Synthesis of five- and six-membered-ring compounds by environmentally friendly radical cyclizations using kolbe electrolysis

Article abstract of DOI:10.1055/s-0028-1083547

Substituted carbocycles, tetrahydrofurans, and tetrahydropyrans can be efficiently obtained from ω-unsaturated carboxylic acids. Our methodology involves a Kolbe decarboxylation followed by an intramolecular radical cyclization and a radical-radical cross-coupling process. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0028-1083547

LETTER
2815
Synthesis of Five- and Six-Membered-Ring Compounds by Environmentally
Friendly Radical Cyclizations Using Kolbe Electrolysis
I
ntramolecular
r
Cycliza
é
tionof
R
ad
d
icals
G
enera
é
tedbyKolb
r
e
Electr
i
olysis c Lebreux, Ferdinando Buzzo, István E. Markó*
Département de Chimie, Université catholique de Louvain, Place Louis Pasteur 1, 1348 Louvain-la-Neuve, Belgium
Fax +32(10)474168; E-mail: istvan.marko@uclouvain.be
Received 30 April 2008
O
Abstract: Substituted carbocycles, tetrahydrofurans, and tetrahy-
dropyrans can be efficiently obtained from w-unsaturated carboxy-
lic acids. Our methodology involves a Kolbe decarboxylation
followed by an intramolecular radical cyclization and a radical–
radical cross-coupling process.
– 2 e–
R
HO
RCO2H, – 2CO2
5
1
CO2
– e–
– e–
RCO2H
Key words: cyclization, electron transfer, green chemistry, hetero-
cycles, radical reactions
O
O
CO2
Over the past decades, the development and synthetic ap-
plications of radical reactions have expanded at a remark-
able rate.1 Nowadays, the generation of five- and six-
membered ring compounds using radical cyclization
methods has become an efficient, well-established, and
reliable synthetic procedure. In many cases, elaborate
polycyclic structures can be readily assembled using rad-
ical-based technologies.2 However, in spite of numerous
efforts, most of these useful radical-mediated transforma-
tions have been carried out using stoichiometric amounts
of toxic reagents, such as organostannanes3 or organomer-
cury derivatives,4 expensive hydride sources, such as
tris(trimethylsilyl)silane,5 or noxious substrates like the
xanthates.6
2
3
4
Scheme 1 Proposed mechanistic sequence: Kolbe decarboxylation–
radical cyclization–radical capture
est yields observed in these reactions. Indeed, initial at-
tempts to enhance the yield of the ring closure of 1 into 5
proved unsuccessful. We envisioned that this misfortune
could be linked to the nucleophilic behavior of radical in-
termediate 314 and, consequently, that higher yields and
wider scope could be obtained by employing substrates
such as 6, bearing an electron-deficient alkenyl substitu-
ent (Scheme 2).
Radical species can be generated by a number of alterna-
tive methods, including electrochemistry.7 Indeed, recent
studies have reported the generation of primary radicals
by electroreductive techniques usually involving a nickel8
or cobalt electrocatalysis.9 Among the oxidative electro-
organic reactions, the Kolbe electrolysis is likely the most
famous.10
R
O
EWG
EWG
– 2e–, – 2CO2
RCO2H, MeOH
HO
6
7
Scheme 2 Electrochemical cyclization of 6
In this communication, we wish to report the successful
implementation of this electrochemically based cycliza-
tion strategy. At the onset of our work, it was decided to
examine the radical cyclization of the linear substrate 8a
(Figure 1), possessing a carboxylic acid function and a
suitably positioned enoate residue. Electrochemically me-
diated ring closure of 8a should deliver a substituted cy-
clopentane derivative. Precursors 8b–d, containing other
electron-withdrawing groups at the e-position or able to
generate a six-membered ring adduct 8e (Figure 1), will
also be investigated.
