Skraup-Doebner-Von Miller quinoline synthesis revisited: Reversal of the regiochemistry for γ-aryl-β,γ-unsaturated α-ketoesters

Article abstract of DOI:10.1021/jo060290n

A reversal of the standard regiochemistry of the Skraup-Doebner-Von Miller quinoline synthesis was observed when anilines were condensed with γ-aryl-β,γ-unsaturated α-ketoesters in refluxing TFA. The reaction is proposed to involve 1,2-addition of the ani

Full text of DOI:10.1021/jo060290n

SCHEME 1
Skraup-Doebner-Von Miller Quinoline
Synthesis Revisited: Reversal of the
Regiochemistry for γ-Aryl-â,γ-unsaturated
r-Ketoesters
Yan-Chao Wu,†,‡ Li Liu,† Hui-Jing Li,†,‡ Dong Wang,† and
Yong-Jun Chen*,†
Center for Molecular Science, Institute of Chemistry, Chinese
Academy of Sciences, Beijing 100080, China, and Graduate
School of Chinese Academy of Sciences, Beijing 100049, China
yjchen@iccas.ac.cn
ReceiVed February 12, 2006
or Lewis acids, gives predominantly 2-substituted quinolines
from the reaction of 3-substituted R,â-unsaturated carbonyl
compounds.1,4 One plausible explanation for this regioselectivity
is that the reaction proceeds via 1,4-addition of anilines to R,â-
unsaturated carbonyl compounds, followed by dehydrative ring
closure and oxidation via route I (Scheme 1).1e-g In this case,
an added oxidant1a,b (such as nitrobenzene) or a Schiff’s base1c
aromatizes the 1,2-dihydro intermediate to the final quinoline.
Eisch and Dluzniewski studied the mechanism of the Skraup-
Doebner-Von Miller quinoline synthesis and proposed that
direct Schiff’s base formation might be the critical step in the
A reversal of the standard regiochemistry of the Skraup-
Doebner-Von Miller quinoline synthesis was observed when
anilines were condensed with γ-aryl-â,γ-unsaturated R-ke-
toesters in refluxing TFA. The reaction is proposed to involve
1,2-addition of the anilines to γ-aryl-â,γ-unsaturated R-ke-
toesters to form Schiff’s base adducts, followed by cycliza-
tion and oxidation. The products were unambiguously shown
to the 2-carboxy-4-arylquinolines by spectroscopy and X-ray
crystallographic analysis.
6
reaction mechanism,5 as suggested by Skraup himself (route
II in Scheme 1). Because directly heating the substrates under
the conditions of the Skraup-Doebner-Von Miller quinoline
synthesis first forms a Schiff’s base, the 1,4-addition of aniline
to the R,â-unsaturated carbonyl component (route I) is probably
only a minor pathway.5 Bischler proposed that the Schiff’s bases
undergo the 1,4-addition to another molecule of aniline, followed
by cyclization and oxidation to 2-substituted quinolines via route
IIa (Scheme 1).4 Eisch and Dluzniewski5 found that heating a
Schiff’s base under strictly anhydrous conditions in DMSO or
acetonitrile led to a putative diazetidinium cation intermediate
which then rearranged rapidly to a 2-substituted quinoline
(Scheme 1, route IIb), a process subsequently supported by
labeling studies.7 In this case, a Schiff’s base aromatizes the
