Radical cyclisation of carbohydrate alkynes: Synthesis of highly functionalised cyclohexanes and carbasugars

Article abstract of DOI:10.1039/a802228c

Carbohydrate alkynes undergo 6-exo radical cyclisation to afford cyclohexanes resulting in the synthesis of carbasugars.

Full text of DOI:10.1039/a802228c

View Article Online / Journal Homepage / Table of Contents for this issue
Radical cyclisation of carbohydrate alkynes: synthesis of highly functionalised
cyclohexanes and carbasugars
Elise Maudru,a Gurdial Singh*a† and Richard H. Wightman*b
a Department of Chemistry, University of Sunderland, Sunderland, UK SR1 3SD
b Department of Chemistry, Heriot-Watt University, Edinburgh, UK EH14 4AS
Carbohydrate alkynes undergo 6-exo radical cyclisation to
afford cyclohexanes resulting in the synthesis of carba-
sugars.
allyl radical is favourable and in fact this can become
synthetically useful.3b
We have been interested in the development of 6-exo
cyclisations of alk-6-ynyl radicals as these would allow the
preparation of heavily substituted hydroxycyclohexanes and of
carbasugars. In this regard we chose to prepare alkynes derived
from d-ribose7 and to investigate their chemistry. Treatment of
the protected silyl-d-ribose derivative 1 with lithium phenyl-
acetylide afforded an inseparable diastereomeric mixture of the
diols 2a in 90% yield (Scheme 1). The protection of both the
hydroxy groups as methoxymethyl (MOM) ethers8 proceeded
in a yield of 89% and allowed chromatographic separation of
the diastereomers 3a and 3b with an anti:syn ratio of 3 : 2.
Alternatively the syn isomer 3b could be prepared almost
exclusively using d-ribonolactone as the starting material.9
Desilylation of 3a was effected with TBAF in THF in 92% yield
and afforded the primary alcohol 6a which was converted to the
corresponding iodide 7a in 76% yield with triphenylphosphine,
imidazole and iodine.10 Similar chemistry with the isomer 3b
gave the alcohol 6b and subsequently the iodide 7b in
comparable yields. With both the iodides 7a and 7b in hand we
investigated their radical cyclisation reactions. Treatment of the
iodide 7a with tri-n-butyltin hydride in refluxing benzene in the
presence of AIBN effected the 6-exo cyclisation affording only
the substituted cyclohexanes 10 in 93% yield (Scheme 2) as an
inseparable mixture of E and Z geometric isomers in a ratio of
1:1 as determined by 1H NMR analysis. Analogous reaction of
the diastereomer 7b afforded the corresponding cyclohexanes
11 and 12 in 70% yield. In this case we were able to separate the
E and Z stereoisomers by chromatography, in a ratio of 2 : 3.
The major isomer 11 was assigned the Z geometry about the
double bond on the basis of NOE experiments.
The use of carbohydrates as synthetic precursors of many
functionalised carbocyclic compounds is widespread.1 The first
conversion of a sugar to a cyclohexane ring involved the
preparation of nitroinositols from 6-deoxy-6-nitrohexose.2 In
the intervening years there have been numerous examples of the
use of radical chemistry for the synthesis of cyclopentane
derivatives,3 with notable work in the carbohydrate area coming
from the Fraser-Reid group.4 It is thus surprising that there are
few examples of the preparation of six-membered aliphatic
rings employing radical cyclisation onto an alkyne.5 Some
reasons for this can be inferred from work on cyclisations of
w-alkenyl radicals. The formation of six-membered rings by
radical chain reactions involving 6-exo cyclisation is some 40
times slower than that of the hex-5-enyl cyclisation.6 The
consequence of this is that chain transfer by attack of the acyclic
radical on the stannane usually present is a much more effective
process than with the hex-5-enyl radical. The second problem is
that 1,5-hydrogen abstraction leading to a resonance stabilised
TBDPSO
OH
O
OH
R'
O
O
R
i or ii
O
O
OH
1
2a R = Ph, R' = OTBDPS
2b R = TMS, R' = OTBDPS
2c R = H, R' = OTBDPS
TBDPS = tert-butyldiphenylsilyl
Having been successful in our initial goal we turned our
attention to the syntheses of compounds with an unsubstituted
exo-methylene group. Thus the protected ribose 1 was treated
with lithium trimethylsilylacetylide in THF and afforded the
diols 2b, in a combined yield of 45% along with the alkynes 2c,
in 28% yield. The diastereomeric mixture 2b was converted to
iii
OMOM
OMOM
R'
R'
O
O
O
O
R
R
+
OMOM
anti-(major)
OMOM
syn-(minor)
Ph
3a R = Ph, R' = OTBDPS
4a R = TMS, R' = OTBDPS
5a R = H, R' = OTBDPS
3b R = Ph, R' = OTBDPS
4b R = TMS, R' = OTBDPS
5b R = H, R' = OTBDPS
OMOM
Bu3SnH, AIBN
7a
iv
iv
C6H6, ∆
MOMO
O
OMOM
OMOM
O
R'
R'
O
O
10
R
R
Ph
Ph
O
O
OMOM
OMOM
OMOM
OMOM
Bu3SnH, AIBN
6a R = Ph, R' = OH
6b R = Ph, R' = OH
7b R = Ph, R' = I
8b R = H, R' = OH
9b R = H, R' = I
v
v
v
+
7b
7a R = Ph, R' = I
C6H6, ∆
8a R = H, R' = OH
MOMO
MOMO
O
O
v
9a R = H, R' = I
O
O
Scheme 1 Reagents and conditions: i, LiC°CPh, THF, 278 °C; ii,
LiC°CTMS, THF, 278 °C; iii, MOMCl, NEt2Pri, CH2Cl2, RT; iv, Bu4NF,
THF, RT; v, I2, PPh3, Im, PhMe, D
11
Scheme 2
12
Chem. Commun., 1998
1505

