J. Am. Chem. Soc. 1997, 119, 5057-5058
5057
Palladium-Catalyzed Carboxylation of Allyl
Stannanes
Min Shi and Kenneth M. Nicholas*,†
Department of Chemistry and Biochemistry
The UniVersity of Oklahoma
Norman, Oklahoma 73019-0370
the reaction temperature (100 °C) and time (72 h) afforded no
improvement. Surprisingly, crotyltributyltin (7, R′ ) Me) failed
to undergo carboxylation entirely (70 °C, 33 atm, 8 mol % Pd-
(PPh3)4).
ReceiVed NoVember 18, 1996
The world’s dwindling petroleum reserves and increasing
atmospheric concentrations of carbon dioxide have stimulated
considerable interest in the capture and chemical conversion of
carbon dioxide. Among several approaches being examined
toward this objective is the activation of carbon dioxide by
transition metal complexes.1 Efforts in our group have centered
on defining and elucidating the reactivity patterns of coordinated
carbon dioxide2 and on the development of new catalytic
reactions of this typically unreactive molecule.3 Especially
synthetically attractive are reactions of CO2 which result in
carbon-carbon bond formation, e.g., the classic carboxylation
of Grignard reagents.4 Indeed, insertions of CO2 into the metal-
carbon bonds of electropositive main group metals1b,5 and many
transition metals1 are common, but corresponding reactions with
less polar organometallics, though still thermodynamically
favorable,6 are rarer. Promoting such transformations would
be highly desirable since the resulting metal carboxylates
(RCO2M) should be more amenable to subsequent transforma-
tion of the weaker M-O bond, enhancing the prospects for
catalytic conversions to organic products. We report herein the
first examples of a transition metal catalyzed insertion of CO2
into otherwise unreactive tin-carbon bonds.
A number of other transition metal complexes, including
Pd(CH3CN)2Cl2, (Ph3P)3RhCl, and (Ph3P)3Ru(CO)H2, were also
tested for their ability to catalyze the carboxylation (70 °C, 33
atm) of allyl stannane 1; in each case, however, 1 was recovered
unchanged. The Lewis acid BF3‚Et2O, an effective catalyst for
the addition of allyl stannanes to aldehydes,10 likewise failed
to induce the carboxylation of 1. Thus, of the systems evaluated
to date, only Pd(0) complexes have been found to be effective
carboxylation catalysts.
Multiple carboxylation of poly(allyl)stannanes also can be
effected with Pd(0) catalysis. Thus, diallyldibutyltin (8) reacted
completely with CO2 under the standard conditions to afford
allyl dicarboxylates 9, 10 (2.3:1),8 and, presumably, 11 (overall
yield >90%, eq 2).11 Surprisingly, no singly inserted product,
Allyltributyltin (1) does not react with CO2 in THF even at
70 °C and 33 atm (24 h). However, under the same conditions
in the presence of 8 mol % Pd(PPh3)4 or Pd(PBu3)4 stannane 1
is quantitatively converted to the carboxylates 2 (90%) and 3
(10%) (eq 1).7,8 The formation of isomer 3 must occur during
the product-forming sequence because neither the starting
material (1) nor the product (2) was isomerized in refluxing
THF in the presence of Pd(PPh3)4. Under the same carboxy-
lation conditions a 30% conversion of allyltriphenyltin (4, R )
Ph) to insertion products 5 and 6 (7:3) was found;8 increasing
i.e., 12, was detected. In the Pd-catalyzed reaction of tetraal-
lyltin (13) with CO2 a complex mixture of carboxylates
(allyl)4-nSn(O2C-allyl)n (n ) 1-4) was produced after 24 h
1
judging by H NMR analysis. However, if the reaction time
was extended to 72 h, the major products (ca. 80 % yield) were
the isomeric tetracarboxylates 14, 15, and 16,11 i.e., all four Sn-
allyl bonds had inserted CO2 (eq 3). Under these conditions
double bond isomerized products (15, 16 n ) 3, 4) dominated
ca. 4:1.
† Phone: 405-325-3696; FAX: 405-325-6111; e-mail: knicholas@ou.edu.
(1) (a) Behr, A. Carbon Dioxide Activation by Metal Complexes; VCH
Publisher: Weinheim, Germany, 1988. (b) Ito, T.; Yamamoto, A. Orga-
nometallic Reactions of Carbon Dioxide. Chapter 3 of Organic and Bio-
organic Chemistry of Carbon Dioxide; Inoue, S., Yamazaki, N., Eds.;
Kodansha Ltd.: Tokyo, Japan, 1982. (c) Gibson, D. H. Chem. ReV. 1996,
96, 2063.
