Synthesis of 3-substituted-4-hydroxyquinoline N-oxides from the Baylis-Hillman adducts of o-nitrobenzaldehydes

Article abstract of DOI:10.1016/S0040-4020(02)01518-1

The reaction of the Baylis-Hillman adducts 1b-f derived from o-nitrobenzaldehydes in trifluoroacetic acid in the presence of triflic acid (0.2equiv.) afforded 3-substituted-4-hydroxyquinoline N-oxides 2b-e and 2a in good to moderate yields. The reaction mechanism was evidenced by the experiment with 1f, the Baylis-Hillman adduct of 2-nitrobenzaldehyde N-tosylimine, as the one involving N-hydroxyisoxazoline as the key intermediate.

Full text of DOI:10.1016/S0040-4020(02)01518-1

Tetrahedron 59 (2003) 385–390
Synthesis of 3-substituted-4-hydroxyquinoline N-oxides from the
Baylis–Hillman adducts of o-nitrobenzaldehydes
*
Ka Young Lee, Jeong Mi Kim and Jae Nyoung Kim
Department of Chemistry and Institute of Basic Science, Chonnam National University, Kwangju 500-757, South Korea
Received 15 October 2002; accepted 19 November 2002
Abstract—The reaction of the Baylis–Hillman adducts 1b–f derived from o-nitrobenzaldehydes in trifluoroacetic acid in the presence of
triflic acid (0.2 equiv.) afforded 3-substituted-4-hydroxyquinoline N-oxides 2b–e and 2a in good to moderate yields. The reaction
mechanism was evidenced by the experiment with 1f, the Baylis–Hillman adduct of 2-nitrobenzaldehyde N-tosylimine, as the one involving
N-hydroxyisoxazoline as the key intermediate. q 2003 Elsevier Science Ltd. All rights reserved.
The Baylis–Hillman reaction is well known as a coupling
reaction of aldehydes and activated alkenes catalyzed by
tertiary amines or tertiary phosphines.1 The versatility of the
functionality have made the Baylis–Hillman adducts
valuable synthetic intermediates.1 Besides the usefulness
of these Baylis–Hillman adducts themselves, further
derivatization with various nucleophilic reagents towards
synthetically useful compounds has been studied in depth by
us2 and other groups.3 Some papers have been reported on
the formation of heterocyclic compounds including quino-
lines and dihydroquinolines from the Baylis–Hillman
adducts.2d,f,3a,b,4
In these respects, we examined the reaction conditions for
the successful formation of quinoline N-oxides containing
acetyl or benzenesulfonyl group at the 3-position, and wish
to report herein the results. After much trials, we have found
appropriate conditions for the formation of 2b in reasonable
yield by increasing the acidity of the reaction medium with
catalytic amounts of triflic acid (Scheme 1).7
As shown in Table 1, the Baylis–Hillman adducts 1b–f,
which were derived from methyl vinyl ketone, phenyl vinyl
sulfone, and ethyl acrylate, respectively, gave the expected
quinoline N-oxide derivatives 2b–e and 2a in good to
moderate yields in trifluoroacetic acid in the presence of
triflic acid (0.2 equiv.). Without triflic acid the reaction did
not afford the corresponding quinoline N-oxides at all or low
yields of products were formed depending on the substrates.
Recently, we have reported on the synthesis of 3-ethoxy-
carbonyl-4-hydroxyquinoline N-oxides such as 2a from the
easily available Baylis–Hillman adduct 1a in trifluoroacetic
acid (entry 1 in Table 1).4a In the reaction we have proposed
a mechanism involving N-hydroxyisoxazoline inter-
mediate.4a,5 Another mechanism involving 6p-electro-
cyclization pathway was also possible and stated
therein.4a,6 However, any experimental evidence for the
mechanism has not been suggested. Moreover, variation of
the substituent at the 3-position other than ethoxycarbonyl
group was impossible. As an example, the Baylis–Hillman
adduct of 2-nitrobenzaldehyde and methyl vinyl ketone, 1b,
did not produce the expected 3-acetyl-4-hydroxyquinoline
N-oxide (2b) at all in the same reaction conditions (60–
708C in trifluoroacetic acid) of our previous paper.4a
Instead, rearranged allylic alcohol derivative 3 was isolated
in 22% yield (Scheme 1), as in our previous report on the
rearrangement of the Baylis–Hillman adducts.2g
The mechanism for the formation of quinoline N-oxides can
be rationalized as we have already proposed (Scheme 2).4a
The remaining possibility involving electrocyclization
pathway can be excluded by the following observations.
