Carbopalladation of nitriles: Synthesis of 3,4-disubstituted 2-aminonaphthalenes and 1,3-benzoxazine derivatives by the palladium-catalyzed annulation of alkynes by (2-Iodophenyl)acetonitrile

Article abstract of DOI:10.1021/jo026229+

Intramolecular carbopalladation of the cyano group has been employed for the synthesis of 3,4-disubstituted 2-aminonaphthalenes. (2-Iodophenyl)acetonitrile reacts with a variety of internal alkynes to afford 2-aminonaphthalenes in high yields with good regioselectivity. The scope and limitations of this process, which proceeds by the intramolecular addition of a vinylpalladium species to the triple bond of the cyano group, have been studied. The annulation of certain hindered propargylic alcohols affords 1,3-benzoxazine derivatives, rather than the expected 2-aminonaphthalenes. The involvement of trialkylamine bases in the formation of these heterocyclic compounds has been established. A proposed mechanism for the synthesis of 1,3-benzoxazine derivatives involves the formation of the expected 2-amino-3-(1-hydroxyalkyl)naphthalenes, followed by their condensation with an iminium ion species formed from the trialkylamine base used in the reaction.

Full text of DOI:10.1021/jo026229+

Ca r bop a lla d a tion of Nitr iles: Syn th esis of 3,4-Disu bstitu ted
2-Am in on a p h th a len es a n d 1,3-Ben zoxa zin e Der iva tives by th e
P a lla d iu m -Ca ta lyzed An n u la tion of Alk yn es by
(2-Iod op h en yl)a ceton itr ile
Qingping Tian, Alexandre A. Pletnev, and Richard C. Larock*
Department of Chemistry, Iowa State University, Ames, Iowa 50011
larock@iastate.edu
Received J uly 23, 2002
Intramolecular carbopalladation of the cyano group has been employed for the synthesis of 3,4-
disubstituted 2-aminonaphthalenes. (2-Iodophenyl)acetonitrile reacts with a variety of internal
alkynes to afford 2-aminonaphthalenes in high yields with good regioselectivity. The scope and
limitations of this process, which proceeds by the intramolecular addition of a vinylpalladium species
to the triple bond of the cyano group, have been studied. The annulation of certain hindered
propargylic alcohols affords 1,3-benzoxazine derivatives, rather than the expected 2-aminonaph-
thalenes. The involvement of trialkylamine bases in the formation of these heterocyclic compounds
has been established. A proposed mechanism for the synthesis of 1,3-benzoxazine derivatives
involves the formation of the expected 2-amino-3-(1-hydroxyalkyl)naphthalenes, followed by their
condensation with an iminium ion species formed from the trialkylamine base used in the reaction.
In tr od u ction
ligands.9 In particular, 1,1′-binaphthyl-2,2′-diamine chiral
auxiliaries have been used for enantioselective reduction
of ketones, asymmetric synthesis of lactones, asymmetric
hydrogenation of R-acylaminoacrylic acids, and asym-
metric alkylation of aromatic aldehydes.10 Additional
substituents at positions 3 and 3′ of binaphthyl systems
have been recognized to impose further steric interac-
tions, which often results in a remarkable increase in
asymmetric induction.9 This underscores the importance
of developing effective and practical routes to 3-substi-
tuted 2-aminonaphthalenes. In another asymmetric ap-
plication, 2-naphthylamines have been used for the
synthesis of naphthyl-Troger’s base, a representative of
a class of chiral structures that have both theoretical and
practical interest as molecular receptors, chiral solvating
agents (e.g., in host-guest complexes) and chiral modi-
fiers in enantioselective reactions.11
Aminonaphthalenes are useful precursors to a variety
of substances that have interesting industrial and phar-
maceutical uses. For example, complex heterocyclic sys-
tems such as benzo[c]phenothiazines,1 benzo[f]quinazo-
lines,2 benzindoles,3 benz[a]- and benz[c]acridines,4
naphtho[1,2-d]imidazoles,5 and others have been synthe-
sized from naphthylamines. For over a century, amino-
naphthalenes have been a staple of the dyestuffs indus-
try, serving as diazo and coupling components in the
preparation of azo dyes.6 Despite the significant carci-
nogenicandmutagenicactivityof1-and2-naphthylamines,6b,7
their derivatives have been explored as spasmolytic,
emetic, and antitumor agents.8
With the emergence of asymmetric synthesis, 2-amino-
naphthalenes have found new uses as starting materials
for the synthesis of binaphthyl C2-symmetric chiral
Naphthylamines can be prepared from naphthalene
and its derivatives by a number of traditional synthetic
organic methods available for the synthesis of aromatic
amines.12 Classical routes to aminonaphthalenes include
the treatment of naphthols with bisulfites and ammonia
(1) (a) Gupta, R. R.; J ain, S. K. Bull. Chem. Soc. J pn. 1976, 49, 2026.
(b) Nomura, Y.; Hayama, T.; Takeuchi, Y.; Tomoda, S.; Kato, Y. Bull.
Chem. Soc. J pn. 1984, 57, 1276.
(2) Rosowsky, A.; Modest, E. J . J . Heterocyclic Chem. 1966, 3, 387.
(3) Rastogi, R.; Zutshi, K. J . Prakt. Chem. 1983, 325, 359.
(4) Haldar, M. K.; Kar, G. K.; Ray, J . K. J . Chem. Res., Synop. 1993,
46.
