Study of a new rate increasing "base effect" in the palladium-catalyzed amination of aryl iodides

Article abstract of DOI:10.1021/jo049774e

Evidence for an interphase deprotonation of Pd(II)-amine complexes with weak carbonate base has been gained for the first time. When a rate-limiting deprotonation step is involved in the catalytic cycle, controlling the structure (shape and size of the particles) and/or molar excess of the carbonate base used can significantly increase the reaction rate of Buchwald-Hartwig aminations. By taking such a "base effect" into account a general protocol for the intermolecular amination of aryl iodides with all types of amines has been developed based on a standard Pd-BINAP catalyst, using cesium carbonate as the base.

Full text of DOI:10.1021/jo049774e

Stu d y of a New Ra te In cr ea sin g “Ba se Effect” in th e
P a lla d iu m -Ca ta lyzed Am in a tion of Ar yl Iod id es
Caroline Meyers, Bert U. W. Maes,* Kristof T. J . Loones, Gunther Bal,
Guy L. F. Lemi e` re, and Roger A. Dommisse
Department of Chemistry, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, Belgium
bert.maes@ua.ac.be
Received February 9, 2004
Evidence for an interphase deprotonation of Pd(II)-amine complexes with weak carbonate base
has been gained for the first time. When a rate-limiting deprotonation step is involved in the catalytic
cycle, controlling the structure (shape and size of the particles) and/or molar excess of the carbonate
base used can significantly increase the reaction rate of Buchwald-Hartwig aminations. By taking
such a “base effect” into account a general protocol for the intermolecular amination of aryl iodides
with all types of amines has been developed based on a standard Pd-BINAP catalyst, using cesium
carbonate as the base.
In tr od u ction
ligands, trialkylphoshine ligands, palladacycles) that
allow the use of bases weaker than the traditional
Arylamines are very important structural entities in
modern organic chemistry. These subunits can be found
in natural products, pharmaceuticals, and agrochemicals,
as well as new materials (small molecules, oligomers, and
polymers) with interesting optical and electronic proper-
“MOtBu” (in order to have a large functional group
compatibility) and by tackling industrially relevant aryl
chloride substrates by using only a low loading of
5
catalyst. If weak bases are used in Buchwald-Hartwig
aminations, rate-limiting deprotonation of the Pd(II)-
1
ties. Traditional routes for their synthesis involve elec-
amine intermediate formed in the catalytic cycle (complex
trophilic nitration followed by reduction and subsequent
reductive amination, Ullmann-type reaction, and nucleo-
philic aromatic substitution of the addition-elimination
2
, Scheme 1) is most likely and consequently often
observed. In the present work we clearly show that in
such cases the reaction rate can be seriously increased
by carefully controlling the structure (particle size and
shape) and/or molar excess of the base used (“base
effect”). We selected the intermolecular amination of aryl
iodides as a case study for this base effect since in ami-
nation reactions, in contrast to C-C bond forming pro-
cesses, the aryl iodides have been shown to be less effec-
tive substrates than aryl bromides or chlorides. Hitherto,
only two studies have been published that describe a
general protocol for the intermolecular amination of aryl
(S
N
Ar) or elimination-addition (via aryne) type.2 All
these traditional methods suffer from a limited substrate
scope due to the requirement of (relatively) harsh reac-
tion conditions or the presence of activating electron
withdrawing groups. Interestingly, in the mid nineties
Buchwald and Hartwig independently discovered the tin
free Pd-catalyzed amination of aryl bromides showing a
3
wide scope. This new procedure has established itself
as the most important method for C-N bond formation
on aromatics currently available.4 Since 1995, major
advances have been made in this field mainly by devel-
oping new palladium catalysts (aryl dialkylphosphine
6
,7
2 3
iodides with use of a weak base (Cs CO ). In 2001,
Buchwald reported that 2-(dicyclohexylphosphanyl)-2′-
dimethylaminobiphenyl (DCPDMAB) is effective for the
coupling of aliphatic amines while 9,9-dimethyl-4,5-bis-
(diphenylphosphanyl)xanthene (XANTPHOS) is a more
(
1) (a) Buckingham, J . Dictionary of Natural Products; University
Press: Cambridge, MA, 1994. (b) Czarnik, A. W. Acc. Chem. Res. 1996,
9, 112. (c) Kleemann, A.; Engel, J .; Kutscher, B.; Reichert, D.
Pharmaceutical Substances, 3rd ed.; Thieme: Stuttgart, Germany,
999. (d) CRC Handbook of Pesticides; Milne, G. W. A., Ed.; CRC
2
(4) For reviews and accounts on the Buchwald-Hartwig reaction
see: (a) Bara n˜ ano, D.; Mann, G.; Hartwig, J . F. Curr. Org. Chem. 1997,
1, 287. (b) Wolfe, J . P.; Wagaw, S.; Marcoux, J .-F.; Buchwald, S. L.
Acc. Chem. Res. 1998, 31, 805. (c) Frost, C. G.; Mendon c¸ a, P. J . Chem.
Soc., Perkin Trans. 1 1998, 2615. (d) Hartwig, J . F. Angew. Chem.,
Int. Ed. Engl. 1998, 37, 2046. (e) Yang, B. H.; Buchwald, S. L. J .
Organomet. Chem. 1999, 576, 125. (f) Hartwig, J . F. In Modern
Amination Methods; Ricci, A., Ed.; Wiley-VCH: Weinheim, Germany,
2000. (g) Muci, A. R.; Buchwald, S. L. Top. Curr. Chem. 2002, 219,
131. (h) Prim, D.; Campagne, J .-M.; J oseph, D.; Andrioletti, B.
Tetrahedron 2002, 58, 2041.