The Baylis–Hillman adduct 1f, derived from N-tosylimine
of 2-nitrobenzaldehyde and ethyl acrylate, afforded
4-hydroxyquinoline derivative 2a in 81% yield. The results
suggested that our proposed reaction mechanism (path a,
Scheme 2)4a would be more plausible than the 6p-electro-
cyclization pathway (path b, Scheme 3).6 As shown in
Scheme 2, 2a can be produced according to the reported
mechanism involving the N-hydroxyisoxazoline inter-
mediate II, whereas the corresponding 3-ethoxycarbonyl-
4-tosylamidoquinoline N-oxide 2f would be formed
according to the electrocyclization pathway.
Keywords: quinoline N-oxides; Baylis–Hillman adducts; o-nitro-
benzaldehydes; triflic acid.
The possibility of formation of allylic alcohol derivative
such as 1a from 1f by the exchange of tosylamido group
with trace amounts of moisture present in the reaction
*
Corresponding author. Tel.: þ82-62-530-3381; fax: þ82-62-530-3389;
e-mail: kimjn@chonnam.chonnam.ac.kr
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)01518-1

386
Table 1. Synthesis of 3-substituted-4-hydroxyquinoline N-oxides 2
Entry B–H adducts (1)
K. Y. Lee et al. / Tetrahedron 59 (2003) 385–390
Conditions
Products (2)
Yield (%)
1
TFA, 60–708C, 20 h
82
4a
2
3
4
TFA, TfOH (0.2 equiv.) 30–408C, 10 h
TFA, TfOH (0.2 equiv.) 40–508C, 20 h
TFA, TfOH (0.2 equiv.) 30–408C, 10 h
78
68
81
5
TFA, TfOH (0.2 equiv.) 30–408C, 10 h
49
6
TFA, TfOH (0.2 equiv.) 30–408C, 10 h
81
medium and subsequent transformation toward quinoline
ring 2a could be ruled out by the supplementary experiment
shown in Scheme 4. The Baylis–Hillman adduct of
benzaldehyde N-tosylimine 1g,2b lack of the ortho-nitro
group, was subjected to the same reaction conditions. As
expected we cannot detect nor isolate the allylic alcohol
derivatives 4 and 5.2g From the results, we could conclude
that the path a is a more reliable one for the formation of
quinoline N-oxides.
In conclusion, we have disclosed the synthesis of some
3-substituted-4-hydroxyquinoline N-oxides 2. We also found
a beneficial experimental evidence for the reaction
mechanism.
1. Experimental
1.1. General procedure
1H NMR (300 MHz) and 13C NMR (75 MHz) spectra were
recorded in CDCl3 and in DMSO-d6. The signal positions
are reported in ppm relative to TMS (d scale) used as an
internal standard. The separations were carried out by flash
column chromatography over silica gel (230–400 mesh
ASTM). Organic extracts were dried over anhydrous
Unfortunately, we could not prepare the corresponding
quinoline N-oxide from the Baylis–Hillman adduct,
2-[hydroxy-(2-nitrophenyl)methyl]acrylonitrile
(1h),
derived from acrylonitrile, in trifluoroacetic acid with or
without triflic acid. Intractable mixtures of products were
observed on TLC.

K. Y. Lee et al. / Tetrahedron 59 (2003) 385–390
387
Scheme 1.
Scheme 2. Mechanism involving N-hydroxyisoxazoline (path a).
Scheme 3. Mechanism involving electrocyclization process (path b).

388
K. Y. Lee et al. / Tetrahedron 59 (2003) 385–390
Scheme 4.
MgSO4 and the solvents were evaporated on a rotary
evaporator under water aspirator pressure. IR spectra are
reported in cm21. Mass spectra were obtained from the
Korea Basic Science Institute (Kwangju branch). Melting
points are uncorrected. The combustion analyses were
carried out at Korea Research Institute of Chemical
Technology, Daejeon, Korea.