(5) Mitra, A.; Chauhan, S. M. S.; George, M. V. J . Org. Chem. 1980,
45, 3182.
(6) (a) Guinot, S. G. R.; Hepworth, J . D.; Wainwright, M. Dyes Pigm.
1998, 36, 387. (b) Ayyangar, N. R.; Bhide, S. R. J . Chromatogr. 1989,
464, 201.
(9) Smrcina, M.; Vyskocil, S.; Maca, B.; Polasek, M.; Claxton, T. A.;
Abbott, A. P.; Kocovsky, P. J . Org. Chem. 1994, 59, 2156, and
references therein.
(10) Rosini, C.; Franzini, L.; Raffaelli, A.; Salvadori, P. Synthesis
1992, 503.
(11) Talas, E.; Margitfalvi, J .; Machytka, D.; Czugler, M. Tetrahe-
dron: Asymmetry 1998, 9, 4151.
(7) (a) Tomkins, B. A.; Ostrum, V. H.; Caton, J . E. Anal. Chim. Acta
1982, 134, 301. (b) Famulok, M.; Bosold, F.; Boche, G. Tetrahedron
Lett. 1989, 30, 321.
(8) (a) Kanao, M.; Hashizuma, T.; Ichikawa, Y.; Irie, K.; Isoda, S. J .
Med. Chem. 1982, 25, 1358. (b) Sprenger, W. K.; Cannon, J . G.;
Barman, B. K.; Burkman, A. M. J . Med. Chem. 1969, 12, 487. (c) J anin,
Y. L.; Bisagni, E. Synthesis 1993, 57.
(12) For comprehensive reviews, see: (a) Gibson, M. S. In The
Chemistry of the Amino Group; Patai, S., Ed.; Interscience: New York,
1968; p 37. (b) Lindsay, R. J . In Comprehensive Organic Chemistry;
Sutherland, I. O., Ed.; Pergamon: Oxford, U.K., 1979; Vol. 2, p 131.
(c) Sauve, G.; Rao, V. S. In Comprehensive Organic Functional Group
Transformations; Katritzky, A. R., Meth-Cohn, O., Rees, C. W., Eds.;
Pergamon: Cambridge, U.K., 1995; Vol. 2, p 752.
10.1021/jo026229+ CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/13/2002
J . Org. Chem. 2003, 68, 339-347
339

Tian et al.
(Bucherer reaction)13 and the acid-catalyzed transforma-
tion of tetralone oximes (Semmler-Wolff reaction).14 Very
specific aminonaphthalene systems, such as substituted
o-aminonaphthols, 2-cyano-1-aminonaphthalenes, and
others, have been prepared with various success by a
variety of aminobenzannulation approaches.15 To the best
of our knowledge, there is no general and efficient
methodology for the synthesis of aminonaphthalenes by
alkyne annulation. Yet, such an approach could be
extremely valuable, as it would allow for rapid construc-
tion of a fairly complex functionalized cyclic system from
two independent components.
Recently, useful palladium-alkyne annulation meth-
odology has been developed in this group, which offers
convenient routes to various carbo- and heterocyclic
compounds.16 These reactions involve the insertion of an
internal alkyne into an arylpalladium intermediate and
subsequent cyclization onto a functional group present
in the ortho position. In continuation of our work, we
have investigated the possibility that a cyanomethyl
group might serve as the neighboring functional group
and that the vinylpalladium intermediate might add
across the carbon-nitrogen triple bond to produce 2-ami-
nonaphthalenes (eq 1). Here, we wish to report full
details of our work on developing the Pd-catalyzed alkyne
annulation of internal alkynes by (2-iodophenyl)acetoni-
trile into general methodology for the synthesis of 3,4-
disubstituted 2-aminonaphthalenes.17,18
to GC-MS analysis, the reaction was complete after 48 h
and higher yields were observed (entries 3 and 4). Other
palladium catalysts, such as Pd(dba)2 and Pd(PPh3)4, did
not afford good yields (entries 5-7). Pd(OAc)2 appears
to be the best catalyst. Somewhat surprising was the
observation that 5 mol % Pd(OAc)2 is as effective as 20
mol % of the catalyst (compare entries 1 and 2, 3 and 4).
The next variable examined was the solvent (entries
8-12). The use of dimethylformamide (DMF) as the
solvent is crucial to the success of this reaction (entry
3). The reaction was also carried out in 9:1 DMF/H2O,
although the yield was considerably lower (entry 8).
Other solvents, including dimethyl sulfoxide (DMSO),
MeNO2, dimethylacetamide (DMA), and MeCN were
inefficient, and none of the desired product was observed
(entries 9-12). The failure of DMA is particularly strik-
ing, since it usually behaves very similar to DMF as one
might expect with such similar structures.
The next task was to find the best base for the reaction.
A series of inorganic bases were examined first (entries
3 and 13-18), and NaOAc was observed to furnish the
highest yield (entry 3). After careful consideration of the
possible mechanism of the annulation (vide infra), we
realized that a hydrogen source is required for the
reaction. With this in mind, NaH and HCOONa were
examined as bases (entries 17 and 18). However, neither
reaction afforded the desired product. It is known that
tertiary amines containing an R-hydrogen can provide a
hydride to palladium through insertion of the palladium
into the C-H bond adjacent to nitrogen.19 Therefore,
Et3N was examined in the reaction, and the yield of
annulation product was significantly improved (entry 19).
