Comparison of the Suzuki cross-coupling reactions of 4,7-dichloroquinoline and 7-chloro-4-iodoquinoline with arylboronic acids using phosphine-free palladium catalysis in water

Article abstract of DOI:10.1139/v03-187

4,7-Dichloroquinoline (1a) and 7-chloro-4-iodoquinoline (1b) undergo Suzuki cross-coupling reactions with arylboronic acids catalyzed by phosphine-free palladium acetate in boiling water. Using phenylboronic acid (2), the reaction of 1a provides 7-chloro-4-phenylquinoline (3) (78%) together with diphenylquinoline (4) (12%), while 1b reacts in a much more regioselective fashion and provides 3 in 98% isolated yield. Although 1b undergoes a more regio-selective Suzuki reaction than 1a, additional important observations are that the overall reaction of 1b with 2 is three times slower than 1a and that the reaction occurs in the absence of tetrabutylammonium bromide. Using optimized reaction conditions, a variety of aryl and vinylboronic acids undergo regioselective Suzuki cross-coupling with 1b to provide the products 7, 10, and 11 in good to excellent yield.

Full text of DOI:10.1139/v03-187

206
Comparison of the Suzuki cross-coupling
reactions of 4,7-dichloroquinoline and 7-chloro-4-
iodoquinoline with arylboronic acids using
phosphine-free palladium catalysis in water1
Richard W. Friesen and Laird A. Trimble
Abstract: 4,7-Dichloroquinoline (1a) and 7-chloro-4-iodoquinoline (1b) undergo Suzuki cross-coupling reactions with
arylboronic acids catalyzed by phosphine-free palladium acetate in boiling water. Using phenylboronic acid (2), the re-
action of 1a provides 7-chloro-4-phenylquinoline (3) (78%) together with diphenylquinoline (4) (12%), while 1b reacts
in a much more regioselective fashion and provides 3 in 98% isolated yield. Although 1b undergoes a more regio-
selective Suzuki reaction than 1a, additional important observations are that the overall reaction of 1b with 2 is three
times slower than 1a and that the reaction occurs in the absence of tetrabutylammonium bromide. Using optimized re-
action conditions, a variety of aryl and vinylboronic acids undergo regioselective Suzuki cross-coupling with 1b to pro-
vide the products 7, 10, and 11 in good to excellent yield.
Key words: palladium, cross-coupling, regioselectivity, quinolines, boronic acids.
Résumé : La 4,7-dichloroquinoléine (1a) et la 7-chloro-4-iodoquinoléine (1b) subissent des réactions de couplage de
Suzuki avec les acides arylboroniques qui sont catalysées par de l’acétate de palladium sans phosphine dans de l’eau
bouillante. Avec l’acide phénylboronique (2), la réaction du composé 1a fournit la 7-chloro-4-phénylquinoléine (3)
(78%) aux côtés de diphénylquinoléine (4) (12%) alors que le composé 1b réagit d’une façon beaucoup plus régiosé-
lective et fournit le composé 3 avec un rendement de 98% en produit isolé. Même si le composé 1b donne lieu à une
réaction de Suzuki plus régiosélective que celle du composé 1a, des observations importantes additionnelles sont que la
vitesse globale du composé 1b avec 2 est trois fois plus lente que celle du composé 1a et que la réaction se produit en
l’absence de bromure de tétrabutylammonium. Utilisant des conditions réactionnelles optimisées, une variété d’acides
aryl- et vinylboroniques donnent lieu à des réactions régiosélectives de couplage croisé de Suzuki avec le composé 1b
et elles fournissent les produits 7, 10 et 11 avec des rendements allant de bons à excellents.
Mots clés : palladium, couplage croisé, régiosélectivité, quinoléines, acides boroniques.
[Traduit par la Rédaction] Friesen and Trimble 214
Introduction
materials are often more readily available than the corre-
sponding bromides or iodides. We have recently described
(12) the synthesis of 2,4,7-trisubstituted quinoline 5-lip-
oxygenase inhibitors using, as one of the key steps, the
regioselective Suzuki cross-coupling reaction of 4,7-
dichloroquinoline (1a) with arylboronic acids. As originally
described by Mitchell and co-workers (6), the reaction of 1a
with 1.1 equiv. of phenylboronic acid (2) catalyzed by
Pd(Ph3P)4 (3%) in refluxing aqueous ethanol – benzene and
in the presence of Na2CO3 provided, after 48 h, 7-chloro-4-
phenylquinoline (3) in 82% isolated yield (Scheme 1). It was
not mentioned whether the bis-addition product 4,7-
diphenylquinoline (4) or the regioisomeric mono-addition
product 4-chloro-7-phenylquinoline (5) were observed in
this reaction.
The palladium-catalyzed cross-coupling of aryl halides
with boronic acids, commonly referred to as the Suzuki re-
action (1–5), is an exceptionally versatile and highly utilized
reaction for the construction of carbon-carbon bonds and, in
particular, for the preparation of heterobiaryls. The use of
heteroaryl chlorides as one of the coupling partners has seen
increasing utility (6–11) and is attractive since the starting
Received 7 July 2003. Published on the NRC Research Press
Web site at http://canjchem.nrc.ca on 20 January 2004.
This paper is dedicated to Professor Ed Piers for his many
contributions to synthetic organic chemistry and to the
chemistry community in Canada.
We have repeated this Suzuki cross-coupling reaction be-
tween 1a and 2 in connection with our medicinal chemistry
program. Dichloride 1a was consumed in 6 h, and in addi-
tion to 3 (81%), a significant amount of bis-addition product
4 (10%) was obtained. Previous reports have demonstrated
that 4,7-dichloroquinoline (1a) undergoes regioselective
palladium-catalyzed Stille (13) and alkoxycarbonylation (14)
R.W. Friesen2 and L.A. Trimble. Department of Medicinal
Chemistry, Merck Frosst Centre for Therapeutic Research,
16711 Trans Canada Hwy., Kirkland, QC H9H 3L1, Canada.
1This article is part of a Special Issue dedicated to Professor
Ed Piers.
2Corresponding author (e-mail: rick_friesen@merck.com).
Can. J. Chem. 82: 206–214 (2004)
doi: 10.1139/V03-187
© 2004 NRC Canada

Friesen and Trimble
207
Scheme 1.
