Synthesis of oxazines and N-arylpyrroles by reaction of unfunctionalized dienes with nitroarenes and carbon monoxide, catalyzed by palladium-phenanthroline complexes

Article abstract of DOI:10.1021/jo0260589

The reaction between an unfunctionalized conjugated diene and a nitroarene under CO pressure and at 100 °C, catalyzed by [Pd(Phen)₂][BF₄]₂ (Phen = 1,10-phenanthroline), affords the corresponding hetero-Diels -Alder adduct (oxazine) in up to 91% yields in one pot. If the reaction mixture is then heated to 200 °C, the oxazines are converted into the corresponding N-arylpyrroles in good yields. Pressures as low as 5 bar can be employed, and 0.08% catalyst is sufficient to effect the transformation. The reaction can be equally run by employing the nitroarene or the diene as limiting agent and works well for nitroarenes bearing either electron-withdrawing or mildly electron-donating substituents. A moderate steric hindrance on the nitroarene (o-methyl) is well tolerated, but 1,4-disubstituted-1,3-dienes are not suitable substrates.

Full text of DOI:10.1021/jo0260589

Syn th esis of Oxa zin es a n d N-Ar ylp yr r oles by Rea ction of
Un fu n ction a lized Dien es w ith Nitr oa r en es a n d Ca r bon Mon oxid e,
Ca ta lyzed by P a lla d iu m -P h en a n th r olin e Com p lexes
Fabio Ragaini,* Sergio Cenini, Daniela Brignoli, Michela Gasperini, and Emma Gallo
Dipartimento di Chimica Inorganica, Metallorganica e Analitica and ISTM-CNR, V. G. Venezian 21,
I-20133 Milano, Italy
fabio.ragaini@unimi.it
Received J une 14, 2002
The reaction between an unfunctionalized conjugated diene and a nitroarene under CO pressure
and at 100 °C, catalyzed by [Pd(Phen)2][BF4]2 (Phen ) 1,10-phenanthroline), affords the corre-
sponding hetero-Diels-Alder adduct (oxazine) in up to 91% yields in one pot. If the reaction mixture
is then heated to 200 °C, the oxazines are converted into the corresponding N-arylpyrroles in good
yields. Pressures as low as 5 bar can be employed, and 0.08% catalyst is sufficient to effect the
transformation. The reaction can be equally run by employing the nitroarene or the diene as limiting
agent and works well for nitroarenes bearing either electron-withdrawing or mildly electron-donating
substituents. A moderate steric hindrance on the nitroarene (o-methyl) is well tolerated, but 1,4-
disubstituted-1,3-dienes are not suitable substrates.
In tr od u ction
nitro compound as the aminating reagent, under reducing
conditions (CO pressure). The reaction was catalyzed by
Ru3(CO)12 in the presence of Ar-BIAN ligands (Ar-BIAN
) bis(arylimino)acenaphthene). It is important to note
that the reaction does not produce salts as byproducts.
The production of such byproducts is becoming increas-
ingly unacceptable for industrial processes. Nicholas and
co-workers have reported on a similar reaction.8 When
we applied the same catalytic system to 1,3-dienes, in
addition to the expected allylic amine, two other products
were formed: the hetero-Diels-Alder adduct derived
from the intermediate nitrosoarene and the diene and
the N-arylpyrrole.9 It was shown that the hetero-Diels-
Alder adduct (oxazine) and the allylic amine are inde-
pendent primary products of the reaction, but the pyrrole
derives from further reaction of the adduct (Scheme 1).
Despite its novelty, the ruthenium-catalyzed reaction
still presents several drawbacks when applied to the
The synthesis of pyrroles1 has been the focus of much
attention in recent years. The synthesis of hetero-Diels-
Alder adducts derived from nitrosoarenes as dienophiles
(oxazines) has also been investigated, since these prod-
ucts have pharmacological activity themselves or can be
easily transformed into other products.2 However, their
usual synthesis requires the intermediate isolation of
nitroso compounds, which is problematic, although an
approach has also been reported in which oxidation of
an aromatic amine in the presence of a diene results in
the trapping of the intermediately formed nitrosoarene
to give the hetero-Diels-Alder adducts in one pot.3,4
Several synthetic approaches are available for the syn-
thesis of pyrroles, but few of them are applicable to
N-arylpyrroles.1,5 Moreover, most synthetic methods give
best results for pyrroles bearing one or two substituents
in the 2 and 5 positions.5 On the contrary, the procedure
here reported is best suited for pyrroles which are only
substituted in the 3 and 4 positions.We have recently
reported6,7 a new synthetic way to produce allylic amines,
employing a simple unactivated olefin and an aromatic
(5) (a) N-arylation of pyrrole by aryl halides, catalyzed by palladium
or copper compounds, has been reported recently,5b-d which gives good
yields in the absence of substituents in the 2 and 5 positions. However,
in this case the pyrrole nucleus must be synthesized before and this
procedure has been applied only to unsubstituted pyrrole. (b) Mann,
G.; Hartwig, J . F.; Driver, M. S.; Ferna´ndez-Rivas, C. J . Am. Chem.
Soc. 1998, 120, 827. (c) Hartwig, J . F.; Kawatsura, M.; Hauck, S. H.;
Shaughnessy, K. H.; Alcazar-Roman, L. M. J . Org. Chem. 1999, 64,
5575. (d) Klapars, A.; Antilla, J . C.; Huang, X.; Buchwald, S. L. J . Am.
Chem. Soc. 2001, 123, 7727.
(6) Cenini, S.; Ragaini, F.; Tollari, S.; Paone, D. J . Am. Chem. Soc.
1996, 118, 11964.
(7) Ragaini, F.; Cenini, S.; Tollari, S.; Tummolillo, G.; Beltrami, R.
Organometallics 1999, 18, 928.
(8) (a) Srivastava, A.; Nicholas, K. M. J . Chem Soc., Chem. Commun.
1998, 2705. (b) Kolev-Veetil, M. K.; Khan, M. A.; Nicholas, K. M.
Organometallics 2000, 19, 3754. (c) Srivastava, A.; Kolev-Veetil, M.