This transformation, which involves radical intermedi-
ates, has been used as a simple and environmentally
friendly procedure to obtain long-chain alkanes from
short-chain carboxylic acids.11,12 Moreover, seminal con-
tributions by Schäfer et al. have demonstrated that cyclic
compounds 5 can be obtained by suitable positioning of
an unsaturation in the starting material (Scheme 1).13 It is
noteworthy that capture of intermediate 4 by a radical de-
rived from a co-acid leads ultimately to the overall cre-
ation of two carbon–carbon bonds.
However, despite obvious synthetic interests, the utility of
the Kolbe electrolysis in promoting these cyclizations has
been somewhat neglected most probably due to the mod-
Substrates 8a–d were synthesized in a straightforward
manner, according to the sequence outlined in Scheme 3.
Various electron-withdrawing groups were introduced by
Horner–Emmons olefination of aldehyde 13 and hydro-
genolysis of benzyl esters 14a–d proceeded quantitatively
and chemoselectively.
SYNLETT 2008, No. 18, pp 2815–2820
1
4
.1
1
.2
0
0
8
Advanced online publication: 15.10.2008
DOI: 10.1055/s-0028-1083547; Art ID: D13908ST
© Georg Thieme Verlag Stuttgart · New York

2816
F. Lebreux et al.
LETTER
mediate radical (3, Scheme 1) by the co-acid competes
significantly with the addition of this radical species onto
the nonactivated alkene. In stark contrast, Kolbe decar-
boxylation–radical cyclization of acid 8a, under identical
conditions, afforded exclusively the five-membered ring
ketal 15a in an excellent 90% isolated yield, albeit as a 1:1
mixture of diastereoisomers (Scheme 4). No trace of the
linear product 16a could be detected in this experiment.
O
EWG
O
O
O
HO
R
HO
OEt
O
O
O
8e
8a EWG = Boc, R = H
8b EWG = CN, R = H
8c EWG = Ac, R = H
8d EWG = CO2Et, R = Me
Figure 1 Five- and six-membered carbocycle precursors
O
– 2 e–
HO
O
O
O
O
OEt
O
O
+
O
O
AcOH
b
a
O
O
O
O
OEt
OEt
O
n = 1, 2
77%
n = 1 74%
n = 2 65%
17
18 35%
19 21%
n
n
O
O
Boc
Boc
10
Boc
11
9
– 2 e–
AcOH
HO
O
+
O
O
O
O
O
O
O
O
d
c
OBn
OBn
O
O
O
n = 1 65%
n = 2 88%
n = 1, 2
93%
8a
15a 90%
16a 0%
n
n
13
12
Scheme 4 Kolbe electrolysis–radical cyclization
O
O
R
O
O
The other substrates 8b–e were then submitted to these
conditions and the results are collected in Table 1. Excel-
lent yields and selectivities are observed for substrates 8a
and 8b in the presence of various co-acids. As can be seen
in entry 1, acetic, propanoic, and isobutyric acids are com-
petent co-acids, leading to adducts 15a–c bearing a meth-
yl, ethyl, and isopropyl substituent, respectively. Whilst
the unsaturated nitrile 8b cyclizes smoothly to 15d (entry
2), the corresponding enone 8c only affords the saturated
product 21. This cathodic reduction takes place faster than
the desired anodic oxidation because of the much lower
reduction potential of the enone 8c as compared to sub-
strate 8a (Figure 2).
O
O
e
f
OBn
OH
99%
R
14a 85%
14b 87%
14c 92%
14d 78%
8a–d
Scheme 3 Synthetic procedure for substrates 8a–d. Reagents and
conditions: (a) NaH (1.1 equiv), n-BuLi (1.1 equiv), allyl bromide
(1 equiv; or butenyl bromide for 8e), THF, 0 °C; (b) ethylene glycol,
PTSA (cat.), benzene, reflux; (c) Ti(Oi-Pr)4, BnOH, 100 °C, 3 h;
(d) NaIO4 (2.5 equiv), OsO4 (1 mol%), THF–H2O, toluene, 1 h;
(e) (MeO)2P(O)CH2R (1.1 equiv), KOt-Bu (1.1 equiv), THF, 0 °C;
(f) HCO2NH4 (2 equiv), Pd/C (10 mol%), MeOH.