1
reactions at the C4 position. However, in the latter reaction,
alkoxycarbonylation at both C4 and C7 produces varying
amounts of diester, depending on the reaction conditions
used. Therefore, the observation of 4 in the reaction of 1a
and 2 was not completely unexpected. Since we required an
efficient cross-coupling procedure that would bring about
regioselective monofunctionalization of a quinoline (6, 7,
13–17), such as 1a, without consuming more elaborate
boronic acids by the production of bis-addition products, we
undertook a more detailed study of the reactions of 1a and
the corresponding iodide 1b to optimize the formation of 3.
The results of this study are presented here.
Table 1. 7-Chloro-4-phenylquinoline (3): H NMR spectrum and
nOe experiments.
Protona
Assignmentb
nOe observedc
H2
8.94 (d, J = 4.4 Hz)
7.34 (d, J = 4.4 Hz)
7.86 (d, J = 9.2 Hz)
7.45 (dd, J = 2.1 9.2 Hz)
8.17 (d, J = 2.1 Hz)
7.48 (m)
H3
H3
H2, H9
H6, H9
H5
H6
H8
None
H9
H10, H11
7.50–7.56 (m)
Note: Spectra obtained in CDCl3 at 600 MHz.
aSee structure 3 for proton labeling.
Results and discussion
bResonances reported in ppm (δ) with multiplicity and coupling con-
stants in parentheses.
Structural assignment
cIn an nOe experiment, irradiation of the resonance assigned to the pro-
ton in column 1 results in an enhancement in intensity of the listed proton
resonances.
In their initial publication (6), Mitchell and co-workers as-
signed regioisomer 3 as the product of the Suzuki cross-
coupling reaction between 1a and 2. This assignment was
made without any structural verification, although one might
reasonably expect 3 to be produced as the major regioisomer
based on precedence with other reactions of 1a (13, 14, 18).
However, we felt that a rigorous structural assignment was
warranted since the production of the bis-addition adduct 4
are typically modified in the Suzuki cross-coupling reaction
(1–5), including, for example, the palladium catalyst
(PdCl2(Ph3P)2, Pd2(dba)3 (with and without P(t-Bu)3),
Pd(dppf)Cl2), solvents (DMF, THF, EtOH), aqueous base
(Na2CO3, Cs2CO3), temperature, and the use of additives
(KF). None of these changes substantially altered the yields
of 3 and 4 that were observed upon reaction of 1a with 2.
Several reports over the last decade have described the suc-
cessful cross-coupling of arylboronic acids and aryl bro-
mides, iodides, or triflates using phosphine-free palladium,
such as Pd(OAc)2, in aqueous–organic co-solvents (19, 20)
or in water alone (21, 22). Employing reaction conditions
optimized by Leadbeater and Marco (22), the reaction of 1a
with 1.1 equiv. of 2 in the presence of 1% Pd(OAc)2 and
0.25 equiv. of nBu4NBr in refluxing water (oil bath at
150 °C) under nitrogen atmosphere provided 3 (78%) and 4
(12%) upon consumption of 1a. Although this reaction dif-
fered little from all the others we had surveyed in terms of
the yields of the two products, it was complete within
30 min (Table 2, entry 1), while the majority of the others,
including the original reaction described by Mitchell, re-
quired extended reaction times (6 h to 2 days). Since the ra-
pidity of the reaction under these conditions was perceived
as an advantage, we chose to further characterize the cross-
coupling reaction of 1a using phosphine-free palladium in
water.
1
reveals that reactivity at both C4 and C7 is possible. H
NMR experiments (CDCl3, 600 MHz) indicated that 3 is in-
deed the predominant mono-substituted regioisomer formed.
Thus, proton resonance assignments (see Table 1) were
made using a 2D-COSY experiment.3 In a subsequent series
of nuclear Overhauser enhancement (nOe) experiments,
clear enhancements in resonance intensity were observed for
H9 upon irradiation of H3 and H5, while irradiation of H8
did not result in nOe of any other resonance.
The results of the two experiments are informative. First,
when the reaction between 1a and 2 was stopped prior to
consumption of 1a (at 15 min), we observed the presence
Reaction optimization studies
To optimize the reaction of 1a for mono-substitution at
C4, we rapidly surveyed many of the reaction conditions that
3 This spectrum is available as supplementary material. Supplementary data may be purchased from the Depository of Unpublished Data,
Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml
for information on ordering electronically).
© 2004 NRC Canada

208
Can. J. Chem. Vol. 82, 2004
Table 2. Pd(OAc)2-catalyzed coupling reactions of 1a and 1b with phenylboronic acid 2.
Entry
Substrate
equiv. PhB(OH)2 (2)
Additive
Time (min)
Yield 3 (%)
Yield 4 (%)
1
2
3
4
5
30
60
1a
1a
1b
1b
1a
1.1
2.2
1.1
1.1
1.1
nBu4NBr
nBu4NBr
nBu4NBr
nBu4NBr
nBu4NI
78
72
98
85
73
12
24
0.2
12
13
Tracea
10
90
180
90
6
7
8
9b
10c
1b
1a
1b
1b
1b
1.1
1.1
1.02
1.02
1.02
None
None
None
None
None
30
10
45
150
420
97
83
96
96
95
0.7
0.3
Tracea
Note: Reaction conditions: 1.0 equiv. 1, 1.02–2.2 equiv. 2, 1% Pd(OAc)2, 0.25 equiv. additive, 1.0 equiv. Na2CO3, H2O
(1 mL/mmol 1), bath temperature 150 °C.
aObserved by TLC but not isolated.
b0.1% Pd(OAc)2.
c0.01% Pd(OAc)2.
of minor amounts of the regioisomeric mono-addition prod-
with 1b (compared with 1a) occurs at an overall reaction
rate that is three times slower than that of 1a (1.5 h vs.
30 min).
uct 4-chloro-7-phenylquinoline 5 (isolated as an inseparable
1
mixture with 3 and identified by H NMR spectroscopy).
Second, using excess phenylboronic acid (2.2 equiv.), the
yield of 4 could be increased to 24% (Table 2, entry 2).
These observations suggest that although the aryl chloride at
C4 is more reactive than the C7 aryl chloride, the latter is
sufficiently reactive that the bis-addition product 4 can be
derived from the initial reaction at either C4 or C7, followed
by a second coupling, which is competitive with the first.
Owing to the inherent reactivity of the two chlorides, it was
expected that it would be difficult to identify reaction condi-
tions for 1a that would eliminate cross-coupling at C7 and
provide 3 exclusively.