K.; Nicholas, K. M. Tetrahedron Lett. 2002, 43, 931. (d) Penoni, A.;
Volkmann, J .; Nicholas, K. M. Org. Lett. 2002, 4, 699. (e) Penoni, A.;
Nicholas, K. M. Chem. Commun. 2002, 484.
(1) Reviews: (a) Gilchrist, T. L. J . Chem. Soc., Perkin Trans. 1 1998,
615. (b) Korostova, S. E.; Mikhaleva, A. I.; Vasil’tsov, A. M.; Trofimov,
B. A. Russ. J . Org. Chem. 1998, 34, 911. (c) Korostova, S. E.;
Mikhaleva, A. I.; Vasil’tsov, A. M.; Trofimov, B. A. Russ. J . Org. Chem.
1998, 34, 1691. (d) Sundberg, R. J . In Comprehensive Heterocyclic
Chemistry II; Katritzky, A. S., Rees, C. W., Scriven, E. F. V., Eds.;
Pergamon: Oxford, U.K., 1996; Vol. 2, p 207. (e) Gilchrist, T. L. J .
Chem. Soc., Perkin Trans. 1 1999, 2849. (f) Bean, G. P. In The
Chemistry of Heterocyclic Compounds; J ones, A. R., Ed.; Wiley: New
York, 1990; Vol. 48, Part I, Chapter. 2, p 105.
(2) Reviews: (a) Boger, D. L.; Weinreb, S. N. Hetero Diels-Alder
Methodology in Organic Synthesis; Academic Press: New York, 1987.
(b) Weinreb, S. M.; Staib, R. R. Tetrahedron 1982, 39, 3087.
(3) Møller, E. R.; J ørgensen, K. A. J . Org. Chem. 1996, 61, 5770.
(4) McClure, K. F.; Danishefsky, S. J . J . Org. Chem. 1991, 56, 850.
(9) Ragaini, F.; Cenini, S.; Borsani, E.; Dompe´, M.; Gallo, E.
Organometallics 2001, 20, 3390.
10.1021/jo0260589 CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/10/2002
460
J . Org. Chem. 2003, 68, 460-466

Oxazines and N-Arylpyrroles
SCHEME 1
ratio of 1200 (rather than 30!) can be currently employed
(corresponding to 0.08 mol % catalyst).
(b) No or only trace amounts of allylic amine are
formed.
synthesis of oxazines and pyrroles: (a) Yields are only
moderate to good (40-60%). (b) Usually a 1 mol %
amount of Ru3(CO)12 has to be employed. (c) A large
excess of diolefin must be employed with respect to the
nitroarene to reach the best results. (d) Steric hindrance
is not tolerated either on the nitroarene or on the diene.
(e) A CO pressure of 40 bar must be applied, requiring
the use of a metal autoclave.
Our group has been active for many years in the field
of the carbonylation reactions of nitroarenes to afford
isocyanates, carbamates, and ureas.10 At the moment, the
most active catalytic system for these last reactions is
based on the use of palladium-phenanthroline com-
plexes.10-12 Here we report that palladium-phenanthro-
line complexes are also very active and selective catalysts
for the reaction of nitroarenes with 1,3-dienes, affording
oxazines and pyrroles.
(c) Solvent: Methanol, the solvent commonly employed
in the carbonylation reactions to yield carbamates, can
be used even for the present reaction, but conversion and
selectivity increase if increasing amounts of the less polar
toluene are added to the solvent mixture (with a corre-
sponding decrease in the methanol amount) up to the
point in which the catalyst become insoluble in the
reaction mixture and the reaction stops completely
(entries 1-6). THF and DMF, however, gave better
results, with the second affording a faster reaction
(entries 13, 15).
(d) Additives: The addition of Et3N to the reaction
mixture increases the rate of the reaction in methanol/
toluene (entries 8, 9, 11) and, to a lower extent, in THF
(entries 13, 16, 32, 37-40), but the effect is not observable
for the reaction in DMF, where conversion is complete
anyway (entries 33, 34). The effect on the selectivity is
small, with a maximum for a 100:1 Et3N/Pd mol ratio
(entries 32, 37-40). Clearly, DMF is a basic enough
solvent not to require any additional base. In the pres-
ence of the amine, the rate of the reaction in THF
becames almost equivalent to the one in DMF, with a
slightly higher selectivity (entries 32, 33). Since DMF is
more difficult to separate after the reaction, the combina-
tion THF/Et3N was preferred in this study. In the
presence of Et3N, diethyl ether gave a selectivity even
slightly higher than THF (entries 32, 35) but at a much
reduced rate. Methylene chloride, on the other hand, is
not a suitable solvent, affording a very low conversion
and only trace amounts of oxazine (entry 36). The
addition of an acid (benzoic acid) to the reaction mixture
slowed the reaction (entries 51, 52), although the effect
on selectivity was small. In the absence of diolefins, acids
are known to increase the rate of carbonylation reactions
of nitroarenes.11,12
Resu lts a n d Discu ssion
Syn th esis of Oxa zin es: Op tim iza tion of th e Ex-
p er im en ta l Con d ition s. The complex [Pd(Phen)2][BF4]2
(Phen ) 1,10-phenanthroline) catalyzes the reaction of
nitrobenzene with 2,3-dimethylbutadiene under CO pres-
sure at 60-120 °C to selectively afford the hetero-Diels-
Alder adduct 2-phenyl-4,5-dimethyl-3,6-dihydro-2H-[1,2]-
oxazine (2c). Byproducts of the reaction are aniline and
azo- and azoxybenzene. Diphenylurea is usually formed
in small amounts but becomes the main byproduct when
the reaction is run in methanol or methanol-containing
solvent mixtures (eq 1).
An extensive optimization of the reaction conditions
has been carried out, employing nitrobenzene and 2,3-
dimethylbutadiene as substrates. The results of all the
experiments are reported in Table 1. Only the general
trends are discussed, and a reference is given to the
relevant entries in Table 1. It should be noted that the
conversion of most reactions was intentionally kept below
100% to be able to evidence any effect on both reaction
rate and selectivity.