1
Press: Boca Raton, FL, 1994. (e) Waring, D. R.; Hallas, G. The
Chemistry and Application of Dyes; Plenum: New York, 1990. (f)
Polymers for Lightwave and Integrated Optics; Hornak, L. A., Ed.;
Marcel Dekker: New York, 1992. (g) D’Aprano, G.; Leclerc, M.; Zotti,
G.; Schiavon, G. Chem. Mater. 1995, 7, 33. (h) Skotheim, T. A.;
Elsenbaumer, R. L.; Reynolds, J . R. Handbook of Conducting Polymers,
2
nd ed.; Marcel Dekker: New York, 1997.
2) (a) Belfield, A. J .; Brown, G. R.; Foubister, A. J . Tetrahedron
999, 55, 11399. (b) Lindley, J . Tetrahedron 1984, 40, 1433. (c) Wenk,
(
1
H. H.; Winkler, M.; Sander, W. Angew. Chem., Int. Ed. 2003, 42, 502.
d) March, J . Advanced Organic Chemistry, 4th ed.; J ohn Wiley &
Sons: New York, 1992.
3) (a) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1348. (b) Louie, J .; Hartwig, J . F. Tetrahedron
Lett. 1995, 36, 3609.
(5) For a review partly dealing with palladium-catalyzed aminations
of aryl chlorides see: Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed.
2002, 41, 4176.
(6) Ali, M. H.; Buchwald, S. L. J . Org. Chem. 2001, 66, 2560.
(7) Urgaonkar, S.; Xu, J . H.; Verkade, J . G. J . Org. Chem. 2003, 68,
8416.
(
(
1
0.1021/jo049774e CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/31/2004
6
010
J . Org. Chem. 2004, 69, 6010-6017

“Base Effect” in Pd-Catalyzed Amination of Aryl Iodides
5-position, respectively.10 In this preliminary study we
SCHEME 1. Gen er a l Ca ta lytic Cycle for th e
Bu ch w a ld -Ha r tw ig Am in a tion w ith Wea k Ba se.
found that a Pd-BINAP catalyst, used in combination
with a large excess of mild carbonate base (Cs CO ), gave
2 3
selective aminations in excellent yields.11 Surprisingly,
we observed that a large excess of base (5 equiv),
although almost completely insoluble in toluene, was
essential to obtain sufficiently fast reactions and full
conversions.10 This and our earlier qualitative observa-
tion that larger excesses of carbonate bases also speed
up Buchwald-Hartwig aminations of chloropyridazin-
3
(2H)-ones and dichloropyridines with use of Pd-BINAP
catalyst in toluene, as well as a similar observation made
by Watanabe’s group when performing aminations of aryl
3
bromides using Pd-P(tBu) in xylene, indicate that this
base effect might be quite general.1
2,13
The first goal of the present work was to investigate if
the combination of a Pd-BINAP catalyst with 5 equiv
of Cs CO is indeed a general protocol for the efficient
2 3
amination of aryl iodides with all types of amines. To
have an idea about the potential of our protocol in
comparison with Buchwald’s method, using a Pd-DCP-
DMAB (aliphatic amines) or Pd-XANTPHOS (anilines)
suitable ligand for C-N bond formation with anilines.6
catalyst in combination with a small excess of Cs
2 3
CO
3
For both catalysts special cosolvents (EtN or tBuOH)
(1.4 equiv), we first investigated the same seven substrate
were needed to obtain rate enhancements, and a high
loading of catalyst is required to couple acyclic secondary
aliphatic amines. Recently, while our work was in
progress, Verkade published an alternative for Buch-
wald’s protocol. By using a bicyclic triaminophosphine
ligand, aryl iodides could be efficiently coupled with
aromatic as well as aliphatic amines under mild reaction
combinations (Table 1, entries 1, 2, 3, 4a, 10, 11, and 12)
Buchwald reported in 2001, but under the reaction
conditions we reported for the regioselective amination
of chloro-iodopyridines with anilines.6 Interestingly,
similar yields were obtained by using the same loading
of palladium in an overnight reaction. For the coupling
of 2-iodoanisole with morpholine some starting material
remained after an overnight reaction under our condi-
tions (Table 1, entry 4a). Therefore we increased the
catalyst loading with 1 mol % and subsequently obtained
a similar yield as under Buchwald’s conditions (Table 1,
entry 4b). To further investigate the applicability of our
new protocol, we studied an additional set of five different
substrate combinations involving the coupling of different
aliphatic amines (primary as well as acyclic and cyclic
secondary amines) (Table 1, entries 5-9). As a starting
point a 1 mol % loading of Pd-BINAP catalyst was
always used. A higher catalyst loading was taken if no
full conversion of starting material was obtained after
an overnight reaction. Clearly, secondary cyclic amines
are coupled more easily than secondary acyclic ones and
the primary aliphatic amines can be situated between
,10
7
conditions. A limitation of this protocol is that the use
of primary aliphatic amines gives only poor results. We
present here our finding that, in contrast to what is
hitherto generally accepted, the use of a standard second
generation catalyst, based on air-stable and commercially
available racemic 2,2′-bis(diphenylphosphanyl)-1,1′-bi-
naphthyl (BINAP) ligand, in combination with a weak
cesium carbonate base can be used as a general catalytic
system for the efficient amination of aryl iodides with
all types of amines if the base effect is taken into
account.8
,9
Resu lts a n d Discu ssion
Recently, our laboratory reported regioselective pal-
ladium-catalyzed aminations of 2-chloro-3-iodopyridine
and 2-chloro-5-iodopyridine with anilines in the 3- and
(compare Table 1, entries 7, 8a, and 9a). In all the cases
studied, secondary cyclic aliphatic amines seem to work
(
8) No successful report on intermolecular aminations on aryl iodides
(10) For a preliminary communication see: Maes, B. U. W.; Loones,
K. T. J .; J onckers, T. H. M.; Lemi e` re, G. L. F.; Dommisse, R. A.;
Haemers, A. Synlett 2002, 1995.
