Beneficial effect of carboxylic acid additives on the Pd-catalyzed intramolecular N-arylation of 2-amino-3-(2-chlorophenylsulfonyl)pyrroles

Article abstract of DOI:10.1016/j.tetlet.2013.09.011

Carboxylic acid additives have been shown to significantly improve the yield of the Pd-catalyzed intramolecular N-arylation of 2-aminopyrroles containing a 2-chlorophenylsulfonyl group at the 3-position to give a novel tricyclic ring system. Pivalic and p

Full text of DOI:10.1016/j.tetlet.2013.09.011

Tetrahedron Letters 54 (2013) 6226–6229
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Beneficial effect of carboxylic acid additives on the Pd-catalyzed
intramolecular N-arylation of 2-amino-3-
(2-chlorophenylsulfonyl)pyrroles
⇑
Perounsack X. Moon, Chad E. Stephens
Department of Chemistry and Physics, Georgia Regents University, Augusta, GA 30904, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Carboxylic acid additives have been shown to significantly improve the yield of the Pd-catalyzed intra-
molecular N-arylation of 2-aminopyrroles containing a 2-chlorophenylsulfonyl group at the 3-position
to give a novel tricyclic ring system. Pivalic and phenylacetic acid provided the best yield enhancement,
while acetic acid showed moderate enhancement. Other acids showed limited to no enhancement or
even a detrimental effect on the reaction.
Received 9 August 2013
Revised 2 September 2013
Accepted 4 September 2013
Available online 11 September 2013
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Buchwald–Hartwig reaction
Intramolecular N-arylation
2-Aminopyrrole
Carboxylic acid additive
Pyrrolo[3,2-b]benzo[1,4]thiazine
Introduction
deprotonation occurs simultaneously with the metallation of that
carbon (concerted metallation–deprotonation pathway).11–14 Other
The transition metal-catalyzed coupling of an amine with an
aryl halide (or pseudohalide) has become the premier method for
preparation of substituted aromatic and heteroaromatic amine
derivatives.1 These so called Buchwald–Hartwig reactions are typ-
ically conducted using a metal catalyst (often Pd), along with a li-
gand (often a phosphine), and a base, which deprotonates the
amine at some point during the reaction.
Over a year ago now, Buchwald and co-workers described the
beneficial effect of acetic acid on the Pd-catalyzed intermolecular
N-arylation of some 2-aminothiazoles.2 Although no exact mecha-
nism for the beneficial effect was proposed, it was shown that the
acid led to an enhancement of the activity of the Pd catalyst
complex [formed from Pd(OAc)2]. That was apparently the first
example of a carboxylic acid additive being used to promote a tran-
sition metal-catalyzed N-arylation reaction. Surprisingly, though,
we have not seen any subsequent examples of this in the literature.
On the other hand, carboxylic acid additives are used extensively to
promote transition metal-catalyzed C–H activation reactions (i.e.,
Heck reactions) of aromatics,3–7 alkenes,8,9 and alkanes.10 In this
case, mechanistic studies suggest that the carboxylate anion, formed
under the basic conditions of the reaction, plays a supporting role in
deprotonation of the carbon that is ultimately arylated, and this
than Heck reactions, there has been a recent report of acetic acid
serving as a beneficial additive in a Pd-catalyzed Hiyama cou-
pling,15 although those authors only speculated that the acid
served to reduce the base-promoted hydrolysis of the mesylate
electrophile used in the reaction.
Previously, we have described both intermolecular16 and intra-
molecular17 N-arylation reactions of certain 2-aminopyrroles cata-
lyzed by Pd complexes. Most recently, we have been exploring an
intramolecular N-arylation approach to a novel tricyclic pyrrol-
o[3,2-b]benzo[1,4]thiazine ring system, again from 2-aminopyr-
roles. We initially found this reaction to proceed quite sluggishly
when using our previously employed conditions for Pd-catalyzed
intramolecular N-arylation, perhaps as we were employing an aryl
chloride as a coupling partner in this case. Being aware of Buch-
wald’s finding that acetic acid can be of benefit to Pd-catalyzed
N-arylation reactions, we decided to explore the use of this addi-
tive. To our delight, the addition of acetic acid led to significant
improvements in yields when using our otherwise standard condi-
tions. Upon further exploring this effect, we found that a few other
carboxylic acids also provided a beneficial effect, with some (such
as pivalic and phenylacetic acid) generally providing higher yields
than acetic acid, while other acids showed no benefit or even ap-
peared to hinder the reaction. Herein, we wish to describe these
findings, which further substantiate the potential benefit of car-
boxylic acid additives on Pd-catalyzed N-arylation reactions.
