A reassessment of the transition-metal free Suzuki-type coupling methodology

Article abstract of DOI:10.1021/jo048531j

(Chemical Equation Presented). We present here a reassessment of our transition-metal free Suzuki-type coupling protocol. We believe that, although the reaction can be run without the need for addition of a metal catalyst, palladium contaminants down to a level of 50 ppb found in commercially available sodium carbonate are responsible for the generation of the biaryl rather than, as previously suggested, an alternative non-palladium-mediated pathway. We present a revised methodology for Suzuki couplings using ultralow palladium concentrations for use with aryl and vinyl boronic acids and discuss the effects of the purity of the boronic acid on the reaction.

Full text of DOI:10.1021/jo048531j

A Reassessment of the Transition-Metal Free Suzuki-Type
Coupling Methodology
Riina K. Arvela,† Nicholas E. Leadbeater,*,† Michael S. Sangi,† Victoria A. Williams,†
Patricia Granados,‡ and Robert D. Singer‡
Department of Chemistry, University of Connecticut, Unit 3060, 55 North Eagleville Road,
Storrs, Connecticut 06269-3060, and Department of Chemistry, Saint Mary’s University,
Halifax, Nova Scotia B3H 3C3, Canada
nicholas.leadbeater@uconn.edu
Received August 20, 2004
We present here a reassessment of our transition-metal free Suzuki-type coupling protocol. We
believe that, although the reaction can be run without the need for addition of a metal catalyst,
palladium contaminants down to a level of 50 ppb found in commercially available sodium carbonate
are responsible for the generation of the biaryl rather than, as previously suggested, an alternative
non-palladium-mediated pathway. We present a revised methodology for Suzuki couplings using
ultralow palladium concentrations for use with aryl and vinyl boronic acids and discuss the effects
of the purity of the boronic acid on the reaction.
Introduction
other groups using palladium salts either in water,2-,5
ionic salts,6,7 or pyridine without the need for addition
of ligands.8
The Suzuki reaction (palladium-catalyzed cross-cou-
pling of aryl halides with boronic acids) is one of the most
versatile and utilized reactions for the selective construc-
tion of carbon-carbon bonds, in particular for the forma-
tion of biaryls.1 As the biaryl motif is found in a range of
pharmaceuticals, herbicides, and natural products as well
as in conducting polymers and liquid crystalline materi-
als, development of improved conditions for the Suzuki
reaction has received much recent attention. A wide
range of metal complexes have been used as catalysts in
these coupling reactions, attention particularly being
focused on palladium. There have been a number of
recent reports of the use of so-called “ligandless” pal-
ladium catalysis of the Suzuki coupling. In 2002, we
showed that, when using Pd(OAc)2, the Suzuki coupling
of a wide range of aryl bromides and aryl boronic acids
can be performed using water as a solvent with the
addition of TBAB as a phase-transfer agent. The coupling
works equally well using either microwave or conven-
tional heating. Similar observations have been made by
In 2003, we reported that, using the appropriate
conditions, it is possible to perform Suzuki-type couplings
without the need for addition of a transition-metal
catalyst.9-11 We have termed this “transition-metal free
Suzuki-type coupling” since the levels of palladium found
in reaction mixtures were below the level of detection of
our apparatus (0.1 ppm). We also analyzed reaction
mixtures for the presence of a wide range of other
elements. Possible catalyst candidates include nickel,12
(2) Zapf, A.; Jackstell, R.; Rataboul, F.; Riermeier, T.; Monsees, A.;
Fuhrmann, C.; Shaikh, N.; Dingerdissen, U.; Beller, M. Chem. Com-
mun. 2004, 38.
(3) Arcadi, A.; Cerichelli, G.; Chiarini, M.; Correa, M.; Zorzan, D.
Eur J. Org. Chem. 2003, 4080.
(4) Bai, L.; Wang, J. X.; Zhang, Y. M. Green Chem. 2003, 5, 615.
(5) Bedford, R. B.; Blake, M. E.; Butts, C. P.; Holder, D. Chem.
Commun. 2003, 466.
(6) Miao, W. S.; Chan, T. H. Org. Lett. 2003, 5, 5003.
(7) Zou, Y.; Wang, Q. R.; Tao, F. G.; Ding, Z. B. Chinese J. Chem.
2004, 22, 215.
(8) Tao, X. C.; Zhao, Y. Y.; Shen, D. Synlett 2004, 359.