(11) For the Pd-catalyzed amination of azaheteroaryl iodides with
strong MOtBu base see: (a) Cheng, J .; Trudell, M. L. Org. Lett. 2001,
3, 1371. Interestingly, a completely opposite regioselectivity has been
observed by Cheng and Trudell when they studied the coupling of
2-chloro-5-iodopyridine with 7-azabicyclo[2.2.1]heptane using a Pd-
carbene complex. (b) Parrot, I.; Ritter, G.; Wermuth, C. G.; Hibert, M.
Synlett 2002, 1123.
(12) Pd-BINAP catalyst in combination with a large excess (5-20
2 3
equiv) of K CO accelerated the amination of azaheteroaryl chloride
substrates with anilines: (a) Chloropyridazin-3(2H)-ones: Ko sˇ mrlj, J .;
Maes, B. U. W.; Lemi e` re, G. L. F.; Haemers, A. Synlett 2000, 1581. (b)
Dichloropyridines: J onckers, T. H. M.; Maes, B. U. W.; Lemi e` re, G. L.
F.; Dommisse, R. Tetrahedron 2001, 57, 7027.
with use of mild bases such as Cs CO in combination with Pd-BINAP
2
3
has been published yet. Only the use of strong NaOtBu in combination
with Pd-BINAP has been reported as a useful general protocol. In
this case also toxic additives were required: Wolfe, J . P.; Buchwald,
S. L. J . Org. Chem. 1997, 62, 6066. There is only one exception reported
in the literature. For the specific case of the palladium-catalyzed
N-arylation of sulfoximines, the use of aryl iodides with Pd-BINAP
as catalyst in combination with Cs
to ensure product formation in acceptable yields the use of additives
LiBr, AgOTf) was essential: Bolm, C.; Hildebrand, J . P. J . Org. Chem.
000, 65, 169.
9) For the Pd-catalyzed amination of aryl iodides with strong
MOtBu base see: (a) Wolfe, J . P.; Buchwald, S. L. J . Org. Chem. 1996,
2 3
CO has been reported. However,
(
2
(
6
1
1, 1133. (b) Driver, M. S.; Hartwig, J . F. J . Am. Chem. Soc. 1996,
18, 7217. (c) See ref 8. (d) Hamann, B. C.; Hartwig, J . F. J . Am. Chem.
Soc. 1998, 120, 7369. (e) Nishiyama, M.; Yamamoto, T.; Koie, Y.
Tetrahedron Lett. 1998, 39, 617. (f) Huang, J .; Grasa, G.; Nolan, S. P.
Org. Lett. 1999, 1, 1307. (g) See ref 6. (h) Urgaonkar, S.; Nagarajan,
M.; Verkade, J . G. J . Org. Chem. 2003, 68, 452.
(13) Pd-P(t-Bu)
3 2 3
catalyst in combination with an excess of K CO
or Rb CO (3 equiv) accelerated the amination of aryl bromide
2
3
substrates with carbazole, indole, and pyrrole: Watanabe, M.; Nish-
iyama, M.; Yamamoto, T.; Koie, Y. Tetrahedron Lett. 2000, 41, 481.
J . Org. Chem, Vol. 69, No. 18, 2004 6011

Meyers et al.
TABLE 1. P d -Ca ta lyzed Am in a tion of Ar yl Iod id es w ith
zonitrile with morpholine and ethyl 3-iodobenzoate with
N-methylaniline in a limited reaction time. When 5 equiv
of cesium carbonate was used in the former coupling 50%
of 3-(morpholin-4-yl)benzonitrile was obtained after re-
fluxing for only 4 h, whereas the use of 2 equiv of the
same base gave a lower isolated yield of 35% in the same
reaction time. Similarly, the coupling of ethyl 3-iodoben-
zoate with N-methylaniline gave 79% of ethyl 3-[methyl-
5
equ iv of Cesiu m Ca r bon a te Typ e 1
(phenyl)amino]benzoate in 3 h of reflux with use of 5
equiv of cesium carbonate, whereas the use of 2 equiv of
the same base in 3 h yielded only 51% reaction product.
Interestingly, in 1997 and 2000 a Pd-BINAP catalyst,
using only a small excess of Cs
2 3
CO (1.4 equiv), has
already been described as a general system for the
1
4
Buchwald-Hartwig amination of aryl bromides. In
contrast, aryl iodides have hitherto only been successfully
aminated with use of a Pd-BINAP catalyst in combina-
8
tion with a strong base (NaOtBu). Clearly, the large
excess of carbonate base makes the difference. Conse-
quently, a further investigation of this new base effect
was our second goal in order to understand its nature.