(e) Olefin amount: The reaction rate is higher if a
larger excess of olefin is present (entries 16-18, 20, 23,
25, 26), but the effect on selectivity is negligible as long
as the olefin is present in at least a 2:1 molar amount
with respect to the nitroarene. In the absence of any
olefin (entry 20), the reaction was very slow; azo- and
azoxybenzene and diphenylurea were formed as principal
products, but a small amount (5% selectivity) of the
trimer of phenyl isocyanate (1,3,5-triphenyl-[1,3,5]triazi-
nan-2,4,6-trione) was also detected and quantified by
HPLC.
The most important observations are the following:
(a) The catalytic system is much more active than the
ruthenium-based one previously reported, and a catalytic
(10) Cenini, S.; Ragaini, F. Catalytic Reductive Carbonylation of
Organic Nitro Compounds; Kluwer Academic Publishers: Dordrecht,
The Netherlands, 1997; and references therein.
(11) (a) Wehman, P.; Borst, L.; Kamer, P. C. J .; van Leeuwen, P.
W. N. M. J . Mol. Catal., A 1996, 112, 23. (b) Wehman, P.; Kamer, P.
C. J .; van Leeuwen, P. W. N. M. Chem. Commun. 1996, 217. (c) Santi,
R.; Romano, A. M.; Panella, F.; Mestroni, G.; Sessanta o Santi, A. J .
Mol. Catal., A 1999, 144, 41.
(f) CO pressure: For reactions run at a 1200:1 catalytic
ratio, the ideal CO pressure is 10 bar (entries 4, 8, 17-
(12) Ragaini, F.; Cenini, S.; Querci, C. It. Pat. Appl. MI2000A
000548, 2000.
J . Org. Chem, Vol. 68, No. 2, 2003 461

Ragaini et al.
a
TABLE 1. Rea ction s betw een 2,3-Dim eth ylbu ta d ien e a n d P h NO2, Ca ta lyzed by [P d (p h en )2](BF 4)2
Phen/ PhNO2/Pd Et3N/Pd diene/
entry Pd mol ratio mol ratio mL
solv
PCO
/
PhNO2
oxazine PhNH2 PhNdNPh PhN(O)NPh (PhNH)2CO
solv
vol/mL T/°C t/h bar conv %b ,c (2c) sel %d,e sel %c, d sel %c,d
sel % c,d
sel %d,e
1
2j
16
16
16
16
16
16
16
16
16
16g
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
32
24
10
16
16
16
16
32
32
16
16
16
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
600
300
600
600
600
600
600
300
600
600
600
600
600
600
600
600
600
1
1
1
1
1
1
2
1
1
1
1
2
2
2
2
2
tol
9
1:8
2.5:6.5 120
4.5:4.5 120
6.5:2.5 120
120
120
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
3
60
60
60
60
60
60
60
30
30
30
30
30
30
30
30
30
30
10
5
10
10
10
10
10
10
10
10
10
10
10
10
1.3
13.7
20.0
54.5
44.3
42.7
72.9
54.6
67.6
59.7
74.8
89.9
75.7
74.4
94.9
94.0
88.5
95.6
86.8
16.2
0
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
8.2
7.5
8.0
12.6
9.9
6.6
6.9
6.2
2.5
7.6
5.6
5.9
7.1
<1
7.7
3.5
18.3
2.4
3.7
3.5
2.1
3.9
f
7.4
4.9
2.5
4.8
3.7
<1
<1
<1
<1
<1
4
2.5
8.6
6.8
5.0
2.7
2.1
1.9
3.5
2.7
1.4
1.4
3.7
2.5
5.1
0.9
<1
<1
MeOH/ tol
MeOH/tol
MeOH/ tol
MeOH/tol
MeOH
MeOH/tol
MeOH/tol
MeOH/tol
MeOH/tol
MeOH/tol
MeOH/tol
THF
66.3
56.7
52.7
51.6
52.2
55.7
54.3
54.8
52.6
57.8
53.8
58.3
56.4
58.6
54.8
54.8
68.7
57.4
36.2
3j
26.5
15.6
10.7
14.2
16.6
9.1
3.5
6.2
7.2
11.2
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
4j
5j
6j
0
4:4
120
120
7j
8j
4.5:4.5 120
4.5:4.5 120
4.5:4.5 120
4.5:4.5 120
4:4
10
9j
100
100
200
200
10j
11j
12j
13
14
15
16
17
18
19
20l
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49k
50k
51h
52h
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
80
3.4
0.9
THF + DMPi 10
DMF
THF
3.4
5.9
3.5
3.7
3.8
3.6
5.2
4.5
3.7
3.1
2.7
4.2
f
3.0
2.0
2.2
2.4
2.0
3.2
8.4
6.4
2.3
<1
<1
1.6
<1
1.8
0.6
10.5
7.6
4.8
3.0
1.8
6.6
f
3.0
1.7
1.9
<1
1.3
<1
<1
1.5
<1
<1
<1
<1
1.3
1.0
2.2
1.2
<1
<1
1.4
1.8
1.1
<1
1.3
<1
1.2
<1
<1
10
10
10
10
10
10
20
30
30
30
30
30
30
30
30
30
30
30
30
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
1.5 THF
1.5 THF
1.5 THF
THF
1.5 THF
1.5 THF
1.5 THF
1.5 THF
0.75 THF
100
66.0
70.8
79.3
80.8
67.0
83.5
73.5
85.5
84.5
67.6
83.1
76.9
72.6
70.2
80.5
<1
75.8
79.0
76.2
76.2
74.4
77.7
76.0
74.5
83.0
86.5
79.5
90.6
80.5
f
100
100
100
100
100
100
99.0
100
86.0
98.8
87.9
100
100
27.5
7.8
86.1
87.7
93.0
85.0
97.8
94.8
71.9
92.5
99.6
99.9
98.5
70.5
99.5
34.8
97.9
51.6
22.5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
THF
THF
THF
THF
THF
THF
THF
DMF
100
110
60
100
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
100 1.5 10
200
200
200
300
100
50
DMF undist 30
Et2O
CH2Cl2
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
<1
3.1
2.2
2.6
3.3
2.9
2.3
1.7
0.9
1.9
1.4
2.2
1.6
2.6
1.3
1.1
2.9
4.6
600
600
600
600
600
600
600
900
1200
1200
600
600
600
100
100
100
100n
100
100
100
200
200
100
100
100
100
100
6
6
6
6
10
5
10
10
10
100 10
100 0.75 10
100 10
100 1.5 10
100 10
6
84.0
65.0
90.2
f
f
53m 18
600
100
6
a
b
Experimental conditions: [Pd(Phen)2][BF4]2 ) 4.5 mg, 7.0 × 10-3 mmol. tol ) toluene. Calculated with respect to the starting
nitroarene. c Measured by GC. Calculated with respect to reacted nitroarene. e Measured by 1H NMR, employing 2,4-dinitrotoluene as
d
an internal standard or by HPLC, with benzophenone as an internal standard. f Not determined. Employing [Pd(TMPhen)2][BF4]2
g
h
(TMPhen ) 3,4,7,8-tetramethyl-1,10-phenanthroline) and TMPhen in place of the unsubstituted compounds. Without Et3N but in the
presence of PhCOOH (mol ratio PhCOOH/Pd ) 100). i DMP ) 2,2-dimethoxypropane (0.5 mL). j In the presence of methanol, small amounts
(1-6% selectivity) of methyl phenylcarbamate were also detected. [Pd(Phen)2][BF4]2 ) 2.25 mg, 3.5 × 10-3 mmol. The trimer of phenyl
k
l
isocyanate was also formed (selectivity. 5%). Pd(dba)2 (7.0 × 10-3 mmol) was employed as catalyst in place of [Pd(Phen)2][BF4]2. Bu3N
m
n
was employed instead of Et3N.
