Intramolecular radical additions to quinolines

Article abstract of DOI:10.1016/S0040-4039(01)00321-5

This paper is concerned with intramolecular radical additions to quinolines. Radical additions to C-2, C-3 and C-4 of a quinoline have all been shown to proceed under neutral conditions. In each case formation of heteroaromatic products, rather than dihydroquinolines, was observed (implicating the so called oxidative tin hydride pathway).

Full text of DOI:10.1016/S0040-4039(01)00321-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2907–2910
Intramolecular radical additions to quinolines
David C. Harrowven,a,* Benjamin J. Suttona and Steven Coultonb
aDepartment of Chemistry, The University, Southampton S017 1BJ, UK
bGlaxoSmithKline, New Frontiers Science Park (North), Third Avenue, Harlow, Essex CM19 5AW, UK
Received 25 January 2001; revised 8 February 2001; accepted 21 February 2001
Abstract—This paper is concerned with intramolecular radical additions to quinolines. Radical additions to C-2, C-3 and C-4 of
a quinoline have all been shown to proceed under neutral conditions. In each case formation of heteroaromatic products, rather
than dihydroquinolines, was observed (implicating the so called oxidative tin hydride pathway). © 2001 Elsevier Science Ltd. All
rights reserved.
We recently completed a total synthesis of the alkaloid
toddaquinoline in which an intramolecular addition of
an aryl radical to a pyridine featured as the key step.1
This success prompted us to consider various extensions
of that method. In particular we were keen to deter-
mine whether radical additions to quinolines were
equally facile when conducted intramolecularly at neu-
tral pH as this could be used to synthesise a host of
interesting condensed heteroaromatics.2–4 In this Letter
we present our preliminary observations which show
that intramolecular radical additions to C-2, C-3 and
C-4 of a quinoline are all facile under standard tin
mediated radical cyclisation conditions and result in the
formation of heteroaromatic products (via the so called
oxidative tin hydride pathway)5 rather than
dihydroquinolines.
Our first objective was to see if radical cyclisations to
C-3 of a quinoline could be effected. To that end
quinolines 2 and 4 were prepared and treated with
tributyltin hydride under standard radical forming con-
ditions.6 Notably, alkene 2 required more forcing con-
ditions to induce cyclisation than alkane 4. In both
cases a benzo[a]acridine derivative was provided, show-
ing that these reactions follow the ‘oxidative tin
hydride’ pathway (Scheme 1).
O
O
O
NaH, THF
PPh3 Br 0 °C, 2 h then
+
-
O
O
O
5 eq. Bu3SnH
I
0.6 eq. AIBN, PhMe
100 °C, 72 h, 42%
I
1
N
CHO
N
N
2
3
5
0 °C, 2 h, 92%
H2N-NHTs, NaOAc
THF-H2O, 70 h, 90%
O
O
O
O
1.2 eq. Bu3SnH
I
0.1 eq. AIBN, PhMe
80 °C, 24 h, 70%
N
N
4
Scheme 1.
Keywords: radicals and radical reactions; quinolines; nitrogen heterocycles; cyclisation.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00321-5