1.2.2. 3-[Hydroxy-(2-nitrophenyl)methyl]but-3-en-2-one
(1b).4b Oil; IR (KBr) 3432, 1723, 1674, 1525, 1350 cm21
;
1H NMR (CDCl3) d 2.37 (s, 3H), 5.80 (s, 1H), 6.18 (s, 1H),
6.23 (s, 1H), 7.48 (t, J¼8.1 Hz, 1H), 7.65 (t, J¼7.8 Hz, 1H),
7.77 (d, J¼7.8 Hz, 1H), 7.96 (d, J¼8.1 Hz, 1H); 13C NMR
(CDCl3) d 24.98, 66.03, 123.58, 125.60, 127.48, 127.80,
132.47, 135.60, 146.89, 147.97, 198.77.
1.2. General procedure for the synthesis of Baylis–
Hillman adducts
1.2.3. 3-[(5-Chloro-2-nitrophenyl)hydroxymethyl]but-3-
en-2-one (1c). Oil; IR (KBr) 3439, 1676, 1524, 1343 cm21
;
1H NMR (CDCl3) d 2.39 (s, 3H), 3.42 (br s, 1H), 5.78 (s,
1H), 6.17 (s, 1H), 6.24 (s, 1H), 7.42 (dd, J¼2.1, 9.0 Hz, 1H),
7.79 (d, J¼2.1 Hz, 1H), 7.96 (d, J¼9.0 Hz, 1H); 13C NMR
(CDCl3) d 25.92, 67.18, 126.22, 126.53, 128.63, 129.11,
138.72, 140,29, 146,08, 148.60, 199.68.
General synthetic method of the starting materials was
reported previously4a and further references cited therein: to
a stirred solution of the corresponding o-nitrobenzaldehyde
(2 mmol) and activated vinyl compounds, ethyl acrylate
(0.6 mL), acrylonitrile (0.6 mL), methyl vinyl ketone
(0.6 mL), or phenyl vinyl sulfone (2 mmol in 3 mL of
THF) was added DABCO (225 mg, 0.2 mmol), and the
solution was stirred at rt–708C for 3–7 days. After the usual
workup, pure products 1a–e and 1h were obtained by
column chromatography on silica gel (hexane/ether, 7:3).
N-Tosylamido derivative 1f was prepared from N-tosyl-
imine of o-nitrobenzaldehyde and ethyl acrylate by the same
procedure. Baylis–Hillman adduct 1g was prepared
according to our previous paper.2b
1.2.4. 2-Benzenesulfonyl-1-(2-nitrophenyl)prop-2-en-1-
ol (1d). Oil; IR (KBr) 3482, 1525, 1347, 1306, 1160,
1
1134 cm21; H NMR (DMSO-d6) d 6.03 (s, 1H), 6.14 (s,
1H), 6.32 (br s, 1H), 6.47 (s, 1H), 7.44–7.98 (m, 9H); 13C
NMR (CDCl3) d 67.15, 124.78, 127.04, 128.24, 129.22,
129.30, 129.64, 133.81, 133.95, 134.01, 138.32, 147.38,
151.09.
1.2.5. 2-Benzenesulfonyl-1-(5-chloro-2-nitrophenyl)-
prop-2-en-1-ol (1e). Oil; IR (KBr) 3479, 1603, 1571,
1525, 1342, 1306, 1163 cm21; 1H NMR (CDCl3) d 3.08 (br
s, 1H), 5.60 (s, 1H), 6.01 (s, 1H), 6.44 (s, 1H), 7.32–7.84
(m, 8H); 13C NMR (CDCl3) d 67.01, 126.38, 127.19,
128.27, 129.36, 129.38, 129.75, 134.15, 136.00, 137.99,
140.82, 145.35, 150.44.