(9) Leadbeater, N. E.; Marco, M. Angew. Chem., Int. Ed. 2003, 42,
1407.
† University of Connecticut.
(10) Leadbeater, N. E.; Marco, M. J. Org. Chem. 2003, 68, 5660.
(11) Li, C. J. Angew. Chem., Int. Ed. 2003, 42, 4856.
(12) (a) Zim, D.; Monteiro, A. L. Tetrahedron Lett. 2002, 43, 4009.
(b) Griffiths, C.; Leadbeater, N. E. Tetrahedron Lett. 2000, 41, 2487.
(c) Lipshutz, B. H.; Sclafani, J. A.; Blomgren, P. A. Tetrahedron 2000,
56, 2139. (d) Leadbeater, N. E.; Resouly, S. M. Tetrahedron 1999, 55,
11889.
‡ Saint Mary’s University.
(1) For reviews, see: (a) Hassan, J.; Se´vignon, M.; Gozzi, C.; Schulz,
E.; Lemaire, M. Chem. Rev. 2002, 102, 1359. (b) Kotha, S.; Lahiri, K.;
Kashinath, D. Tetrahedron 2002, 58, 9633. (c) Suzuki, A. J. Organomet.
Chem. 1999, 576, 147. (d) Miyaura, N.; Suzuki, A. Chem. Rev. 1995,
95, 2457.
10.1021/jo048531j CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/08/2004
J. Org. Chem. 2005, 70, 161-168
161

Arvela et al.
platinum,13 copper,14,15 and ruthenium14 since all of these
metals have been shown to act as catalysts for the Suzuki
reaction either individually or together. However, none
of these metals were present in the product mixture in
concentrations above 1 ppm.
SCHEME 1
There have been a number of reports that have
recently appeared in the literature that suggest that C-C
coupling reactions can be catalyzed by trace quantities
of palladium complexes. De Vries and co-workers have
shown that the Heck reaction can be run with the
addition of what they term “homeopathic” quantities of
palladium catalysts, but find that when using very low
metal concentrations the rate of reaction is too slow to
be practical.16,17 Choudary18 and co-workers have ob-
tained very high turnover numbers in Heck couplings
with aryl chlorides when using layered double hydroxide
supported nanopalladium catalysts. They suggest that
the reaction occurs at the surface of the nanoparticles.
Ko¨hler and co-workers have recently studied the Heck
reaction using a range of solid catalysts.19 Their results
suggest that there is in situ generation of highly active
dissolved palladium species and thus that the catalysis
is in effect homogeneous with palladium dissolution and
reprecipitation being crucial and inherent parts of the
catalytic cycle. Very high turnover numbers have also
been reported for Suzuki coupling reactions, mostly
involving the use of ligated palladium complexes such
as those bearing N-heterocyclic carbene ligands,20 bulky
phosphines,2 or else using palladacycles.21 As with the
Heck reaction, Suzuki couplings have been performed
using solid catalysts.18,22,23 Again, the in situ generation
of small quantities of highly active dissolved palladium
species is suggested.
Results and Discussion
In our reports of 2003, we presented a general method
for microwave-promoted transition-metal free Suzuki-
type couplings of aryl bromides and aryl boronic acids in
water using sodium carbonate as a base and TBAB as
an additive. Central to the success of the methodology
are the criteria that the reaction must be performed in
water at 150 °C in a sealed vessel using microwave
heating. Caution.24 We found that while a wide range
of functional groups on the aryl halide component are
tolerated in the reaction, only the coupling with phenyl-
boronic acid worked well (Scheme 1). Representative aryl
iodides were also screened in the coupling reaction using
our methodology but product yields were lower than their
bromo counterparts. Aryl chlorides could not be coupled.
Our first objective in reassessing the methodology was
to determine why the boronic acid scope was so limited.
We found that the success of the reaction is dependent
on the purity of the boronic acid used. Impurities in the
boronic acid, even in very small amounts, can effectively
shut down the reaction. We wanted to revisit the reaction
to look more thoroughly at the scope of the boronic acid
component since we felt that it may be the case that our
previous screening of the boronic acids may have led to
“false negatives”. Therefore we embarked upon the
purification of a range of boronic acids. We used three
methods to achieve this. The boronic acids can be purified
by column chromatography using hexane/ethyl acetate
or, in certain cases, hexane/methanol as eluent. We start
by eluting with pure hexane and then gradually increase
the ratio of the more polar solvent. Any inorganic
impurities will remain on the silica and traces of aryl
chloride and other organic impurities can be removed
before the polar boronic acid elutes from the column. An
alternative strategy is to adsorb the boronic acid on to
silica (7.5 g silica for each 1 g boronic acid) and then wash
the silica with hexane followed by elution with pure ethyl
acetate or methanol to obtain the purified boronic acid.