Thus, we turned our attention to the reactions of 7-chloro-
4-iodoquinoline (1b) (15, 18). Since the order of reactivity
of aryl halides in the Suzuki cross-coupling reaction is gen-
erally accepted to be I > Br >> Cl (1–5), it was thought that
introduction of a C4-iodide might enhance the inherent C4
regioselectivity by virtue of the additional chemoselectivity
that would be imparted by the more reactive iodide.
A subsequent reaction of 1b with 1.1 equiv. of 2 under the
conditions described above provided 3 (98%) and 4 (0.2%)
upon consumption of 1b at 1.5 h (Table 2, entry 3). How-
ever, if the reaction was allowed to progress for 3 h, the
yield of 4 increased to 12% at the expense of 3 (85%) (Ta-
ble 2, entry 4). These observations are consistent with the re-
sults shown in Fig. 2.
Subjection of 1b and 2 to the reaction conditions origi-
nally described by Mitchell also provided 3 (93%) as the
sole product upon consumption of 1b (24 h), suggesting that
the regioselectivity observed in the Suzuki reactions of 1b is
not limited to reactions catalyzed by phosphine-free palla-
dium in water but is inherent to the reactivity of 1b. Simi-
larly, the overall rate of reaction of 1b is approximately three
times slower than that of 1a under the Mitchell conditions
(see introduction), suggesting that the solely aqueous condi-
tions do not significantly impact the relative overall rates of
reaction of 1a and 1b.
The difference in the reactivity of the two dihaloquino-
lines 1a and 1b was clearly seen upon monitoring the time
course of their reactions with 2 (1.1 equiv. 2, 0.25 equiv.
nBu4NBr, 1% Pd(OAc)2, 1 equiv. Na2CO3, H2O, reflux) by
1H NMR spectroscopy. At suitable time points, aliquots of
each reaction mixture were removed and analyzed for 1a or
1b, 3, 4, and 5 by the integration of characteristic and well-
separated proton resonances (see Experimental section). The
data are plotted in Figs. 1 and 2 for the reactions of 1a and
1b, respectively. In Fig. 1, it is clearly seen that 3, 4, and 5
were formed throughout the time-course of the reaction.
When the starting material 1a was consumed at 30 min, the
reaction mixture was composed of 3 and 4 in a ratio of ap-
proximately 9:1. In contrast, the reaction of 1b with 2
(Fig. 2) revealed that 3 was the only product formed until
starting material was consumed (at 1.5 h), after which time
the bis-addition product 4 was produced at the expense of 3.
The regioisomeric mono-adduct 5 was not observed in this
latter reaction. Of note, and somewhat surprising, is the fact
that the dramatic improvement in regioselectivity observed
Thus, the cross-coupling reaction of the chloro-iodide 1b
catalyzed by phosphine-free Pd(OAc)2 in water is slower but
more regioselective than the dichloride 1a. The difference in
reactivity was further exemplified by the results obtained
from conducting a competition experiment between 1a and
1b for phenylboronic acid (2) (1.0 equiv. of each of the
quinolines and 1.02 equiv. of phenylboronic acid) using 1%
Pd(OAc)2 (Scheme 2). After 2 h, only the chloro-iodide 1b
was consumed to afford 3 (97%), and the dichloride 1a was
recovered in good yield (93%).
The differences that are observed in the reaction
regioselectivity and overall reaction rate of 1a and 1b are
readily rationalized. First, the regioselectivity of each reac-
tion is reflective of the rate of irreversible oxidative addition
of Pd(0) (generated from Pd(2+) and phenylboronic acid
(19, 23)) to the C4 and C7 aryl halide bonds. In the case of
dichloride 1a, although the reactivity of the two chlorides is
in the order C4—Cl > C7—Cl, the C7—Cl is of sufficient
reactivity that the bis-addition product 4 derives from the
initial reaction at either site, followed by a second coupling
© 2004 NRC Canada

Friesen and Trimble
209
Fig. 1. Time course of the Pd(OAc)2-catalyzed coupling reaction of 1a and 2. The mol% of 1a (ٗ), 3 (᭹), 4 (ꢀ), and 5 (᭛), as mea-
1
sured by H NMR spectroscopy (see Experimental section), are represented as a function of time.
that is competitive with the first. For chloro-iodide 1b, the
difference in reactivity between C4—I and C7—Cl is >
100:1 (Table 2, entry 3). This chemoselectivity dictates that
the bis-addition product 4 is produced only when the starting
material is essentially completely consumed (see Fig. 2).
Second, the more rapid overall reaction of 1a compared with
1b suggests that the oxidative addition step is not rate limit-
ing in the overall catalytic cycle of 1b. In contrast to those
Suzuki reactions, in which the oxidative addition step is rate
determining (leading to the typical ordering of overall
reactivity I > Br >> Cl (1–5)), here transmetalation of the
aryl-Pd-I species that is derived from 1b appears to be rate
determining. There are previous reports suggesting that
transmetalation, rather than oxidative addition, can be rate
determining in Suzuki cross-coupling reactions using both
phosphine-free (21) and phosphine-ligated palladium cata-
lysts (24, 25). Subtle differences in reaction conditions also
have a dramatic impact on the mechanism since both oxida-
tive addition (19) and transmetalation (21) steps have been
found to be rate determining when using ligandless palla-
dium catalysts. Thus, even though 1b undergoes a rapid, ir-
reversible, and chemoselective oxidative addition of Pd(0) to
the aryl iodide bond at C4 (which results in very high regio-
selectivity compared with 1a), the decreased rate of trans-
metalation results in the overall rate of reaction of chloro-
iodide 1b becoming slower than that of dichloride 1a (note
that the rate-determining step in the overall catalytic cycle of
1a may be the oxidative addition or transmetalation steps).
In addition, in spite of the fact that the overall rate of cross-
coupling of 1a is more rapid than 1b, the highly chemo-
selective, rapid, and irreversible oxidative addition of Pd(0)
to 1b rationalizes the selective consumption of 1b over that
of 1a in the competition experiment outlined in Scheme 2.
Similar observations were made by Fu and co-workers (25)
in their study of the Suzuki cross-coupling reactions cata-
lyzed by Pd2(dba)3–P(t-Bu)3. In these examples, the cross-
couplings of aryl bromides with phenylboronic acid pro-
ceeded more rapidly than the corresponding aryl iodides,
and in competition experiments analogous to those outlined
in Scheme 2, only the aryl iodides underwent cross-
coupling.