It was also noticed that employing n-Bu4NCl as the
chloride source afforded better yields than LiCl (compare
entries 3 and 20, 19 and 21). Another tertiary amine,
i-Pr2NEt, was also examined in the reaction, and a high
yield was obtained (entry 22). However, the correspond-
ing secondary amines did not furnish good yields (entries
23 and 24). We also confirmed again that a reaction time
of 48 h is enough for the reaction, since no improvement
in yield was observed after 48 h (compare entries 21, 25,
and 26).
Resu lts a n d Discu ssion
We chose the reaction of diphenylacetylene and (2-
iodophenyl)acetonitrile as a model system for our initial
investigation (eq 1; R1, R2 ) Ph). First of all, the standard
reaction conditions used in much of our previous pal-
ladium annulation chemistry16 were employed in the
reaction. As hoped for, the carbon-nitrogen triple bond
of the cyano group participated in the reaction and
2-amino-3,4-diphenylnaphthalene was obtained in a 34%
yield (Table 1, entry 1).
Since the yield of the reaction was low, considerable
effort has been carried out to optimize the reaction
conditions so as to improve the yields. First, different
reaction times and palladium catalysts were examined
in the reaction (Table 1, entries 1-7). Using Pd(OAc)2,
GC-MS analysis indicated that the reaction had not
reached completion even after 11 h (entries 1 and 2).
Thus, longer reaction times were employed. According
In a continued effort to optimize the reaction condi-
tions, the effects of phosphines and Lewis acids were
explored (entries 27-32). It was observed that the
addition of the Lewis acids Zn(OAc)2 and ZnCl2 only made
the reaction worse (compare entries 27-29). The addition
of catalytic amounts of phosphines, such as PPh3, P(o-
tolyl)3, and tris(2,6-dimethoxyphenyl)phosphine, did not
improve the yield, and longer reaction times were re-
quired for the reaction to reach completion (entries 30-
32).
Having established the optimal solvent, reaction time,
catalyst, and base, we turned our attention toward the
amount of the base Et3N. Varying amounts of Et3N, from
1 to 5 molar equiv, were examined (entries 21 and 33-
35). The results indicate that 2 equiv of Et3N afforded
the best yield of naphthylamine (entry 34).
(13) (a) Drake, N. L. Org. React. 1942, 1, 105. (b) Seeboth, H. Angew.
Chem., Int. Ed. Engl. 1967, 6, 307. (c) Canete, A.; Melendrez, M. X.;
Saitz, C.; Zanocco, A. L. Synth. Commun. 2001, 31, 2143.
(14) Newman, M. S.; Hung, W. M. J . Org. Chem. 1973, 38, 4073.
(15) (a) Sommer, M. B.; Begtrup, M.; Boegesoe, K. P. J . Org. Chem.
1990, 55, 4822. (b) Herndon, J . W.; Zhang, Y.; Wang, K. J . Organomet.
Chem. 2001, 634, 1. (c) Grotjahn, D. B.; Kroll, F. E. K.; Schaefer, T.;
Harms, K.; Do¨tz, K. K. Organometallics 1992, 11, 298. (d) Merlic, C.
A.; Xu, D.; Gladstone, B. G. J . Org. Chem. 1993, 58, 538.
(16) (a) Larock, R. C. J . Organomet. Chem. 1999, 576, 111. (b)
Larock, R. C. Pure Appl. Chem. 1999, 71, 1435.
(19) For mechanistic discussion of this process, see: (a) McCrindle,
R.; Ferguson, G.; Arsenault, G. J .; McAlees, A. J . J . Chem. Soc., Chem.
Commun. 1983, 571, and references therein. (b) Trzeciak, A. M.;
Ciunik, Z.; Ziolkowski, J . J . Organometallics 2002, 21, 132. (c)
Murahashi, S.-I.; Hirano, T.; Yano, T. J . Am. Chem. Soc. 1978, 100,
348. (d) Pletnev, A. A.; Tian, Q.; Larock, R. C. J . Org. Chem. 2002, 67,
9276-9287.
(17) For a preliminary communication, see: Larock, R. C.; Tian, Q.;
Pletnev, A. A. J . Am. Chem. Soc. 1999, 121, 3238.
(18) See also: Pletnev, A. A.; Larock, R. C. Tetrahedron Lett. 2002,
43, 2133.