1.2.1. 2-[Hydroxy-(2-nitrophenyl)methyl]acrylic acid
ethyl ester (1a).4a Oil; IR (KBr) 3465, 1713, 1528,
1
1351 cm21; H NMR (CDCl3) d 1.22 (t, J¼7.2 Hz, 3H),
3.55 (br s, 1H), 4.17 (qd, J¼1.2, 7.2 Hz, 2H), 5.74 (s, 1H),
6.19 (s, 1H), 6.38 (s, 1H), 7.47 (t, J¼7.8 Hz, 1H), 7.65 (t,
J¼7.8 Hz, 1H), 7.76 (d, J¼7.8 Hz, 1H), 7.95 (d, J¼7.8 Hz,
1H); 13C NMR (CDCl3) d 13.99, 61.19, 67.74, 124.57,
126.29, 128.69, 128.92, 133.49, 136.21, 140.88, 148.39,
165.96.
1.2.6. 2-[(2-Nitrophenyl)(toluene-4-sulfonylamino)-
methyl]acrylic acid ethyl ester (1f). White solid; mp
;
99–1008C; IR (KBr) 1720, 1528, 1349, 1161 cm21 1H

K. Y. Lee et al. / Tetrahedron 59 (2003) 385–390
389
1
NMR (CDCl3) d 1.07 (t, J¼7.1 Hz, 3H), 2.31 (s, 3H), 3.90–
4.01 (m, 2H), 5.61 (s, 1H), 5.97 (d, J¼8.8 Hz, 1H), 6.03 (d,
J¼8.8 Hz, 1H), 6.14 (s, 1H), 7.14 (d, J¼8.5 Hz, 2H), 7.30 (t,
J¼8.1 Hz, 1H), 7.43 (t, J¼6.8 Hz, 1H), 7.57 (d, J¼6.8 Hz,
1H), 7.61 (d, J¼8.5 Hz, 2H), 7.74 (d, J¼8.1 Hz, 1H); 13C
NMR (CDCl3) d 13.84, 21.45, 54.47, 61.28, 124.84, 127.16,
128.65, 128.73, 129.50, 130.13, 133.00, 133.38, 137.23,
137.50, 143.53, 148.09, 165.00.
1525 cm21; H NMR (DMSO-d6) d 7.54–7.66 (m, 3H),
7.87–7.90 (m, 2H), 8.00–8.05 (m, 3H), 8.85 (s, 1H);13C
NMR (DMSO-d6) d 67.01, 126.38, 127.19, 128.27, 129.36,
129.38, 129.75, 134.15, 136.00, 137.99, 140.82, 145.35,
150.44; FAB Mass 336 (Mþþ1). Anal. calcd for C15H10-
ClNO4S: C, 53.66; H, 3.00; N, 4.17. Found: C, 53.41; H,
3.15; N, 4.20.
1.3.5. 3-Hydroxymethyl-4-(2-nitrophenyl)but-3-en-2-one
1
1.2.7. 2-[Hydroxy-(2-nitrophenyl)methyl]acrylonitrile
1
(3). Oil; IR (KBr) 3486, 1661, 1517, 1340, 1233 cm21; H
(1h).4c Oil; IR (KBr) 3439, 2228, 1526, 1348 cm21; H
NMR (CDCl3) d 2.54 (s, 3H), 2.74 (br s, 1H), 4.24 (s, 2H),
7.56–7.77 (m, 3H), 7.97 (s, 1H), 8.24 (d, J¼8.7 Hz, 1H);
13C NMR (CDCl3) d 25.90, 57.77, 125.17, 130,06, 130.64,
131.83, 134.03, 139.52, 140.15, 147.28, 201.31.
NMR (CDCl3) d 2.70 (br s, 1H), 6.02 (s, 1H), 6.14 (s, 1H),
6.17 (s, 1H), 7.56 (t, J¼7.7 Hz, 1H), 7.74 (t, J¼7.7 Hz, 1H),
7.86 (d, J¼8.1 Hz, 1H), 8.04 (d, J¼8.1 Hz, 1H); 13C NMR
(CDCl3) d 68.48, 116.56, 124.06, 124.73, 128.70, 129.41,
132.34, 134.01, 134.27, 147.31.
Acknowledgements
1.3. General procedure for the synthesis of 4-hydroxy-
quinoline N-oxides
This work was supported by the grant (No. R05-2000-000-
00074-0) from the Basic Research Program of the Korea
Science and Engineering Foundation.