This has the advantage of being procedurally easier and
also, in some cases, leads to a greater recovery of the
boronic acid then when using column chromatography.
Boronic acids can be recrystallized from water25 to give
pure samples although we find that this can lead to
significant material loss and does not always remove all
the impurities effectively.
These bodies of work made us critically reassess our
transition-metal free protocol. We present our findings
here. We believe that, although the reaction can be run
without the need for addition of a metal catalyst, pal-
ladium contaminants down to a level of 50 ppb found in
commercially available sodium carbonate are responsible
for the generation of the biaryl rather than, as previously
suggested, an alternative non-palladium-mediated path-
way.
(13) Bedford, R. B.; Hazelwood, S. L.; Albisson, D. A. Organometal-
lics 2002, 21, 2599.
(14) Thathagar, M. B.; Beckers, J.; Rothenberg, G. J. Am. Chem.
Soc. 2002, 124, 2159.
(15) Liu, X. X.; Deng, M. Z. Chem. Commun. 2002, 622. (b) Savarin,
C.; Liebeskind, L. S. Org. Lett. 2001, 3, 2149.
(16) de Vries, A. H. M.; Mulders, J. M. C. A.; Mommers, J. H. M.;
Henderickx, H. J. W.; de Vries, J. G. Org. Lett. 2003, 5, 3285.
(17) de Vries, A. H. M.; Parlevliet, F. J.; Schmieder-van de Vond-
ervoort, L.; Mommers, J. H. M.; Henderickx, H. J. W.; Walet, M. A.
M.; de Vries, J. G. Adv. Synth. Catal. 2002, 344, 996.
(18) Choudary, B. M.; Madhi, S.; Chowdari, N. S.; Kantam, M. L.;
Sreedhar, B. J. Am. Chem. Soc. 2002, 124, 14127.
(19) Prockl, S. S.; Kleist, W.; Gruber, M. A.; Kohler, K. Angew.
Chem., Int. Ed. 2004, 43, 1881.
(20) Palencia, H.; Garcia-Jimenez, F.; Takacs, J. M. Tetrahedron
Lett. 2004, 45, 3849.
(24) Caution: The water is heated well above its boiling point so
all necessary precautions should be taken when performing such
experiments. Vessels designed to withhold elevated pressures must
be used. The microwave apparatus used here incorporates a protective
cage around the microwave vessel in case of explosion. After completion
of an experiment, the vessel must be allowed to cool to a temperature
below the boiling point of the solvent before removal from the
microwave cavity and opening to the atmosphere.
(21) Alonso, D. A.; Najera, C.; Pacheco, M. C. J. Org. Chem. 2002,
67, 5588.
(22) Smith, M. D.; Stepan, A. F.; Ramarao, C.; Brennan, P. E.; Ley,
S. V. Chem. Commun. 2003, 2652.
(23) Conlon, D. A.; Pipik, B.; Ferdinand, S.; LeBlond, C. R.; Sowa,
J. R.; Izzo, B.; Collins, P.; Ho, G. J.; Williams, J. M.; Shi, Y. J.; Sun, Y.
K. Adv. Synth. Catal. 2003, 345, 931.
(25) Gilman, H.; Moore, L. O. J. Am. Chem. Soc. 1958, 80, 3609.
162 J. Org. Chem., Vol. 70, No. 1, 2005

Transition-Metal Free Suzuki-Type Coupling Methodology
TABLE 1. Suzuki-Type Coupling of Aryl Bromides and
Aryl Boronic Acidsa
TABLE 2. Suzuki-Type Coupling of Aryl Bromides with
1a
a Reactions run on 1 mmol scale with no added palladium.
Microwave irradiation of 150 W used. Temperature ramped from
rt to 150 °C and held for 5 min. b Significant quantities of styrene
observed. c Decomposition observed.
To build on this, we then screened the coupling of 1
with a wider range of aryl bromides, the results being
shown in Table 2. Yields vary depending on the nature
and position of the substituents. When an ortho-substi-
tuted aryl halide is used, yields tend to be lower than
with the para-substituted analogue due to competitive
deboronation of 1 (Table 2, entries 3 and 4). A problem
of the methodology seems to be the stability of the
products in the hot basic aqueous media, with significant
decomposition being observed in a number of cases.