340 J . Org. Chem., Vol. 68, No. 2, 2003

Carbopalladation of Nitriles
TABLE 1. Op tim iza tion of th e Rea ction Con d ition s for P a lla d iu m -Ca ta lyzed An n u la tion of Dip h en yla cetylen e by
(2-Iod op h en yl)a ceton itr ile (eq 1, R1, R2 ) P h )a
entry
catalyst
solvent
base (equiv)
additive (equiv)
chloride (equiv)
time (h)
% isolated yield
1
2
3
4
5
6
7
8
9
5% Pd(OAc)2
20% Pd(OAc)2
5% Pd(OAc)2
20% Pd(OAc)2
5% Pd(dba)2
5% Pd(dba)2
5% Pd(PPh3)4
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
5% Pd(OAc)2
DMF
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
NaOAc (2)
Na2CO3 (2)
K2CO3 (2)
KOAc (2)
NaHCO3 (2)
NaH (2)
HCOONa (2)
Et3N (3)
NaOAc (2)
Et3N (3)
i-Pr2NEt (3)
Et2NH (3)
i-Pr2NH (3)
Et3N (3)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
LiCl (1)
11
11
48
48
11
72
32
24
21
21
21
96
48
48
48
48
48
48
48
48
48
48
48
48
72
120
24
21
21
72
72
96
48
48
48
48
48
48
34
40
52
51
19
30
32
37
DMF
DMF
DMF
DMF
DMF
DMF
9:1 DMF/H2O
DMSO
MeNO2
DMA
MeCN
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
b
-
b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
-
b
-
c
-
41
35
42
32
d
LiCl (1)
LiCl (1)
LiCl (1)
-
-
73
63
76
75
d
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (3)
n-Bu4NCl (3)
n-Bu4NCl (3)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
n-Bu4NCl (1)
d
-
51
76
77
35
23
24
77
78
66
63
78
71
74f
83g
74h
Et3N (3)
NaOAc (2)
NaOAc (2)
NaOAc (2)
Et3N (3)
Et3N (3)
Et3N (3)
Et3N (1)
Et3N (2)
Et3N (5)
Et3N (2)
Zn(OAc)2 (1)
ZnCl2 (1)
PPh3 (0.1)
P(o-tolyl)3 (0.1)
PAr3e (0.1)
Et3N (2)
Et3N (2)
a
All reactions were run in the presence of 2 equiv of diphenylacetylene (unless indicated otherwise) at 100 °C. b (2-Iodophenyl)acetonitrile
disappeared, but none of the desired product was obtained. c (2-Iodophenyl)acetonitrile did not react completely, and only a trace of the
desired product was observed. Only a trace of the desired product was observed. e Ar ) 2,6-dimethoxyphenyl. f A 1 equiv amount of
d
g
h
diphenylacetylene was employed. A 3 equiv amount of diphenylacetylene was employed. A 5 equiv amount of diphenylacetylene was
employed.
SCHEME 1
The final variable examined was the stoichiometry of
the diphenylacetylene. Different amounts of diphenyl-
acetylene were employed in the reaction (entries 34 and
36-38). The highest yield was obtained when 3 equiv of
diphenylacetylene was utilized (entry 37).
On the basis of the above investigation, the optimal
conditions for this reaction are as follows: 5 mol %
Pd(OAc)2, 3 equiv of diphenylacetylene, 2 equiv of Et3N,
and 1 equiv of n-Bu4NCl in DMF are heated at 100 °C
for 48 h. i-Pr2NEt is also a suitable choice of base.
The proposed mechanism of 2-aminonaphthalene for-
mation is shown in Scheme 1. This process presumably
starts with reduction of the Pd(OAc)2 to the actual
catalyst Pd(0). The oxidative addition of (2-iodophenyl)-
acetonitrile to Pd(0) produces an arylpalladium interme-
diate I, which rapidly adds across the triple bond of the
alkyne to afford a vinylic palladium species II. A priori,
two different paths for intramolecular carbopalladation
of the cyano group in II appear plausible. The vinylic
palladium moiety II may undergo addition to the neigh-
boring CN triple bond to generate the iminopalladium
intermediate III, which undergoes rapid tautomerization
to the aminopalladium species IV (path 1). An alternative
path might involve base-induced formation of ketenimine
V20 and subsequent syn addition of the vinylpalladium
J . Org. Chem, Vol. 68, No. 2, 2003 341

Tian et al.
TABLE 2. P d -Ca ta lyzed An n u la tion of In ter n a l Alk yn es by (2-Iod oa r yl)a ceton itr iles (eq 1)a
342 J . Org. Chem., Vol. 68, No. 2, 2003

Carbopalladation of Nitriles
Ta ble 2 (Con tin u ed )
a
b
See the Experimental Section for the reaction conditions. A 2 equiv amount of H2O was employed in the reaction; without water,
the yield was 61%. c Isolated as an inseparable mixture with about 6% of the other regioisomer as determined by 1H NMR spectral
analysis.
To study the scope of this annulation, a variety of
internal alkynes have been introduced into the reaction
(Table 2). 2-Amino-3,4-diarylnaphthalenes 1 and 2 were
obtained in very good yields from annulation of the
corresponding diarylacetylenes (entries 1 and 2). The
current methodology is expected to be readily applicable
to the synthesis of various 2-amino-3,4-diarylnaphtha-
lenes from symmetrical diarylacetylenes. An excellent
overall yield of the annulation product was also obtained
in the reaction of an unsymmetrical alkyne, 2-(phenyl-
ethynyl)toluene (entry 3). However, both possible regio-
isomers 3 and 4 were isolated in approximately equal
amounts. Apparently, the difference between the steric
demands of the two aryl substituents in this diarylacety-
lene is not large enough to command better regioselec-
tivity in arylpalladium addition across the triple bond of
the alkyne (see below). The attempted annulation of
2-(phenylethynyl)phenol afforded 2-phenylbenzofuran
moiety to the C-N double bond to generate IV (path 2).
On the basis of the success of our previous annulation of
diphenylacetylene with 2-(2-iodophenyl)-2-methylpro-
panenitrile (eq 2),17 which presumably proceeds via a very
similar mechanism, we favor path 1. In the next step,
aminopalladium complex IV is reduced to the final
product, accompanied by regeneration of the Pd(0) cata-
lyst. Although we have evidence indicating that the
Pd(II) moiety in IV is reduced by Et3N (vide infra), we
cannot exclude some involvement of DMF, since the yield
of the annulation product was sharply reduced when
other solvents, such as DMSO and DMA, were used in
the reaction (entries 9 and 11, Table 1). Also, traces of
water inadvertently present in DMF or n-Bu4NCl may
be the hydrogen source in the reduction21 or simple
protonation of IV to generate Pd(II), which is subse-
quently reduced to Pd(0) by other species.
