TEMPO free radical derived β-aminoalkoxyketolactones: Substrates for base-, acid- and radical-induced fragmentation reactions

Article abstract of DOI:10.1039/a909750c

β-Aminoalkoxyketolactones 2 and 3 obtained from the TEMPO free radical substitution of iodolactones 1 undergo fragmentations under basic, acidic and free radical conditions.

Full text of DOI:10.1039/a909750c

TEMPO free radical derived b-aminoalkoxyketolactones: substrates for base-,
acid- and radical-induced fragmentation reactions
Arthur G. Schultz† and Xuqing Zhang*
Department of Chemistry, Rensselaer Polytechnic Institute, Troy, NY 12180-3590, USA. E-mail: zhangx4@rpi.edu
Received (in Corvallis, OR, USA) 10th December 1999, Accepted 2nd February 2000
b-Aminoalkoxyketolactones 2 and 3 obtained from the
TEMPO free radical substitution of iodolactones 1 undergo
fragmentations under basic, acidic and free radical condi-
tions.
Fragmentation reactions of iodoketolactones of general struc-
ture 1, available as single enantiomers,1 with LiOH under
aqueous conditions afford butenolide carboxylic acids and
4-hydroxy-2-cyclohexenones by way of competing additions of
hydroxide ion to the ketone and lactone carbonyl groups.2 The
bicyclo[3.2.1] ring system in 1 locks the conformation of the
two carbonyl groups with respect to the iodo substituent. To
examine the importance of stereoelectronic factors3 in these
fragmentation processes it was desirable to have access to not
only the axial iodo substituent in 1 but also an equatorial leaving
group. We have found that it is possible to carry out an exchange
of iodide in 1 with 2,2,6,6-tetramethyl-1-piperidinyloxy free
radical (TEMPO) to provide both the axial and equatorial
substituted b-aminoalkoxyketolactones 2 and 3 (Scheme 1).4
Herein we describe the fragmentation reactions of 2 and 3 under
basic, acidic and free radical conditions.
Mixtures of b-aminoalkoxyketolactones 2 and 3 were
produced from 1 and were separated by flash chromatography
on silica gel. Treatment of 2a with 2 equiv. of LiOH in a 5+1
mixture of THF and water gave butenolide carboxylic acid 4 in
78% isolated yield with no trace of the 4-hydroxycyclohex-
enone 5 (Scheme 2). Addition of aqueous LiOH to the isomer 3a
with an axial aminoalkoxy group gave mainly the 4-hydroxy-
cyclohexenone 5 along with 18% of 4 (Scheme 2).
That 2a undergoes fragmentation to give 4 rather than 5 is a
result of the antiperiplanar relationship of the leaving group X
and the cyclohexanone C–C bond in the tetrahedral intermediate
generated by addition of hydroxide ion to the ketone carbonyl
group. The formation of 5 is explained by addition of hydroxide
ion to the lactone carbonyl group of 3a followed by a relatively
fast fragmentation of the lactone C–C bond along with
elimination of the antiperiplanar aminoalkoxy group. A com-
peting fragmentation of 3a occurs by addition of hydroxide ion
to the ketone carbonyl group followed by a relatively slow
cleavage of the cyclohexanone C–C bond to give a lactone
enolate; elimination of the poorly oriented aminoalkoxy group
is probably not in concert with C–C bond cleavage, but must
Scheme 2
await a conformational adjustment in the enolate to bring
orbitals into proper alignment.3 It is noteworthy that iodo-
lactone 1a also gave a mixture of 4 and 5 on treatment with
aqueous LiOH.4
We were interested in the development of fragmentation
processes initiated by intramolecular carbonyl addition reac-
tions. Hydrogenolysis of the benzyl ether 2c in the presence of
dilute hydrochloric or acetic acids provided the hemiketal 6, and
treatment of 6 with moderately concentrated HCl in ethanol at
25 °C resulted in fragmentation to give the medium ring
lactone-butenolide 7 in 90% yield (Scheme 3).5 Hemiketal 6
also was converted to 7 (88%) by fragmentation under basic
reaction conditions (Et3N in CH2Cl2, 25 °C). By contrast,
hemiketal 8 obtained from hydrogenolysis of the isomeric
benzyl ether 3c did not undergo fragmentation to 7 under
comparable acidic or basic reaction conditions (Scheme 3).
The utilization of aminoalkoxy substituents to initiate acid-
catalyzed fragmentation reactions appears to be without
precedent in the chemical literature. It is thought that fragmenta-
tion of 6 occurs from the N-protonated aminoalkoxy substituent
to provide the oxide of a sec-amine as the leaving group; the
amine oxide would be expected to rearrange to 1-hydroxy-
Scheme 1
† Deceased January 20th, 2000.
Scheme 3
DOI: 10.1039/a909750c
Chem. Commun., 2000, 399–400
This journal is © The Royal Society of Chemistry 2000
399