To probe the coupling further, we screened the reaction
of 4-bromoacetophenone with the less reactive 1-hexenyl-
boronic acid and found that only a 20% yield of product
is obtained (Scheme 2). The reaction between â-bro-
mostyrene and purified phenylboronic acid yields stilbene
in 82% yield (Scheme 2), this showing that coupling of
an alkenyl bromide with an aryl boronic acid is possible
without the need for addition of a catalyst. It is also
possible to couple â-bromostyrene with 1 to yield a diene
product.
a Reactions run on 1 mmol scale with no added palladium.
Microwave irradiation of 150 W used. Temperature ramped from
rt to 150 °C and held for 5 min.
With a representative range of purified arylboronic
acids in hand, we have screened them in the coupling
protocol. The results are shown in Table 1. Using bro-
moacetophenone as the aryl halide substrate, we find
that it is possible to couple this with pure boronic acids
bearing electron-withdrawing or electron-donating groups.
The presence of an o-methyl functionality on the boronic
acid does not lead to a lowering in product yield (Table
1, entry 6). We have investigated the coupling of 4-meth-
oxybenzeneboronic acid with 4-bromoacetophenone, 4-bro-
motoluene, and 4-bromoanisole (Table 1, entries 3-5).
We find that in the case of the latter, no product is
obtained as compared to high yields with the other
examples where excellent yields of the desired biaryl are
observed.
Building on our results with the boronic acids, we
wanted to probe the effects of purity of the other reagents
on the coupling reaction. In light of the reports of
“homeopathic” palladium catalysis of the Heck reaction,17
we in particular wanted to reanalyze reagents and whole
product mixtures for sub-ppm concentrations of pal-
ladium using ICP-MS analysis to enable us to measure
lower concentrations than previously possible. One source
of palladium could be the water used in the reaction.
Since starting our program, we have been using ultra-
pure water, purified to a specific resistance of >16 mΩ‚
cm on every occasion. We analyzed this for palladium and
found levels in the region of 0.24 ppb which we believe
to be too low to be catalytically active. The organic
reagents used in the reactions are all purified prior to
use and we believe them to be palladium free. The only
reagent we cannot easily purify is the sodium carbonate.
In an attempt to develop the methodology further, we
wanted to investigate the coupling of alkenyl boronic
acids with aryl halides and alkenyl halides with aryl
boronic acids, the product in both cases being substituted
alkenes. Suzuki couplings of these types of substrates are
known using Pd catalysts. We first screened the reaction
between trans-2-phenylvinylboronic acid (1) and 4-bro-
moacetophenone and 4-bromoanisole. We find that in
both cases, the desired disubstituted alkene product is
formed. We observe only the trans isomer of the stilbene
products.
J. Org. Chem, Vol. 70, No. 1, 2005 163

Arvela et al.
SCHEME 2a
a Reactions run on 1 mmol scale with no added palladium. Microwave irradiation of 150 W used. Temperature ramped from rt to 150
°C and held for 5 min.
Analysis of solutions of so-termed “ultrapure” sodium
carbonate in ultrapure water at a concentration equal
to half that used in the reaction showed a palladium
concentration of between 10 and 24 ppb. Therefore the
expected palladium concentration in our reaction vessels
would be from 20 to 50 ppb. Study has shown that these
levels of palladium are found in many commercially
available sources of sodium carbonate but not potassium
carbonate. Analysis of aqueous solutions of “ultrapure”
potassium carbonate showed levels of palladium corre-
sponding to 0.09 ppb in the concentrations used in our
experiments. We therefore wanted to investigate in more
detail why, as reported in our earlier papers, when the
coupling methodology is performed using sodium carbon-
ate as a base the reaction works well, but with the
potassium analogue it does not. We wanted to see if the
ppb levels of palladium in the Na2CO3 proved sufficient
to catalyze the coupling reaction.
To probe this we performed the reaction of 4-bromoac-
etophenone and phenylboronic acid in the presence of 100
ppb palladium acetate (corresponding to approximately
50 ppb Pd). The reaction was performed on a 1 mmol
scale, using potassium carbonate as the base, 2 mL of
water as the solvent, and TBAB as a phase-transfer
agent. The reaction mixture was heated in a sealed vessel
to 150 °C in a scientific microwave and held at this
temperature for 5 min. We obtained a 75% yield of
product, this comparing to <5% product formation with
potassium carbonate in the absence of the added pal-
ladium. This corresponds to a palladium loading of
0.0000008 mol % and a turnover number of 1 250 000.