(20) Whittaker, D. In The Chemistry of Triple-Bonded Functional
Groups; Patai, S., Ed.; J ohn Wiley & Sons: New York, 1994; p 513.
(21) Coperet, C.; Sugihara, T.; Wu, G.; Shimoyama, I.; Negishi, E.
J . Am. Chem. Soc. 1995, 117, 3422.
J . Org. Chem, Vol. 68, No. 2, 2003 343

Tian et al.
instead of the target aminonaphthalene (eq 3). Examples
of this Pd-catalyzed cyclization are well-known in the
literature.22
regioselectivity (entries 10 and 11). The regiochemistry
of these and other products in Table 1 could be deter-
mined by 1D and 2D H NMR spectral analysis. We also
1
observed the formation of the tetracyclic amine 11, which
presumably arises from the in situ cyclization of 10 by a
mechanism that is unclear (entry 11). The reaction of
1-phenyl-2-(trimethylsilyl)acetylene afforded the simple
coupling product 12 in a 58% yield (entry 12). None of
the desired aminonaphthalene product was observed.
This reaction presumably proceeds by desilylation of the
alkyne to produce phenylacetylene, which undergoes
coupling with (2-iodophenyl)acetonitrile to give 12. The
same product 12 was also obtained in a similar yield
when phenylacetylene was used as the alkyne (entry 13).
A significant amount of 1,4-diphenylbutadiyne was also
detected by GC-MS in this reaction mixture. The more
hindered silyl group in 1-triisopropylsilyl-1-propyne was
stable to desilylation, and this alkyne afforded a 54%
yield of the aminonaphthalene 13 (entry 14).
The annulation works reasonably well for internal
alkynes other than diarylacetylenes. Thus, 2-amino-4-
methyl-3-phenylnaphthalene (5) was isolated in a 65%
yield as a single isomer from the reaction of (2-iodophen-
yl)acetonitrile and 1-phenyl-1-propyne (entry 4). The
regioselectivity of this reaction can be nicely explained
by addition of the aryl group of the arylpalladium
intermediate I (Scheme 1) to the less hindered end of the
alkyne, placing the palladium moiety on the more
hindered end of the original triple bond. Such regio-
selectivity for the addition has been frequently observed
in our previous research.23,24 As a result, the more
sterically hindered group present in the alkyne ends up
in the 3 position of the naphthalene product and the less
hindered group in the 4 position, which is indeed ob-
served. However, for this regioselectivity to be pro-
nounced, the two groups must be significantly different
sterically (entries 4, 10, 11, 14, and 19). In other cases,
formation of both possible regioisomers may be expected
(entries 3 and 5). When 1-phenyl-1-butyne was employed
as the alkyne, the anticipated aminonaphthalene 6 was
isolated in a 37% yield along with an unexpected product
7, which obviously arose from the second regioisomer
(entry 5). While we observed only a single isomer 8 in
the annulation of 3-phenyl-2-propyn-1-ol (entry 6), we
cannot rule out the possibility that an aldehyde or
carboxylic acid product similar to 7 was formed in the
reaction, but was subsequently lost to side reactions, such
as oxidation or condensation.
The unusual product 14 was formed as the sole product
in the reaction of 4-octyne (entry 15). The (E)-stereo-
1
chemistry of 14 was established from H NMR coupling
constants between the olefinic hydrogens (J ) 16.2 Hz)
and fully confirmed by 2D NOESY spectroscopy. Internal
alkynes bearing electron-withdrawing groups did not
undergo annulation, possibly because of competing Michael
addition-like processes (entries 16 and 17). Also unsuc-
cessful was the annulation of 1,4-diphenylbutadiyne
(entry 18). It has been reported, however, that diaryl-
butadiynes may easily undergo Pd-catalyzed formation
of 1,2,3-butatriene derivatives under conditions similar
to ours.25 Another 1,3-diyne, 2,2,7,7-tetramethyl-3,5-
octadiyne, afforded the expected annulation product 15
in excellent yield and with very good regioselectivity
(entry 19).
An electron-rich (2-iodophenyl)acetonitrile derivative
16 was prepared and used in the annulation of diphen-
ylacetylene and 4-octyne (entries 20 and 21). In both
cases, the corresponding 2-aminonaphthalenes 17 and 18
were obtained in good yields. Interestingly, we did not
observe any unsaturated product similar to 14 in the
reaction of 4-octyne (entry 21). No electron-poor deriva-
tives of (2-iodophenyl)acetonitrile were examined in the
annulation, since our previous research on nitrile carbo-
palladation indicated that arylpalladium and vinyl-
palladium intermediates (I and II, Scheme 1) formed
from electron-deficient substrates are usually not nu-
cleophilic enough for successful attack on the cyano
group.18,19d
We have also prepared a number of protected cyano-
hydrins 19 (Scheme 2) in order to investigate the pos-
sibility of synthesizing 3,4-disubstituted 2-amino-1-
naphthols by this methodology. Unfortunately, only
2-acetoxy-2-(2-iodophenyl)acetonitrile (19c) proved stable
enough to survive our reaction conditions, cleanly afford-
ing the oxazole derivative 20 in a 53% yield (entry 22).