1,2-dihydro intermediate to the final quinoline.7 The lack of
The Skraup-Doebner-Von Miller quinoline synthesis, which
generally refers to the reaction of R,â-unsaturated carbonyl
compounds with anilines to give quinolines, has been of great
value for constructing the quinoline system since its discovery
one and a quarter centuries ago.1 Many protocols for this
reaction have been developed because of the importance of
quinolines as pharmaceuticals, 2 ligands, and functional materi-
als.3 It is well documented that the Skraup-Doebner-Von
Miller synthesis, which is often carried out using protic acids
1,4-6
formation of 4-substituted quinolines in these reactions
(2) (a) Robert, A.; Dechy-Cabaret, O.; Cazelles, J.; Meunier, B. Acc.
Chem. Res. 2002, 35, 167. (b) Siim, B. G.; Atwell, G. J.; Anderson, R. F.;
Wardman, P.; Pullen, S. M.; Wilson, W. R.; Denny, W. A. J. Med. Chem.
1997, 40, 1381. (c) Sadana, A. K.; Mirza, Y.; Aneja, K. P.; Prakash, O.
Eur. J. Med. Chem. 2003, 38, 533. (d) Oliva, A.; Meepagala, K. M.; Wedge,
D. E.; Harries, D.; Hale, A. L.; Aliotta, G.; Duke, S. O. J. Agric. Food
Chem. 2003, 51, 890. (e) Chen, J. J.; Drach, J. C.; Townsend, L. B. J. Org.
Chem. 2003, 68, 4170.
(3) (a) Francio`, G.; Faraone, F.; Leitner, W. Angew. Chem., Int. Ed. 2000,
39, 1428. (b) Blaser, H. U.; Jalett, H. P.; Lottenbach, W.; Studer, M. J.
Am. Chem. Soc. 2000, 122, 12675. (c) Ferri, D.; Bu¨rgi, T. J. Am. Chem.
Soc. 2001, 123, 12074. (d) Vayner, G.; Houk, K. N.; Sun, Y.-K. J. Am.
Chem. Soc. 2004, 126, 199. (e) Bojinov, V. B.; Grabchev, I. K. Org. Lett.
2003, 5, 2185.
† Center for Molecular Science, Institute of Chemistry, Chinese Academy of
Sciences.
‡ Graduate School of Chinese Academy of Sciences.
(1) (a) Skraup, Z. H. Ber. Dtsch. Chem. Ges. 1880, 13, 2086. (b) Doebner,
O.; von Miller, W. Ber. Dtsch. Chem. Ges. 1881, 14, 2812. (c) Mills, W.
H.; Harris, J. E. G.; Lambourne, H. J. Chem. Soc. 1921, 119, 1294. (d)
Manske, R. H. F.; Kulka, M. Organic Reactions; Adams, R., Ed.; John
Why & Sons: New York; 1953, Vol. 7, p 59. (e) Weissberger, A.; Taylor,
E. C. In The Chemistry of Heterocyclic Compounds- Quinolines; Gurnos,
J., Ed.; Stonebridge Press: Bristol, 1977; p 93. (f) Matsugi, M.; Tabusa,
F.; Minamikawa, J. Tetrahedron Lett. 2000, 41, 8523. (g) Ranu, B. C.;
Hajra, A.; Dey, S. S.; Jana, U. Tetrahedron 2003, 59, 813. (h) Strell, M.;
Kopp, E. Chem. Ber. 1958, 91, 2854. (i) Martin, von, P.; Winkler, T. HelV.
Chim. Acta 1994, 77, 100. (j) For the latest leading reference on the
mechanism of Skraup-Doebner-Von Miller quinoline synthesis, see:
Denmark, S. E.; Venkatraman, S. J. Org. Chem. 2006, 71, 1668.
(4) Bischler, A. Ber. Dtsch. Chem. Ges. 1892, 25, 2864.
(5) Eisch, J. J.; Dluzniewski, T. J. Org. Chem. 1989, 54, 1269.
(6) Skraup, Z. H. Ber. Dtsch. Chem. Ges. 1882, 15, 897.
10.1021/jo060290n CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/19/2006
6592
J. Org. Chem. 2006, 71, 6592-6595