Another interesting observation was made when the cross-
coupling reaction of 1a and 2 was conducted in the presence
of 0.25 equiv. nBu4NI rather than nBu4NBr (compare Ta-
ble 2, entries 1 and 5). This modification had little effect on
the isolated yield of 3 (73%) and 4 (13%), but the reaction
now required 1.5 h to reach completion, similar to the rate
that was observed using 1b and nBu4NBr (Table 2, entry 3).
The presence of ammonium salts (21, 26) and the variation
of halides (27–29) have been shown to have profound effects
on the rate and mechanism of palladium-catalyzed processes
in general. Although the reason behind this rate change is
not completely understood, one can simplistically speculate
that the aryl-Pd-Cl intermediate (formed upon reaction of 1a
with Pd(0)) is converted to aryl-Pd-I in the presence of
nBu4NI (the bond strength for palladium halides is Pd—Cl <
Pd—Br < Pd—I (30, 31)). The reaction regioselectivity
would still be governed by the relative rate of oxidative addi-
© 2004 NRC Canada

210
Can. J. Chem. Vol. 82, 2004
Fig. 2. Time course of the Pd(OAc)2-catalyzed coupling reaction of 1b and 2. The mol% of 1b (᭝), 3 (᭹), and 4 (ꢀ), as measured by
1H NMR spectroscopy (see Experimental section), are represented as a function of time.
Scheme 2.
tion of Pd(0) to the aryl chlorides at C4 and (or) C7, but the
overall rate would depend on the transmetalation of
phenylboronic acid (2) to the aryl-Pd-I intermediate.
portant, since the reaction of 1b still proceeds to completion,
but monitoring for the disappearance of starting material 1b
becomes unnecessary. Recall that monitoring for the disap-
pearance of starting material 1b is necessary when using 1.1
equiv. of 2 since prolongation of the reaction after consump-
tion of starting material (from 1.5 to 3 h) results in a signifi-
cant increase in the formation of 4 (Table 2, entries 3 and 4).
Finally, although the amount of Pd(OAc)2 used can be de-
creased from 1% to 0.01%, this reduction in catalyst loading
occurs with a corresponding increase in the time required for
the reaction to reach completion (Table 2, entries 9 and 10).
Since 1b displays superb C4-regioselectivity and provides
3 in excellent yield upon reaction with 2, we set out to fur-
ther characterize the cross-coupling reactions of 1b using 2
as substrate (Table 2, entries 6–10). Surprisingly, when the
nBu4NBr additive was removed (Table 2, entry 6), 1b was
consumed within 30 min, a threefold increase in overall rate.
The yield of 3 (97%) was not affected. A similar observation
of threefold rate enhancement in the absence of nBu4NBr
was made, using 1a as substrate (Table 2, entry 7). These
observations are surprising since they are in contrast to liter-
ature data (21), which suggest that nBu4NBr is required for
successful Suzuki cross-couplings using phosphine-free
Pd(OAc)2 in water. It may further imply a possible change in
mechanism between the two reaction conditions (27, 28).
The yield of 3 (96%) was not affected by decreasing the
amount of phenylboronic acid (2) from 1.1 to 1.02 equiv.
(Table 2, entry 8). This seemingly minor modification is im-
Reaction scope
As illustrated in Scheme 3 and Table 3 (entries 1–14), re-
action of 1b using the optimized reaction conditions (A: 1.0
equiv. 1b, 1.02 equiv. 6 or 8 or 9, 1% Pd(OAc)2, 1.0 equiv.
Na2CO3, H2O (1 mL/mmol 1b), bath temperature 150 °C)
with a variety of boronic acids afforded the expected prod-
ucts 7, 10, and 11 in good to excellent yields (84%–97%)
without observation of the corresponding bis-addition prod-
© 2004 NRC Canada

Friesen and Trimble
Table 3. Palladium-catalyzed coupling reactions of 1b with arylboronic acids 6, 8, and 9.
211
Entry
Boronic acid
X
Reaction conditionsa
Time (min)
Product
Yield (%)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
45
15
150
150
35
120
150
30
60
90
60
120
90
2
—
2-Me
2-OMe-5-Me
2-Cl
3-F
3-NHAc
3-CN
4-F
4-OMe
4-Ac
3-F-4-OBn
2,6-diMe
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
3
96
93
91c
58 (85)
97
76 (84)
88c
96
97
89c
93
0
89c
95
43 (79)
94
6a
6b
6c
6d
6e
6f
6g
6h
6i
6j
6k
8
9
6c
6g
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
10
11
7c
7g
15
2880
1440
2-Cl
4-F
aReaction conditions. A: 1.0 equiv. 1b, 1.02 equiv. 6 or 8 or 9, 1% Pd(OAc)2, 1.0 equiv. Na2CO3, H2O (1 mL/mmol
1b), bath temperature 150 °C. B: 1.0 equiv. 1b, 1.1 equiv. 6, 3% Pd(Ph3P)4, 1.5 equiv. aq Na2CO3
(2 mol/L), benzene (3 mL/mmol 1b), ethanol (0.3 mL/mmol 1b), reflux.
bIsolated yield of purified compound. Yield in parentheses based on unrecovered starting material 1b.
cA trace of starting material observed by TLC.
or isolation of starting material 1b. In each of these exam-
ples, the black precipitate that signals the completion of the
reaction (and which is typically seen with those substrates
that do consume all of 1b) is not produced even if additional
arylboronic acid or palladium catalyst is added. For several
cases, the quantity of residual 1b is minor (not isolated), and
products 7 are obtained in yields over 88% (Table 3, entries
3, 7, 10, 13). In two cases (Table 3, entries 4 and 6), signifi-
cant amounts of 1b remain, although yields of 7 based on re-
covered starting material are still good. There does not
appear to be any correlation between the steric or electronic
nature of the substituent on the arylboronic acid and the ex-
tent of consumption of 1b. Since these reactions are hetero-
geneous and take place in the absence of organic co-solvent,
it is possible that the aqueous solubility or some other physi-
cal property of the boronic acid substrates may be responsi-
ble for these observed differences. However, this cannot be
the sole explanation. Reactions of two representative
arylboronic acids (6c and 6g) with 1b were conducted under
the biphasic Mitchell reaction conditions (Pd(Ph3P)4, aque-
ous ethanol – toluene) for comparison purposes, and the re-
activity patterns observed are similar to those observed
under the phosphine-free palladium conditions in water
(compare Table 3, entries 4 and 15, and 8 and 16, respec-
tively). In the case of 6g, 7g was obtained in good yield
(94%), but the reaction was complete after only 24 h. In the
case of 6c, a significant amount of starting material re-
mained after an extended reaction time (2 days).