19) both in term of conversion and selectivity. However,
the worse results obtained at 5 bar have turned out to
result from an excessive CO consumption with respect
to the available amount. When two reactions are com-
pared, conducted with a lower nitrobenzene amount
(entries 45, 46), the results obtained at 5 and 10 bar are
almost indistinguishable, with the lower pressure afford-
ing an even slightly better performance. This point is very
important, since 5-10 bar is a pressure that can be
withstood even by glass autoclaves of the kind commonly
employed to effect hydrogenation reactions and that are
much more widespread in common organic synthesis
laboratories than are metal autoclaves. Thus, the syn-
thetic utility of the method is greatly enhanced. Use of 1
atm of CO (in a common flask) was attempted. However,
for this specific reaction we also had to lower the reaction
temperature to 60 °C to avoid boiling of the solvent and
of the olefin, and at this low temperature and pressure
the reaction did not proceed at all. The increase in
conversion on lowering the pressure below 60 bar is
surprising, since exactly the reverse occurs in the absence
of the diene.12 At the moment, the best explanation for
462 J . Org. Chem., Vol. 68, No. 2, 2003

Oxazines and N-Arylpyrroles
TABLE 2. Syn th esis of Oxa zin es fr om
this observation is that the diene must coordinate to the
palladium center and a competition exists between the
diene and CO.
(g) Solvent amount: Increasing the solvent amount
from 10 to 30 mL slightly increases rate and selectivity,
but the effect is at the limits of the experimental error
(entries 18, 21, 22).
(h) Temperature: The temperature at which the best
selectivity is achieved is 100 °C, but temperatures up to
120 °C are also suitable. Higher temperatures must be
avoided, since under these conditions the oxazine starts
to decompose into the pyrrole. At 60 °C the conversion
was not complete even if the amount of starting ni-
trobenzene was halved (entries 26-30).
(i) Ligand: All palladium-phenanthroline-based cata-
lysts need an excess of ligand to be stable under the
reaction conditions. Under the conditions reported, the
best molar ratio Phen/Pd is 16 (entries 38, 41-43), but
the amount is not crucial, provided it is not decreased
too much. We note that, on the basis of our experience
in using the same catalytic system for carbonylation
reactions, the absolute concentration of the phenanthro-
line ligand is more relevant than the molar ratio with
palladium. So the value for the phenanthroline concen-
tration should be kept constant if the palladium amount
or the volume are changed rather than the Phen/Pd ratio.
Use of 3,4,7,8-tetramethyl-1,10-phenanthroline as ligand
gave very similar results, with an even slightly lower
conversion (entries 9, 10) and was not investigated
further. In the absence of the diene, tetramethylphenan-
throline affords better conversion than simple Phen.10-12
(j) Base: Apart from Et3N, Bu3N was also tried (entries
38, 44). The results were similar, with Et3N giving a
slightly better selectivity.
2,3-Dim eth ylbu ta d ien e a n d Differ en t Nitr oa r en esa
ArNO2
ArNH2
sel %c,d
oxazine
sel %d,e
nitroarene
PhNO2
PhNO2
4-ClC6H4NO2
3,4-Cl2C6H3NO2
4-MeC6H4NO2
2-MeC6H4NO2
4-MeOC6H4NO2
conv %b,c
99.6
99.0
96.6
92.0
90.8
91.3
7.5
2.1
2.0
4.0
7.6
4.6
2.9
1.7
83.0 (2c)
85.5 (2c)
81.9 (2b)
80.5 (2a )
91.4 (2d )
83.3 (2f)
traces (2e)
f
a
Experimental conditions: [Pd(Phen)2][BF4] ) 4.5 mg, 7.0 ×
10-3 mmol, molar ratios Pd/Phen/Et3N/ArNO2 ) 1:16:100:600, 2,3-
dimethylbutadiene ) 2 mL, in THF (30 mL) under CO (10 bar) at
100 °C for 6 h. The main byproducts are azo- and azoxyarenes
and diarylureas, which were detected but not quantified by GC-
b
MS. Calculated with respect to the starting nitroarene. c Mea-
d
sured by GC. Calculated with respect to reacted nitroarene.
e Measured by 1H NMR. f Et3N/Pd ) 200.
oxazine. We have failed to identify this intermediate, but
it is apparently not nitrosobenzene, since it was not
detected (by GC) in the reaction mixture after a 45 min
reaction (entry 50).