We then performed the same no-added palladium reac-
tion using our sodium carbonate and obtained a 95% yield
of the desired biaryl. This leads us to believe that,
although the reaction occurs without the direct addition
of a transition-metal complex, it is catalyzed by ppb levels
of palladium originating from the commercially available
sodium carbonate. The fact that the reaction is reproduc-
ible with a wide range of sources of commercially avail-
able sodium carbonate confirms our initial observation
that palladium is a very low level contaminant found in
this material.
etophenone and phenylboronic acid in the presence of a
1 000 000-fold excess of EDTA had no effect on the
product yield. This added credence to our hypothesis that
the reaction was taking place via a metal-free mecha-
nism. However, we now find that running the same
reaction using potassium carbonate as base leads to
product formation but only when 100 ppb palladium is
added to the reaction mixture. This indicates that EDTA
does not efficiently sequester Pd from the reaction
mixture at elevated temperatures or that the complex
formed between the palladium acetate and the EDTA is,
itself, catalytically active. We managed to obtain a sample
of sodium carbonate containing only 0.21 ppb Pd and
found that, when using this as the base in the reaction
of 4-bromoacetophenone and phenylboronic acid, no
product was obtained. Addition of 100 ppb palladium
acetate to an identical reaction mixture leads to a 90%
yield of the desired biaryl. This acts as definitive proof
that the reaction is catalyzed by the ppb levels of Pd
found in commercially available sodium carbonate.
A further interesting observation we made was that
the coupling of 4-bromoacetophenone and phenylboronic
acid was higher yielding when the reaction mixture was
not stirred. In the absence of stirring, the reaction
mixture forms two distinct phases: a lower aqueous layer
containing the base and an upper organic layer contain-
ing the organic substrates. We believe that the TBAB sits
in both the organic and aqueous phases. This made us
probe the role of the water and TBAB. Running the
reaction in the absence of water does not yield product,
even with the addition of higher loadings of palladium.
We believe that one of the key roles of the water is simply
to dissolve the base and that the reaction takes place at
the aqueous/organic interface. Badone and co-workers
have investigated the effects of solvent, including water,
on the rate of the ligandless palladium acetate catalyzed
Suzuki reaction of a range of aryl bromides, iodides, and
triflates.26 They report that when using water as a
solvent the addition of 1 equiv of tetrabutylammonium
bromide (TBAB) to the reaction mixture greatly acceler-
ates the reaction. They find that activated aryl bromides
can be coupled with phenylboronic acid to yield biaryls
fairly rapidly (1 h) and in good yields whereas with aryl
iodides the reaction does not reach completion. The role
To probe more thoroughly whether the reaction was
Pd-mediated or not we revisited some earlier experiments
we had performed using EDTA to scavenge ppb levels of
palladium from the reaction mixture. In our initial
studies we found that running the reaction of 4-bromoac-
(26) Badone, D.; Baroni, M.; Cardamone, R.; Ielmini, A.; Guzzi, U.
J. Org. Chem. 1997, 62, 7170.
164 J. Org. Chem., Vol. 70, No. 1, 2005

Transition-Metal Free Suzuki-Type Coupling Methodology
TABLE 3. Suzuki Coupling of Aryl Bromides with Phenylboronic Acid
a Reactions run on 1 mmol scale. Pd(NO3)2 used as catalyst. Microwave irradiation of 150 W used. Temperature ramped from rt to 150
°C and held for 5 min. b Reactions run on 1 mmol scale. Pd(OAc)2 used as catalyst. Reaction mixture placed into an oil bath preheated to
150 °C and held there for 7 min before being removed.
