Phenyl acetylene (Aldrich),
Carbon Monoxide (BOC),
2-pyridyl-diphenylphosphine (Aldrich),
Palladium acetate (Alfa Aesar),
para-toluene sulfonic acid (Aldrich)

The reaction was conducted in a basic stainless steel autoclave equipped with a glass liner, pressure gauge, and gas inlet and outlet valve. Pd(OAc)₂ (15.0 mg, 6.85 x 10-5 mols, 0.05 % ), Ph₂Py (0.541g, 2.055 mmols, 1.5 %), and a stirring bead were added to a Schlenk flask which was then flushed with nitrogen prior to the addition of dry, degassed methanol (40 ml). This solution was stirred vigorously for 15 minutes to ensure the catalyst dissolved. The flask was then briefly opened and p-toluene-sulfonic acid (0.391g, 2.055 mmols), and phenyl acetylene (13.95 g, 0.137 mols) were added (At which point the last traces of insoluble Pd residues dissolve. The catalyst solution should be orange). The reaction solution was added to the autoclave, in air, which was then rapidly sealed and flushed twice with CO (CAUTION! TOXIC GAS). The autoclave was then pressurised to 40 bar, sealed and stirred on a conventional stirrer hotplate at 60 ^oC for 4 hours. At the end of this time, the carbon monoxide was released (CAUTION!), and the autoclave dismantled. The reaction solution could then be directly analysed for conversion and selectivity by diluting in CDCl₃ and running an ¹H NMR spectrum. In this case, both conversion and selectivity were essentially 100%. Addition of diethyl ether (50 ml), and washing the solution with ammonium chloride (30 ml) and water (30 ml), followed by drying (MgSO₄) of the organic extracts gave a solution from which solvent was removed by rotary evaporation. The resulting oil was distilled under vacuum, the product (colourless oil) being collected at _{60 ^oC/ 2 mm Hg. For the application we required, it was essential that the product did not contain even trace impurities of (sublimed) Ph}2~PPy, so as a precaution the distillation was halted as soon as most of the product had been collected. As a result of this, the isolated yields were considerably lower than the conversion, in this case (14g, 0.086 mols, 63%). The product should be stored in the freezer, since prolonged storage in air and daylight results in polymerisation.

The alkoxy carbonylation procedure described represents an atom efficient method to prepare 1,1-disubstituted unsaturated esters. The reaction is easy to carry out, does not require special care to exclude air or moisture, and works well using a standard autoclave. The catalyst is extremely reactive: Drent and co-workers developed the reaction as an industrial synthesis of methyl methacrylate from propyne, and although few details are reported in the publication cited, they note that even lower catalyst loadings can be used if desired, and that the reaction also works well with a range of alkynes and alcohols (We have obtained similar results to those described here with isopropanol as alcohol and hexyne as substrate). CAUTION: Carbon monoxide is toxic and odourless, so the reactions must be kept in a well ventilated fume cupboard at all times, and preferably carried out in the presence of an electronic CO detector. Researchers should ensure that the autoclave apparatus can withstand the pressure used for the reaction. The unsaturated esters are interesting substrates for a further carbonylation reaction, since this reaction delivers synthetically useful compounds with unusually high branched regioselectivity.

¹H NMR (299.94 MHz, CDCl₃): 3.78 (3H, s), 5.88 (1H, d, J= 2 Hz), 6.36 (1H, d, J = 2 Hz), 7.2-7.5 (5H, m)

(a) M. L. Clarke, Tetrahedron Lett., 2004, 45, 4043. (b) E. Drent, P. Arnoldy, P. H. M. Budzelaar, J. Organomet. Chem., 1993, 455, 247.

Other reference: R. A. Baber, M. L. Clarke, K. M. Heslop, A. C. Marr, A. G. Orpen, P. G. Pringle, A. Ward, D. E. Zambrano-Williams, Dalton Trans., 2005, 1079.