(o) When Pd(dba)2 was employed as catalyst in place
of [Pd(Phen)2][BF4]2 (PhNO2:Pd ) 600, 6 h) the reaction
stopped at a 22.5% conversion of nitrobenzene and
formation of metallic palladium was clearly observed. The
selectivity in oxazine remained high, 90.2% (entry 53).
When [Pd(Phen)2][BF4]2 is employed as catalyst, the
initial palladium(II) complex is surely reduced to the
zerovalent state during the reaction.10 Thus, it appears
that the BF4- anion has a role in stabilizing the so-formed
complex, although at the moment is not clear how it
operates.
Use of Differ en t Nitr oa r en es. The scope and limita-
tions of the reaction in eq 1 were examined by using a
series of different nitroarenes, with 2,3-dimethylbutadi-
(k) In some cases, we purified the diene by distillation
over sodium before use, but the results are essentially
indistinguishable from those obtained with the unpuri-
fied diene.
1
ene as the reference diene, employing H NMR spectros-
copy to quantify the oxazine and gas chromatography for
the analysis of the nitroarene and aniline. The results
are reported in Table 2.
(l) Addition of an internal drying agent (2,2-dimeth-
oxypropane) to the reaction mixture also gave indistin-
guishable results, indicating that the source of the
hydrogen giving rise to the aniline and urea byproducts
is not adventitious water (entries 13, 14).
Both electron-withdrawing and mildly electron-donat-
ing substituents are tolerated, but with the strongly
electron-donating p-methoxy group, no oxazine was
obtained. Quite importantly, a moderate steric hindrance
on the nitroarene (o-methyl group) is well tolerated. More
severe steric hindrance has not been examined yet. In
the case of the ruthenium-based system, use of o-
nitrotoluene resulted in only trace amounts of the ox-
azine. The ruthenium catalyst also gave a much de-
creased selectivity even for p-nitrotoluene. There are two
reasons for the poor results obtained with p-methoxyni-
trobenzene. The first is that the first step of the activation
of a nitroarene by a transition metal complex is always
an electron transfer from the metal to the nitroarene10
and this transfer is more difficult in the presence of
electron-donating substituents. The second is that under
CO pressure a competition exists between the hetero-
Diels-Alder reaction and a carbonylation reaction to
afford diarylureas (or even carbamates, when the reaction
is run in methanol). The hetero-Diels-Alder reaction is
known to be disfavored by electron-donating substituents
on the aryl ring,13 whereas the reverse occurs for the
carbonylation reaction.
(m) Catalytic ratio: The selectivity increases if the
catalytic ratio is lowered from 1200 to 600 but is
essentially unaffected by further lowering of the ni-
trobenzene amount (entries 22-24). The effect is due to
the high nitrobenzene concentration in the experiment
run at the 1200:1 catalytic ratio. If the same ratio is
obtained by halving the catalyst amount, with respect
to the experiment with a 600:1 ratio, rather than
doubling the nitrobenzene amount, the selectivities are
comparable, even if the reaction slows down and a
complete conversion is not achieved in 6 h. To reach the
same conversion, the reaction time must be increased to
10 h, but the selectivity decreases at such a long reaction
time, probably because the oxazine starts to decompose
(see also later; entries 45, 47-49)
(n) When the reaction is carried out for 1.5 h instead
of 6 h, the conversion is often very high, but the
selectivity in oxazine is lower than for the corresponding
reaction run for a longer time. This implies that either a
relatively stable intermediate or a reversibly formed
byproduct is initially obtained that later converts to the
J . Org. Chem, Vol. 68, No. 2, 2003 463

Ragaini et al.
TABLE 3. Syn th esis of Oxa zin es fr om Differ en t Dien esa
PhNO2/Pd
mol ratio
diene/
mmol
ArNO2
ArNH2
oxazine
sel %d,e
nitroarene
PhNO2
diene
catal/mg
4.5
THF/mL
30
t/h
conv %b,c
sel %c, d
1.1
isoprene
600
17.7
6
93.0
70.5 (4A)
11.8 (4B)
f
42.4 (5A)
15.1 (5B)
traces
4-ClC6H4NO2
PhNO2
1,4-diphenylbutadiene
myrcene
1.5
4.5
10
30
300
600
5.9
17.7
10
6
75.5
94.1
8.8
2.0
4-ClC6H4NO2
cyclohexadiene
1.5
10
600
5.9
6
88.1
15.0
a
Experimental conditions: catalyst [Pd(Phen)2][BF4]2, molar ratios Pd/Phen/Et3N/ArNO2/diene ) 1:16:100:600:2520, under CO (10
b
bar) at 100 °C for 6 h. 4-Chloroazo- and 4-chloroazoxybenzene were detected, but not quantified, by GC-MS. Calculated with respect to
the starting nitroarene. c Measured by GC. Calculated with respect to reacted nitroarene. e Measured by 1H NMR. f Not detected.
d
Use of Differ en t Olefin s. Apart from 2,3-dimethyl-
butadiene, isoprene, myrcene, 1,4-diphenylbutadiene,
and 1,3-cyclohexadiene were also tested as diolefins
(Table 3).
Use of 1,4-diphenylbutadiene gave low conversions and
only trace amounts of hetero-Diels-Alder adduct, which
is at least partly due to a reversibility in the formation
of the hetero-Diels-Alder adduct at high temperature.
We have previously observed this reversibility by heating
the adduct obtained by reaction of nitrosobenzene with
1,4-diphenylbutadiene.9 With cyclohexadiene a high con-
version was observed, but again only trace amounts of
the desired adduct were observed by GC-MS.
Two adducts (A and B) are possible from the reaction
between nitrobenzene and isoprene or myrcene. The
formation of one isomer with respect to the other in the
reaction of nitrosoarenes with dienes has been previously
investigated both from an experimental and a theoretical
point of view.2,14
NMR signals could be made by bidimensional NMR
spectroscopy (see Experimental Section). The minor
isomer is about 7% of the major one in both the products
of the catalytic reaction and in the one employing
nitrosobenzene as substrate. In the case of myrcene, both
isomers 5A,B have been reported15 and were indeed
observed even in our reaction, in a closer ratio with
respect to the isoprene-derived compounds.17 The lower
yields in the myrcene-derived oxazine is due to the
formation of relevant amounts of phellandrene, the
corresponding furane, which was detected, although not
quantified, by GC-MS. We have previously observed the
formation of 2,3-dimethylfurane in the thermal decom-
position of 2,3-dimethylbutadiene derived oxazines,7 but
this product is usually formed in very low yields and at
higher temperatures. Why loss of aniline from 5 is so easy
is not obvious at this stage.