2908
D. C. Harrow6en et al. / Tetrahedron Letters 42 (2001) 2907–2910
Cyclisations to C-3 of the quinoline could also be
envisioned when a radical precursor is tethered at C-4
of the quinoline. To probe that reaction course the
alkenes 6 and 7 were synthesised from 4-quinolinecar-
boxaldehyde. Attempts to separate these diastereoiso-
mers proved intractable so the mixture was treated with
tributyltin hydride under standard radical forming con-
ditions. One major component, trans-alkene 12, was
given in 83% yield together with traces (15%) of the
desired cyclised product 11. This suggests that addition
O
I
O
O
NaH, 0 °C, THF, 2 h then
1
+
4-quinolinecarboxaldehyde
THF, 0 °C, 2 h, 65%
O
I
N
1:1 - 6:7 inseparable
N
6
7
1.2 eq. Bu3SnH, 0.2 eq. AIBN PhMe, 80 °C, 18 h
I
O
O
O
O
O
•
•
+
+
•
O
Bu3Sn
N
N
N
10
8
9
O
O
O
O
N
N
11, 15%
12, 83%
Scheme 2.
O
O
O
1.4 eq. Bu3SnH
8 eq. H2N-NHTs
6 + 7
8 eq. NaOAc
THF-H2O, 72 h, 87%
0.2 eq. AIBN, PhMe
80 °C, 48 h, 67%
O
I
N
N
14
13
Scheme 3.
O
O
O
O
Br
1.2 eq. Bu3SnH
N
N
+
+
0.9 eq. AIBN
PhMe, 80 °C
18 h, 88%
O
O
O
O
N
N
8:4:3 - 16:17:18
15
16
17
18
THF-H2O
reflux, 72 h, 59%
H2N-NHTs
NaOAc
O
O
O
O
Br
1.2 eq. Bu3SnH
N
N
+
+
0.1 eq. AIBN
PhMe, 80 °C
72 h, 72% (+25% 19)
3:2:2 - 20:21:22
O
O
O
O
N
N
19
20
21
22
Scheme 4.

D. C. Harrow6en et al. / Tetrahedron Letters 42 (2001) 2907–2910
2909
OH
OH
Br
1 eq. Bu3SnH
n-BuLi, Et2O, -100 ˚C, 2 h
then 2-bromobenzaldehyde
-100 to -60 ˚C, 2 h, 77%
0.1 eq. AIBN, PhMe
80 ˚C, 18 h, 50%
N
N
Br
N
23
24
25
27
MnO2, DCM RT, 24 h, 94%
O
O
1.2 eq. Bu3SnH
0.2 eq. AIBN, PhMe
80 ˚C, 48 h, 75%
N
Br
N
26
Scheme 5.
of tributyltin radical to the alkene 6 outpaces homolysis
of the carbon to iodine bond in this case (Scheme 2).
Acknowledgements
Pleasingly, cyclisation of the corresponding alkane 13
proceeded smoothly to give dihydrobenzo[i ]phen-
anthridine 14 in 67% yield (Scheme 3).
The authors thank GlaxoSmithKline and the EPSRC
for a CASE award (to B.J.S.), the EPSRC National
Crystallographic Service Centre at The University of
Southampton for the X-ray analysis of 5 and Joan
Street for some invaluable NOE studies.
Our next objective was to tether a radical precursor at
C-3 of a quinoline to see if such substrates gave rise to
products derived from ipso-attack or cyclisation to
C-2/C-4. To that end alkene 15 was synthesised and
treated with tributyltin hydride under the aforemen-
tioned conditions.6 Though cyclisation to ben-
zo[k]phenanthridine 16 and benzo[c]acridine 17 was
achieved, considerable quantities of recovered starting
material accompanied these products. Indeed, only
when the reaction was conducted with a near stoichio-
metric quantity of AIBN was all the starting material
consumed! Importantly, and in contrast to radical addi-
tions to pyridines, cyclisation to C-4 of the quinoline
was favoured over cyclisation to C-2.1 This regiochemi-
cal preference was mirrored with alkane 19: treatment
with tributyltin hydride giving a complex mixture of
products including dihydrobenzo[k]phenanthridine 20,
dihydrobenzo[c]acridine 21 and quinoline 22 (Scheme
4).
References
1. (a) Harrowven, D. C.; Nunn, M. I. T.; Blumire, N. J.;
Fenwick, D. R. Tetrahedron Lett. 2000, 41, 6681; (b)
Harrowven, D. C.; Nunn, M. I. T. Tetrahedron Lett. 1998,
39, 5875.
2. For an overview of radical additions to heteroaromatic
bases, see: (a) Minisci, F.; Vismara, E.; Fontana, F. Hete-
rocycles 1989, 28, 489. For recent examples, see: (b) Rus-
sell, G. A.; Rajaratnam, R.; Wang, L. J.; Shi, B. Z.; Kim,
B. H.; Yao, C. F. J. Am. Chem. Soc. 1993, 115, 10596; (c)
Russell, G. A.; Wang, L. J.; Yao, C. F. J. Org. Chem.
1995, 60, 5390.
3. For recent reviews on the synthesis of heterocycles by
radical chemistry, see: (a) Bowman, W. R.; Bridge, C. F.;
Brookes, P. J. Chem. Soc., Perkin Trans. 1 2000, 1; (b)
Aldabbagh, F.; Bowman, W. R. Contemp. Org. Synth.
1997, 4, 261.
4. For related work from our group, see: (a) Harrowven, D.
C.; Lucas, M. C.; Howes, P. D. Tetrahedron Lett. 1999, 40,
4443; (b) Harrowven, D. C.; Hannam, J. C. Tetrahedron
1999, 55, 9341; (c) Harrowven, D. C.; Lucas, M. C.;
Howes, P. D. Tetrahedron Lett. 1999, 40, 1187; (d) Har-
rowven, D. C.; Browne, R. Tetrahedron Lett. 1995, 36,
2861; (e) Harrowven, D. C. Tetrahedron Lett. 1993, 34,
5653.
Finally, it is worth noting that all our attempts to effect
a radical cyclisation leading to the creation of a five
membered ring met with failure. For example, bromides
24 and 26 were each reduced to the corresponding
arene on exposure to tributyltin hydride—suggesting
that the cyclisation pathway is more akin to a 5-endo-
trig than a 5-exo-trig process in such cases (Scheme 5).
In conclusion, we have shown that intramolecular radi-
cal additions to quinolines are facile processes that can
be used to effect the synthesis of various condensed
heterocycles. Radical additions to C-2, C-3 and C-4 of
a quinoline have all been demonstrated and in each
case formation of a heteroaromatic product, rather
than dihydroquinoline, was observed. A rapid, stannyl
radical mediated cis to trans isomerisation of a C-4
tethered styrene has also been uncovered.
5. ‘Oxidative tin hydride reaction’ is a term used to describe
reactions mediated by a trialkyltin hydride in which a
radical cyclisation step is followed by an oxidation step in
the absence of an added oxidising agent. For examples,
see: (a) Bowman, W. R.; Mann, E.; Parr, J. J. Chem. Soc.,
Perkin Trans. 1 2000, 2991; (b) Bowman, W. R.; Heaney,
H.; Jordan, B. M. Tetrahedron 1991, 47, 10119; (c)