An almost identical yield (52%) of 20 was achieved when
the reaction was run at 130 °C. Silyl-protected cyanohy-
drins 19a ,b underwent a retroreaction to produce 2-io-
dobenzaldehyde as the major product. Even the piperi-
dine derivative 19d was hydrolyzed to 2-iodobenzaldehyde,
Surprisingly, symmetrical propargylic diols failed to
afford annulation products (entries 7 and 8). Protecting
the hydroxy groups in 2-butyne-1,4-diol did not rectify
the problem, as the corresponding dimethoxy derivative
did not undergo annulation either (entry 9). The reasons
behind these results are unclear and may include coor-
dination of these electron-rich functionalized alkynes to
the palladium catalyst prior to the oxidative addition
step, which diverts the catalyst from the annulation
process.
Internal alkynes bearing a bulky tert-butyl group led
to the expected 2-aminonaphthalenes 9 and 10 with good
(22) Kabalka, G. W.; Wang, L.; Pagni, R. M. Tetrahedron 2001, 57,
8017, and references therein.
(23) (a) Larock, R. C.; Yum, E. K. J . Am. Chem. Soc. 1991, 113, 6689.
(b) Larock, R. C.; Doty, M. J .; Cacchi, S. J . Org. Chem. 1993, 58, 4579.
(c) Larock, R. C.; Yum, E. K.; Doty, M. J .; Sham, K. K. C. J . Org. Chem.
1995, 60, 3270.
(24) Interestingly, the regiochemistry of the Pd-catalyzed hydro-
stannylation of unsymmetrical o-substituted diphenylacetylenes ap-
pears to be controlled by the electronic polarization of the carbon-
carbon triple bond, see: Alami, M.; Liron, F.; Garvais, M.; Peyrat, J .-
F.; Brion, J .-D. Angew. Chem., Int. Ed. 2002, 41, 1578. In that report,
the palladium moiety of H-PdSnR3 adds to the more electron-rich end
of the triple bond, which also happens to be the more sterically
hindered end.
(25) Dyker, G.; Borowski, S.; Henkel, G.; Kellner, A.; Dix, I.; J ones,
P. G. Tetrahedron Lett. 2000, 41, 8259.
344 J . Org. Chem., Vol. 68, No. 2, 2003

Carbopalladation of Nitriles
SCHEME 2a
mediate X during the normal annulation cycle (Scheme
1), the palladium moiety may insert into a benzylic C-H
bond of the alkyl substituent in the 3 position of the
naphthalene and eventually migrate to the alkyl chain
to furnish an alkylpalladium intermediate IX by a
mechanism that differs from the one in Scheme 3. After
a palladium â-hydrogen elimination, IX can produce the
product 14. We have no evidence that allows us to choose
between these two mechanisms. The absence of the side
chain elimination product in entry 21 (Table 2) may
probably be explained by the assumption that the benz-
ylic C-H bond in the more electron-rich analogue of X
is much less activated toward oxidative addition to the
aminopalladium species, thereby precluding its migration
to the alkyl chain.
a
(a) R′R2SiCN, cat. KCN, cat. 18-crown-6, CH2Cl2, r.t.; (b)
Using hindered propargylic alcohols for the annulation
with (2-iodophenyl)acetonitrile, we have made an un-
usual observation that 1,3-benzoxazine derivatives 21 are
produced instead of the anticipated 2-aminonaphthalenes
(eq 4, R ) Ph or Me). Thus, annulation of (2-iodophenyl)-
NaCN, piperidine, aq NaHSO3, 0 °C to r.t.; (c) KCN, aq NaHSO3,
0 °C to r.t.; (d) Ac2O, pyridine, r.t.; (e) 3,4-dihydropyran, cat.
p-TsOH, CH2Cl2, r.t.
SCHEME 3
acetonitrile onto 2-methyl-3-pentyn-2-ol afforded 2,4,4,5-
tetramethyl-1,4-dihydro-2H-naphtho[2,3-d][1,3]oxazine
(21a ) in a 25% isolated yield (entry 1, Table 3). Analogous
1,3-benzoxazine derivative 21b was obtained from 2-meth-
yl-4-phenyl-3-butyn-2-ol in a 35% yield under our original
annulation conditions. When the reaction was run at 130
°C, the yield of 21b increased to 50% (entry 2). Substitut-
ing tri-n-butylamine for triethylamine led to the forma-
tion of 4,4-dimethyl-5-phenyl-2-propyl-1,4-dihydro-2H-
naphtho[2,3-d][1,3]oxazine (21c) in a 38% yield (entry 3).
This result provides clear and unambiguous evidence
that the C2-carbon of the 1,3-oxazine ring comes from
the trialkylamine base, one of the alkyl groups of which
is incorporated into the structure of the final product. It
has also been established that primary alkyl groups are
transferred preferentially from the trialkylamine because
no incorporation of the isopropyl unit was found when
i-Pr2NEt was employed (entry 4).
