Solvent dependent regioselective hydrogenation of substituted quinolines

Article abstract of DOI:10.1055/s-2004-835654

Various substituted quinolines have been reduced under H₂ using Rh/Al₂O₃. Using methanol as solvent leads selectively to the 1,2,3,4-tetrahydroquinoline derivatives whereas in hexafluoroisopropanol the decahydro compounds are obtained.

Full text of DOI:10.1055/s-2004-835654

LETTER
2827
Solvent Dependent Regioselective Hydrogenation of Substituted Quinolines
R
egioselective
a
H
ydroge
n
b
S
ubs
i
tituted
e
Q
uinolinesnne Fache*
Laboratoire de Chimie Organique, Photochimie et Synthèse, Université Claude Bernard Lyon I, UMR CNRS 5181, Bât. Raulin,
43 bd du 11 novembre 1918, 69622 Villeurbanne cedex, France
Fax +33(4)72448136; E-mail: fache@univ-lyon1.fr
Received 21 September 2004
We wish to report herein the selective hydrogenation of
Abstract: Various substituted quinolines have been reduced under
various substituted quinolines into either 1,2,3,4-tetra-
H2 using Rh/Al2O3. Using methanol as solvent leads selectively to
hydroquinoline derivatives or decahydro compounds, de-
pending on the reaction solvent (Figure 1).
the 1,2,3,4-tetrahydroquinoline derivatives whereas in hexafluoro-
isopropanol the decahydro compounds are obtained.
Key words: hydrogenation, substituted quinolines, rhodium, sol-
vent effects, HFIP
50 bar H2, r.t.
R
R
catalyst solvent
N
N
H
or
a
b
1,2,3,4-Tetrahydroquinoline derivatives are important in-
termediates for the synthesis of drugs, agrochemicals and
dyes.1 Numerous examples of direct access to these com-
pounds by reduction of functionalized quinolines have
been already published. Hydrogenation with molecular
hydrogen using transition metal homogeneous2 or
heterogeneous3 catalysis or hydride reduction4 is among
the most widely developed methods.
Figure 1
In the course of our study on the reduction of aromatic
compounds, we showed that hexafluoroisopropanol
(HFIP) was a good solvent for such a reaction with the
colloidal system RuCl3/TOA (trioctylamine) and thus
investigated its potential in the hydrogenation of quino-
lines.8 The results are reported in Table 1.
The decahydro skeleton is also found in biologically ac-
tive molecules such as antihypertensive5 or anticholin-
ergic agents.6 Nevertheless, direct access from quinoline
derivatives is difficult and usually requires acidic solvents
such as acetic- or trifluoroacetic acid.7
We first tested several heterogeneous systems on quino-
line. In methanol–H2O (entry 1) RuCl3 was inactive but in
HFIP–H2O 1,2,3,4-tetrahydroquinoline was selectively
obtained after 2 hours whereas after 24 hours the fully
Table 1 Choice of the Catalyst for the Hydrogenation of Quinoline
Entry
Catalyst
Solvent
Time (h)
Conversion (%)
Products
N
N
H
H
b
a
Composition of the reaction mixture
(%, isolated yield)
1
RuCl3/TOAa
PtO2
MeOH–H2O
HFIP–H2O
6
2
24
0
95
100
–
100
0
–
0
100 (61)
2
3
HFIP
19
100
72
28
Rh/Al2O3
(5%)
MeOH
HFIP
1
14
1
14
19
100
100
50
100
100
100 (92)12
77
100
5
0
23
0
95
0
100 (90)
a TOA = trioctylamine; 2.5% catalyst/quinoline.
SYNLETT 2004, No. 15, pp 2827–2829
0
7
.1
2
.2
0
0
4
Advanced online publication: 08.11.2004
DOI: 10.1055/s-2004-835654; Art ID: D28404ST
© Georg Thieme Verlag Stuttgart · New York