Scheme 3.
ucts analogous to 4. Degassing of the water (19, 21) is not
required for a successful cross-coupling reaction.
The range of arylboronic acids 6 that serve as suitable
coupling partners include those bearing electron donating
and withdrawing substituents at the ortho-, meta-, and para-
positions. 3-Thienylboronic acid 8 (Table 3, entry 13) and
(E)-1-hexenylboronic acid 9 (Table 3, entry 14) are also suit-
able substrates. Typically, these reactions are complete
within 1 h, but there are some exceptions. No reaction is ob-
served after 2 h when using the sterically encumbered 2,6-
dimethylphenylboronic acid (6k) (Table 3, entry 12). In sev-
eral other examples, the reactions do not proceed to comple-
tion even after 2.5 h, as evidenced by the observation (TLC)
We have demonstrated that 7-chloro-4-iodoquinoline (1b)
undergoes highly regioselective Suzuki cross-coupling reac-
tions at C4 with a variety of boronic acids (aryl and vinyl),
catalyzed by phosphine-free palladium in water solvent in
the absence of nBu4NBr. The regioselectivity that is ob-
served is superior to that observed with the dichloride 1a,
but the overall reaction rate of 1b is three times slower than
1a (using phenylboronic acid). These results can be inter-
© 2004 NRC Canada

212
Can. J. Chem. Vol. 82, 2004
preted by considering that transmetalation of the intermedi-
ate aryl-Pd-I species, rather than the formation of this
species by the rapid and chemoselective oxidative addition
of Pd(0) to 1b, is the rate-determining step in the overall cat-
alytic cycle of 1b.
Typical procedure (Table 3)
A mixture of 1b (1.46 g, 5.04 mmol), phenylboronic acid
2 (97%, 647 mg, 1.02 equiv.), Pd(OAc)2 (11 mg, 0.01
equiv.), Na2CO3 (2.5 mL of a 2 mol/L solution, 1.0 equiv.),
and water (2.5 mL) was placed in a 50 mL round-bottom
flask under a nitrogen atmosphere and immersed in an oil
bath at 150 °C. After stirring for 45 min, the mixture was
cooled to room temperature; water (50 mL) was added, and
the mixture was extracted twice with ethyl acetate (50 mL).
The combined organics were washed with water (50 mL)
and brine (50 mL), dried (MgSO4), filtered, and concen-
trated. The residue was subjected to flash column chroma-
tography (SiO2, eluting with hexane–ether 3:1 volume
fraction) to provide 3 (1.16 g, 96%) as a white solid and 4
(9 mg, 0.7%) as a white solid (Table 2, entry 8 and Table 3,
entry 1).
Experimental section
All reagents were obtained from commercial suppliers ex-
cept 1b, which was prepared from 1a and aqueous HI (57%)
as described (15, 18). Melting points are uncorrected. FT-IR
spectra were recorded as solutions in chloroform unless
stated otherwise, and wavenumbers (ν) are reported in cm–1.
1H and 13C NMR spectra were recorded as solutions in ace-
tone-d6, unless stated otherwise, at field strengths of
400 MHz (1H) or 100 MHz (13C). Chemical shifts are re-
ported in parts per million (ppm) relative to the signals of re-
sidual protonated solvent as internal standard. 13C NMR
spectra were acquired with proton decoupling, and peak
multiplicities refer to JCF. Elemental analyses were per-
formed at the Laboratoire d’Analyse Élémentaire in the De-
partment of Chemistry, Université de Montréal, Montréal.
7-Chloro-4-phenylquinoline (3)
mp 73 to 74 °C (lit. (6) value mp 73.5–74.5 °C). 1H NMR
δ: 7.37 (d, 1H, J = 4.4 Hz), 7.47–7.58 (m, 6H), 7.86 (d, 1H,
J = 9.0 Hz), 8.11 (d, 1H, J = 2.1 Hz), 8.92 (d, 1H, J =
4.4 Hz). 13C NMR δ: 122.1, 125.4, 127.7, 127.9, 129.0,
129.1, 129.2, 129.9, 134.9, 137.8, 148.5, 149.6, 151.7.
Reactions for Figs. 1 and 2
4,7-Diphenylquinoline (4)
mp 128 to 129 °C. IR: 3023. H NMR δ: 7.44 (d, 1H, J =
A mixture of 1a (1.00 g, 5.04 mmol) (Fig. 1) or 1b
(1.46 g, 5.04 mmol) (Fig. 2), phenylboronic acid (2) (97%,
647 mg, 1.02 equiv.), nBu4NBr (406 mg, 0.25 equiv.),
Pd(OAc)2 (11 mg, 0.01 equiv.), Na2CO3 (2.5 mL of a 2
mol/L solution, 1.0 equiv.), and water (2.5 mL) was placed
in a 50 mL round-bottom flask under a nitrogen atmosphere
and immersed in an oil bath at 150 °C. After the time inter-
vals indicated in the figures, a small aliquot of the reaction
mixture was removed and dissolved in ethyl acetate. The
mixture was dried (MgSO4), filtered, and concentrated. The
1
4.3 Hz), 7.48 (m, 1H), 7.54–7.63 (m, 7H), 7.88 (m, 2H),
7.93 (dd, 1H, J = 1.9, 8.8 Hz), 8.02 (d, 1H, J = 8.8 Hz), 8.39
(d, 1H, J = 1.8 Hz), 8.99 (d, 1H, J = 4.4 Hz). 13C NMR δ:
122.1, 126.4, 126.8, 127.1, 128.1, 128.8, 129.3, 129.5,
129.9, 130.3, 138.8, 140.7, 142.4, 148.8, 150.1, 151.5. Anal.
calcd. for C21H15N: C 89.65, H 5.37, N 4.98; found: C
89.47, H 5.38, N 4.99.
The remainder of the examples cited in Table 3 were pre-
pared on the same scale in a manner similar to 3, with reac-
tion times as listed in the table. The eluting solvents used for
column chromatography are indicated below.