TABLE 1. Selected Results of Reaction Conditionsa
knowledge, this is the first example of regiochemical reversal
for the standard Skraup-Doebner-Von Miller quinoline syn-
thesis.
A series of substituted anilines 1a-m and γ-aryl-â,γ-
unsaturated R-ketoesters 2a-g were examined to better under-
stand the reaction mechanism as well as to examine the scope
and limitation of the method. Listed in Table 2 (entries 1-19)
are representative results. Either anilines or γ-aryl-â,γ-unsatur-
ated R-ketoesters bearing strong electron-withdrawing groups
such as nitro resulted in decreased yields (entries 5, 12, and
17). There were decreases in yields with 2-substituted anilines
in comparison to 4-substituted analogues (entries 13-18),
indicating the steric effects are greater than the electronic effects.
When 4-aminophenol (1m) was used, 6-hydroxy quinoline 3s
was obtained in 79% yield (entry 19), demonstrating excellent
chemoselectivity. Although this approach was ineffective for
the simple R,â-unsaturated aldehyde and ketone (entry 20), the
use of γ-aryl-â,γ-unsaturated R-ketoesters appears to be a
general and direct route to quinoline derivatives that might not
be easily accessible otherwise. As in the case of 3a and 4a, the
structures of 3o and 3s were assigned from spectral and single-
crystal data.
yields (%)b
entry
reaction conditions
3a
4a
1
2
3
4
5
6
7
8
9
10 mol % of Hf(OTf)4, CH2Cl2, rt, 48 h
HCl (37%, H2O), rt, 24 h
18
<2
<2
<2
<2
<2
18
44
<2
<2
<2
<2
<2
46
hydrogen chloride, -CH2Cl2, rt, 24 h
hydrogen chloride, -CH2Cl2, reflux, 24 h
hydrogen chloride, -PhMe, reflux, 24 h
1 equiv of H2SO4, CH2Cl2, reflux, 24 h
1 equiv of TFA, CH2Cl2, reflux, 24 h
1 equiv of TFA, PhMe, reflux, 24 h
TFA, reflux, 12 h
TFA, reflux, 12 h
HCO2H, reflux, 12 h
AcOH, reflux, 12 h
29
33
61
80
76
<2
<2
<2
<2
<2
10c
11c
12c
a 1a (0.2 mmol) and 2a (0.2 mmol) in the solvents indicated (2 mL).
b Isolated yields based on 1a. c 1a (0.2 mmol) and 2a (0.4 mmol) were
used.
To better understand the regiochemical reversal in this
quinoline synthetic method, we thought it was important to
confirm that Schiff’s base 5 was indeed involved as an
intermediate. For this purpose, we screened various reaction
conditions to obtain 5 by increasing the concentration of
reactants and shortening the reaction time at different temper-
atures. In one case, for example, a mixture of 2,3-dimethylaniline
(1a, 10 mmol) and (3E)-2-oxo-4-phenylbut-3-enoate methyl
ester (2a, 20 mmol) in 10 mL of TFA was stirred at reflux for
6 h to give Schiff’s base 5a (2%) and 2-carboxy-4-arylquinoline
3a (51%, Scheme 2). It is worth mentioning that the intermediate
suggested that the direct cyclization of Schiff’s bases to
4-substituted 1,4-dihydroquinolines (Scheme 1, route IIc) might
be a less-favored pathway.
In the present study, we wished to obtain 2-carboxy-4-
substituted quinolines by increasing the intramolecular cycliza-
tion reactivity of Schiff’s base intermediates via route IIc
(Scheme 1), thereby causing a reverse of the regiochemistry of
the standard Skraup-Doebner-Von Miller quinoline process.
We envisioned that introducing an electron-withdrawing group
on the R,â-unsaturated carbonyl component might increase the
electrophilic reactivity of the Schiff’s base CdC double bond
and facilitate subsequent cyclization.8 Accordingly, (3E)-2-oxo-
4-phenylbut-3-enoate methyl ester (2a) was chosen to react with
2,3-dimethylaniline (1a) under various reaction conditions.
Several representative results are summarized in Table 1. It was
found that the 2-phenyl-4-carboxy quinoline 4a was produced
in 44% yield when a solution of 1a (0.2 mmol), 2a (0.2 mmol),
and Hf(OTf)4 (10 mol %) in 2 mL of dichloromethane was
stirred at room temperature for 48 h. Surprisingly, a small
amount of 2-carboxy-4-phenylquinoline 3a (18%) was also
isolated (entry 1). Hydrochloric acid (entry 2), hydrogen chloride
(entries 3-5), and sulfuric acid (entry 6), often effective for
the Skraup-Doebner-Von Miller quinoline synthesis, were
ineffective for this reaction. It was noted that TFA/solvent
systems were effective for the formation of 3a and 4a in
different ratios (entries 7 and 8). Dramatically, the formation
of 4a was inhibited and the yield of 3a increased to 61% when
TFA alone was used (entry 9) and to 80% when the molar ratio
of 2a to 1a increased from 1:1 to 2:1 (entry 10, isolated yields
were based on 1a). Formic acid was also an effective solvent
for this reaction (entry 11) and might be useful for a large-
scale process. However, TFA was chosen in our investigations
because it led to the best overall isolated yield (entries 7-10).
The structures of 3a and 4a were determined from their spectra
and by single-crystal X-ray analysis. To the best of our
SCHEME 2
corresponding to aza-Michael addition as sometimes proposed
was not detectable. Schiff’s base 5a was isolated and structurally
characterized and was converted to the corresponding quinoline
3a by cyclization and oxidation (air was tentatively proposed
to partly be the oxidant) in refluxing TFA for 2 h (Scheme 2).
Although a more detailed mechanistic investigation is in
progress, these preliminary results suggest that the reaction
might involve 1,2-addition of anilines 1 to γ-aryl-â,γ-unsatur-
ated R-ketoesters 2 in TFA to form the intermediate 5, followed
by direct intramolecular cyclization and oxidation to give
2-carboxy-4-arylquinoline products 3 (Scheme 3).
In contrast to the Schiff’s bases obtained from γ-aryl-â,γ-
unsaturated R-ketoesters in Table 2, Schiff’s base 5t, prepared
according to the reported procedure,5 could not be converted
to 1,4-dihydroquinoline 6t or the corresponding oxidized product
3t in TFA (Scheme 4), confirming the findings of Eisch and
(7) McKillop, A. In ComprehensiVe Heterocyclic Chemistry II; Katritzky,
A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon Press: Tarrytown,
1996; Vol 5, p 219.
(8) Linderman, R. J.; Kirollos, K. S. Tetrahedron Lett. 1990, 31, 2689.
J. Org. Chem, Vol. 71, No. 17, 2006 6593