2828
F. Fache
LETTER
reduced product was exclusively obtained. Nevertheless, deca-hydrogenated products whereas in HFIP all the
difficulties in isolating the final product from the reaction quinoline was transformed into the decahydroquinoline.
mixture due to the presence of trioctylamine (TOA) in the We tested our system on various substituted quinolines
medium led us to test other catalytic systems. PtO2 gave and the results are reported in Table 2.11
interesting results (entry 2) but it was Rh/Al2O3 (5%)9
which turned out to be the most efficient catalyst (entry 3).
the tetrahydro stage, whereas HFIP led to a mixture of
Using MeOH caused the reduction to stop selectively at
Thus in MeOH, after only one hour, all the quinoline was decahydro isomers. Trifluoroethanol (TFE) was also test-
converted into 1,2,3,4-tetrahydroquinoline while in HFIP ed but it led to the same results as methanol (Table 2, entry
the conversion was of only 50%. On the contrary, with a 2). These results are of high interest because HFIP avoids
longer reaction time, MeOH led to a mixture of tetra- and the use of strongly acidic conditions and the product can
Table 2 Different Substituted Quinolines Hydrogenated Using Rh/Al2O3 System
Entry Substrate
Solvent
Time (h)
Conversion (%) Products
R
R
N
H
N
H
a
b
Composition of the reaction mixture
(%, 1H NMR)
(isolated yield)
1
1
MeOH
TFE
1
14
1
16
1
100
100
100
100
50
100 (74)13
90
100 (70)
50
50
0
0
10
0
50
50
100 (81)
N
OH
HFIP
14
100
2
2
3
MeOH
HFIP
1
18
1
83
100
55
100
0
0
0
100 (80)13
N
100
0
24
100
100 (80)15
3
MeOH
1
100
82a
0
Br
N
4
4
5
MeOH
HFIP
22h
20h
40
10
10012
100
0
0
N
5
MeOH
1
4
50
70
100
0
0
100 (70)16
N
COOH
22
1
24
100
4
100
50
100
0
50
HFIP
0
100 (65)17
6
6
7
MeOH
1
7
18
1
18
48
23
93
100
19
100
100
10012
74
71
0
0
0
0
0
0
b
HFIP
0
N
0
6014
7
MeOHc
HFIPc
24
24
0
100
0
0
0
100 (65)7
HO
N
N
a Debrominated tetrahydroquinoline was also isolated in 18% yield.
b The rest of the converted starting material was transformed into 4-methyl-5,6,7,8-tetrahydro-
quinoline 6c.
c 4 Mol% catalyst used.
N
6c
Synlett 2004, No. 15, 2827–2829 © Thieme Stuttgart · New York