2910
D. C. Harrow6en et al. / Tetrahedron Letters 42 (2001) 2907–2910
Escolano, C.; Jones, K. Tetrahedron Lett. 2000, 41, 8951;
(d) de Mata, M. L. E. N.; Motherwell, W. B.; Ujjainwalla,
F. Tetrahedron Lett. 1997, 38, 141; (e) Murphy, J. A.;
cal characteristics. The Wittig reagent 1 and bromoarene
15 were synthesised according to the following sequences
(Scheme 6).
+
-
O
O
O
O
O
O
O
O
a.
b.
c.
OH
OH
Br
PPh3 Br
I
I
I
1
d.
Br
O
+
PPh3 Br
-
O
O
e.
f.
Br
+
15
O
O
O
Br
Br
separable
N
28
Scheme 6. Reagents and conditions: (a) I2, AgOCOCF3, CHCl3, 5 min, 83%;7 (b) PBr3, Et3N, 0°C, DCM–THF, then 40°C, 30 min,
83%;7 (c) PPh3, xylene, 100°C, 24 h, 95%; (d) Br2, AcOH, 0°C, 2 h, 67%;8 (e) PPh3, xylene, 80°C, 5 h, 100%; (f) NaH, THF, 0°C,
2 h, then 3-quinolinecarboxaldehyde, 84%, 3:1 - 15:28.
Sherburn, M. S. Tetrahedron 1991, 47, 4077; (f) Har-
rowven, D. C.; Nunn, M. I. T.; Newman, N. A.; Fenwick,
D. R. Tetrahedron Lett. 2001, 42, 961.
7. Cossy, J.; Tresnard, L.; Pardo, D. G. Eur. J. Org. Chem.
1999, 1925.
8. Padwa, A.; Cochran, J. E.; Kappe, C. O. J. Org. Chem.
6. All new compounds gave satisfactory spectral and analyti-
1996, 61, 3706.
.