tertiary alcohols as nucleophiles (Scheme 1). The ability to
generate tertiary esters directly from aromatic bromides
and an alcohol in the presence of a base would provide a
simple access to this class of compounds.
We initiated a catalyst screening for the carbonylation of
6-bromo-1,4-benzodioxane (1) with sodium tert-butoxide
using a slight excess (1.5 equiv) of carbon monoxide (CO).
As the source of CO, the two-chamber technique recently
developed in our laboratories was applied employing a
carbon monoxide precursor (see Supporting Information
From Table 1, PPF-tBu and DiPrPF were selected to test the
reaction of sodium tert-butoxide with 4-bromobenzonitrile to
ensure that the developed method could be applied to both
electron-rich and -poor aromatic systems. Whereas PPF-tBu
afforded a 70% conversion of 4-bromobenzonitrile, DiPrPF
provided a more useful 90% conversion with a 76% isolated
yield of 4-cyanobenzoic acid tert-butyl ester. DiPrPF was
chosen for further optimization, and next the amount of CO
required was investigated. Decreasing the loading of CO to
1.1 equiv led to a drop in the yield (entry 7).9 Lowering the
catalytic loading in the CO-releasing chamber A to 1 mol % of
both Pd(dba)2 and HBF4P(tBu)3 did not affect the isolated
yield of 2 (entries 8ꢀ10). Increasing the concentration of
reactants had no effect on the reaction outcome (entry 10).
Changing the solvents to dioxane or toluene led to a decrease
in isolated yield. Finally, applying potassium tert-butoxide as
the nucleophile only afforded trace amounts of the desired
alkoxycarbonylated coupling product.
Table 1. Optimization of the Alkoxycarbonylation Reaction of
Aryl Bromide 1 with Sodium tert-Butoxidea
Having established the optimum reaction conditions,
the scope of the palladium-catalyzed alkoxycarbonylation
of functionalized aryl bromides with sodium tert-butoxide
was explored (Table 2). Simple substituted aryl bromides
all provided the desired products in yields attaining 90%
(entries 2ꢀ6) except bromobenzene, which only afforded a
60% isolated yield (entry 1). This lower yield was assigned to
the product’s relatively low boiling point.10 Both electron-
rich and -poor systems proved reactive toward the cataly-
tic system, and the desired tert-butyl ester derivatives were
obtained in yields ranging from 44% to 86% (entries
7ꢀ13).11 Next, a handful of heteroaromatic bromides
was tested, which satisfyingly afforded the desired pro-
ducts in good yields (entries 14ꢀ18). In the case of 4-
bromoisoquinoline and 5-bromoisoquinoline, debromina-
tion was also observed in the crude reaction mixture
leading to an overall decrease in isolated yield (entries 16
and 17).12 Furthermore, ortho-substituents impeded the
alkoxycarbonylation reaction, leading to trace amounts of
the desired products (results not shown).
entry
ligands
yield [%]b
1
BINAP
Xantphos
DPPF
87
78
87
88
90
0c
2
3
4
PPF-tBu
DiPrPF
DtBuPF
DiPrPF
DiPrPF
DiPrPF
DiPrPF
5
6
7d
8e
9e,f
10e,f,g
77
89
87
88
a Chamber A: 9-Methyl-fluorene-9-carbonyl chloride (0.45 mmol),
Pd(dba)2 (5 mol %), HBF4P(tBu)3 (5 mol %), DIPEA (1.5 equiv) in THF
(2 mL). Chamber B: 6-Bromo-1,4-benzodioxane 1 (0.3 mmol), Pd(dba)2
(5 mol %), L (6 mol %) in THF (2 mL). b Isolated yield after column
chromatography. c Determined by 1H NMR analysis. d CO (0.33 mmol)
e Pd(dba)2 (1 mol %), HBF4P(tBu)3 (1 mol %). f DIPEA (1.2 equiv).
g CO (0.75 mmol), 1 (0.5 mmol). PPF-tBu: (R)-1-[(SP)-2-(Diphenyl-
phosphino)ferrocenyl]ethyldi-tert-butylphosphine. DiPrPF: 1,10-Bis-
(diisopropylphosphino)ferrocene.
Attention wasnext turnedtotestthe application of other
sodium tertiary alkoxides as nucleophlies. Suspecting that
several of the isolated esters in Table 2 were volatile, it was
expected that an increase in overall molecular weight
would be accompanied by an increase in isolated yield.
Toward this end, the sodium salt of 1-adamantol was cho-
sen, resembling the bulk of sodium tert-butoxide (Table 3).
Not only did the catalytic system prove to be adaptable to
other bulky alkoxides, but higher yields were obtained in
all reactions except with 4-trifluoromethyl bromobenzene
for further details).5ꢀ7 Initial screenings revealed that
bidentate ligands were superior in comparison to their
monodentate counterparts such as PPh3, PCy3, and cat-
aCXium A (results not shown).8 All but one of the tested
bidentate ligands led to desired product 2 in high isolated
yields except DtBuPF (entry 6) which, surprisingly resulted
in no conversion at all (Table 1, entries 1ꢀ5 and 6).
(9) Decreasing the catalytic loading in the CO-releasing chamber of
Pd(dba)2 and HBF4P(tBu)3 to 1 mol % did not influence the yield
(entries 8 and 9).
(10) The boiling point of 3 is 54 °C (1Torr). See: Kazuo, N.; Keiko, O.;
Asami, Y.; Keiichi, I. Chem. Lett. 1994, 2, 209.
(11) The boiling point of 13 is 60 °C (1 Torr). See: Amin, H. B.;
Taylor, R. J. Chem. Soc., Perkin Trans. 2: Phys. Org. Chem.
(1972ꢀ1999) 1975, 1802.
(6) Hermange, P.; Gøgsig, T. M.; Lindhardt, A. T.; Taaning, R. H.;
Skrydstrup, T. Org. Lett. 2011, 13, 2444.
(7) Nielsen, D. U.; Taaning, R. H; Lindhardt, A. T.; Gøgsig, T. M.;
(12) 18 was obtained along with approximately 30% of the debro-
minated isoquinoline according to crude 1H NMR analysis. 19 and its
debrominated isoquinoline were obtained as an inseparable mixture (see
Supporting Information).
Skrydstrup, T. Org. Lett. 2011, 13, 4456.
€
(8) Magerlein, W.; Indolese, A. F.; Beller, M. Angew. Chem., Int. Ed.
2001, 40, 2856. cataXCium A: Di(1-adamantyl)-n-butylphosphine.
Org. Lett., Vol. 14, No. 1, 2012
285
Xin, Zhuo