LETTER
Regioselective Hydrogenation of Substituted Quinolines
2829
be isolated in its pure form by simple filtration and eva-
poration of the solvent (bp 59 °C). Surprisingly, in HFIP,
irrespective of the substrate, the reaction seemed to begin
slowly but after around 24 hours the entire product was to-
tally reduced, whereas in MeOH even extended reaction
times did not lead to complete conversion of the tetrahy-
dro- into the decahydro derivative. Products substituted
on the benzene ring were easily reduced following this
pattern (entries 1 and 2). In MeOH (entry 3) after one hour
6-bromoquinoline was fully reduced into its tetrahydro-
derivative accompanied with only 18% of the dehaloge-
nated product. Longer reaction times led only to dehalo-
genated products. In HFIP, after 1 hour, a complex
mixture of type a and b products plus their dehalogenated
analogs were obtained. Products substituted on the pyri-
dine ring at the C-2 were more difficult to reduce and re-
quired longer reaction times. Steric effects do not appear
to play a role as the carboxylic acid derivative 5 (entry 5)
was more easily hydrogenated than the methyl one 4 (en-
try 4). As for the 4-methylquinoline (6), it is noteworthy
that the 5,6,7,8-tetrahydroderivative was selectively ob-
tained in HFIP after 18 hours and required very long reac-
tion time (>48 h) to be converted into the decahydro
derivative. This product is known to be difficult to re-
duce.18 Moreover, it was the first time we obtained such
an isomer. This reactivity is notable as, usually, the same
acidic conditions are required to obtain either 5,6,7,8-tet-
rahydro or decahydro derivatives. Thus, we had expected
to obtain mixtures of both the 5,6,7,8-tetrahydroquino-
lines and the decahydro compounds in HFIP and not the
1,2,3,4-isomers. Thus, we assume that the acidic proper-
ties of HFIP are not the main explanation for the selectiv-
ity measured in this solvent.
References
(1) Katritzky, A. R.; Rachwal, S.; Rachwal, B. Tetrahedron
1996, 52, 15031; and references cited therein.
(2) (a) Wang, W.-B.; Lu, S.-M.; Yang, P.-Y.; Han, X.-W.; Zhou,
Y.-G. J. Am. Chem. Soc. 2003, 125, 10536. (b)Yang, P.-Y.;
Zhou, Y.-G. Tetrahedron: Asymmetry 2004, 15, 1145.
(c) Lu, S.-M.; Han, X.-W.; Zhou, Y.-G. Adv. Synth. Catal.
2004, 346, 909.
(3) Hönel, M.; Vierhapper, F. W. J. Chem. Soc., Perkin Trans. 1
1980, 1933.
(4) (a) Fujita, K.-I.; Kitatsuji, C.; Furukawa, S.; Yamaguchi, R.
Tetrahedron Lett. 2004, 45, 3215. (b) Srikrishna, A.;
Reddy, T. J.; Viswajanani, R. Tetrahedron 1996, 52, 1631.
(5) Wright, G. C.; White, R. E. U.S. Patent 4291163, 1981;
Chem. Abstr. 1981, 96, 19986.
(6) Efange, S. M. N.; Khare, A. B.; Mach, R. H.; Parsons, S. M.
J. Med. Chem. 1999, 42, 2862.
(7) Blaser, H. U.; Jalett, H. P.; Lottenbach, W.; Studer, M. J.
Am. Chem. Soc. 2000, 122, 12675.
(8) Fache, F.; Piva, O. Synlett 2004, 1294.
(9) (a) Campanati, M.; Casagrande, M.; Fagiolino, I.; Lenarda,
M.; Storaro, L.; Battagliarin, M.; Vaccari, A. J. Mol. Catal.
A: Chem. 2002, 184, 267. (b) Campanati, M.; Vaccari, A.;
Piccolo, O. J. Mol. Catal. A: Chem. 2002, 179, 287.
(10) Rh/Al2O3 (5%) was purchased from Acros and used as
received.
(11) All the products have been fully characterized by 1H NMR
(300 MHz) and 13C NMR (75 MHz) and analyses are in
agreement with already published data. Data not fond in the
literature: Product 1b (mixture of diastereoisomers): 1H
NMR (300 MHz, CDCl3): d = 0.94–1.90 (m, 11 H), 2.18 (t,
J = 10.3 Hz) and 2.37 (dd, J = 2.2 Hz, J = 10.7 Hz) for 1H in
the ratio 1:3, 2.69 (m, 2 H), 3.18 (m, 2 H), 3.50 (m)and 3.71
(dt, J = 4.5, J = 11.9 Hz) for 1H in the ratio 1:3. 13C NMR (75
MHz, CDCl3): d = 19.3, 20.4, 23.4, 23.6, 23.7, 24.9, 28.2,
28.9, 30.9, 32.2, 32.4, 34.2 (Ct) and 34.6 (Ct), 45.6 (Cs) and
46.2 (Cs), 60.2 (Ct) and 64.0 (Ct) and 66.9 (Ct), 69.9 (Ct) and
70.4 (Ct) and 70.8 (Ct) and 70.9 (Ct). Product 3a: 1H NMR
(300 MHz, CDCl3): d = 1.83 (t, 2 H, J = 6.0 Hz), 2.63 (t, 2
H, J = 6.0 Hz), 3.19 (t, 2 H, J = 6.0 Hz), 6.34 (d, 1 H, J = 8.1
Hz), 6.97 (m, 2 H). 13C NMR (75 MHz, CDCl3): d = 21.7,
26.9, 42.1, 109.6 (Cq), 116.6, 124.5 (Cq), 129.7, 132.2, 142.7
(Cq).
(12) Pitts, M. R.; Harrison, J. R.; Moody, C. J. J. Chem. Soc.,
Perkin Trans. 1 2001, 955.
(13) Hönel, M.; Vierhapper, F. W. J. Chem. Soc., Perkin Trans. 1
1980, 1933.
(14) Booth, H.; Varghan Griffiths, D.; Jozefowicz, M. L. J.
Chem. Soc., Perkin Trans. 2 1976, 6, 751.
(15) Vierhapper, F. W.; Eliel, E. L. J. Org. Chem. 1975, 40, 2734.
(16) Shuman, R. T.; Ornstein, P. L.; Paschal, J. W.; Gesellden, P.
D. J. Org. Chem. 1990, 55, 738.
Finally, cinchonidine (7), a natural product widely used in
asymmetric synthesis, was fully reduced with this system
in HFIP leading to a mixture of isomers (entry 7).7 In
MeOH no activity was detected, probably for solubility
reasons.
In conclusion, we have shown that it is possible to hydro-
genate selectively substituted quinolines in presence of
Rh/Al2O3 into either 1,2,3,4-tetrahydroderivatives or
decahydroquinolines by a judicious choice of solvent.
Here again, HFIP19 shows interesting properties and
avoids the use of more forcing conditions.20
(17) Swarbrick, M. E.; Lubell, W. D. Chirality 2000, 12, 366.
(18) Uchida, M.; Chihiro, M.; Morita, S.; Kanbe, T.; Yamashita,
H.; Yamasaki, K.; Yabuuchi, Y.; Nakagawa, K. Chem.
Pharm. Bull. 1989, 37, 2109.
Acknowledgment
The author thanks Central Glass Company-Japan, for a generous
gift of HFIP and O. Piva for scientific support.
(19) For a recent review on the use of HFIP in organic synthesis
see: Bégué, J. P.; Bonnet-Delpon, D.; Crousse, B. Synlett
2004, 18.
(20) In a Typical Procedure: Quinoline derivative (1 mmol) and
Rh/Al2O310 (25 mg, 0.01 mmol) were stirred in 1.5 mL of the
selected solvent under 50 bar H2 at r.t. for the requisite time.
The reaction mixture was filtered through celite, the solvent
evaporated and the crude product analyzed by NMR without
further purification.
Synlett 2004, No. 15, 2827–2829 © Thieme Stuttgart · New York