⇑
Corresponding author. Tel.: +1 706 667 4995; fax: +1 706 434 5597.
E-mail address: cstephe7@gru.edu (C.E. Stephens).
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.09.011

P. X. Moon, C. E. Stephens / Tetrahedron Letters 54 (2013) 6226–6229
6227
O
S
t-BuOH, 100 °C] gave only a 7% isolated yield of novel tricyclic 2a
after heating for 18 h and recrystallizing the crude product from
MeOH (our standard isolation protocol due to the insoluble nature
of the tricyclic products) (entry 1, Table 1). However, with the
addition of various carboxylic acids (4.2 M equiv) to the otherwise
identical reaction, isolated yields ranging from 29–53% were ob-
tained, with phenylacetic and pivalic acid giving the highest yields
(both 53%, entries 3 and 4), acetic acid giving 34% yield (entry 2),
and benzoic acid giving 29% yield (entry 5).20 In contrast, no product
was obtained when 2-nitrobenzoic acid,5 2-picolinic acid,21 or suc-
cinic acid (a diacid) were used as additives (entries 6–8). This could
be due to the potentially stronger ligating effects of these acids,
which may disrupt the catalyst activity. This could also be due to
the greater relative acidity of these other acids (especially 2-nitro-
benzoic and 2-picolinic acid), which may decrease the ability of
their conjugate bases to effectively promote the concerted metalla-
tion–deprotonation pathway (if operable in this case),7,13 although
2-nitrobenzoic acid has been shown to be better than acetic acid
at enhancing a Pd-catalyzed Heck reaction using Ag2O as base.5
With these favorable results in hand for the 18 h reaction, we
wanted to see if the use of the acid additive would allow the reac-
tion to be accomplished in a shorter time. Moving forward, we kept
the amount of additive used in this study constant at the original
O
Me
Me
O
Me
Me
1. RNH2, Δ
O
S
2.
O
Cl
OH
N
R
NH2
CN
Cl
EtOH, Δ, 2 h
1a (R = n-Bu) 57%
1b (R = CH2Ph) 53%
Scheme 1. Synthesis of 2-aminopyrroles (1a–b).
Results and discussion
As shown in Scheme 1, the 2-aminopyrroles (1a–b) used in this
study were prepared by the solvent-free reaction of acetoin with a
primary amine, followed by reaction of the intermediate
a-ami-
noketone with (2-chlorophenyl)sulfonylacetonitrile in refluxing
EtOH.17 Although 2-aminopyrroles have been prepared with
phenylsulfones at the 3-position,18 these appear to be the first
ortho-halogenated derivatives.19
The Pd-catalyzed cyclization of 2-aminopyrroles 1a–b is illus-
trated in Table 1, along with the yield data. Cyclization of 1a (with
R = n-Bu) using our previously described conditions for N-arylation
of other 2-aminopyrroles [Pd2dba3 (Catalyst A), X-Phos, Cs2CO3,
Table 1
Yields for Pd-catalyzed cyclization of 2-aminopyrroles 1a–b to tricyclics 2a–b using various carboxylic acid additives
O
S
Pd2dba3 (Cat. A) (1.8 mole%)
or
O
O
O
Me
Me
Me
Me
S
Pd(OAc)2 (Cat. B) (3.6 mole%)
Cl
N
R
NH2
N
R
N
H
X-Phos (9.6 mole%), Cs2CO3 (1.4 equiv.)
carboxylic acid additive (4.2 equiv.)
