-catalysed reactions, see: T. Kobayashi and M. Tanaka, J. Organomet. Chem., 1982, 231, C12.
7. Methyl 3-diethylamino-2-heptenoate was formed as undesired side product in small quantities. See ref. 6..
8. Cyclopropane ring formation from γ-haloganoalkanoates and related compounds having a γ-halogenoalkyl group bound to electron-withdrawing groups has been well documented. See: B. Zwanenburg and N. De Kimpe, in Methods of Organic Chemistry, ed. A. de Meijere, Georg Thieme Verlag, Stuttgart, 1997, Vol. E17a, p. 45..
9. The reaction also produced dimers and trimers of 1-hexyne as main by-products, along with the recovery of compound 1a in 12% yield..
10. (a) Carbon–oxygen bond formation [italic v]ia reductive elimination from palladium complexes has been reported: G. Mann and J. F. Hartwig, J. Am. Chem. Soc., 1996, 118, 13109; (b) R. A. Widenhoefer, H. A. Zhong and S. L. Buchwald, J. Am. Chem. Soc., 1997, 119, 6787.
11. NEt₃ also is likely to hinder the nucleophilic attack of the carbonyl oxygen through its coordination to the palladium center of complex Z-4. At the same time, however, its presence in the system facilitates the reaction by trapping chloromethane generated. Accordingly the effect of NEt₃ on the reaction rate is ambiguous..

---