With the desired precursors in hand, typical Kolbe
conditions15 were employed initially in the electrochemi-
cal step, resulting in modest yields that were rapidly im-
proved by modulating the main parameters pertaining to
this reaction. Hence, the current density, which has a sig-
nificant influence on the overall yield, should be kept be-
tween 80 and 120 mA/cm2. Furthermore, the use of
smooth platinum electrodes, a weakly acidic medium
(partial neutralization of the acids with 5 mol% of KOH),
a high dilution in MeOH (40–80 mM in substrate) and an
excess of the co-acid (5 equiv) provided the best results.
Eventually, it was also noticed that a switching electrical
supply was useful in preventing the fading out of the cur-
rent due to the formation of a layer of black deposit on the
electrodes.16 These optimized conditions were then ap-
plied to substrates 8a and 17.
Figure 2 Reduction potentials of acrylate 8a and enone 8c
The Kolbe decarboxylation–radical cyclization of the
substituted ester 8d proceeded efficiently, affording 15e
and 20 in a combined yield of 80%. The methyl ether 20
originates from the oxidation of the intermediate tertiary
radical akin to 4, into the corresponding carbocation, fol-
Whilst cyclization of 17 proceeded smoothly, affording
the substituted cyclopentane 18 in 35% yield, a significant
amount of linear product 19 (21%) was also generated in
this process. It thus appears that the capture of the inter-
Synlett 2008, No. 18, 2815–2820 © Thieme Stuttgart · New York

LETTER
Intramolecular Cyclization of Radicals Generated by Kolbe Electrolysis
2817
Table 1 Carbocycle Synthesis
Entry
1
Substrate
Co-acid
Products
Yield (%)
O
Boc
Boc
Boc
O
HO
O
90
78
O
OH
O
O
8a
15a
O
O
OH
O
15b
Boc
O
75
65
80
OH
O
O
15c
O
CN
CN
O
O
O
HO
O
2
3
O
OH
OH
OH
O
O
8b
15d
O
O
Ac
HO
O
HO
O
Ac
O
O
21
8c
O
CO2Et
MeO
O
CO2Et
CO2Et
50 (15e)
30 (20)
HO
O
4
5
O
O
O
O
8d
20
15e
O
Boc
O
HO
O
O
64 (15f)
16 (16e)
O
O
O
OH
O
Boc
Boc
15f
16e
8e
lowed by capture by methanol. Finally, six-membered conditions to substrates 22a–d, possessing an ether func-
carbocyles can also be generated under these electro- tion (Figure 3).
chemical conditions, as shown in Table 1, entry 5. In this
case, and for the first time, a small amount of the linear
isomer 16e has also been produced, no doubt due to the
lower rate of cyclization.
The ring closure of 22a and 22b should lead to substituted
tetrahydrofurans whilst 22c and 22d should generate the
corresponding tetrahydropyrans. Substrate 22b was se-
lected in order to verify if an electrogenerated tertiary rad-
In order to expand the scope of this electrochemically me- ical would undergo cyclization whereas 22c would enable
diated radical cyclization methodology, we applied these us to test the importance of the Thorpe–Ingold effect. Car-
Synlett 2008, No. 18, 2815–2820 © Thieme Stuttgart · New York

2818
F. Lebreux et al.
LETTER
boxylic acids 22a–d (Figure 3) were synthesized in 70– results are collected in Table 2. As can be seen, electroly-
80% overall yields, according to general procedures that sis of 22a proceeded smoothly and delivered the desired
shall be reported elsewhere.