The observation that the use of 5 equiv in comparison
with 2 equiv of cesium carbonate significantly speeds up
the coupling reactions of amines with aryl iodides is quite
odd since also 2 equiv of this base scarecely dissolve in
solvents commenly used in Buchwald-Hartwig amina-
tions.14 The well-known low solubility of cesium carbon-
ate in toluene and the obtained experimental evidence
that larger excesses increase the reaction rates suggest
that an interphase mechanism is operating. There seems
to be a direct solid-liquid phase interaction between the
solid carbonate and the five-coordinated Pd(II)-amine
intermediate in the toluene phase. If the deprotonation
of such a Pd(II)-amine intermediate is the rate-limiting
step of the catalytic cycle, one can indeed imagine that
by supplying a larger excess of base a substantial effect
on the reaction rate of the coupling will be observed. Such
an interphase mechanism has hitherto never been taken
into consideration in Buchwald-Hartwig aminations and
might be a generally useful parameter for the further
optimization of aminations with a rate-limiting depro-
tonation once a catalyst giving a significant conversion
of starting material has been selected.15 To gain further
evidence for an interphase process we investigated the
influence of the shape and size of the cesium carbonate
particles on the coupling rate. We selected two batches
a
All reactions were run overnight (17 h); reaction times were
not minimized; yields refer to the average of several runs; x mol
of Pd(OAc)2, x mol % of BINAP, aryl iodide (1.5 mmol), amine
%
(
14) For Pd-BINAP-catalyzed aminations of aryl bromides with a
(
(
1.8 mmol), Cs2CO3 “type 1” (7.5 mmol), toluene (15 mL), oil bath
120 °C). 0.5x mol % of Pd2(dba)3, 2x mol % of XANTPHOS or
small excess of Cs CO (1.4 equiv) as base see: (a) Wolfe, J . P.;
2
3
b
Buchwald, S. L. Tetrahedron Lett. 1997, 38, 6359. (b) Wolfe, J . P.;
Buchwald, S. L. J . Org. Chem. 2000, 65, 1144. In the description of
the general experimental procedure of both manuscripts it is indicated
DCHPDMAB, aryl iodide (1.5 mmol), amine (1.8 mmol), Cs2CO3
1.4 equiv), mixture of dioxane/NEt3 (2:1) or dioxane/tBuOH (2:1
(
c
2 3
that the Cs CO used was standardly ground prior to its use. The
or 1:1), ∆. DCHPDMAB was used as the ligand for the Pd
catalyst. XANTPHOS was used as the ligand for the Pd catalyst.
authors gave no specific reason for this grinding and no effect on the
reaction rate was reported. The supplier and quality of the base were
not specified.
d
(
15) Interestingly, Hartwig and co-workers investigated the effect
of phase-transfer catalysts, in palladium-catalyzed aminations of aryl
bromides and chlorides using Pd[P(tBu) as the catalyst, to increase
the solubility of base in toluene. The phase-transfer catalyst should
assist in the rate of deprotonation of the intermediate Pd(II)-amine
complex. Potassium hydroxide in combination with water and a
catalytic amount of cetyltrimethylammonium bromide was found to
be superior: Kuwano, R.; Utsunomiya, M.; Hartwig, J . F. J . Org. Chem.
2002, 67, 6479. In 2003, Buchwald reported that phase-transfer
catalysts had no effect when solid potassium hydroxide without added
very well with a low catalyst loading (Table 1, entries 5,
3 2
]
6
, and 7; see also Table 1, entry 2). Table 1 indicates that
the general profile of our catalytic system is equal to the
Pd-DCPDMAB (aliphatic amines) and Pd-XANTPHOS
(anilines) system reported by Buchwald. The use of a
large excess of base (5 equiv of cesium carbonate), when
Pd-BINAP is used, is essential to obtain sufficiently high
reaction rates in the coupling of amines with aryl iodides
as exemplified by the study of the coupling of 3-iodoben-
water was used with Pd(OAc)
DTBPB) as precatalyst for the amination of aryl chlorides in toluene:
Zim, D.; Buchwald, S. L. Org. Lett. 2003, 5, 2413.
2
/2-(di-tert-butylphosphanyl)biphenyl
(
6
012 J . Org. Chem., Vol. 69, No. 18, 2004

“Base Effect” in Pd-Catalyzed Amination of Aryl Iodides
F IGURE 1. Study of the effect of the structure (particle size and shape) of the base on the coupling rate of 2-chloro-3-iodopyridine
with p-toluidine.
of cesium carbonate from different suppliers and inves-
tigated their structure with electron microscopy (SEM)
1 for this amination reaction, using a 2 mol % loading of
Pd-BINAP catalyst. Two equivalents of cesium carbon-
ate were used for both reactions, only the type differed.
Figure 1 clearly shows that cesium carbonate type 2 gives
the fastest reaction. The observed effect is quite large
since the reaction rate in the linear area shown increases
with a factor 5.5 changing from cesium carbonate type 1
to type 2. This observed difference in rate is in agreement
with the isolated yields after a limited reaction time of 8
h since the coupling of 2-chloro-3-iodopyridine with
p-toluidine yielded 91% of 2-chloro-3-(4-methylpheny-
lamino)pyridine with use of 2 equiv of cesium carbonate
type 2 while the same excess of type 1 gave an incomplete
reaction and an isolated yield of only 68% in the same
reaction time. We decided to follow also the coupling of
2-chloro-3-iodopyridine with an aliphatic amine using an
HPLC-UV system. The intermediate Pd(II)-amine com-
plex containing an aliphatic amine is expected to be less
acidic, which must be reflected in a decrease of the
observed reaction rate. In Figure 2 the obtained result
for the amination of 2-chloro-3-iodopyridine with benzyl-
amine using a 2 mol % loading of Pd-BINAP catalyst
and 2 equiv of cesium carbonate type 1 is shown. When
we compare the rate of this coupling with that previously
discussed for the amination with p-toluidine using the
same type and excess of base, a rate decrease is indeed
observed (factor 1.5). We decided to double the catalyst
loading from 2 mol % to 4 mol % to speed up the coupling
rate since after 3.5 h reaction time the amination of
2-chloro-3-iodopyridine with benzylamine yielded less
(
see Supporting Information). The electron microscopic
image of the cesium carbonate of Acros (99.5%) (“type
”), which we used in all the experiments summarized
1
in Table 1, revealed uniformely bulbed particles with a
smooth surface and narrow size distribution while the
image of the cesium carbonate of Aldrich (99%) (“type
2
”) showed rough irregularly shaped particles with a
broad size distribution (containing many small par-
ticles).16 If an interphase mechanism is indeed involved
in the rate-limiting step one expects that running two
identical reactions only differing in the structure of the
cesium carbonate particles used should give a different
reaction rate. To get a good idea about the magnitude of
this particle effect on the reaction rate, we decided to
select a specific amination reaction and followed the
appearance rate of reaction product with an HPLC-UV
system. We chose the coupling of 2-chloro-3-iodopyridine
with p-toluidine since our first discovery of the base effect
in the amination of aryl iodides was done with this
1
0
specific substrate combination. The percentage of reac-
tion product is plotted versus the reaction time in Figure
(
16) Unfortunately, attempts to obtain the specific surface area (m2/
g) of the cesium carbonate of Acros (99.5%) (type 1) and Aldrich (99%)
type 2) via classical adsorption experiments with liquid N were
(
2
unsuccessful since the contact surface area was too small: Rouquerol,
F.; Rouquerol, J .; Sing, K. Adsorption by powders & porous solids;
Academic Press: London, UK, 1999. Consequently, only the difference
in particle shape and size of the cesium carbonate particles, observed
by recording SEM images, could be taken into account.