We have previously mentioned that use of an atmo-
spheric CO pressure with 2,3-dimethylbutadiene limited
the temperature to 60 °C and no reaction was observed.
Since myrcene has a relatively high boiling point, a
reaction was also run at atmospheric CO pressure and
100 °C (DMF as solvent) by using this olefin and
nitrobenzene as substrates. However, the reaction was
very slow and even after 25 h only 33% nitrobenzene had
been converted, with a 35% selectivity in oxazine.
Use of Excess Nitr oa r en e. During this and previous
works in this field, we have always operated in the
presence of an excess of diolefin with respect to the
nitroarene. However, in most cases the diene is more
expensive than the nitroarene and the number of com-
mercially available 1,3-dienes is much more limited.
Thus, it would be important to be able to maximize the
yield with respect to the diolefin. To this aim, we have
performed a reaction employing a 2-fold molar amount
of nitroarene with respect to the diene.18 Since the excess
diene is evaporated with the solvent, values for conver-
sion and selectivity cannot be separately given, but the
oxazine yield with respect to the starting 2,3-dimethyl-
butadiene was 90.5%. Note that the selectivity value is
surely higher, since no dimers or oligomers of the diene
Only isomer A has been said to be obtained by reaction
of isoprene (R1 ) H) with either nitrosobenzene15 or
p-chloronitrosobenzene.16 Under our conditions, reaction
of nitrobenzene with isoprene gave indeed 4A as the
largely dominating isomer, as determined by comparison
1
of the H NMR spectrum with the one reported in the
literature for this compound, but some low-intensity
signals were also present, which could not be attributed
to any known compound and which are close to the ones
of 4A and can be attributed to the previously undetected
minor isomer 4B. We thus reexamined the uncatalyzed
hetero-Diels-Alder reaction of nitrosobenzene with 2,3-
dimethylbutadiene and again observed the same pattern
of signal as for the catalytic reaction. Separation of 4A
from 4B could not be achieved by column chromatogra-
phy, but a complete attribution of the corresponding
(17) A similar situation is observed when p-ClC6H4NO2 is employed
as substrate instead of PhNO2. Preliminary data indicate that both
isomers are again formed from isoprene (selectivity %: 66.5 (A); 6.7
(B)) and myrcene (selectivity %: 45.2 (A); 17.5 (B)). However, the
corresponding compounds have not been isolated and fully character-
ized and evidence is based only on the 1H NMR spectra of the crude
mixtures and their similarity with the ones of the nitrobenzene-derived
products.
(18) PhNO2, 1.027 g, 8.342 mmol; 2,3-dimethylbutadiene, 0.476 mL,
348.8 mg, 4.218 mmol; [Pd(Phen)2][BF4]2, 4.5 mg, 7.03 × 10-3 mmol;
mol ratios 1:16:100:600 Pd/Phen/Et3N/butadiene; 100 °C; PCO ) 10 bar;
in 30 mL of THF for 6 h.
(13) (a) Hamer, J .; Ahmad, M.; Holliday, R. E. J . Org. Chem. 1963,
28, 3034. (b) Kresze, G.; Firl, J .; Zimmer, H.; Wollnik, U. Tetrahedron
1974, 20, 1605.
(14) (a) Leach, A. G.; Houk, K. N. J . Org. Chem. 2001, 66, 5192. (b)
McCarrick, M. A.; Wu, Y.-D.; Houk, K. N. J . Org. Chem. 1993, 58, 3330.
(15) Sasaki, T.; Eguchi, S.; Ishii, T.; Yamada, H. J . Org. Chem. 1970,
35, 4273.
(16) Kresze, G.; Korpiun, O. Tetrahedron 1966, 22, 2493.
464 J . Org. Chem., Vol. 68, No. 2, 2003

Oxazines and N-Arylpyrroles
TABLE 4. Syn th esis of N-Ar ylp yr r olesa
could be detected. Conversion of the nitroarene was
85.8%, and the amount reacted in excess with respect to
the diene was transformed mainly into azoxybenzene.
This result is very interesting, since it opens the way to
the use of very expensive or difficult to synthesize dienes.
Such a development would be inconceivable for the
ruthenium-catalyzed reaction, where a large excess of
diene is necessary for the reaction to proceed.
ArNO2
ArNH2
sel %c,d
pyrrole (3)
nitroarene
PhNO2
4-ClC6H4NO2
4-ClC6H4NO2
3,4-Cl2C6H3NO2
4-MeC6H4NO2
2-MeC6H4NO2
conv %b,c
sel %d,e
100
14.1
12.2
13.1
21.2
10.0
2.7
66.0 (3c)
55.6 (3b)
53.2 (3b)
43.8 (3a )
48.4 (3d )
51.5 (3f)
92.9
90.2
98.3
95.3
96.9
f
Syn th esis of N-Ar ylp yr r oles. In our previous paper,9
we reported that when the catalytic reaction between a
nitroarene and 2,3-dimethylbutadiene in the presence of
Ru3(CO)12/Ph-BIAN was performed at 200 °C, the initially
formed oxazine 2 was transformed into the corresponding
N-arylpyrrole 3 and in some cases this transformation
was almost quantitative. The formation of pyrroles by
decomposition of hetero-Diels-Alder adducts is not
unprecedented.1f,2a,19 When we performed a reaction
under the same conditions of the experiments in Table
1, but at 200 °C, extensive oligomerization of the diene
occurred. Palladium-phenanthroline complexes contain-
ing a coordinated alkyl group are known to be active
catalysts for olefin polymerization.20 Apparently, at 200
°C a species active in this reaction can be formed under
our reaction conditions. It should be stressed that at 100
°C, on the other hand, only trace amounts of a dimer of
dimethylbutadiene and no higher oligomer can be de-
tected by GC-MS in the crude reaction mixture. The best
procedure we found to obtain the pyrrole consists of a
two steps-one pot protocol (eq 2).
a
Experimental conditions: [Pd(Phen)2][BF4]2 ) 4.5 mg, 7.0 ×
10-3 mmol, molar ratios Pd/Phen/Et3N/ArNO2 ) 1:16:100:600, 2,3-
dimethylbutadiene ) 1 mL, in THF (30 mL) under CO (10 bar) at
100 °C for 6 h, followed by evaporation of the solvent and then
200 °C for 3 h in benzene (30 mL) under either CO or N2 (10 bar).