tetrahydrofurans 23a–e in good to excellent yields (entry
1). A wide range of functionalized co-acids could be em-
ployed in this transformation, leading to variously substi-
The carboxylic acids 22a–d were then submitted to the
Kolbe electrolysis under our optimized conditions. The
Table 2 Heterocycle Synthesis
Entry
1
Substrate
Co-acid
Products
Yield (%)
86
O
CO2Et
CO2Et
O
HO
O
OH
O
23a
22a
O
CO2Et
72
90
89
O
OH
23b
MeO2C
CO2Et
MeO2C
O
CO2H
23c
CO2Et
CO2H
O
23d
O
O
87
O
O
CO2Et
CO2H
O
23e
F3C
O
CO2Et
CO2Et
0
O
F3C
OH
23f
O
CO2Et
O
HO
2
3
16–24
O
OH
OH
O
O
22b
O
24
CO2Et
CO2Et
O
O
HO
89
25
22c
O
CO2Et
MeO2C
O
O
57a
MeO2C
4
CO2Et
HO
CO2Et
CO2H
O
O
27
26
22d
a Yield of cyclized product 26 (yield of 27: 35–40%).
Synlett 2008, No. 18, 2815–2820 © Thieme Stuttgart · New York

LETTER
Intramolecular Cyclization of Radicals Generated by Kolbe Electrolysis
2819
(3) (a) Davies, A. G. J. Chem. Res. 2006, 3, 141. (b) Büchi, G.;
Wüst, H. J. Org. Chem. 1979, 44, 546. (c) Stork, G.; Mook,
R. J. Am. Chem. Soc. 1983, 105, 3720.
(4) (a) Danishefsky, S.; Chackalamannil, S.; Uang, B.-J. J. Org.
Chem. 1982, 47, 2231. (b) Corey, E.; Shih, C.; Shih, N.-Y.;
Shimoji, K. Tetrahedron Lett. 1984, 25, 5013.
(5) (a) Chatgilialoglu, C.; Ferreri, C.; Gimisis, T. The
Chemistry of Organic Silicon Compounds, Vol. 2;
Rappoport, S.; Apeloig, Y., Eds.; Wiley: London, 1998,
1539. (b) Arya, P.; Samson, C.; Lesage, M.; Griller, D.
J. Org. Chem. 1990, 55, 6248. (c) Gandon, L. A.; Russell,
A. G.; Guveli, T.; Brodwolf, A. E.; Kariuki, B. M.; Spencer,
N.; Snaith, J. S. J. Org. Chem. 2006, 71, 5198.
O
OEt
O
OEt
O
O
HO
O
HO
O
O
22a
22b
O
O
O
HO
OEt
HO
OEt
O
O
22c
22d
(d) Varlamov, V. T.; Denisov, E. T.; Chatgilialoglu, C.
J. Org. Chem. 2001, 66, 6317. (e) Wnuk, S. F.; Garcia, P. I.;
Wang, Z. Org. Lett. 2004, 6, 2047.
Figure 3 Five- and six-membered heterocyclic precursors
(6) (a) Zard, S. Z. Angew. Chem., Int. Ed. Engl. 1997, 36, 673.
(b) Walter, W.; Bode, K. D. Angew. Chem., Int. Ed. Engl.
1967, 6, 281.
tuted heterocycles. Only electron-deficient co-acids, such
as trifluoroacetic acid, failed to couple with 22a, most
probably because of identical polarity of the in situ gener-
ated radicals.
(7) (a) Grimshaw, J. Electrochemical Reactions and
Mechanisms in Organic Chemistry; Elsevier Science:
Amsterdam, 2000. (b) Duñach, E. A.; Esteves, P.; Freitas,
A. M.; Medeiros, M. J.; Olivero, S. Tetrahedron Lett. 1999,
40, 8693. (c) Toyota, M.; Ilangovan, A.; Kashiwagi, Y.;
Ihara, M. Org. Lett. 2004, 6, 3629. (d) Torii, S.; Inokuchi,
T.; Yukawa, T. J. Org. Chem. 1985, 50, 5875. (e) Miranda,
J. A.; Wade, C. J.; Little, R. D. J. Org. Chem. 2005, 70, 8017.
(8) (a) Duñach, E.; Esteves, A. P.; Medeiros, M. J.; Olivero, S.