1
crude residue was analyzed by H NMR spectroscopy. One
or more of the following characteristic resonances were inte-
grated, and for each compound present, the mol% composi-
tion of the sample was calculated. 1a: 7.74 (d, 1H), 7.78 (dd,
1H), 8.30 (d, 1H), 8.89 (d, 1H); 1b: 7.75 (dd, 1H), 8.20 (d,
1H), 8.53 (d, 1H); 3: 7.37 (d, 1H), 7.86 (d, 1H), 8.11 (d,
1H), 8.92 (d, 1H); 4: 8.02 (d, 1H, J = 8.8 Hz), 8.39 (d, 1H),
8.99 (d, 1H); 5: 8.34 (d, 1H), 8.36 (d, 1H), 8.87 (d, 1H). The
resulting data were used to generate Figs. 1 and 2.
7-Chloro-4-(2-methylphenyl)quinoline (7a)
Chromatography (hexane–ether 4:1 volume fraction) pro-
vided 7a as a white solid (1.19 g, 93%), mp 86 to 87 °C. IR:
1
3017. H NMR δ: 2.07 (s, 3H), 7.26 (d, 1H, J = 7.5 Hz),
7.36–7.49 (m, 5H), 7.55 (dd, 1H, J = 2.1, 9.0 Hz), 8.15 (d,
1H, J = 2.1 Hz), 9.00 (d, 1H, J = 4.3 Hz). 13C NMR δ: 19.6,
122.4, 126.1, 126.4, 127.8, 128.1, 129.0, 129.1, 129.8,
130.7, 135.0, 136.1, 137.3, 148.5, 149.4, 151.9. Anal. calcd.
for C16H12ClN: C 75.74, H 4.77, N 5.52; found: C 75.67, H
4.78, N 5.55.
Competition experiment (Scheme 2)
A mixture of 1a (1.00 g, 5.04 mmol), 1b (1.46 g,
5.04 mmol), phenylboronic acid (2) (97%, 647 mg, 1.02
equiv.), nBu4NBr (406 mg, 0.25 equiv.), Pd(OAc)2 (11 mg,
0.01 equiv.), Na2CO3 (2.5 mL of a 2 mol/L solution, 1.0
equiv.), and water (2.5 mL) was placed in a 50 mL round-
bottom flask under a nitrogen atmosphere and immersed in
an oil bath at 150 °C. After stirring for 2 h, the mixture was
cooled to room temperature; water (50 mL) was added, and
the mixture was extracted twice with ethyl acetate (50 mL).
The combined organics were washed with water (50 mL)
and brine (50 mL), dried (MgSO4), filtered, and concen-
trated. The residue was subjected to flash column chroma-
tography (SiO2, eluting with hexane–ether 4:1 volume
fraction) to provide 1a (927 mg, 93%) as a white solid and 3
(1.17 g, 97%) as a white solid.
7-Chloro-4-(2-methoxy-5-methylphenyl)quinoline (7b)
Chromatography (hexane – ethyl acetate 6:1 volume frac-
tion) provided 7b as a white solid (1.30 g, 91%), mp 133 to
1
134 °C. IR: 2961, 1502, 1250. H NMR δ: 2.35 (s, 3H), 3.69
(s, 3H), 7.12 (m, 2H), 7.34 (dd, 1H, J = 1.8, 8.4 Hz), 7.39
(d, 1H, J = 4.4 Hz), 7.50 (dd, 1H, J = 2.2, 9.0 Hz), 7.62 (d,
1H, J = 9.0 Hz), 8.10 (d, 1H, J = 2.2 Hz), 8.95 (d, 1H, J =
4.4 Hz). 13C NMR δ: 20.4, 55.8, 112.2, 123.3, 126.7, 127.6,
129.1, 129.3, 130.8, 131.5, 132.2, 135.0, 146.9, 149.7,
152.2, 155.6. Anal. calcd. for C17H14ClNO: C 71.96, H 4.97,
N 4.94; found: C 71.96, H 5.04, N 4.86.
© 2004 NRC Canada

Friesen and Trimble
213
7-Chloro-4-(2-chlorophenyl)quinoline (7c)
7-Chloro-4-(4-methoxyphenyl)quinoline (7h)
Chromatography (hexane – ethyl acetate 6:1 volume frac-
tion) provided 1b (466 mg, 32%) and 7c as a white solid
(803 mg, 58%), mp 89 to 90 °C. IR: 2976. H NMR δ: 7.43
(d, 1H, J = 4.4 Hz), 7.44 (dd, 1H, J = 1.8, 7.3 Hz), 7.49–
7.59 (m, 4H), 7.64 (dd, 1H, J = 1.4, 7.8 Hz), 8.16 (d, 1H,
J = 1.8 Hz), 9.02 (d, 1H, J = 4.4 Hz). 13C NMR δ: 122.7,
125.6, 127.8, 128.0, 129.0, 130.2, 130.8, 131.9, 133.1,
135.1, 136.5, 146.0, 149.3, 151.8. Anal. calcd. for
C15H9Cl2N: C 65.72, H 3.31, N 5.11; found: C 65.67, H
3.24, N 5.08.
Chromatography (hexane – ethyl acetate 3:1 volume frac-
tion) provided 7h as a white solid (1.32 g, 97%), mp 90 to
91 °C. IR: 2963. 1H NMR δ: 3.92 (s, 3H), 7.15 (br d, 2H, J =
8.7 Hz), 7.42 (d, 1H, J = 4.4 Hz), 7.50 (br d, 2H, J =
8.7 Hz), 7.56 (dd, 1H, J = 2.2, 9.0 Hz), 7.98 (d, 1H, J =
9.0 Hz), 8.11 (d, 1H, J = 2.2 Hz), 8.93 (d, 1H, J = 4.4 Hz).
13C NMR δ: 55.3, 114.6, 122.0, 125.6, 127.5, 128.1, 128.9,
129.9, 131.2, 134.8, 148.3, 149.8, 151.8, 160.7. Anal. calcd.
for C16H12ClNO: C 71.25, H 4.48, N 5.19; found: C 71.16,
H 4.47, N 5.13.