(2) Tsai, J.-C.; Khan, M. A.; Nicholas, K. M. Organometallics 1989, 8,
2967. Tsai, J.-C.; Khan, M. A.; Nicholas, K. M. Organometallics 1991,
10, 29. Fu, P.-F.; Khan, M. A.; Nicholas, K. M. Organometallics 1991,
10, 382. Ziegler, W.; Nicholas, K. M. J. Organomet. Chem. 1992, 423,
C35. Fu, P.-F.; Khan, M. A.; Nicholas, K. M. Organometallics 1992, 11,
2607. Fu, P.-F.; Khan, M. A.; Nicholas, K. M. J. Am. Chem. Soc. 1992,
114, 6579. Fu, P.-F.; Rahman, A. K. F.; Nicholas, K. M. Organometallics
1994, 13, 413. Fu, P.-F.; Khan, M. A.; Nicholas, K. M. J. Organomet.
Chem. 1995, 506, 49.
(8) Authentic samples of previously unreported carboxylates 2, 3, 5, 6,
9 and 10 were prepared from the reaction of 3- or 2-propenoic acid with
bis(tributyltin) oxide, bis(triphenyltin) oxide, or dibutyltin oxide, respectively
(ref 9, toluene/110 °C/H2O azeotrope). These were identical spectroscopi-
cally with products from the reactions of 1, 4, and 8 with CO2. Spectral
(3) Rahman, A. K. F. Tsai, J.-C.; Nicholas, K. M. J. Chem. Soc., Chem.
Commun. 1992, 1334. Tsai, J.-C.; Nicholas, K. M. J. Am. Chem. Soc. 1992,
114, 5117. Nicholas, K. M. J. Organomet. Chem. 1980, 188, C10.
(4) March, J. L. AdVanced Organic Chemistry, 4th ed.; Wiley: New
York, 1992; pp 933-934, and references therein.
1
data (IR, H and 13C NMR, and MS) for all products are compiled in the
Supporting Information.
(9) Okawara, R.; Wada, M. In Organotin Compounds; Sawyer, A. K.,
Ed.; M. Dekker: New York, 1971; Vol. 2, p 253.
(10) Pereyre, M.; Quintard, J.-P. Quintard; Rahm, A. Tin in Organic
Synthesis; Butterworths: London, G.B., 1987; pp 216-229.
(11) Precise product ratios in eqs 2 and 3 could not be determined because
the 1H and 13C NMR spectra of the “mixed” isomers 11 and 16 are
indistinguishable from mixtures of the “symmetrical” isomers, 9 + 10 and
14 + 15. Thus, NMR spectra (1H and 13C) of a presumably authentic
mixture of 9, 10, and 11 obtained by reaction of Bu2SnO with a 1:1 mixture
of 2- and 3-propenoic acid were the same as a 1:1 mixture of 9 and 10.
Integration of the 1H NMR spectra of the product mixtures from eqs 2 and
3, however, allowed determination of the reported allyl to crotyl isomer
ratios. The hydrolytic and thermal lability of the stannyl esters precluded
separation of the isomers by flash or gas chromatography.
(5) Sanderson, R. T. J. Am. Chem. Soc. 1955, 77, 4531.
(6) Bond enthalpy considerations for M-CR3 + CO2 f M-OC(dO)-
CR3 indicate an exothermic process for virtually all metals since a strong
M-O bond is formed at the expense of a weaker M-C bond and a C-C
σ-bond is formed at the expense of a comparably strong C-O π-bond.
(7) The following procedure is representative. Allyltributyltin (0.33 g,
1.0 mmol), Pd(Ph3P)4 (0.093 g, 0.08 mmol), 20 mL of dry THF, and a
magnetic stir bar were placed in the 50 mL glass liner of a stainless steel
autoclave under a nitrogen purge. After the autoclave was purged several
times with CO2, it was pressurized with CO2 (33 atm), sealed, and heated
at 70 °C for 24 h with stirring. After cooling and release of the pressure,
the solvent was removed under reduced pressure, and the residue was passed
through a short column of flash silica (1:1 ethyl acetate/hexane) to afford
a spectroscopically pure mixture of the tin carboxylates 2 and 3.
S0002-7863(96)03983-2 CCC: $14.00 © 1997 American Chemical Society