J . Org. Chem, Vol. 69, No. 18, 2004 6013

Meyers et al.
F IGURE 2. Study of the effect of the structure (particle size and shape) of the base on the coupling rate of 2-chloro-3-iodopyridine
with benzylamine.
than 15% conversion of starting material to reaction
product. As expected, doubling the catalyst loading
significantly improved the reaction rate. When we sub-
sequently changed the type of cesium carbonate from 2
equiv of type 1 to 2 equiv of type 2 the rate in the linear
area shown increased with a factor 1.5. This rate increase
is less pronounced for the coupling of benzylamine than
for the reaction with p-toluidine and can be rationalized
Since cesium carbonate is very hygroscopic and since
we always weighed it in the air it most probably took up
some water during handling. Thermogravimetric analysis
(TGA) of the batches cesium carbonate type 1 and type
2
used revealed a different water content. The weight
loss on heating to 188 °C was 0.38% for type 1 and 1.63%
for type 2 base. When type 1 and type 2 bases were
exposed to the air for 1 h the weight loss on heating at
the same temperature was 1.65% for the former and
by taking into account that the pK
benzylamine complex is much higher than the pK
a
value of the Pd(II)-
value
a
3
.15% for the latter. These data confirm that the base
of the Pd(II)-p-toluidine complex. The observed differ-
ence for the coupling of 2-chloro-3-iodopyridine with
benzylamine using type 1 and type 2 base respectively
is also reflected in the isolated yields after a limited
reaction time of 8 h with 4 mol % of Pd-BINAP. The
amination with 2 equiv of type 1 yielded 52% of 3-ben-
zylamino-2-chloropyridine while the same excess of type
has indeed taken up water from the air and therefore
we investigated the influence of the amount of water on
the amination rate. For the amination of 2-chloro-3-
iodopyridine with p-toluidine we compared the yields of
the reactions using 2 equiv of cesium carbonate of both
types that had been exposed to the air for 1 h with those
previously obtained under standard conditions (see previ-
ous paragraph). These standard conditions involve the
use of cesium carbonate rapidly weighed in air as all
other reagents. For the coupling of cesium carbonate type
2
carbonate base gave a higher yield (76%) in the same
1
7
reaction time. Not surprisingly, on the basis of the rate
difference observed in the linear area (factor 2.2) between
the coupling of p-toluidine (2 mol % Pd-BINAP) and
benzylamine (4 mol % of Pd-BINAP) with 2-chloro-3-
iodopyridine using 2 equiv of cesium carbonate type 2,
in this case both couplings gave an incomplete reac-
tion.
1
a yield of 64% and an incomplete reaction was obtained
in 8 h of reflux, which is similar to the result obtained
under standard conditions (68%) in the same reaction
time. Similarly, the use of base type 2 exposed to the air
for 1 h yielded 85% of 2-chloro-3-(4-methylphenylamino)-
pyridine and 91% under standard conditions both per-
formed in a reaction time of 8 h. These results indicate
that the water content, in the studied range, does not
have a significant influence on the coupling rate.
(
17) The regioselective amination of 2-chloro-3-iodopyridine with
aliphatic amines is not restricted to primary amines. Amination of
-chloro-3-iodopyridine with morpholine, under the standard condi-
tions, using 4 mol % of Pd-BINAP and 5 equiv of Cs CO of Acros
99.5%) (“type 1”) gave 84% isolated yield of 2-chloro-3-(morpholin-4-
yl)pyridine in a reaction time of 17 h.
2
2
3
(
6
014 J . Org. Chem., Vol. 69, No. 18, 2004

“Base Effect” in Pd-Catalyzed Amination of Aryl Iodides
TABLE 2. P d -Ca ta lyzed Am in a tion of Ar yl Iod id es w ith Cesiu m Ca r bon a te Typ e 2 a n d Com p a r ison w ith th e Yield s
Obta in ed w ith 5 equ iv of Cesiu m Ca r bon a te Typ e 1
a
All reactions were run overnight (17 h); reaction times were not minimized; yields refer to the average of several runs; x mol % of
Pd(OAc)2, x mol % of BINAP, aryl iodide (1.5 mmol), amine (1.8 mmol), Cs2CO3, toluene (15 mL), oil bath (120 °C).