of the ammonium salts is thought to be 2-fold. First, they
facilitate solvation of the organic substrates in the sol-
vent medium. Second, they are thought to enhance the
rate of the coupling reaction by activating the boronic
acid to reaction by formation of a boronate complex
[ArB(OH)3]-[R4N]+. TBAB has been used recently in
conjunction with a palladium oxime catalyst for the
Suzuki coupling of aryl chlorides with phenylboronic acid
in water.27 We wanted to explore whether it was the fact
that we were using TBAB as an additive in our reactions
that made the coupling so efficient at the very low
palladium loadings. To test this, we needed to find an
alternative phase-transfer agent to TBAB. In our initial
research, we had screened a range of organic solvent/
water mixtures as media for the coupling without the
addition of TBAB or a metal catalyst and found that
either the reaction was inhibited or else considerably
lower yields of the desired biaryl were formed. We re-
screened mixtures of water and, respectively, NMP,
DMF, and methanol using the 100 ppb palladium acetate
solution but no TBAB and found the same results as
previously. We then focused on mixtures of water and
alcohols and found that although methanol/water and
butanol/water mixtures inhibited the reaction, a 1:1
ethanol/water mixture was an excellent solvent mixture
giving comparable yields to when water alone is used in
the presence of TBAB. This shows that the TBAB is not
essential to the success of the coupling reaction.
ppm quantities of metal complex, even small amounts of
undissolved material will have a profound effect on the
bulk concentration. For this reason, and also because the
palladium contaminant in sodium carbonate is not likely
to be Pd(OAc)2, we wanted to turn our attention to a
water-soluble simple Pd salt. Our attention has focused
on palladium nitrate since this is readily soluble in water
and hence solutions of exact and reliable concentration
can be prepared. We wanted to develop a general
methodology using ultralow levels of palladium and,
initially, TBAB as phase-transfer agent and then using
water/ethanol as the reaction medium. We decided to
screen selected reactions of 4-bromoacetophenone, 4-bro-
motoluene, and 4-bromoanisole with phenylboronic acid.
We used both sodium and potassium carbonate and
worked at three palladium concentrations, namely 100,
250, and 2.5 ppm. These correspond to catalyst loadings
of 0.0000016, 0.000004, and 0.00004 mol %, respectively.
The reactions were run using a microwave power of 150
W we ramped the temperature from rt to 150 °C, this
taking around 1-2 min and then held at this tempera-
ture for 5 min. Product yields are shown in Tables 3 and
4. Not surprisingly, product yields increase with increas-
ing palladium concentration. We find that reactions run
using water and ethanol are higher yielding than those
using water and TBAB, especially at low palladium
concentrations. Product yields obtained using K2CO3 are
generally lower than those observed with Na2CO3.
A problem with using ultralow concentrations of pal-
ladium acetate in water for the reactions is the confidence
that can be put on the exact metal concentration of the
solution. Palladium acetate is only sparingly soluble in
water, even when using high dilution. Working with sub-
We repeated the coupling between 4-bromoacetophe-
none and phenylboronic acid for different reaction times
using water/TBAB and find that the reaction reaches
completion within a total microwave irradiation time of
90 s. This corresponds to turnover frequencies of
(27) Botella, L.; Na´jera, C. Angew. Chem., Int. Ed. 2002, 41, 179.
J. Org. Chem, Vol. 70, No. 1, 2005 165

Arvela et al.
TABLE 4. Suzuki Coupling of Aryl Bromides with Phenylboronic Acid Using Water/Ethanol as Solvent
a Reactions run on 1 mmol scale. Pd(NO3)2 used as catalyst. Microwave irradiation of 150 W used. Temperature ramped from rt to 150
°C and held for 5 min. b Reactions run on 1 mmol scale. Pd(OAc)2 used as catalyst. Reaction mixture placed into an oil bath preheated to
150 °C and held there for 7 min before being removed.
28 000 000, 47 000 000, and 50 000 000 for palladium
loadings of 100, 250, and 2.5 ppm, respectively.
We dipped the tube into an oil bath preheated to 150 °C
and chose a reaction time of 7 min since this equated to
the microwave heating conditions (2 min heat up and 5
min hold). Caution.29 Using water and TBAB as reaction
medium, a yield of 4-acetylbiphenyl of 8% was obtained.
This compares to a yield of 56% when using microwave
heating. We then screened other reaction conditions and
reagents, looking primarily at the effects of changing the
aryl halide substrate and palladium concentration. Prod-
uct yields are shown in Tables 3 and 4 for water/TBAB
and water/ethanol reaction mixtures, respectively. At low
palladium concentrations, product yields obtained using
conventional heating are significantly lower than those
observed when using microwave heating. However at a
palladium concentration of 2.5 ppm yields are much
closer, particularly when a water and ethanol mixture
is used as the reaction medium.
We re-screened the coupling of 1 with representative
aryl bromides using 2.5 ppm Pd(OAc)2, the results being
shown in Table 2. As with the no-catalyst added studies,
yields vary depending on the nature of the substituents.