t-BuOH, 100°C, N2
2a (R = n-Bu)
2b (R = CH2Ph)
1a (R = n-Bu)
1b (R = CH2Ph)
Entry
Starting 2-aminopyrrole
Pd catalyst
Carboxylic acid additive
Time (h)
%Yield of tricyclica (isolated)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1b
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
—
Acetic
Phenylacetic
Pivalic
Benzoic
2-Nitrobenzoic
2-Picolinic
Succinicb
—
Acetic
Phenylacetic
Pivalic
Benzoic
—
Acetic
Benzoic
Phenylacetic
Pivalic
18
18
18
18
18
18
18
18
3
3
3
3
3
7
34
53
53
29
0
0
0
0
0
26
30
0
3
3
3
3
0
49c
20
37
53
3
19
20
21
22
23
24
1a
1a
1a
1a
1a
1b
1b
1b
1b
1b
1b
B
B
B
B
B
B
B
B
B
B
B
—
3
3
3
3
3
3
0.5
0.5
0.5
0.5
0.5
0
51
61
54
0
58
41
81
84
83
0e
Acetic
Phenylacetic
Pivalic
Benzoic
—
25d
26d
27d
28d
29d
—
Acetic
Phenylacetic
Pivalic
Benzoic
a
b
c
Yield of pure product after recrystallization from MeOH.
2.1 M equiv used.
The yield was 50% when K2CO3 was used as base.
K2CO3 used as base.
d
e
Reaction was performed three times with consistent results.

6228
P. X. Moon, C. E. Stephens / Tetrahedron Letters 54 (2013) 6226–6229
4.2 equiv (3 equiv based on the carbonate base). For comparison,
Buchwald and coworkers used just 3 mol % of acetic acid in their
N-arylation of 2-aminothiazoles,2 although others have used as
much as 10 M equiv of acid additive for Heck reactions.8
(entry 25). However, the yield was substantially increased to
81–84% with the addition of acetic, phenylacetic, or pivalic acid
(entries 26–28). Surprisingly though, no product could be isolated
when benzoic acid was used as the additive (entry 29) (this reac-
tion was performed three times; TLC analysis showed only trace
amounts of the tricyclic each time). Thus, benzoic acid clearly
showed a detrimental effect to the 30 min reaction using Pd(OAc)2
as catalyst. This negative result is difficult to explain, especially
considering that benzoic acid showed moderate beneficial effects
in reactions using Pd2dba3 as catalyst (with longer reaction times;
entries 5 and 16). Also, several substituted benzoic acids have
reportedly shown beneficial effects in Heck reactions.5
With a shorter reaction time of 3 h, tricyclic 2a was not isolated
when the acid additive was omitted (entry 9). However, with the
addition of phenylacetic (entry 11) or pivalic acid (entry 12), yields
of 26 and 36%, respectively, were obtained. In contrast to the 18 h
reaction, no product could be isolated when acetic or benzoic acid
was used (entries 10 and 13). Thus, phenylacetic and pivalic acid
appear to be the most active of the acid additives employed for
the cyclization of 1a. The better activity of pivalic acid compared
to acetic acid for this reaction is consistent with a number of re-
ports on the variable activity of different carboxylic acid additives
for Heck reactions, which often show pivalic acid to be
superior.7,9,13 On the other hand, we could find no reference to
phenylacetic acid being used as an additive in a transition metal-
catalyzed coupling reaction.
Next, we examined the cyclization of benzyl-substituted pyr-
role 1b, starting with a 3 h reaction time. Tricyclic 2b was also
not isolated when the acid additive was omitted (entry 14). How-
ever, with the addition of acetic, phenylacetic, pivalic, or benzoic
acid, 2b was obtained in yields ranging from 20–53% (entries 15–
18), with pivalic acid again giving the highest yield (53%, entry
18). Thus, in contrast to the cyclization of n-Bu pyrrole 1a at 3 h,
the cyclization of benzyl pyrrole 1b at 3 h was promoted by acetic
and benzoic acid, although these again did not perform as well as
pivalic acid. Also, in contrast to the cyclization of n-Bu pyrrole 1a, a
higher yield of tricyclic was obtained when using acetic acid (entry
15) compared to phenylacetic acid (entry 17), a result that does not
fit the general trend observed in this study. Consistently, though,
no tricyclic product was isolated when 2-nitrobenzoic, 2-picolinic,
or succinic acid were used in the attempted cyclization of 1b (data
not shown).