To a stirred solution of 1b (221 mg, 1 mmol) in trifluoro-
acetic acid (2 mL) was added triflic acid (30 mg, 0.2 mmol)
at rt, and warmed to 30–408C during 10 h. After cooling to
rt, the reaction mixture was poured into water and extracted
with chloroform (2£30 mL). The organic layers were dried
(MgSO4) and evaporated to give crude 2b. Column
chromatography on silica gel (CH2Cl2/MeOH, 14:1)
afforded analytically pure 2b as a white solid, 158 mg
(78%). Other compounds were synthesized analogously.
References
1. Ciganek, E. Organic Reactions, Wiley: New York, 1997; Vol.
51, pp 201–350. (b) Drewes, S. E.; Roos, G. H. P. Tetrahedron
1988, 44, 4653. (c) Basavaiah, D.; Rao, P. D.; Hyma, R. S.
Tetrahedron 1996, 52, 8001. (d) Langer, P. Angew. Chem. Int.
Ed. 2000, 39, 3049. (e) Kim, J. N.; Lee, K. Y. Curr. Org. Chem.
2002, 6, 627.
1.3.1. 1-(4-Hydroxy-1-oxyquinolin-3-yl)ethanone (2b).
78%; mp 209–2108C; IR (KBr) 3099, 2528, 1672, 1609,
1
1527, 1451 cm21; H NMR (DMSO-d6) d 2.61 (s, 3H),
2. (a) Im, Y. J.; Lee, K. Y.; Kim, T. H.; Kim, J. N. Tetrahedron
Lett. 2002, 43, 4675. (b) Kim, J. N.; Lee, H. J.; Lee, K. Y.;
Gong, J. H. Synlett 2002, 173. (c) Chung, Y. M.; Gong, J. H.;
Kim, T. H.; Kim, J. N. Tetrahedron Lett. 2001, 42, 9023.
(d) Kim, J. N.; Kim, H. S.; Gong, J. H.; Chung, Y. M.
Tetrahedron Lett. 2001, 42, 8341. (e) Kim, J. N.; Im, Y. J.;
Gong, J. H.; Lee, K. Y. Tetrahedron Lett. 2001, 42, 4195.
(f) Kim, J. N.; Lee, H. J.; Lee, K. Y.; Kim, H. S. Tetrahedron
Lett. 2001, 42, 3737. (g) Kim, H. S.; Kim, T. Y.; Lee, K. Y.;
Chung, Y. M.; Lee, H. J.; Kim, J. N. Tetrahedron Lett. 2000, 41,
2613. (h) Lee, H. J.; Kim, H. S.; Kim, J. N. Tetrahedron Lett.
1999, 40, 4363. (i) Lee, H. J.; Seong, M. R.; Kim, J. N.
Tetrahedron Lett. 1998, 39, 6223.
7.50–7.56 (m, 1H), 7.83–7.86 (m, 2H), 8.27 (d, J¼8.1 Hz,
1H), 8.60 (s, 1H); 13C NMR (DMSO-d6) d 31.22, 115.61,
115.84, 125.85, 126.36, 128.32, 133.30, 139.16, 143.79,
173.77, 195.50; ESI MS m/z calcd for C11H9NO3 203.0,
positive mode 226.0 [MþNa]þ. Anal. calcd for C11H9NO3: C,
65.02; H, 4.46; N, 6.89. Found: C, 65.12; H, 4.48; N, 6.78.
1.3.2. 1-(6-Chloro-4-hydroxy-1-oxyquinolin-3-yl)etha-
none (2c). 68%; mp 228–2308C; IR (EtOH) 3427, 2975,
1
2928, 1674, 1599 cm21; H NMR (CDCl3) d 2.60 (s, 3H),
7.88 (d, J¼1.5 Hz, 2H), 8.18 (t, J¼1.5 Hz, 1H), 8.60 (s, 1H),
12.94 (br s, 1H); 13C NMR (DMSO-d6) d 31.05, 116.22,
118.14, 125.31, 129.40, 130.62, 133.19, 137.84, 143.97,
172.50, 195.15; FAB Mass 238 (Mþþ1). Anal. calcd for
C11H8ClNO3: C, 55.60; H, 3.39; N, 5.89. Found: C, 55.57;
H, 3.41; N, 5.80.