With the knowledge of the importance of the particle
size and shape, resulting from the kinetic studies, we
reinvestigated all the coupling reactions summarized in
Table 1 using base type 2 (Table 2). The same loading of
catalyst that gave optimal results for the aminations with
no general trend could be observed. Sometimes they give
low yields and in these cases large amounts of starting
material remained after an overnight reflux (Table 2,
entries 1a-5a). However, for the coupling of morpholine
with ethyl 3-iodobenzoate (Table 2, entry 6) and mor-
pholine, benzylamine, and diethylamine with 3-iodoben-
zonitrile (Table 2, entries 7, 8, and 9) the aminations
worked very well with only 2 equiv of base type 2. To
rationalize this, one has to realize that the acidity of the
hydrogen atom on the complexed nitrogen is determined
by the structure of the whole Pd(II)-amine complex: the
5
equiv of base type 1 was used. We started with only 2
equiv of cesium carbonate type 2 since we knew that this
type gives faster reactions. As can be seen in Table 2, in
general, only anilines can be coupled efficiently with aryl
iodides under these reaction conditions in an overnight
reaction (Table 2, entries 10-12). For aliphatic amines
J . Org. Chem, Vol. 69, No. 18, 2004 6015

Meyers et al.
under otherwise identical conditions (see higher). Ground
cesium carbonate yielded 85% of ethyl 3-[methyl(phenyl)-
amino]benzoate in 3 h of reflux while unground base
yielded only 51% in the same reaction time. This obser-
vation also supports an interphase deprotonation mech-
anism.
In conclusion, based on kinetic data, we have proven
for the first time that an interphase deprotonation of Pd-
(II)-amine complexes with weak bases is possible. Con-
trolling the structure (shape and size of the particles)
and/or molar excess of the base used can significantly
speed up Buchwald-Hartwig aminations of aryl halides
with rate-limiting deprotonation. In the last years, the
research in palladium-catalyzed aminations was mainly
focused on the development of new ligands that allow the
tackling of industrially relevant aryl chlorides by using
a weak base and a low catalyst loading. Taking into
account a possible base effect might further improve some
of these existing protocols. Our successful application of
the base effect for the intermolecular amination of aryl
iodides (using a standard second generation Pd-BINAP
catalyst in combination with weak cesium carbonate
base), which historically have been shown to be less
effective substrates, clearly demonstrates the power of
this base effect.
F IGURE 3. Proposed structure of the Pd(II)-amine complex
formed in the catalytic cycle.
ligand (kept constant in this study), the halide (kept
constant in this study), and the amine, as well as the
aryl group from the aryl iodide substrate used (Figure
4
b
3
).
The main factor determining the Br o¨ nsted acidity of
the complexed amine is the Lewis acidity of the Pd(II)
center in the oxidative addition complex to which the
amine (Lewis base) is coordinated. If an electron poor aryl
iodide is used as a substrate the Lewis acidity of the Pd-
(II) metal will increase and consequently also the acidity
of the hydrogen atom on the nitrogen atom that com-
plexes with the oxidative addition complex. Therefore
good results with aliphatic amines can be obtained by
using only 2 equiv of base type 2 depending on the specific
aliphatic amine used and the position and electron with-
drawing effect of the substituents located on the aryl
iodide. Using this model one can also understand why
all anilines react very well with use of only 2 equiv of
cesium carbonate type 2 since the aryl group of the ani-
line will significantly contribute to the Br o¨ nsted acidity,
Exp er im en ta l Section
Gen er a l P r oced u r e for th e P d -Ca ta lyzed Am in a tion s
of Ar yl Iod id es. A round-bottom flask was charged with Pd-
2
(OAc) , (()-BINAP, and toluene (5 mL). The mixture was
flushed with nitrogen for 10 min under magnetic stirring. In
another round-bottom flask aryl iodide (1.5 mmol), amine (1.8
mmol), and Cs CO were weighed. Then, the Pd(OAc) /BINAP
a
resulting in a lower pK value. As expected, doubling the
2
3
2
amount of cesium carbonate type 2 in the coupling reac-
tions with aliphatic amines, giving only poor yields and
no full conversions of aryl iodide (Table 2, entries 1a-
solution was added, and the flask was rinsed well with an
additional amount of toluene (10 mL). The resulting mixture
was flushed with nitrogen for 5 min under magnetic stirring
and subsequently heated in a preheated oil bath under
5
a), improved the yields to a level similar to that obtained
earlier with 5 equiv of base type 1 (compare entries 1a-
a with entries 1b-5b in Table 2). No attempts were
vigorous magnetic stirring (oil bath temperature 120 °C) (N
2
atmosphere). After reflux overnight the mixture was cooled
to room temperature. The solid material was filtered off and
washed well with CH Cl (200 mL). The filtrate was evapo-
5
made to search for the optimal amount of base type 2 for
these specific coupling reactions. The optimal excess for
these aminations will be situated between 2 and 4 equiv
depending on the acidity of the specific Pd(II)-amine
complex formed. Since the structure (particle size and
shape) of a specific type of cesium carbonate might differ
from batch to batch this optimal excess is also of no real
practical use. It has to be emphasized that the producers
do not guarantee that the structure of the cesium
carbonate particles of Acros (99.5%) and Aldrich (99%)
will be equal in different batches. Consequently, for all
the aminations published in this study cesium carbonate
from the same batches was used.
In principle, grinding should also have an effect on the
coupling rate if an interphase mechanism is operating.14
Unfortunately, simple grinding in a mortar cannot be
reproducibly performed and this is a serious problem
when larger quantities of base with the same particle
profile are required in order to be able to quantitatively
compare different reactions. To investigate the effect of
grinding we performed only one test experiment and
compared the yield of the amination of ethyl 3-iodoben-
zoate with N-methylaniline using 2 equiv of ground
cesium carbonate type 1 with the yield previously ob-
tained using 2 equiv of unground base of the same type
2
2
rated and the resulting crude product was purified by flash
column chromatography.
The characterization data of (2,5-dimethylphenyl)hexy-
lamine, ethyl 3-(1,4-dioxa-8-azaspiro[4.5]dec-8-yl)benzoate,
[
4-(dipropylamino)phenyl](phenyl)methanone, 4-(2-methox-
yphenyl)morpholine, ethyl 2-[(3-cyanophenyl)amino]benzoate,
ethyl 3-[methyl(phenyl)amino]benzoate, and ethyl 4-[methyl-
(
phenyl)amino]benzoate were identical with those reported in
6
the literature.