Next, we examined the cyclization when using Pd(OAc)2 (Cata-
lyst B) as the Pd source, using the equivalent amount of catalyst
based on Pd and keeping all other conditions the same. A 3 h reac-
tion of n-Bu pyrrole 1a under these conditions and without acid
additive gave no isolated product (entry 19). However, when ace-
tic, phenylacetic, or pivalic acid were added, product 2a was ob-
tained in good yields ranging from 51–61% (entries 20–22), with
phenylacetic giving the highest yield in this case (entry 21). Thus,
pivalic and phenylacetic acid promoted the cyclization of 1a in 3 h
regardless of which catalyst was used, but acetic acid only pro-
moted the reaction when Pd(OAc)2 was used. Benzoic acid, on
the other hand, failed to promote the cyclization of 1a at 3 h
regardless of which catalyst (A or B) was used (entries 13 and
23). These results further suggest that pivalic and phenylacetic acid
are the most active of the acid additives used in this study, that
acetic acid is a slightly less active, and that benzoic acid is consid-
erably less active.
Upon cyclization of benzyl pyrrole 1b at 3 h using Pd(OAc)2, tri-
cyclic product 2b was surprisingly obtained in 58% yield even
when without an acid additive (entry 24). This was in stark
contrast to the analogous reaction using Pd2dba3, which gave no
isolated product (entry 14). This is also in contrast to the analogous
reaction of butyl pyrrole 1a using Pd(OAc)2, which gave no product
at 3 h when the acid was omitted (entry 19). As a side note, these
results thus clearly show that the cyclization of the benzyl pyrrole
1b is more efficient than 1a, and that Pd(OAc)2 is a better Pd source
for this cyclization. The enhanced reactivity of 1b, which has the
aromatic-containing R group, may be another example of the ‘aro-
matic effect’ which has been observed with some transition metal-
catalyzed reactions.22
Conclusion
In conclusion, we have found that the Pd-catalyzed intramolec-
ular N-arylation of 2-aminopyrroles containing a 2-chlorophenyl-
sulfone at the 3-position is significantly enhanced by the
addition of certain carboxylic acid additives. Pivalic and phenylace-
tic acids, both aliphatic acids, were the best of the acid additives
evaluated, while acetic acid showed, in general, a lower beneficial
effect. Benzoic acid showed only a modest enhancement of the
reaction when using Pd2dba3 as catalyst, but clearly showed a det-
rimental effect when using Pd(OAc)2 and a short reaction time. The
three other carboxylic acids tested under our conditions, which
were mostly more acidic than the active acids, showed no benefi-
cial effect, or even a detrimental effect, compared to the use of no
additive. This work supports Buchwald’s original finding that a car-
boxylic acid additive can promote the Pd-catalyzed N-arylation
reaction, and shows that pivalic acid and phenylacetic acid (a more
easily handled free-flowing solid acid) may potentially be more
active as an additive for the N-arylation reaction than acetic acid.
Finally, although carboxylic acid additives are now routinely used
for Heck reactions, this appears to be just the second report of their
use for Buchwald–Hartwig reactions. Further work is thus still
needed to determine the generality of the effect, the precise role
of the acid additive, and the optimal amount of acid additive to
be used in the reaction.
Acknowledgments
We would like to thank Dr. Iryna Lebedyeva (Dept. of Chemis-
try, University of Florida) for acquisition of some of the NMR spec-
tra. We also thank our Dept. for partial funding of this work.