The following mechanism accounts for the Pd-catalyzed
synthesis of 1,3-benzoxazine derivatives from (2-iodo-
phenyl)acetonitrile, hindered propargylic alcohols, and
trialkylamines (Scheme 5). The marked difference in
steric bulk between the two substituents on the triple
bond of the alkyne causes regioselective formation of the
vinylpalladium intermediate XI shown in Scheme 5,
which proceeds to add to the cyano group and eventually
form the aminopalladium species XII. It is possible that
the Pd(II) in XII is reduced at this point by the trialkyl-
presumably by traces of water present in the DMF or
n-Bu4NCl. Attempted annulation of the THP-protected
cyanohydrin 19e afforded a messy reaction mixture
containing numerous products that could not be identi-
fied.
A possible mechanism for the formation of unsaturated
products 7 and 14 is depicted in Scheme 3. Oxidative
addition of (2-iodophenyl)acetonitrile to Pd(0) and sub-
sequent insertion of 4-octyne furnishes intermediate VI.
This species may undergo â-hydrogen elimination to
produce an allene intermediate VII. Addition of the
Pd-H to the cyano group will generate an acyl-like
organopalladium intermediate VIII, which in turn can
add to the allene to furnish (after imine tautomerization)
a σ- or π-benzylic intermediate IX, which might be
expected to eliminate HPdX to generate the new carbon-
carbon double bond and eventually regenerate the Pd(0)
catalyst. When electron-rich acetonitrile 16 is used (entry
21, Table 2), none of the similar unsaturated product is
observed presumably because the vinylic palladium
intermediate corresponding to VI is more nucleophilic
and attacks the C-N triple bond faster then it undergoes
â-hydrogen elimination.
We also propose an alternative mechanism shown in
Scheme 4. Upon formation of the aminopalladium inter-
SCHEME 4
J . Org. Chem, Vol. 68, No. 2, 2003 345

Tian et al.
SCHEME 5
TABLE 3. Syn th esis of 1,3-Ben zoxa zin e Der iva tives
(eq 4)a
and a Pd(II) species at some other point in the reaction.
The iminium intermediate then undergoes nucleophilic
attack by the two nucleophilic groups present in the
hydroxyl-substituted aminonaphthalene, which results
in formation of the 1,3-oxazine ring. This mechanism may
explain the exclusive transfer of primary alkyl groups
from the amine (entry 4, Table 3), as the iminium ion
species formed from a secondary alkylamine may be too
hindered for facile condensation with the bulky alcohol
moiety.
The reaction between (2-iodophenyl)acetonitrile and
2-methyl-3-pentyn-2-ol afforded another unusual product
besides 21a (eq 5). A complex furan derivative 22 was
isolated in a 26% yield. This compound is clearly a
product of a double alkyne insertion, which presumably
proceeds by the mechanism shown in Scheme 6. Instead
of adding to the C-N triple bond of the cyano group, the
vinylpalladium intermediate XIII apparently inserts a
second molecule of the alkyne to furnish dienylpalladium
species XIV. Intramolecular attack of an OH group then
leads to the final product 22.
a
See the Experimental Section for the reaction conditions.
b
This reaction was run at 130 °C; the yield at 100 °C was 35%.
c 96% conversion of the starting material after 48 h.
amine. This reduction may involve Pd(II) insertion into
the activated R-C-H bond of the amine, reductive
elimination, and fragmentation of the resulting (R-
aminoalkyl)palladium species, which leads to regenera-
tion of Pd(0) and formation of the anticipated 2-amino-
naphthalene product and an iminium moiety (Scheme 5).
It is also possible that this iminium intermediate, which
has been proposed in other Pd-catalyzed transformations
of triethylamine,19a,26 may be formed from triethylamine
Con clu sion s
The palladium-catalyzed annulation of alkynes by (2-
iodophenyl)acetonitrile provides a general route to 3,4-
disubstituted 2-aminonaphthalenes, which are formed in
moderate to very good yields from a variety of internal
alkynes. In many cases, the annulation exhibits excellent
(26) Murahashi, S.-I.; Watanabe, T. J . Am. Chem. Soc. 1979, 101,
7429.
346 J . Org. Chem., Vol. 68, No. 2, 2003

Carbopalladation of Nitriles
SCHEME 6
113.8, 122.3, 125.5, 126.0, 126.9, 127.6, 129.8, 131.5, 131.6,
131.9, 134.3, 139.7, 142.7, 157.9, 158.2 (1 sp3 carbon missing
due to overlap); IR (neat) 3465, 3365, 3052, 3012, 2957, 2837
1613, 1240 cm-1; HRMS m/z 355.15795 (calcd for C24H21NO2,
355.15723).
regioselectivity. The unusual formation of 1,3-benzox-
azine derivatives from certain hindered propargylic al-
cohols has also been observed. This reaction apparently
proceeds with involvement of the trialkylamine base
present in the reaction, which transfers one of its alkyl
groups to the final product.
2-Am in o-6,7-dim eth oxy-3,4-di-n -pr opyln aph th alen e (18).