The main byproducts are azo- and azoxyarenes and diarylureas,
b
which were detected but not quantified by GC-MS. Calculated
d
with respect to the starting nitroarene. c Measured by GC. Cal-
culated with respect to reacted nitroarene. e Measured by 1H NMR.
f 2,3-Dimethylbutadiene ) 2 mL.
transformation of the oxazine into the pyrrole. The
present palladium system, however, does not appear to
do the same. In an attempt to increase the yield of the
transformation from 2 to 3, we have added Ru3(CO)12 and
Ph-BIAN to the reaction mixture either since the begin-
ning or after the first step and at different CO pressures,
but worse results were obtained in any case. The presence
of the palladium-phenanthroline system appears to
interfere with the ruthenium-catalyzed reaction.
Con clu sion s
In this paper we have reported a new catalytic system
for the synthesis of hetero-Diels-Alder adducts (ox-
azines) and N-arylpyrroles from nitroarenes, unactivated
diolefins, and CO. Selectivities are generally good, and
it should also been considered that alternative synthetic
approaches usually require several steps. In the case of
hetero-Diels-Alder adducts, the traditional synthesis
requires the isolation of a nitrosoarene, whose prepara-
tion often affords lower yields than the global reaction
reported here and is much more experimentally demand-
ing. Moreover, even the anilines employed in most
syntheses are normally prepared by reduction of the
corresponding nitroarenes, so that a further step is
avoided. Last but not least, the present synthesis does
not produce salt byproducts, a feature that is very
important in the light of possible industrial applications.
In this work we employed commercially available
dienes, but it is useful to recall that much progress has
recently been done in the synthesis of a range of 2,3-
disubstituted-1,3-dienes by the metathesis reactions of
mono- or disubstituted alkynes with ethylene, catalyzed
by the Grubbs’ catalyst, RuCl2(PCy3)2(dCHPh), or by
related complexes.21
The reaction was first performed at 100 °C similarly
to what is reported in Table 1; then the autoclave was
vented and the solution evaporated in vacuo in the same
glass liner in which the reaction is performed. At this
point toluene was added as solvent and the reaction
mixture was placed again in the autoclave and heated
at 200 °C for 3 h. Pressure has to be applied even in this
second stage to avoid boiling of the solvent, but it is
indifferent to apply a CO or a dinitrogen pressure. Since
the excess diene has been eliminated in the evaporation
step, no oligomers are formed. The results of a series of
reactions performed by this protocol are reported in Table
4.
Contrary to ruthenium-catalyzed reactions, pyrrole
selectivities are lower than the corresponding oxazine
ones, although it must be noted that the absolute yields
reported in this paper are anyway higher than the
previously reported ones. We have previously observed
that the ruthenium/Ar-BIAN system also catalyzes the
At the moment, we have not yet performed a mecha-
nistic study of this reaction. However, the fact that no
allylic amine is formed and the similar ratio 4A/4B
found for the catalytic reaction of nitrobenzene with 2,3-
(21) (a) Kinoshita, A.; Sakakibara, N.; Mori, M. J . Am. Chem. Soc.
1997, 119, 12388. (b) Kinoshita, A.; Sakakibara, N.; Mori, M. Tetra-
hedron 1999, 55, 8155. (c) Mori, M.; Tonogaki, K.; Nishiguchi, N. J .
Org. Chem. 2002, 67, 224. (d) Smulik, J . A.; Driver, S. T. J . Org. Chem.
2000, 65, 1788. (e) Smulik, J . A.; Driver, S. T. Org. Lett. 2000, 2, 2271.
(f) Smulik, J . A.; Giessert, A. J .; Driver, S. T. Tetrahedron Lett. 2002,
43, 209. (g) Tonogaki, K.; Mori, M. Tetrahedron Lett. 2002, 43, 2235.
(19) Okuro, K.; Dang, T. D.; Khumtaveeporn, K.; Alper, H. Tetra-
hedron Lett. 1996, 37, 2713.
(20) Ittel, S. D.; J ohnson, L. K.; Brookhart, M. Chem. Rev. 2000,
100, 1169.
J . Org. Chem, Vol. 68, No. 2, 2003 465

Ragaini et al.
3f,28 4A,15 5A,B15), with the exception of the minor isomer 4B.
Some additional spectroscopic data can be found in ref 9,
together with some hints about products separation. In the
dimethylbutadiene and for the uncatalyzed reaction of
the same diene with nitrosobenzene strongly point to-
ward a mechanistic picture in which nitrobenzene is
reduced to nitrosobenzene and the latter then reacts with
the diene in an off-metal reaction. Dehydration of the
oxazine to pyrrole also appears to be a purely organic
reaction under the present conditions.
With respect to the previously reported ruthenium-
catalyzed reaction: (a) The selectivity in oxazine has been
improved from 40-60% to 80-90%. The selectivity in
pyrrole has also been improved, although not to such a
large extent. (b) The metal loading has been decreased
from 3% to 0.08%. (c) Pressure has been lowered from
40 to 5 bar, thus allowing for the use of a glass autoclave.
(d) A moderate steric hindrance on the nitroarene is now
tolerated, whereas it was not in the case of the ruthenium
catalyzed reaction. Substrates with a larger steric hin-
drance have not been tested yet. (e) A lower molar excess
of diene is required with respect to the nitroarene, and
equally good results are obtained if the situation is
reversed by operating with an excess of nitroarene. (f)
The only aspect in which no progress was made is the
use of 1,4-disubstituted dienes, which are still unsuitable
substrates for this reaction.