2 2
4-(1-Na p h th yl)m or p h olin e (1): Eluent CH Cl ; mp 82 °C;
δ (CDCl ) 8.22-8.17 (m, 1H, H-5 or H-8), 7.82-7.77 (m, 1H,
H 3
H-5 or H-8), 7.54 (br d, J ) 8.2 Hz, 1H, H-4), 7.48-7.41 (m,
2
7
δ
1
1
H, H-6,7), 7.37 (dd, J ) 8.2, 7.5 Hz, 1H, H-3), 7.04 (dd, J )
.5, 1.1 Hz, 1H, H-2), 3.86 (br t, 4H, H-2′), 3.18 (br t, 4H, H-3′);
C
3
(CDCl ) 149.4, 134.8, 128.8, 128.4, 125.8, 125.8, 125.3, 123.7,
23.3, 114.6, 67.4, 53.5; νmax (KBr) 2853, 1453, 1397, 1258,
113, 1011, 918, 801, 776; MS (ESI) 214, 170, 168, 167, 141;
+
HRMS (ESI) for C14
214.1226.
H
16NO [M + H ] calcd 214.1232, found
Eth yl 3-(m or p h olin -4-yl)ben zoa te (2): Eluent CH
EtOAc 95:5; δ (CDCl ) 7.58 (br s, 1H, H-2), 7.54 (d, J ) 7.6
Hz, 1H, H-6), 7.31 (dd, J ) 8.1, 7.6 Hz, 1H, H-5), 7.07 (d, J )
.1 Hz, 1H, H-4), 4.37 (q, J ) 7.2 Hz, 2H, CH ), 3.84 (br t, 4H,
H-2′), 3.19 (br t, 4H, H-3′), 1.38 (t, J ) 7.2 Hz, 3H, CH ); δ
CDCl ) 166.6, 151.1, 131.3, 128.9, 120.7, 119.7, 116.2, 66.6,
2 2
Cl /
H
3
8
2
3
C
(
3
60.7, 49.0, 14.2; νmax (liquid film) 1715, 1601, 1581, 1445, 1367,
1302, 1265, 1228, 1122, 959, 757; MS (ESI) 236, 208, 190, 164,
6
016 J . Org. Chem., Vol. 69, No. 18, 2004

“Base Effect” in Pd-Catalyzed Amination of Aryl Iodides
+
1
20; HRMS (ESI) for C13
found 236.1275.
-(Mor p h olin -4-yl)ben zon itr ile (3): Eluent CH
8:2; mp 105 °C; δ (CDCl ) 7.34 (dd, J ) 9.3, 7.6 Hz, 1H, H-5),
.13 (dt, J ) 7.8, 1.2 Hz, 1H, H-4 or H-6), 7.12-7.09 (m, 2H,
H
18NO
3
[M + H ] calcd 236.1287,
coefficient of the starting material and reaction product at the
used wavelength (254 nm), has been taken into account.
Ch r om a togr a p h ic Con d ition s. The analytical column
(Lichrospher RP-select B, C8, 5-µm particles, 120 mm × 4 mm)
was purchased from Merck (Darmstadt, Germany). For the
amination of 2-chloro-3-iodopyridine with p-toluidine the
HPLC analysis was executed with a gradient elution at 1 mL/
3
2
Cl /EtOAc
2
9
7
H
3
H-2, H-4 or H-6), 3.86 (br t, 4H, H-2′), 3.18 (br t, 4H, H-3′); δ
CDCl ) 151.1, 129.7, 122.5, 119.2, 119.0, 117.8, 112.8, 66.3,
3
C
(
4
9
C
8.2; νmax (KBr) 2226, 1600, 1573, 1293, 1270, 1244, 1116, 994,
min starting at 50:50 (v/v) NH
composition of the mobile phase was altered to 17:83 (v/v) NH -
OAc (0.1 M)-MeOH in 10 min. For the amination of 2-chloro-
3-iodopyridine with benzylamine the HPLC analysis was
executed with a gradient elution at 1 mL/min starting at 70:
4
OAc (0.1 M)-MeOH. The
60, 864, 776, 679; MS (ESI) 189, 145, 143; HRMS (ESI) for
4
+
11
H
3
13
N
2
O [M + H ] calcd 189.1028, found 189.1026.
-(Ben zyla m in o)ben zon itr ile (4): Eluent CH
0:30; mp 68-69 °C; δ (CDCl ) 7.35-7.23 (m, 5H, H-2′, H-3′,
H-4′), 7.15 (td, J ) 7.6, 0.9 Hz, H-5), 6.88 (dt, J ) 7.6, 1.2 Hz,
2 2
Cl /hexane
7
H
3
30 (v/v) NH OAc (0.1 M)-MeOH. The composition of the
4
1
6
1
1
1
H, H-4 or H-6), 6.77 (dd, J ) 2.4, 0.9 Hz, 1H, H-4 or H-6),
.74 (br s, 1H, H-2), 4.33 (d, J ) 5.2 Hz, 2H, CH ), 4.31 (br s,
(CDCl ) 148.3, 138.2, 129.8, 128.7, 127.5, 127.3,
mobile phase was altered to 38:62 (v/v) NH
MeOH in 20 min.
4
OAc (0.1 M)-
2
H, NH); δ
C
3
2-Ch lor o-3-(4-m eth ylp h en yla m in o)p yr id in e (6): Eluent
CH Cl /heptane/EtOAc 70:25:5; δ (CDCl ) 7.78 (dd, J ) 4.6,
20.7, 119.4, 117.2, 115.0, 112.8, 47.7; νmax (KBr) 3387, 2229,
2
2
H
3
601, 1583, 1531, 1340, 777; MS(ESI) 209, 91; HRMS (ESI)
1.7 Hz, 1H, H-6), 7.34 (dd, J ) 8.1, 1.7 Hz, 1H, H-4), 7.14 (d,
J ) 8.1 Hz, 2H, H-3′, H-5′), 7.03 (d, J ) 8.1 Hz, 2H, H-2′, H-6′),
7.00 (dd, J ) 8.1, 4.6 Hz, 1H, H-5), 6.08 (br s, 1H, NH), 2.32
+
for C14
H
13
N
2
[M + H ] calcd 209.1079, found 209.1084.