Tetrahedron Lett. 2004, 45, 7935. (b) Toyota, M.;
Ilangovan, A.; Kashiwagi, Y.; Ihara, M. Org. Lett. 2004, 6,
3629. (c) Torii, S.; Inokuchi, T.; Yukawa, T. J. Org. Chem.
1985, 50, 5875.
(9) (a) Scheffold, R. In Transition Metals in Organic Synthesis;
Scheffold, R., Ed.; Wiley: New York, 1983, 355.
(b) Scheffold, R.; Dike, M.; Dike, S.; Herold, T.; Walder, L.
J. Am. Chem. Soc. 1980, 102, 3642. (c) Scheffold, R.;
Abrecht, S.; Orlinski, R.; Ruf, H.-R.; Stamouli, P.;
Tinembart, O.; Walder, L.; Weymuth, C. Pure Appl. Chem.
1987, 59, 363.
(10) (a) Utley, J. Chem. Soc. Rev. 1997, 26, 157. (b) Torii, S.;
Tanaka, H. In Organic Electrochemistry, 4th ed; Lund, H.;
Hammerich, O., Eds.; Marcel Dekker: New York, 2001, 499.
(11) For the preparation of Kolbe dimers, see: (a) Brown, A. C.;
Walker, J. Justus Liebigs Ann. Chem. 1891, 261, 107.
(b) Fichter, F.; Lurie, S. Helv. Chim. Acta 1933, 16, 885.
(12) For Kolbe cross-coupling reactions, see: (a) Seebach, D.;
Renaud, P. Helv. Chim. Acta 1985, 68, 2342. (b) Kubota,
T.; Aoyagi, R.; Sando, H.; Kawasumi, M.; Tanaka, T. Chem.
Lett. 1987, 1435.
(13) (a) Garwood, R. F.; Weedon, B. C. L. J. Chem. Soc., Perkin
Trans. 1 1973, 2714. (b) Weiguny, J.; Schäfer, H. J.
Electroorg. Synth. 1993, 57, 235. (c) Matzeit, A.; Schäfer,
H.; Amatore, C. Synthesis 1995, 1432.
In contrast, the tertiary radical derived from 22b cyclized
only poorly, probably owing to its steric hindrance
(Table 2, entry 2). Whereas acid 22c was smoothly trans-
formed into tetrahydropyran 25 (Table 2, entry 3), the re-
moval of the gem-dimethyl substituent resulted in a
reduced yield of 26 (entry 4).17
In summary, we have developed an efficient electrochem-
ical method for the ecologically benign and safe synthesis
of substituted carbocycles, tetrahydrofurans, and tetrahy-
dropyrans from w-unsaturated carboxylic acids. Our ap-
proach embodies a Kolbe decarboxylation followed by a
radical cyclization. The intermediate radical is finally cap-
tured by another radical, generated by the concomitant de-
carboxylation of the co-acid, leading to the generation of
two carbon–carbon bonds in a single operation. The sim-
plicity, broad functional-group tolerance and good yields,
provided by this electrochemical oxidation, makes this
procedure an attractive methodology for the synthesis of
variously functionalized cyclic structures.
Acknowledgment
Financial support for this research by the Fonds pour la Formation
à la Recherche dans l’Industrie et dans l’Agriculture (FRIA, stu-
dentship to F. Lebreux), Astra Zeneca, Syngenta (studentship to F.
Buzzo), the Université catholique de Louvain and Merck Sharp and
Dohme (Academic Development Program Award to I.E. Markó) is
gratefully acknowledged.
(14) Although a radical adsorbed on an electrode surface might
display altered behavior, we believe that the slightly
nucleophilic character of the alkyl radical 3 is preserved.
See, for example: (a) Chkir, M.; Lelandais, D. J. Chem.
Soc., Chem. Commun. 1971, 1369. (b) Giese, B. Angew.
Chem., Int. Ed. Engl. 1983, 22, 753. For a scale measuring
the nucleophilicity of radicals, see: (c) De Vleeschouwer,
F.; Van Speybroeck, V.; Waroquier, M.; Geerlings, P.; De
Proft, F. Org. Lett. 2007, 9, 2721.
