Formation of substituted macrocyclic ethers by radical cyclisation

Article abstract of DOI:10.1039/a608494j

ω-Iodopolyoxaalkyl acrylates and related compounds undergo radical cyclisation when treated with tributyl-stannane to afford substituted macrocylic polyethers.

Full text of DOI:10.1039/a608494j

View Article Online / Journal Homepage / Table of Contents for this issue
Formation of substituted macrocyclic ethers by radical cyclisation
Athelstan L. J. Beckwith,*a Kitty Drok,a Bernard Maillard,*b Marie Degueil-Castaingb and Annie Philipponb
a Research School of Chemistry, Australian National University, Canberra, Australia 0200
b Laboratoire de Chimie Organique et Organome´tallique, associe´ au CNRS URA35, Universite´ Bordeaux 1, F-33405
Talence-Cedex, France
w-Iodopolyoxaalkyl acrylates and related compounds
undergo radical cyclisation when treated with tributyl-
stannane to afford substituted macrocylic polyethers.
equivalents of tributylstannane (0.01 mol dm23) and AIBN as
catalyst in benzene under reflux for 2 h gave the corresponding
cyclic polyethers 3b–j (and in some cases 4g–j), but no cyclised
product could be obtained from 2a. Uncyclised products formed
by direct hydrogen atom transfer to the initially formed radicals
were also detected. Their formation allowed the estimation of
the rate constants for cyclisation on the basis of the mechanism
shown in Scheme 1. Accurate GC analysis of the product
mixtures and treatment of the data in the usual way6 gave values
of kc/kH which, when combined with the rate constant for
hydrogen transfer from tributylstannane to a primary radical
(kH = 6.3 3 106 dm3 mol21 s21 at 80 °C),7 gave kc at 80 °C. The
results presented in Table 1 show that the radicals 5b and 5c,
respectively, cyclise 30 and 10 times faster than their all-
carbon-chain analogues (kc = 4.8 3 103 and 1.2 3 104 s21 at
80 °C, calculated from data in the literature8).
Each of the substrates 2b–f undergoes exclusive endo
cyclisation to give 3b–f, but substrates 2g–j bearing an
activating substituent at the remote end of the acrylate double
bond also undergo some exo cyclisation. Interestingly, the
maleate ester 2h and the corresponding fumarate 2g cyclise with
slightly different endo:exo ratios of 43:57 and 58:42,
respectively. As measured approximately by NMR the ratio was
1:1 for cyclisation of 2i. In an attempt to improve the endo:exo
selectivity we examined the behaviour of the ester 2j9 bearing
the bulky tert-butyl group. However, once again the 3j:4j ratio
was about 1:1.
.
3-Oxahex-5-enyl radicals (CH2NCHCH2OCH2CH2 ) undergo
exo cyclisation about 25 times more rapidly at 80 °C than do
hex-5-enyl radicals.1,2 Similarly the 2-allyloxyphenyl radical
cyclises much more rapidly than its all carbon analogue.3 The
effect of an oxygen atom in the chain on the rates of cyclisation
of these and related species has been attributed to the decrease
in the strain energy of the cyclisation transition structures
resulting from the replacement of a carbon atom by oxygen,1,4
an hypothesis supported by molecular mechanics calculations.4
These observations suggested that the ring closure of radicals
derived from suitable polyether precursors might proceed
relatively rapidly. Here we show that such radicals do indeed
cyclise more readily than their all-carbon analogues, and that
this method provides a viable alternative to ionic reactions as a
route to macrocyclic polyethers of potential utility as ion-
ophores.
Suitable radical precursors 2a–j were readily prepared from
ethylene glycol oligomers. In a typical example, treatment of
the monotosylate of tetraethylene glycol5 with acryloyl chloride
and Hu¨nig’s base gave the acrylate 1 (92%) which was
converted into the iodide 2c (95%) on stirring with sodium
iodide in acetone. In some cases the oligomeric ethylene glycol
was first esterified by treatment with an acid chloride and
triethylamine and the iodide was then prepared via the mesylate.
Heating of each of the compounds 2b–j with 1.1 molar
Consideration of the competing reactions available to the
various radicals involved in these cyclisations suggested that the
O
O
OTs
O
2
O
1
CH2R
O
R
O
O
R
O
+
O
O
n
O
O
n
I
O
Bu3SnH
n
O
O
O
2
a n = 0 R = H
f n = 5 R = H
b n = 1 R = H g n = 1 R = E-CO2Et
c n = 2 R = H h n = 1 R = Z-CO2Et
3
4
d n = 3 R = H
e n = 4 R = H
i
j
n = 2 R = E-CO2Et
n = 2 R = Z-CO2But
Scheme 1
•
O
O
CH2•
R
R
O
O
O
O
O
n
n
O
O
O
O
n
5
kc
Bu3SnH
O
O
Bu3SnH kH
n
6
3
O
a n = 0 d n = 3
b n = 1 e n = 4
c n = 2
O
O
f n = 5
7
Chem. Commun., 1997
499