TABLE 2. Syntheses of 2-Carboxy-4-arylquinolinesa
a 1 (0.2 mmol) and 2 (0.4 mmol) in TFA (2 mL). b Isolated yields based on 1.
6594 J. Org. Chem., Vol. 71, No. 17, 2006

SCHEME 3
8-18 h, after which TFA was distilled out for reuse. The residue
was redissolved in 20 mL of CH2Cl2, and the solution was washed
with 5 mL of saturated aqueous NaHCO3, dried over anhydrous
Na2SO4, filtered, and evaporated under reduced pressure. The
products 3 were isolated by flash chromatography on silica gel
(200-300 mesh) with an ethyl acetate-petroleum ether mixture
(1:5, v/v).
Ethyl 6-Chloro-4-(4-chlorophenyl)quinoline-2-carboxylate (3o):
83.0% yield; white solid; mp ) 220-221 °C; 1H NMR (300 MHz,
CDCl3) δ 8.31 (d, J ) 9.0 Hz, 1H), 8.12 (s, 1H), 7.87 (d, J ) 2.3
Hz, 1H), 7.73 (dd, J ) 9.0, 2.3 Hz, 1H), 7.56 (d, J ) 8.4 Hz, 2H),
7.47 (d, J ) 8.4 Hz, 2H), 4.58 (q, J ) 7.1 Hz, 2H), 1.50 (t, J ) 7.1
Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 165.0, 148.0, 147.7, 146.5,
135.4, 135.2, 135.1, 132.8, 131.2, 130.8, 129.2, 128.2, 124.2, 122.0,
62.5, 14.4; FTIR (KBr) 3031, 2985, 1716, 1597, 1485, 1450, 1377,
1273, 1253, 1141, 1087, 1020, 847, 829, 789 cm-1. Anal. calcd
for C18H13Cl2NO2: C, 62.45; H, 3.78; N, 4.05. Found: C, 62.18;
H, 3.89; N, 3.93.
SCHEME 4
Dluzniewski in DMSO or acetonitrile.5 It appears that when
γ-aryl-â,γ-unsaturated esters 2 are used, as opposed to R,â-
unsaturated ketones or aldehydes, the greater electron-withdraw-
ing character of the ester groups in refluxing TFA facilitates
electrophilic ring closure, and the following oxidation leads to
regiochemical reversal in comparison with the standard Skraup-
Doebner-Von Miller quinoline protocol. Further mechanistic
investigations as well as applications of this method are in
progress in our laboratory.
Acknowledgment. We thank the National Natural Science
Foundation of China (No. 20232010, 20332030) and the Chinese
Academy of Sciences for financial support. We appreciate the
suggestions and corrections from the reviewers.
Supporting Information Available: Experimental procedures;
spectral data for γ-aryl-â,γ-unsaturated R-ketoesters (2), quinolines
(3 and 4), and Schiff’s bases 5; copies of 1H NMR and 13C NMR;
and X-ray crystallographic data of 3a, 3o, 3s, and 4a in CIF format.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Experimental Section
General Procedure to Synthesize 2-Carboxy-4-arylquinolines.
A mixture of an aniline 1 (0.2 mmol) and an γ-aryl-â,γ-unsaturated
R-ketoester 2 (0.4 mmol) in 2 mL of TFA was stirred at reflux for
JO060290N
J. Org. Chem, Vol. 71, No. 17, 2006 6595