filtration and isolation of the polymer, 9 could be redissolved
in CH2Cl2 and used again. It was used through four cycles
in eq 1 with no loss in activity. Complete retention of activity
of the catalysts in these subsequent cycles was determined
by comparison of the initial rates of each reaction with one
another. An advantage of the catalyst 9 in these comparisons
was that the small physical losses expected for filtration and
handling of these catalysts (ca. 2% per cycle) do not
obfuscate the analysis of catalyst recyclability since the
amount of catalyst in each cycle can be easily and accurately
quantified by UV spectroscopy.
with a second heating and cooling cycle allowed us to repeat
the process. This process was successfully used five times
using 2,6-diisopropylphenol in a DMF-heptane thermomor-
phic system. Similar chemistry using a 90% ethanol-water/
heptane thermomorphic system produced >90% yields of
phenyl carbonate in the fifth cycle using (Boc)2O with both
phenol and 2,6-dimethylphenol. Our recent report shows that
a similar scheme using N-octadecylacrylamide polymers
allows us to recycle catalysts in a nonpolar phase; so
extension of this chemistry should provide a way to recycle
catalysts that form polar or nonpolar products.13
Titration of the 4-dialkylaminopyridine groups of 9 shows
that the basic groups in this polymer have a pKa of 9.7. This
pKa was estimated using the software program BEST18
because polymer 9 is insoluble in water at pH values above
10. The same software and titrimetric analysis yielded a pKa
of 9.9 for DMAP, providing further evidence of the similarity
of the DAAP groups in 9 to their low molecular weigh
analogue DMAP. This experiment also shows that protona-
tion/deprotonation and dissolution/precipitation is another
way to recover this soluble polymeric catalyst.
Recycling of catalyst 9 in carbonate formation using
dialkylphenols as substrates was also successful. Using
solvent precipitation, catalyst 9 was recycled three times in
eq 2 providing a 90% yield of both 2,6-dimethyl- and 2,6-
diisopropylphenyl carbonate in each cycle (R ) Me, 95%,
i
91%, 94%; R ) Pr, 91%, 93%, 89%). We have also been
able to use alternative separation schemes (Figure 1) to
Besides incorporating a dye into a polymer as an innocent
bystander to facilitate studies of catalyst isolation/separation/
reuse, we also used dyes as both labels and ligands. In this
case, we took advantage of the fact that an azo arene dye
can be designed to serve as a precursor of a Pd(II) catalyst
ligand. Nitrogen-ligated palladacycles have been used previ-
ously in reactions such as aromatic C-H activation, aromatic
hydroxylation and in carbon-carbon coupling reactions.19
As shown below, a similar azo dye bound Pd(II) species is
useful in Heck chemistry. In this case, the dye serves three
roles. It serves as a ligand, as a reporter for catalyst/polymer
recovery, and as a probe of catalyst stability.
O-Methyl 4-(4-nitrophenylazo)phenol, an azo dye with a
strong electron-donating group on one aromatic ring and an
strong electron-withdrawing group on the other, was chosen
as a substrate for C-Pd palladation to avoid non-ring specific
aromatic palladation. First, a low molecular weight catalyst,
10, was prepared using the chemistry in eq 3. Then a similar
Figure 1. Thermomorphic catalysis using heptane insoluble 9. The
polymeric catalyst is separated as a N,N-dimethylformamide-rich
phase. The heptane phase contains the acylation product.
recover these acylation catalysts.13,17
Specifically, when we dissolved the catalyst 9 (5 mol %)
in a liquid/liquid biphasic mixture of N,N-dimethylformamide
(DMF) and heptane (C7) at 25 °C, we found that this polar
poly(N-isopropylacrylamide)-bound catalyst dissolved se-
lectively in the polar DMF-rich phase. However, 9 was also
soluble in the homogeneous solvent mixture that resulted
on heating this solvent mixture to 70 °C. On cooling of this
thermomorphic mixture back to 25 °C, this catalyst stayed
in the DMF-rich phase. The polymer was present in a resting
biphasic mixture of heptane (C7) and dimethylacetamide
(DMF) exclusively (>99.5%) in the DMF-rich phase as
measured by UV analysis of the DMF- and C7-rich phases.
This makes it possible to use 9 in a homogeneous reaction
and to separate it with a liquid/liquid separation after the
reaction. For example, when a substrate phenol and (tBoc)2O
reagent were added to the DMF-C7 mixture containing 9
and the biphasic mixture was heated, a monophasic solution
formed and 9 catalyzed the acylation of 2,6-dialkylphenols
by (Boc)2O. After 1 h, the mixture was cooled. The phases
separated and the C7-rich phase (the nonpolar phase)
azo dye ligand was prepared and coupled to either a poly-
(N-isopropylacrylamide) or a poly(N-octadecylacrylamide)
polymer as shown in Scheme 2.
Catalyst 10 was shown to be active at 0.1 mol % in Heck
chemistry using acrylate Heck acceptors and iodoarenes and
t
containing the relatively nonpolar Boc-derivative of 2,6-
diisopropyl-phenol was separated from the catalyst-contain-
ing DMF-rich phase. Addition of fresh heptane and substrates
(18) Martell, A. E.; Motekaitis, R. J. The Determination and Use of
Stability Constants, 2nd ed.; VCH Publishers: New York, 1992; p 171.
(19) Dupont, J.; Pfeffer, M.; Spencer, J. Eur. J. Inorg. Chem. 2001, 8,
1917.
(17) Bergbreiter, D. E.; Osburn, P. L.; Sink, E.; Wilson, A. J. Am. Chem.
Soc. 2000, 122, 9058.
Org. Lett., Vol. 4, No. 5, 2002
739