References and notes
1. Magano, J.; Dunetz, J. R. Chem. Rev. 2011, 111, 2177.
2. McGowan, M. A.; Henderson, J. L.; Buchwald, S. L. Org. Lett. 2012, 14, 1432.
3. Vamvounis, G.; Gendron, D. Tetrahedron Lett. 2013, 54, 3785.
4. Gaulier, S. M.; McKay, R.; Swain, N. A. Tetrahedron Lett. 2011, 52, 6000.
5. Lebrasseur, N.; Larrosa, I. J. Am. Chem. Soc. 2008, 130, 2926.
6. Caron, L.; Campeau, L.-C.; Fagnou, K. Org. Lett. 2008, 10, 4533.
7. Lafrance, M.; Fagnou, K. J. Am. Chem. Soc. 2006, 128, 16496.
8. Zhang, G.; Li, Z.; Huang, Y.; Xu, J.; Wu, X.; Yao, H. Tetrahedron 2013, 69, 1115.
9. Kim, K. H.; Lee, S.; Kim, S. H.; Lim, C. H.; Kim, J. N. Tetrahedron Lett. 2012, 53,
5088.
10. Lafrance, M.; Gorelsky, S. I.; Fagnou, K. J. Am. Chem. Soc. 2007, 129, 14570.
11. Gorelsky, S. I.; Lapointe, D.; Fagnou, K. J. Org. Chem. 2012, 77, 658.
12. Lapointe, D.; Fagnou, K. Chem. Lett. 2010, 39, 1118.
13. Lafrance, M.; Lapointe, D.; Fagnou, K. Tetrahedron 2008, 64, 6015.
14. Phenol additives have also been used: Rutherford, J. L.; Rainka, M. P.;
Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 15168.
15. So, C. M.; Lee, H. W.; Lau, C. P.; Kwong, F. Y. Org. Lett. 2009, 11, 317.
16. Griswold, C. P.; Suthiwangcharoen, N.; Pochini, S. M.; Stephens, C. E. Synth.
Commun. 2011, 41, 1672.
17. Suthiwangcharoen, N.; Pochini, S. M.; Sweat, D. P.; Stephens, C. E. J. Heterocycl.
Chem. 2011, 48, 706.
18. Mattson, R. J.; Wang, L.-C.; Walter Sowell, J. J. Heterocycl. Chem. 1980, 17, 1793.
19. Data for butyl pyrrole 1a: coarse light yellow crystals, mp 135–136 °C (MeOH).
IR (cmÀ1): 3443, 3346, 2959, 2934, 2865, 1622 (s), 1544, 1492, 1449, 1275 (s),
With a reaction time of 30 min, the cyclization of 1b using
Pd(OAc)2 and no acid additive still gave a reasonable 41% yield

P. X. Moon, C. E. Stephens / Tetrahedron Letters 54 (2013) 6226–6229
6229
1137 (s), 1095, 1035, 750, 709, 669, 578, 562. 1H NMR (300 MHz, DMSO-d6):
0.88 (t, 3H), 1.27 (m, 2H), 1.48 (m, 2H), 1.62 (s, 3H), 1.92 (s, 3H), 3.66 (t, 2H),
5.68 (s, 2H), 7.50–7.57 (m, 3H), 8.07 (d, J = 7.4 Hz, 1H). 13C NMR (75 MHz,
DMSO-d6): 8.8, 9.3, 13.7, 19.3, 31.0, 41.4, 92.3, 108.7, 118.5, 127.3, 129.5, 131.7,
131.7, 133.4, 141.2, 144.7. Anal. Calcd for C16H21ClN2O2S (340.87): C, 56.38; H,
6.21; N, 8.22. Found: C, 56.36; H, 6.22; N, 8.10. Data for benzyl pyrrole 1b: light
yellow crystals/cubes, mp 164–166.5 °C (MeOH). IR (cmÀ1): 3409, 3329, 3083,
3004, 2947, 2917, 1607, 1536, 1485, 1449, 1292, 1280 (s), 1142, 1117, 1097 (s),
1035, 769, 728 (s), 707, 692. 1H NMR (300 MHz, DMSO-d6): 1.64 (s, 3H), 1.78 (s,
3H), 5.02 (s, 2H), 5.85 (s, 2H), 7.02-7.04 (m, 2H), 7.23-7.35 (m, 3H), 7.52–7.60
(m, 3H), 8.12 (m, 1H). 13C NMR (75 MHz, DMSO-d6): 8.8, 9.3, 44.7, 92.5, 109.1,
109.5, 118.8, 126.0, 127.0, 127.4, 128.5, 129.6, 131.7, 133.5, 137.4, 141.1, 145.3.