and fragmentation to a 2,4-lactone fused-2-cyclohexenone
(structure not shown; cf. 3a ? 5) is not competitive with
reduction of the alkyl radical with Bu3SnH.
In summary, the 2,2,6,6-tetramethyl-1-piperidinyloxy group
has been found to be an effective leaving group in fragmentation
reactions performed under basic, acidic and free radical
conditions. It is important to note that elimination of 2,2,6,6-
tetramethylpiperidine from 2 or 3 to give the corresponding b-
diketone did not occur under any of these reaction conditions.10
The availability of b-aminoalkoxyketolactones 2 and 3 as single
enantiomers1 suggests that these fragmentation reactions ought
to have substantial utility in organic synthesis; the development
of methods to selectively generate 2 and 3 and related substrates
are under investigation.
We thank Dr P. R. Guzzo and Dr L. Pettus for early
contributions to this project and the National Institutes of Health
(GM 26568) for generous financial support.
Scheme 4
2,2,6,6-tetramethylpiperidine.6 Ionization of the protonated
axial aminoalkoxy substituent in 8 does not occur because of the
synclinal relationship of the C–ONR2 bond and the cyclohexane
C–C bond.7 Fragmentation of 6 with Et3N occurs by deprotona-
tion of the hemiketal with concomitant elimination of
R2NO2.
Perhaps the most interesting application of the aminoalkoxy
substituent as a regulator of fragmentation reactions is in the
area of free radical-mediated ring expansions.8 Treatment of the
alkyl iodide 9 with AIBN and Bu3SnH (added over 6 h) in
refluxing benzene gave the nine-membered-ring ketone 10 in
77% yield (88% based on recovered 9) (Scheme 4).9 However,
the isomer 11 with an axial aminoalkoxy substituent gave the n-
propyl derivative 12 with no trace of 10 (Scheme 4).
Notes and references
1 A. G. Schultz, Chem. Commun., 1999, 1263.
2 A. G. Schultz, M. Dai, S.-K. Khim, L. Pettus and K. Thakkar,
Tetrahedron Lett., 1998, 39, 4203.
3 P. Deslongchamps, Stereoelectronic Effects in Organic Chemistry,
Pergamon, Oxford, 1984; C. A. Grob, Angew. Chem., Int. Ed. Engl.,
1969, 8, 535.
4 A. G. Schultz, M. Dai, F. S. Tham and X. Zhang, Tetrahedron Lett.,
1998, 39, 6663; D. L. Boger and J. A. McKie, J. Org. Chem., 1995, 60,
1271.
5 Benzyl ether 2c also was converted to 7 (91%) in one experimental
operation by way of hydrogenolysis under more strongly acidic
conditions (HCl in EtOH).
6 Ionization initiated by O-protonation or a bridging N,O-protonation
should also be considered.
These data suggest that the primary alkyl radical generated
from 9 undergoes addition to the ketone carbonyl group,
followed by a relatively fast alkoxy radical-induced fragmenta-
tion to give 10 and the TEMPO free radical. The primary radical
generated from 11 probably also undergoes addition to the
ketone carbonyl group (reversible), but without proper align-
ment of the aminoalkoxy substituent this addition is non-
productive; reversion to the alkyl radical and eventual reduction
with Bu3SnH gives the n-propyl derivative 12. It should be
noted that alkyl radical addition to the lactone carbonyl group
7 P. S. Wharton and G. A. Hiegel, J. Org. Chem., 1965, 30, 3254.
8 P. Dowd and W. Zhang, Chem. Rev., 1993, 93, 2091; G. H. Posner, K. S.
Webb, E. Asirvatham, S. Jew and A. Degl’Innocenti, J. Am. Chem. Soc.,
1988, 110, 4754.
9 For related free radical ring expansions that give cyclononenones, see:
J. E. Baldwin, R. M. Adlington and J. Robertson, J. Chem. Soc., Chem.
Commun., 1988, 1404.
10 D. H. Hunter, D. H. R. Barton and W. J. Motherwell, Tetrahedron Lett.,
1984, 25, 603 and references therein.
Communication a909750c
400
Chem. Commun., 2000, 399–400