1
7-Chloro-4-(3-fluorophenyl)quinoline (7d)
4-(4-Acetylphenyl)-7-chloroquinoline (7i)
Chromatography (hexane–ether 4:1 volume fraction) pro-
vided 7d as a white solid (1.26 g, 97%), mp 88 to 89 °C. IR:
2979. 1H NMR δ: 7.32–7.41 (m, 3H), 7.49 (d, 1H, J =
4.4 Hz), 7.59 (dd, 1H, J = 2.2, 9.0 Hz), 7.65 (br q, 1H, J =
7.8 Hz), 7.92 (d, 1H, J = 9.0 Hz), 8.13 (d, 1H, J = 2.1 Hz),
Chromatography (hexane – ethyl acetate 3:2 volume frac-
tion) provided 7i as a white solid (1.27 g, 89%), mp 143 to
1
144 °C. IR: 3015, 1688. H NMR δ: 2.70 (s, 3H), 7.52 (d,
1H, J = 4.4 Hz), 7.61 (dd, 1H, J = 2.2, 9.0 Hz), 7.73 (d, 2H,
J = 8.4 Hz), 7.91 (d, 1H, J = 9.0 Hz), 8.16 (d, 1H, J =
2.2 Hz), 8.21 (d, 2H, J = 8.4 Hz), 9.02 (d, 1H, J = 4.4 Hz).
13C NMR δ: 26.4, 122.1, 125.1, 127.8, 128.0, 128.98,
129.01, 130.2, 135.1, 137.7, 142.2, 147.5, 149.6, 151.8,
197.1. Anal. calcd. for C17H12ClNO: C 72.47, H 4.29, N
4.97; found: C 72.36, H 4.34, N 4.90.
8.98 (d, 1H, J = 4.4 Hz). 13C NMR δ: 115.9 (d, JCF
=
21.1 Hz), 116.8 (d, JCF = 22.5 Hz), 122.2, 125.2, 126.0 (d,
JCF = 2.7 Hz), 127.8, 128.0, 129.0, 131.2 (d, JCF = 8.5 Hz),
135.1, 140.1 (d, JCF = 7.9 Hz), 147.1, 149.6, 151.8, 163.2 (d,
JCF = 245.6 Hz). Anal. calcd. for C15H9ClFN: C 69.91, H
3.52, N 5.44; found: C 69.70, H 3.50, N 5.45.
4-(4-Benzyloxy-3-fluorophenyl)-7-chloroquinoline (7j)
Chromatography (hexane – ethyl acetate 3:1 volume frac-
tion) provided 7j as a white solid (1.70 g, 93%), mp 150 to
151 °C. IR: 3014. 1H NMR (CDCl3) δ: 5.26 (s, 2H), 7.18 (m,
2H), 7.27 (br d, 1H, J = 11.5 Hz), 7.30 (d, 1H, J = 4.4 Hz),
7.39 (m, 1H), 7.43–7.53 (m, 5H), 7.89 (d, 1H, J = 9.0 Hz),
8.19 (d, 1H, J = 1.9 Hz), 8.93 (d, 1H, J = 4.4 Hz). 13C NMR
(CDCl3) δ: 71.3, 115.5, 117.4, 117.6, 121.3, 125.0, 125.41,
125.45, 127.0, 127.4, 127.7, 128.3, 128.7, 130.5, 130.6,
135.4, 136.1, 147.0, 147.1, 147.3, 149.0, 150.8, 151.3,
153.8. Anal. calcd. for C22H15ClFNO: C 72.63, H 4.16, N
3.85; found: C 72.21, H 4.12, N 3.83.
4-(3-Acetamidophenyl)-7-chloroquinoline (7e)
Chromatography (hexane – ethyl acetate 1:2 volume frac-
tion) provided 1b (142 mg, 10%) and 7e as a white solid
1
(1.14 g, 76%), mp 184 to 185 °C. IR: 3435, 3015, 1692. H
NMR (CDCl3) δ: 2.23 (s, 3H), 7.22 (br d, 1H, J = 7.6 Hz),
7.33 (d, 1H, J = 4.4 Hz), 7.44–7.50 (m, 2H), 7.63–7.69 (m,
3H), 7.89 (d, 1H, J = 9.0 Hz), 8.17 (d, 1H, J = 2.0 Hz), 8.94
(d, 1H, J = 4.4 Hz). 13C NMR (CDCl3) δ: 24.6, 120.1, 120.8,
121.4, 125.1, 125.2, 127.3, 127.9, 128.1, 129.4, 135.7,
138.0, 138.5, 148.4, 148.7, 150.4, 168.7. Anal. calcd. for
C17H13ClN2O: C 68.81, H 4.42, N 9.44; found: C 68.86, H
4.49, N 9.12.
7-Chloro-4-(3-thienyl)quinoline (10)
7-Chloro-4-(3-cyanophenyl)quinoline (7f)
Chromatography (hexane – ethyl acetate 4:1 volume frac-
tion) provided 10 as a white solid (1.10 g, 89%), mp 74 to
Chromatography (hexane – ethyl acetate 3:1 volume frac-
tion) provided 7f as a white solid (1.18 g, 88%), mp 160 to
1
1
75 °C. IR: 2975. H NMR δ: 7.43 (dd, 1H, J = 1.2, 4.9 Hz),
161 °C. IR: 2982, 2234. H NMR (CDCl3) δ: 7.26 (d, 1H,
7.49 (d, 1H, J = 4.4 Hz), 7.56 (dd, 1H, J = 2.2, 9.1 Hz), 7.74
(dd, 1H, J = 3.0, 4.9 Hz), 7.80 (m, 1H), 8.10 (m, 2H), 8.92
(d, 1H, J = 4.4 Hz). 13C NMR δ: 121.9, 125.4, 126.2, 127.3,
127.7, 128.0, 129.0, 129.2, 134.9, 138.2, 143.2, 149.7,
151.8. Anal. calcd. for C13H8ClNS: C 63.54, H 3.28, N 5.70;
found: C 63.53, H 3.20, N 5.66.
J = 4.4 Hz), 7.43 (dd, 1H, J = 2.1, 9.0 Hz), 7.62–7.78 (m,
5H), 8.11 (d, 1H, J = 2.1 Hz), 8.91 (d, 1H, J = 4.4 Hz). 13C
NMR (CDCl3) δ: 113.1, 118.1, 121.3, 124.4, 126.3, 128.1,
128.8, 129.6, 132.1, 132.6, 133.6, 135.6, 138.6, 145.8,
148.9, 150.8. Anal. calcd. for C16H9ClN2: C 72.60, H 3.43,
N 10.58; found: C 72.50, H 3.39, N 10.55.