View Article Online
formation of cyclised products should be improved by conduct-
ing the experiments with very low concentrations of both the
iodoacrylate and Bu3SnH, thus respectively decreasing of the
formation of oligomers and the rate of hydrogen atom transfer
to uncyclised radicals. Because such high dilution conditions
require the use of very large quantities of solvent for the
production of small amounts of products, they are of no interest
for preparative purposes. However, the same effect was
obtained by the separate slow addition with syringe pumps of
both Bu3SnH and iodoacrylate to a solution of AIBN in benzene
(method A). Under these conditions cyclised products were
isolated in modest to good yield (Table 2). In an alternative
procedure the precursors were heated with 1.1 molar equiva-
lents of Bu3SnH (0.01 mol dm23), AIBN and sodium tosylate
(method B). The role of the sodium tosylate is still uncertain but
it certainly promotes cyclisation. Each of these methods is
suitable for the convenient preparation of macrocyclic poly-
ethers on a gram scale.
Finally we examined the preparation of macrocycles that do
not contain a lactone group within the ring and hence should be
able to survive robust manipulation of the side chain. Thus
treatment of the ester 8 with Bu3SnH in the usual way gave 10
(64%) with a rate constant for cyclisation of kc = 2 3 104 s21
at 80 °C, while 9 gave 11 (87%). An advantage of products such
as 10 and 11 is that the side chain can be readily converted into
functionality suitable for tethering the macrocycle to other
ionophores or to fluorescent agents. Thus 11 readily underwent
hydrolysis with KOH in methanol to afford the acid 12 (84%)
while reduction with LAH in THF gave the alcohol 13 (79%).
In conclusion it appears that, as in the case of simple hexenyl
species, the introduction of oxygen atoms into suitable long
chain precursors enhances the rate of radical cyclisation
sufficiently to allow the preparation of substituted macrocyclic
polyethers in modest to good yield under convenient experi-
mental conditions. These results therefore complement those
describing other applications of radical macrocyclisation in
synthesis.10,11 Preliminary experiments indicate that some of
the macrocyclic polyethers described above may be useful as
complexing agents.
R
CO2Et
We gratefully acknowledge technical assistance from Mr
Robert Longmore and the award of a Commonwealth Post-
graduate Scholarship to K. D. and of a BDI Scholarship by
CNRS and Re´gion Aquitaine to A. P.
O
O
O
O
n
I
O
n
8 n = 1
9 n = 2
O
10 n = 1 R = CO2Et
11 n = 2 R = CO2Et
12 n = 2 R = CO2H
13 n = 2 R = CH2OH
References
1 A. L. J. Beckwith and S. A. Glover, Aust. J. Chem., 1987, 40, 157.
2 A. L. J. Beckwith, V. W. Bowry and G. Moad, J. Org. Chem., 1988, 53,
1632.
3 A. N. Abeywickrema and A. L. J. Beckwith, J. Chem. Soc., Chem.
Commun., 1986, 464.
4 A. L. J. Beckwith and C. H. Schiesser, Tetrahedron, 1985, 41, 3925.
5 The monotosylate was prepared by the general method previously used
for the preparation of the ditosylate: J. Dale and P. O. Kristiansen, Acta
Chem. Scand., 1972, 26, 1471.
6 A. L. J. Beckwith and G. Moad, J. Chem. Soc., Chem. Commun., 1974,
472.
7 C. Chatgilialoglu, K. U. Ingold and J. C. Scaiano, J. Am. Chem. Soc.,
1981, 103, 7739.
Table 1 Cylisation rate constants for polyether radicals in benzene at
80 °C
Radical
kc 3 1024/s21
Radical
kc 3 1024/s21
5b
5d
5f
15 ± 4
5.1 ± 0.8
3.0 ± 04
5c
5e
13 ± 1
10 ± 2
8 N. A. Porter and V. H.-T. Chang, J. Am. Chem. Soc., 1987, 109,
4976.
Table 2 Formation of macrocyclic polyethers by radical cyclisation
9 2j was obtained via DCC coupling of mono tert-butyl maleate with
tetraethylene glycol monotosylate. tert-Butyl maleate was prepared by
the same general method as that previously used for the preparation of
tert-butyl phthalate: W. A. Fessler and R. L. Schriner, J. Am. Chem.
Soc., 1936, 58, 1384.
10 N. A. Porter, D. R. Magnin and B. T. Wright, J. Am. Chem. Soc., 1986,
108, 2787; N. A. Porter, B. Lacher, V. H.-T. Chang and D. R. Magnin,
J. Am. Chem. Soc., 1989, 111, 8309.
11 For some recent examples see: J. Ullrich and D. P. Curran, Tetrahedron
Lett., 1995, 36, 8921; E. I. Troyansky, R. F. Ismagilov, V. V. Samoshin,
Y. A. Strelenko, D. V. Demchuk, G. I. Nikishin, S. V. Lindeman,
V. V. Khrustalyov and Y. T. Struchkov, Tetrahedron, 1995, 51, 11 431;
I. Ryu, K. Nagahara, H. Yamazaki, S. Tsunoi and N. Sonoda, Synlett,
1994, 643; M. J. Begley, G. Pattenden, A. J. Smithies and D. S. Walter,
Tetrahedron Lett., 1994, 35, 2417; K. S. Feldman, H. M. Berven,
A. L. Romanelli and M. Parvez, J. Org. Chem., 1993, 58, 6851.
Substrate
Ring size
Products [yield (%)]
2a
2b
2c
2d
2e
2f
2g
2h
2i
9
12
15
18
21
24
12/11
12/11
15/14
15/14
3a
3b (78a)
3c (72a, 78b)
3d (70a, 87b)
3e (63a, 90b)
3f (30a)
3g (37a 4g (21a)
3h (43a,c) 4h (57a,c
3i (36d) 4i (36d)
3j (29d) 4j (29d)
)
2j
a
b
Method A (see text); isolated yields. Method B (see text); yields
determined by GC with an internal reference. Relative yield; for this
reaction the absolute yield was not determined. Method B; total yield
c
d
determined by GC; relative yields by NMR.
Received, 19th December 1996; Com. 6/08494J
500
Chem. Commun., 1997