However, not unexpectedly, the product yields were
higher when using the 2.5 ppm Pd(OAc)2 solution. We
find that yields obtained when using TBAB as phase-
transfer agent (Table 5, entries 1-5) are higher than
those when using a mixture of water and ethanol as
reaction medium (Table 5, entry 6). A combination of
water, ethanol, and TBAB as a reaction medium gives
very similar results as with water and TBAB (Table 5,
entries 7-9). This would suggest that TBAB plays a more
Having undertaken a range of experiments using
microwave heating, we wanted to compare our data with
that obtained using conventional heating. Since the start
of the field of microwave-assisted synthesis, far shorter
times for reactions as compared to those using “conven-
tional” heating have been reported, and this has sparked
debate into the nature of the microwave heating.28 The
acceleration of reactions could simply be an effect of the
thermal energy generated by the microwaves interacting
with the substrates or could be an effect specific to
microwave heating. In most cases, the observed differ-
ences between microwave and conventional heating can
be attributed to simple thermal effects. We were inter-
ested in trying to ascertain whether the rates of the
reactions we were observing in our Suzuki reaction were
simply due to efficient thermal heating. In their report
of cross couplings of boronic acids and poly(ethylene
glycol) supported aryl iodides in water using Pd(OAc)2
as catalyst, Schotten and co-workers do not comment
specifically on whether their conditions for conventional
heating are optimized or what temperature their micro-
wave reactions reach but do report that reaction times
for the microwave-assisted reactions are a matter of 2-4
min as compared to 2 h for the conventional thermal
reactions. We performed the coupling of 4-bromoac-
etophenone and phenylboronic acid in water/TBAB with
addition of 100 ppb Pd using conventional heating. We
kept the quantities of reagents and solvent exactly the
same as in the microwave-promoted protocol and per-
formed the reaction in a sealed 10 mL microwave tube.
(29) Caution: A blast shield should be in place and vessels designed
to withhold elevated pressures must be used. After completion of an
experiment, the vessel must be allowed to cool before opening to the
atmosphere.
(28) For a review see: Perreux, L.; Loupy, A. Tetrahedron 2001, 57,
9199.
166 J. Org. Chem., Vol. 70, No. 1, 2005

Transition-Metal Free Suzuki-Type Coupling Methodology
TABLE 5. Suzuki Coupling of Aryl Bromides with 1a
a Reactions run on 1 mmol scale. Pd(OAc)2 used as catalyst. Microwave irradiation of 150 W used. Temperature ramped from rt to 150
°C and held for 5 min. b Decomposition observed.
TABLE 6. Effects of Palladium Concentration on the
Coupling of 4-bromoacetophenone and Un-purified
Phenylboronic Acida
tion to 1 ppm leads to a product conversion of 93%. These
results show that the contaminants in boronic acids must
either form a catalytically inactive complex with the
palladium or else inhibit the reaction by some other
pathway.
entry
Pd concentration
product yield (%)
1
2
3
4
50 ppb
100 ppb
1 ppm
18
48
93
94
In summary, we have shown that when the Suzuki
reaction is performed without the addition of a transition-
metal catalyst but using sodium carbonate as a base,
palladium contaminants down to a level of 50 ppb found
in the commercially available base are responsible for the
generation of the biaryl rather than, as previously
suggested, an alternative non-palladium-mediated path-
way. In light of this, we have presented here a revised
methodology for Suzuki couplings using ultralow pal-
ladium concentrations for use with aryl and vinyl boronic
acids. The methodology can be summarized as follows:
The reaction is run using commercially avaliable aryl
halides and boronic acids. Water is used as a solvent.
Either ethanol or TBAB is used as a phase-transfer
agent. For arylboronic acids both ethanol and TBAB are
effective while for vinylboronic acids TBAB is the phase-
transfer agent of choice. A palladium catalyst loading of
1 ppm Pd is recommended for optimal yields. The
reaction is run for 7 min using microwave heating (2 min
heating to temperature and 5 min at temperature). When
using conventional heating, increasing the catalyst load-
ing to 2.5 ppm is recommended for good yields of product
after the 7 min reaction time. Regardless of the heating
method, a reaction temperature of 150 °C is optimal.
2.5 ppm
a Reactions run on 1 mmol scale. Pd(OAc)2 used as catalyst.
Microwave irradiation of 150 W used. Temperature ramped from
rt to 150 °C and held for 5 min.
important role in the reaction of aryl bromides with 1 as
compared with arylboronic acids.