(E)-7-Chloro-4-(1-hexenyl)quinoline (11)
7-Chloro-4-(4-fluorophenyl)quinoline (7g)
Chromatography (hexane – ethyl acetate 4:1 volume frac-
tion) provided 7g as a white solid (1.25 g, 96%), mp 105 to
106 °C. IR: 2978. H NMR δ: 7.39 (br t, 2H, J = 8.8 Hz),
Chromatography (hexane–ether 4:1 volume fraction) pro-
vided 11 as a colorless oil (1.05 g, 95%). IR (film): 2957,
2928, 1579. H NMR δ: 0.95 (t, 3H, J = 7.3 Hz), 1.42 (m,
1
1
7.47 (d, 1H, J = 4.4 Hz), 7.59 (dd, 1H, J = 2.2, 9.0 Hz),
7.61–7.66 (m, 2H), 7.92 (d, 1H, J = 9.0 Hz), 8.14 (d, 1H, J =
2.2 Hz), 8.98 (d, 1H, J = 4.4 Hz). 13C NMR δ: 116.1 (d,
JC-F = 21.8 Hz), 122.3, 125.4, 127.81, 127.85, 129.0, 132.0
(d, JC-F = 8.5 Hz), 134.0, 135.0, 147.5, 149.6, 151.8, 163.4
(d, JC-F = 246.5 Hz). Anal. calcd. for C15H9ClFN: C 69.91,
H 3.52, N 5.44; found: C 69.73, H 3.40, N 5.44.
2H), 1.53 (m, 2H), 2.36 (br q, 2H, J = 7.0 Hz), 6.59 (dt, 1H,
J = 15.6, 7.0 Hz), 7.15 (d, 1H, J = 15.6 Hz), 7.54 (m, 2H),
8.03 (d, 1H, J = 2.2 Hz), 8.21 (d, 1H, J = 9.0 Hz), 8.81 (d,
1H, J = 4.6 Hz). 13C NMR δ: 13.8, 22.5, 31.4, 33.3, 117.7,
124.3, 125.0, 126.1, 127.1, 128.9, 134.7, 139.3, 143.7,
149.6, 151.8. Anal. calcd. for C15H16ClN: C 73.31, H 6.56,
N 5.70; found: C 73.15, H 6.66, N 5.64.
© 2004 NRC Canada

214
Can. J. Chem. Vol. 82, 2004
16. T. Shiota and T. Yamamori. J. Org. Chem. 64, 453 (1999).
17. J.M. Nolan and D.L. Comins. J. Org. Chem. 68, 3736 (2003).
18. R.M. Peck, R.K. Preston, and H.J. Creech. J. Am. Chem. Soc.
81, 3984 (1959).
19. T.I. Wallow and B.M. Novak. J. Org. Chem. 59, 5034 (1994).
20. C.R. LeBlond, A.T. Andrews, Y. Sun, and J.R. Sowa, Jr. Org.
Lett. 3, 1555 (2001).
References
1. N. Miyaura and A. Suzuki. Chem. Rev. 95, 2457 (1995).
2. S.P. Stanforth. Tetrahedron, 54, 263 (1998).
3. A. Suzuki. J. Organomet. Chem. 653, 83 (2002).
4. J. Hassan, M. Sévignon, C. Gozzi, E. Schulz, and M. Lemaire.
Chem. Rev. 102, 1359 (2002).
5. N. Miyaura. Topics Curr. Chem. 219, 5 (2002).
6. N.M. Ali, A. McKillop, M.B. Mitchell, R.A. Rebelo, and P.J.
Wallbank. Tetrahedron, 48, 8117 (1992).
21. D. Badone, M. Baroni, R. Cardamone, A. Ielmini, and U.
Guzzi. J. Org. Chem. 62, 7170 (1997).
22. N.E. Leadbeater and M. Marco. J. Org. Chem. 68, 888 (2003).
23. R.F. Heck. Palladium reagents in organic syntheses. Academic
Press, London. 1985. p. 184.
7. M.A. Ciufolini, J.W. Mitchell, and F. Roschangar. Tetrahedron
Lett. 37, 8281 (1996).
8. O. Lohse, P. Thevenin, and E. Waldvogel. Synlett, 45 (1999).
9. Y. Gong and H.W. Pauls. Synlett, 829 (2000).
10. M. Allegretti, A. Arcadi, F. Marinelli, and L. Nicolini. Synlett,
609 (2001).
11. A.F. Littke and G.C. Fu. Angew. Chem. Int. Ed. 41, 4176
(2002).
12. D. Dubé, M. Blouin, C. Brideau, C.-C. Chan, S. Desmarais, D.
Ethier, J.-P. Falgueyret, R.W. Friesen, M. Girard, Y. Girard, J.
Guay, D. Riendeau, P. Tagari, and R.N. Young. Bioorg. Med.
Chem. Lett. 8, 1255 (1998).
13. J.-Y. Legros, G. Primault, and J.-C. Fiaud. Tetrahedron, 57,
2507 (2001).
24. G.B. Smith, G.C. Dezeny, D.L. Hughes, A.O. King, and T.R.
Verhoeven. J. Org. Chem. 59, 8151 (1994).
25. A.F. Littke, C. Dai, and G.C. Fu. J. Am. Chem. Soc. 122, 4020
(2000).
26. A.-S. Castanet, F. Colobert, J.-R. Desmurs, and T. Schlama. J.
Mol. Catalysis A Chem. 182–183, 481 (2002).
27. C. Amatore and A. Jutand. J. Organomet. Chem. 576, 254
(1999).
28. C. Amatore and A. Jutand. Acc. Chem. Res. 33, 314 (2000).
29. K. Fagnou and M. Lautens. Angew. Chem. Int. Ed. 41, 26
(2002).
30. P.M. Maitlis. The organic chemistry of palladium. Vol 1. Aca-
demic Press, New York. 1971. Chap 1.
31. S. Tasler and B.H. Lipshutz. J. Org. Chem. 68, 1190 (2003).
14. D. Najiba, J.-F. Carpentier, Y. Castanet, C. Biot, J. Brocard,
and A. Mortreux. Tetrahedron Lett. 40, 3719 (1999).
15. A.S. Bhanu Prasad, T.M. Stevenson, J.R. Citineni, V. Nyzam,
and P. Knochel. Tetrahedron, 53, 7237 (1997).
© 2004 NRC Canada