Obtained as an amber oil in a 62% yield from the reaction of
16 and 4-octyne after purification by column chromatography
using 1:1 hexanes/ethyl acetate: 1H NMR (CDCl3) δ 1.09 (t, J
) 7.6 Hz, 3H), 1.11 (t, J ) 7.6 Hz, 3H), 1.57-1.70 (m, 4H),
2.64-2.69 (m, 2H), 2.93-2.97 (m, 2H), 3.70 (br s, 2H), 3.96 (s,
3H), 3.97 (s, 3H), 6.84 (s, 1H), 6.88 (s, 1H), 7.17 (s, 1H); 13C
NMR (CDCl3) δ 14.8, 14.9, 22.9, 24.0, 30.2, 31.3, 55.7, 55.8,
104.0, 105.2, 108.1, 122.1, 125.7, 129.4, 135.7, 141.7, 146.9,
Exp er im en ta l Section
Gen er a l Meth od s. 1H and 13C NMR spectra were recorded
at 300 and 75 MHz or 400 and 100 MHz, respectively. Thin-
layer chromatography was performed using commercially
prepared 60-mesh silica gel plates (Whatman K6F), and
visualization was effected with short-wavelength UV light (254
nm) and basic KMnO4 solution [3 g of KMnO4 + 20 g of K2-
CO3 + 5 mL of NaOH (5%) + 300 mL of H2O]. All melting
points are uncorrected. All reagents were used directly as
obtained commercially unless otherwise noted. Pd(OAc)2 was
donated by J ohnson Matthey, Inc. and Kawaken Fine Chemi-
cals Co., Ltd.
Gen er a l P r oced u r e for t h e P a lla d iu m -Ca t a lyzed
R ea ct ion of (2-Iod op h en yl)a cet on it r ile a n d In t er n a l
Alk yn es. Palladium acetate (0.0028 g, 0.0125 mmol), Et3N
(0.070 mL, 0.5 mmol), n-Bu4NCl (0.070 g, 0.25 mmol), (2-
iodophenyl)acetonitrile (0.061 g, 0.25 mmol), the alkyne (0.75
mmol), and 5 mL of DMF were placed in a 4 dram vial, which
was heated in an oil bath at 100 °C for 48 h unless indicated
otherwise. The reaction mixture was cooled, diluted with ether,
washed with saturated aqueous NH4Cl, dried over Na2SO4 or
MgSO4, and filtered. The solvent was removed on a rotary
evaporator, and the product was isolated by column chroma-
tography on silica gel. The following compounds, prepared by
the above procedure, have been previously described by us:17
2-amino-3,4-diphenylnaphthalene (1), 2-amino-4-methyl-3-
phenylnaphthalene (5), 2-amino-3-tert-butyl-4-methylnaph-
thalene (9), and 2-amino-3-((E)-1-propenyl)-4-n-propylnaph-
thalene (14).
148.9; IR (neat) 3457, 3370, 3004, 2956, 2869, 2826, 1627 cm-1
;
HRMS m/z 287.18906 (calcd for C18H25NO2, 287.18853).
2,4,4-Tr im eth yl-5-p h en yl-1,4-d ih yd r o-2H-n a p h th o[2,3-
d ][1,3]oxa zin e (21b). Obtained as a white solid in a 50% yield
from the reaction (conducted at 130 °C) of (2-iodophenyl)-
acetonitrile and 2-methyl-4-phenyl-3-butyn-2-ol after purifica-
tion by column chromatography using 4:1 hexanes/ethyl
1
acetate: mp 179-180 °C (EtOH); H NMR (CDCl3) δ 1.22 (s,
3H), 1.44 (d, J ) 5.4 Hz, 3H), 1.54 (s, 3H), 4.43 (br s, 1H), 5.07
(q, J ) 5.4 Hz, 1H), 6.86 (d, J ) 8.7 Hz, 1H), 6.97-7.02 (m,
2H), 7.25-7.32 (m, 3H), 7.43-7.46 (m, 3H), 7.55 (d, J ) 8.1
Hz, 1H); 13C NMR (CDCl3) δ 16.4, 22.2, 29.6, 30.4, 73.5, 110.4,
122.3, 125.1, 125.8, 126.9, 127.3, 127.4, 127.6, 130.9, 132.4,
132.9, 137.2, 140.2, 140.3; IR (neat) 3398, 3062, 3026, 2982,
2931, 1618 cm-1; HRMS m/z 303.16293 (calcd for C21H21NO,
303.16231).
Characterization of all other new annulation products
prepared in this study can be found in the Supporting
Information.
Ack n ow led gm en t. We thank the donors of the
Petroleum Research Fund, administered by the Ameri-
can Chemical Society, for partial support of this re-
search, and J ohnson Matthey, Inc. and Kawaken Fine
Chemicals Co., Ltd., for donating the palladium acetate
and triphenylphosphine.
The following new compounds were prepared by the above
procedure.
2-Am in o-3,4-bis(4-m eth oxyph en yl)n aph th alen e (2). Ob-
tained as a yellow solid in an 80% yield from the reaction of
(2-iodophenyl)acetonitrile and bis(4-methoxyphenyl)acetylene
after purification by column chromatography using 2:1 hex-
anes/ethyl acetate: mp 194-196 °C (EtOH); 1H NMR (CDCl3)
δ 3.76 (s, 3H), 3.77 (br s, 2H), 3.78 (s, 3H), 6.73-6.79 (m, 4H),
7.00-7.05 (m, 4H), 7.10-7.12 (m, 2H), 7.35-7.41 (m, 2H), 7.64
(d, J ) 8.0 Hz, 1H); 13C NMR (CDCl3) δ 55.1, 108.0, 112.9,
Su p p or tin g In for m a tion Ava ila ble: Preparation and
characterization of the starting materials 16 and 19c and
characterization data for all new compounds (including copies
of 1H and 13C NMR spectra). This material is available free of
charge via the Internet at http://pubs.acs.org.
J O026229+
J . Org. Chem, Vol. 68, No. 2, 2003 347