We finally wanted to return to the issue of the purity
of the boronic acid. All our experiments above had been
run using purified phenylboronic acid. As discussed
earlier, impurities in the boronic acid, even in very small
amounts, can effectively shut down the coupling reaction
when no palladium catalyst is added. We wanted to see
what level of palladium was required to effect the
coupling in good yield when an impure batch of boronic
acid was used. To do this, we took a sample of phenyl-
boronic acid that we knew did not work in the no-added
catalyst methodology and ran a series of experiments
where we varied the palladium concentration and moni-
tored the product yield in the coupling reaction with
4-bromoacetophenone. The results are shown in Table 6.
We found that at a 50 ppb concentration we obtained only
an 18% conversion. Increasing the palladium concentra-
J. Org. Chem, Vol. 70, No. 1, 2005 167

Arvela et al.
General Procedure for the Palladium-Added Coupling
Reactions Using TBAB as Phase-Transfer Agent. The
protocol was as with the no-catalyst-added methodology but
instead of adding 2 mL of water, aliquots of either a 10 or 1
ppm solution of Pd(OAc)2 or Pd(NO3)2 were added, and the
volume was made up to 2 mL with ultrapure water to give
the desired metal concentration.
General Procedure for the Palladium-Added Coupling
Reactions Using Ethanol as a Cosolvent. The protocol was
as with the no-catalyst-added methodology but instead of
adding 2 mL of water and 1 mmol of TBAB, 1 mL of EtOH
was added, aliquots of either a 10 or 1 ppm solution of Pd-
(OAc)2 or Pd(NO3)2 were added, and the volume was made up
to 2 mL with ultrapure water to give the desired metal
concentration.
General Procedure for the Conventional Heating
Experiments. The reaction mixtures were prepared as in the
case of the microwave heating experiments. After the tube was
sealed with a septum it was placed into an oil bath preheated
to 150 °C and held there for 7 min before being removed and
allowed to cool, and then the product extracted and purified
in a manner identical to that in the microwave protocol.
Experimental Section
General Experimental Procedures. All reactions were
carried out in air. Microwave reactions were conducted using
a focused microwave unit. The machine consists of a continu-
ous focused microwave power delivery system with operator
selectable power output from 0 to 300 W. Reactions were
performed in glass vessels (capacity 10 mL) sealed with a
septum. The pressure is controlled by a load cell connected to
the vessel via a 14-gauge needle which penetrates just below
the septum surface. The temperature of the contents of the
vessel was monitored using a calibrated infrared temperature
control mounted under the reaction vessel. H and 13C NMR
spectra were recorded at 293 K on a 400 MHz spectrometer.
Mass spectroscopy was performed at Notre Dame University,
IN.
1
General Procedure for the No-Catalyst-Added Cou-
pling Reactions. In a 10 mL glass tube were placed aryl
halide (1.0 mmol), arylboronic acid (1.3 mmol), Na2CO3 (392
mg, 3.7 mmol), tetrabutylammonium bromide (322 mg, 1.0
mmol), 2 mL of water, and a magnetic stir bar (optional). The
vessel was sealed with a septum, shaken vigorously (essential),
and placed into the microwave cavity. Microwave irradiation
of 150 W was used, the temperature being ramped from rt to
150 °C. Once this temperature was reached, the reaction
mixture was held at this temperature for 5 min. After allowing
the mixture to cool to room temperature, the reaction vessel
was opened and the contents poured into a separating funnel.
Water (30 mL) and ethyl acetate (30 mL) were added, and the
organic material was extracted and removed. After further
extraction of the aqueous layer with ether, combining the
organic washings and drying them over MgSO4, the ethyl
acetate was removed in vacuo leaving the crude product. The
product was purified and isolated by chromatography using
hexane/ethyl acetate as eluent.
Acknowledgment. The University of Connecticut
(R.K.A.), NSF REU program (M.S.S.), and CEM Micro-
wave Technology (V.A.W.) are thanked for financial
support. Saint Mary’s University, CFI, NSERC-Discov-
ery, and NSERC-CRD programs (R.D.S., P.G.) are
thanked for support of the ICP-MS analyses.
Supporting Information Available: Spectral data for the
cross-coupling products. This material is available free of
charge via the Internet at http://pubs.acs.org.
JO048531J
168 J. Org. Chem., Vol. 70, No. 1, 2005