Anal. calcd for C19H19ClN2O2S (374.88): C, 60.87; H, 5.11; N, 7.47. Found: C,
60.76; H, 5.09; N, 7.34.
pure products given in Table 1). The yield was recorded as zero when no
product was obtained from the recrystallization. In these cases, TLC generally
showed that only a trace amount of product, with starting pyrrole being the
major component. Data for butyl tricyclic 2a: off white/tan fibrous needles, mp
249–251 °C (MeOH). IR (cmÀ1): 3267, 3190, 3139, 2955, 2930, 2873, 1603,
1552 (s), 1486, 1242 (s), 1135, 1091 (s), 1049, 759 (s). 1H NMR (300 MHz,
DMSO-d6): 0.89 (t, 3H), 1.29 (m, 2H), 1.55 (m, 2H), 2.12 (s, 3H), 2.13 (s, 3H),
3.94 (t, 2H), 7.13 (t, 1H), 7.39 (d, J = 8.1 Hz, 1H), 7.50 (m, 1H), 7.82 (d, J = 8.1 Hz,
1H). 13C NMR (75 MHz, DMSO-d6): 8.7, 9.5, 13.6, 19.2, 31.8, 42.1, 98.8, 107.0,
117.0, 120.6, 121.2, 122.8, 123.6, 131.4, 131.5, 135.6. Anal. Calcd for
C16H20N2O2S (304.41): C, 63.13; H, 6.62; N, 9.20. Found: C, 63.19; H, 6.55; N,
9.10. Data for benzyl tricyclic 2b: off white matted solid, mp 288–291 °C
(MeOH). IR (cmÀ1): 3284, 3195, 3143, 2955, 2937, 1602, 1550 (s), 1482, 1243
(s), 1167, 1094 (s), 1052, 767, 730, 694. 1H NMR (300 MHz, DMSO-d6): 1.98 (s,
3H), 2.16 (s, 3H), 5.30 (s, 2H), 6.98 (d, J = 7.4 Hz, 2H), 7.15 (t, 1H), 7.25–7.37 (m,
20. General procedure for cyclization: To a 100 mL flask were added (in open air) the
2-aminopyrrole (0.6 mmol), palladium catalyst [0.0108 mmol/1.8 mol % of
Pd2dba3 or 0.0216 mmol/3.6 mol % of Pd(OAc)2], X-Phos ligand (0.0575 mmol,
9.6 mol %), cesium or potassium carbonate (0.84 mmol, 1.4 molar equiv), tert-
butanol (7 mL), and, finally, the acid additive (2.52 mmol, 4.2 molar equiv). The
mixture was then stirred at reflux at 100 °C (oil bath temperature) under N2 for
the indicated time (Table 1). The reaction was then cooled and diluted with
excess water to give a crude solid that was filtered and recrystallized from
MeOH (with standing overnight in the cold box) to give product (yields for
4H), 7.48 (dd, J = 1.4, 7.0 Hz, 1H), 7.85 (dd, J = 1.2, 8.1 Hz, 1H), 10.67 (s, NH). 13
C
NMR (75 MHz, DMSO-d6): 8.8, 9.5, 45.4, 99.0, 107.7, 117.0, 120.9, 121.7, 122.9,
123.7, 125.8, 127.3, 128.8, 131.6, 132.2, 135.6, 137.2. Anal. Calcd for
C19H18N2O2S (338.42): C, 67.43; H, 5.36; N, 8.28. Found: C, 67.17; H, 5.31; N,
8.07.
21. Lam, M. S.; Lee, H. W.; Chan, A. S. C.; Kwong, F. Y. Tetrahedron Lett. 2008, 49,
6192.
22. Perdriau, S.; Harder, S.; Heeres, H. J.; de Vries, J. G. ChemSusChem 2012, 5, 2427.