References and Notes
(1) (a) Giese, B. Radicals in Organic Synthesis: Formation of
Carbon–Carbon Bonds; Pergamon: Oxford, 1986.
(b) Curran, D. P. In Comprehensive Organic Synthesis, 4;
Trost, B. M.; Fleming, I.; Semmelhack, M. F., Eds.;
Pergamon: Oxford, 1991, 715. (c) Beckwith, A. L. J. Chem.
Soc. Rev. 1993, 143.
(15) Heinhorn, J.; Soulier, J.-L.; Bacquet, C.; Lelandais, D. Can.
(2) (a) Rossi, R. A.; Penenory, A. B. Curr. Org. Synth. 2006, 3,
121. (b) Majumdar, K. C.; Basu, P. K.; Chattopadhyay, S. K.
Tetrahedron 2006, 63, 793. (c) Walton, J. C. Top. Curr.
Chem. 2006, 264, 163.
J. Chem. 1983, 61, 584.
(16) Representative Procedure for the Electrochemical
Synthesis of Carbocycles and Heterocycles
In an undivided beaker-type cell (100 mL) bearing two
Synlett 2008, No. 18, 2815–2820 © Thieme Stuttgart · New York

2820
F. Lebreux et al.
LETTER
electrodes of platinum foil (1.3 cm × 1.3 cm × 0.5 mm), acid
22a (500 mg, 2.47 mmol) and a fivefold excess of AcOH
(707 mL, 12.38 mmol) were dissolved in MeOH (50 mL).
The acids were partially neutralized by NaOMe to obtain a
current density of 100 mA/cm2. The reaction was monitored
by TLC or GC and stopped when the pH of the solution
changed from 5 to 8; normally 1.2–1.4 F/mol had been
consumed. The reaction mixture was then concentrated
under reduced pressure. The residue was treated with a sat.
aq solution of NaHCO3, and extracted with CH2Cl2 (3 × 10
mL). The organic layers were collected, dried on Na2SO4,
and concentrated under reduced pressure. The crude material
was purified by flash chromatography (PE–Et2O, 4:1) to
afford 366 mg of compound 23a (86%, mixture of two
inseparable diastereomers, A/B = 1:1) as a colorless liquid
smelling of fruit. GC [60 °C (3 min), 15 °C/min → 290 °C,
290 °C (1 min)]: tR = 10.90, 11.02 min. IR (neat): 647, 732,
915, 1047, 1076, 1179, 1262, 1377, 1459, 1725, 2873, 2978.
1H NMR (300 MHz, CDCl3): d = 1.15 (d, 3 H, A, J = 6.6
Hz), 1.20 (d, 3 H, B, J = 6.6 Hz), 1.24 (t, 3 H, A, J = 7.2 Hz),
1.26 (t, 3 H, B, J = 7.2 Hz), 1.50–1.65 (m, 1 H, A + B), 1.95–
2.20 (m, 1 H, A + B), 2.25–2.55 (m, 2 H, A + B), 3.40–3.46
(dt, 1 H, A + B, J = 7.8, 2.2 Hz), 3.60–3.95 (m, 3 H, A + B),
4.08–4.18 (m, 2 H, A + B). 13C NMR (75 MHz, CDCl3):
d = 14.20 (A), 14.23 (B), 16.1 (A), 16.2 (B), 30.3 (B), 30.8
(A), 42.39 (A), 42.41 (B), 42.90 (B), 42.93 (A), 60.4 (A + B),
68.0 (A), 68.1 (B), 71.1 (A), 71.9 (B), 175.7 (A + B). MS
(APCI): m/z (%) = 82.9 (15), 126.9 (95), 144.9 (15),
173.0(10). HRMS (sodium complex): m/z calcd: 195.0997;
found: 195.0998.
(17) Compound 27 originates from a possible rearrangement of
either the primary radical or the derived carbocation. Studies
are currently ongoing to elucidate this intriguing
transformation.
Synlett 2008, No. 18, 2815–2820 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.