1
following, the H NMR spectrum is anyway reported for some
of these compounds when these data were not available in the
literature or the present spectrum is of significantly better
quality than the published one.
2f: 1H NMR (CDCl3, 298 K) δ, ppm, 1.67 (s, 3 H), 1.75 (s, 3
H), 2.36 (s, 3H), 3.54 (s, 2 H), 4.32 (s, 2 H ), 7.0-7.5 (m, 4 H).
3f: 1H NMR (CDCl3, 298 K) δ, ppm, 2.19 (s, 6 H), 2.35 (s,
3H), 6.61 (s, 2 H), 7.1-7.42 (m, 4 H).
4A,B: Anal. Calcd for C11H13NO: C, 75.40; H, 7.48; N, 7.99.
Found: C, 75.11; H, 7.59; N, 7.67. The attribution of the NMR
signals of the two inseparable isomers has been made by a
two-dimensional 13C-1H NMR spectrum (HMQC with BIRD).
The carbon atoms are numbered as in the following scheme.
The signals due to the aryl groups of the two isomers in both
1
the H and 13C spectra overlap completely. The signal due to
the C5 carbon of 4B was too weak to be confidently assigned.
4A: 1H NMR (CDCl3, 298 K) δ, ppm, 1.83 (s, 3 H, CH3),
3.71 (s, 2 H, -CH2N-), 4.50 (pss, br, 2 H, -CH2O-), 5.65 (pss
br, 1 H, CdC-H), 7.02 (t, J ) 7.3 Hz, 1 H, H-para), 7.15 (d,
J ) 8.6 Hz, 2 H, H-ortho), 7.33 (dd, J 1 ) 8.6, J 2 ) 7.3 Hz, 2 H,
H-meta); 13C NMR (CDCl3, 298 K) δ, ppm, 20.6 (C1), 56.5 (C2),
68.9 (C3), 120.2 (C4), 131.1 (C5), 116.2 (C6), 129.2 (C7), 122.8
(C8), 150.8 (C9).
Exp er im en ta l Section
Gen er a l P r oced u r e. For the general procedure see ref 9.
In addition, [Pd(Phen)2][BF4]2 was prepared by either of two
methods reported in the literature.22
Ca ta lytic Rea ction s. Catalytic reactions were performed
as previously described.7 Reagents amounts are given in the
tables. Those reactions that were repeated showed nitroarene
conversion and selectivities to be reproducible within (1% and
(2%, respectively. When both HPLC and 1H NMR were
employed to quantify the oxazine, the two independently
obtained values always agreed within (2%.
Id en tifica tion of th e Or ga n ic P r od u cts of Ca ta lysis.
The byproducts azo- and azoxybenzene, diphenylurea, and all
anilines are commercial products and were identified by
comparison of their CG or HPLC chromatograms and GC-MS
spectra with those of authentic samples. 1,3,5-Triphenyl-[1,3,5]-
triazinan-2,4,6-trione23 and all oxazines and pyrroles prepared
in this work have been previously reported in the literature
(2a ,9 2b,3 2c,24 2d ,25 2e,18 2f,25 3a ,9 3b,26,27 3c,26,27 3d ,26,27 3e,26,27
4B: 1H NMR (CDCl3, 298 K) δ, ppm, 1.75 (s, 3 H, CH3),
3.80 (pss br, 2 H, -CH2N-), 4.40 (s, 2 H, -CH2O-), 5.65 (pss,
br, 1 H, CdC-H), 7.02 (t, J ) 7.3 Hz, 1 H, H-para), 7.15 (d,
J ) 8.6 Hz, 2 H, H-ortho), 7.33 (dd, J 1 ) 8.6, J 2 ) 7.3 Hz, 2 H,
H-meta); 13C NMR (CDCl3, 298 K) δ, ppm, 18.5 (C1), 52.4 (C2),
72.5 (C3), 117.5 (C4), 116.2 (C6), 129.2 (C7), 122.8 (C8), 150.8
(C9).
5A: 1H NMR (CDCl3, 298 K) δ, ppm, 1.64 (d, J ) 1.4 Hz, 3
H), 1.72 (s, 3H), 2.0-2.3 (m, 4H), 3.75 (s, 2H), 4.53 (d, J ) 1.8
Hz, 2H), 5.14 (m, 1H), 5.67 (s br, 1H), 6.9-7.4 (m, 5H).
5B: 1H NMR (CDCl3, 298 K) δ, ppm, 1.64 (d, J ) 1.4 Hz, 3
H), 1.72 (s, 3H), 2.0-2.3 (m, 4H), 3.84 (s br, 2H), 4.45 (s, 2H),
5.14 (m, 1H), 5.67 (s br, 1H), 6.9-7.4 (m, 5H).
Ack n ow led gm en t. We thank the MURST (Pro-
grammi di Ricerca Scientifica di Rilevante Interesse
Nazionale) for financial support and M. Dompe´ for
performing initial experiments.
(22) (a) Bontempi, A.; Alessio, E.; Chanos, G.; Mestroni, G. J . Mol.
Catal. 1997, 42, 67. (b) Milani, B.; Anzilutti, A.; Vicentini, L.; Sessanta
o Santi, A.; Zangrando, E.; Geremia, S.; Mestroni, G. Organometallics
1997, 16, 5064.
J O0260589
(23) Vorbrueggen, H.; Krolikiewicz, K. Tetrahedron 1994, 50, 6549.
(24) Augelmann, G.; Streith, J .; Fritz, H. Helv. Chim. Acta 1985,
68, 95.
(26) J ones, A., R.; Spotswood, T. McL.; Cheuychit, P. Tetrahedron
1967, 23, 4469.
(27) J ones, R. A. Aust. J . Chem. 1966, 19, 289.
(28) Rocek, J . Collect. Czech. Chem. Commun. 1954, 19, 275.
(25) Wichterle, O. Collect. Czech. Chem. Commun. 1947, 12, 292.
466 J . Org. Chem., Vol. 68, No. 2, 2003