3. (a) Basavaiah, D.; Satyanarayana, T. Tetrahedron Lett. 2002,
43, 4301. (b) Basavaiah, D.; Reddy, R. M.; Kumaragurubaran,
N.; Sharada, D. S. Tetrahedron 2002, 58, 3693. (c) Basavaiah,
D.; Sarma, P. K. S. J. Chem. Soc., Chem. Commun. 1992, 955.
ˆ ´
(d) Charette, A. B.; Cote, B.; Monroc, S.; Prescott, S. J. Org.
1.3.3. 3-Benzenesulfonyl-1-oxyquinolin-4-ol (2d). 81%;
mp 289–2908C; IR (KBr) 3101, 2740, 1607, 1532, 1311,
Chem. 1995, 60, 6888. (e) Basavaiah, D.; Krishnamacharyulu,
M.; Hyma, R. S.; Pandiaraju, S. Tetrahedron Lett. 1997, 38,
2141. (f) Chavan, S. P.; Ethiraj, K. S.; Kamat, S. K. Tetrahedron
Lett. 1997, 38, 7415. (g) Perlmutter, P.; Tabone, M.
Tetrahedron Lett. 1988, 29, 949. (h) Lawrence, R. M.;
Perlmutter, P. Chem. Lett. 1992, 305. (i) Foucaud, A.; El
Guemmout, F. Bull. Soc. Chim. Fr. 1989, 403.
1
1291, 1138 cm21; H NMR (DMSO-d6) d 7.50–7.66 (m,
4H), 7.85–7.90 (m, 2H), 8.02–8.12 (m, 3H), 8.84 (s, 1H),
12.90 (br s, 1H); 13C NMR (DMSO-d6) d 116.16, 116.72,
125.88, 126.44, 127.15, 128.26, 129.12, 133.55, 134.13,
139.95, 141.38, 143.17, 170.20; ESI MS m/z calcd for
C15H11NO4S 300.9, positive mode 323.9 [MþNa]þ. Anal.
calcd for C15H11NO4S: C, 59.79; H, 3.68; N, 4.65. Found:
C, 59.59; H, 3.80; N, 4.64.
4. (a) Kim, J. N.; Lee, K. Y.; Kim, H. S.; Kim, T. Y. Org. Lett.
2000, 2, 343. (b) Familoni, O. B.; Kaye, P. T.; Klaas, P. J.
J. Chem. Soc., Chem. Commun. 1998, 2563. (c) Kim, J. N.; Lee,
K. Y.; Ham, H.-S.; Kim, H. R.; Ryu, E. K. Bull. Kor. Chem. Soc.
2001, 22, 135. (d) Bode, M. L.; Kaye, P. T. J. Chem. Soc.,
Perkin Trans. 1 1993, 1809. (e) Bode, M. L.; Kaye, P. T.
1.3.4. 3-Benzenesulfonyl-6-chloro-1-oxyquinolin-4-ol
(2e). 49%; mp 294–2968C; IR (EtOH) 3446, 1725,

390
K. Y. Lee et al. / Tetrahedron 59 (2003) 385–390
J. Chem. Soc., Perkin Trans. 1 1990, 2612. (f) Amri, H.; El
of the reviewers during the course of evaluation of Ref. 4a. The
possibility of thermal electrocyclization pathway was examined
with 1a and 1b at around 2208C in diphenyl ether. However, no
reaction was observed.
Gaied, M. M.; Ben Ayed, T.; Villieras, J. Tetrahedron Lett.
1992, 33, 6159. (g) Amri, H.; El Gaied, M. M.; Ben Ayed, T.;
Villieras, J. Tetrahedron Lett. 1992, 33, 7345. (h) Jungheim,
L. N. Tetrahedron Lett. 1989, 30, 1889.
7. Scrutinizing the reaction mechanism that have proposed in our
previous paper,4a we think the acidity of the reaction medium
might play an important role for the formation of quinoline
N-oxides.
5. Woodrell, C. D.; Kehayova, P. D.; Jain, A. Org. Lett. 1999, 1,
619.
6. The possibility of 6p-electrocyclization was suggested by one