3
-(Dip r op yla m in o)ben zon itr ile (5): Eluent heptane/ether
0:10; δ (CDCl ) 7.22 (dd, J ) 9.4, 7.4 Hz, 1H, H-5), 6.85 (dt,
H, J ) 7.4, 1.2 Hz, H-4 or H-6), 6.81-6.77 (m, 2H, H-2, H-4
9
1
H
3
(s, 3H, CH
3
C 3
); δ (CDCl ) 138.5, 138.3, 138.0, 137.6, 133.7, 130.0,
122.9, 121.9, 120.4, 20.6; νmax (liquid film) 3405, 1589, 1516,
1486, 1322, 810; MS(ESI) 219, 184, 183, 168; HRMS (ESI) for
or H-6), 3.23 (t, J ) 7.7 Hz, 4H, H-1′, H-1′′), 1.59 (h, J ) 7.5
+
Hz, 4H, H-2′, H-2′′), 0.94 (t, J ) 7.5 Hz, 6H, H-3′, H-3′′); δ
C
C
12
H
12
N
2
Cl [M + H ] calcd 219.0689, found 219.0682.
(CDCl
3
) 148.2, 129.8, 119.9, 118.3, 115.7, 114.3, 112.9, 52.7,
2-Ch lor o-3-(ben zyla m in o)p yr id in e (7): Eluent CH
2
Cl
2
;
2
2
0.1, 11.3; MS (ESI) 203, 161, 119; νmax (liquid film) 2961, 2874,
δ
H
(CDCl
3
) 7.70 (dd, J ) 4.6, 1.7 Hz, 1H, H-6), 7.40-7.28 (m,
227, 1596, 1497, 1467, 1367, 1245, 1194, 1009, 772, 682; MS-
5H, H-2′, H-3′, H-4′, H-5′, H-6′), 7.01 (ddd, J ) 8.1, 4.6, 0.6
Hz, 1H, H-5), 6.82 (dd, J ) 8.1, 1.7 Hz, 1H, H-4), 4.84 (br s,
+
(ESI) 203, 161, 119; HRMS (ESI) for C13
H
19
N
2
[M + H ] calcd
2
03.1548, found 203.1540.
Gen er a l P r oced u r e for th e Kin etic Exp er im en ts. A
two-necked round-bottom flask was charged with Pd(OAc)
1H, NH), 4.38 (br d, J ) 5.7 Hz, 2H, CH
2
C 3
); δ (CDCl ) 140.6,
137.7, 137.1, 136.6, 128.8, 127.6, 127.1, 123.3, 117.8, 47.4); νmax
2
,
(liquid film) 3408, 1582, 1559, 1494, 1447, 1380, 1357, 1050,
(
()-BINAP, and toluene (5 mL). The mixture was flushed with
795, 732; MS(ESI) 219, 183, 128, 91; HRMS (ESI) for C12
Cl [M + H ] calcd 219.0689, found 219.0679.
H
12
N
2
-
+
nitrogen for 10 min under magnetic stirring. In another round-
bottom flask 2-chloro-3-iodopyridine (1.5 mmol, 0.356 g), amine
(
2 3 2
1.8 mmol), and Cs CO were weighed. Then, the Pd(OAc) /
Ack n ow led gm en t. Prof. Dr. B. U. W. Maes thanks
the University of Antwerp (RAFO-RUCA) and the Fund
for Scientific Research-Flanders (FWO-Vlaanderen)
for financial support. The authors wish to thank the
technical staff of the research group (J . Aerts, G.
Rombouts, J . Schrooten, W. Van Lierde, and J . Ver-
reydt) for their assistance, Prof. Dr. P. Cool for perform-
ing N2 adsorption experiments, EMAT (Prof. Dr. G. Van
Tendeloo and J . Eysermans) for recording SEM images,
and Prof. Dr. H. Desseyn and J . J anssens for TGA
measurements.
BINAP solution was added, and the flask was rinsed well with
an additional amount of toluene (10 mL). The resulting
mixture was flushed with nitrogen under magnetic stirring.
At that time the first sample was taken. After flushing with
nitrogen for 5 min the second sample was taken from the
reaction mixture. Next the reaction mixture was heated in a
preheated oil bath under vigorous magnetic stirring (oil bath
temperature 120 °C) (N
sample was taken every 5 min, during the second hour every
5 min, and then every 30 min. The samples were taken with
2
atmosphere). During the first hour a
1
a GC-syringe. Herewith 50 µL of fluid was taken from the flask
via a septum, brought into a 25 mL volumetric flask, and
2
immediately diluted with a 50:50 H O/MeOH solution (HPLC
Su p p or tin g In for m a tion Ava ila ble: General experimen-
grade solvents). The solutions were filtered (0.2 mm; Nylon)
and subsequently analyzed with HPLC-UV. The percentage
reaction product is determined by dividing the HPLC peak
area of the reaction product by the sum of the peak areas of
the starting material and reaction product after a correction
factor on the peak areas, based on the difference in extinction
tal considerations, SEM images of cesium carbonate “type 1”,
1
“
type 2”, and ground “type 1” as well as H NMR spectra for
compounds 1-7. This material is available free of charge via
the Internet at http://pubs.acs.org.
J O049774E
J . Org. Chem, Vol. 69, No. 18, 2004 6017