View Article Online
Me
OH
OMOM
i
+
i
9b
ii
9a
14
MOMO
MOMO
O
O
MOMO
O
O
O
O
13
14
18
ii
iii
HO
RO
Me
Me
OMOM
OR
OH
OH
OAc
iii
iv
MOMO
RO
O
OR
HO
AcO
OAc
O
OR
OH
OAc
19
20
16 R = H
15
iv
17 R = Ac
Scheme 4 Reagents and conditions: i, Bu3SnH, PhH, AIBN, D; ii, BH3–
Me2S, THF; 0 °C, H2O2, NaOH; iii, 6 m HCl, MeOH; iv, Ac2O, Py, RT
Scheme 3 Reagents and conditions: i, Bu3SnH, AIBN, PhH, D; ii, BH3–
Me2S, THF; 0 °C, H2O2, NaOH; iii, 6 m HCl, MeOH; iv, Ac2O, Py, RT
rhamnose 19 in 99% yield. Further structural integrity of 19 was
established by acetylation with excess acetic anhydride and
pyridine which resulted in formation of the tetraacetate 20 in
99% yield.
We thank the EPSRC for access to central facilities for high
resolution mass spectrometric data at the University of Wales,
Swansea (Director, Dr J. A. Ballantine) and Professor W. T.
Borden (University of Washington) for enlightening discus-
sions regarding vinyl radicals. We thank Dr K. M. Morgan
(Heriot-Watt University) for Monte Carlo Calculations.
the MOM ethers 4a and 4b that could be separated in a
combined yield of 71% with an anti:syn ratio of 3.5 : 1. Both of
these diastereomers were processed separately. Desilylation of
4a and 4b gave the corresponding primary alcohols 8a and 8b
in 91 and 96% yields respectively. These compounds were
identical to those obtained from 2c after the diastereoisomers
had been subjected to protection by MOM chloride to afford 5a,
5b and desilylation, the anti:syn ratio being 3 : 1. The syn-
product was correlated with material prepared from the pure
syn-isomer 2c, which we have described in our earlier report.9
The alcohols 8a and 8b were converted smoothly to the
corresponding iodides 9a and 9b in 71 and 78% yields
respectively. At this juncture we were in a position to study the
6-exo radical cyclisation of these iodides. Thus (Scheme 3) the
syn isomer 9b was treated with tri-n-butyltin hydride and AIBN
in refluxing benzene and afforded the expected exo-methylene-
cyclohexane 13 in 49% yield along with the cyclohexene 14 in
39% yield where the MOM group had been lost. The structure
of 14 was clearly evident from the 1H NMR spectrum which had
resonances due to only one MOM group and in addition there
was a resonance at d 5.33 due to the vinylic proton and a
resonance at d 1.75 due to a methyl group. We next subjected
the iodide 9a to these conditions (Scheme 4) and we observed
that in this case the cyclisation reaction was appreciably slower,
taking 24 h to reach completion, but much cleaner in that only
14 was obtained in 99% yield. The formation of 14 can be
rationalised by cyclisation of a primary radical in a 6-exo mode
onto the alkyne, resulting in the formation of a vinyl radical
which then abstracts a hydrogen from the methylene carbon of
the MOM group followed by b-scission11 resulting in an allylic
radical which subsequently affords the observed product. The
formation of 14 is a reflection of the geometry of the vinyl
radical in that these are bent with a bond angle of ca. 135° whilst
the a-phenyl substituted vinyl radical is linear.12
Notes and References
† E-mail: gurdial.singh@sunderland.ac.uk
1 R. J. Ferrier and S. Middleton, Chem. Rev., 1993, 93, 2779.
2 H. O. L. Fischer and J. M. Grosheintz, J. Am. Chem. Soc., 1948, 70,
1476; H. O. L. Fischer and B. Iselin, J. Am. Chem. Soc., 1948, 70,
3946.
3 For excellent reviews on radical cyclisation reactions see: (a) B. Giese,
B. Kopping, T. Gobel, J. Dickhaut, G. Thoma, K. J. Kulicke and F.
Trach, Org. React., 1996, 48, ch. 2; (b) W. B. Motherwell and D. Crich,
Free Radical Chain Reactions in Organic Synthesis, Academic Press,
New York, 1992.
4 B. Fraser-Reid and R. Tsang, Strategies and Tactics in Organic
Synthesis, Academic Press, New York, 1989, vol. 2.
5 G. Bu¨chi and H. Wu¨est, J. Org. Chem., 1979, 44, 546; M. D. Bachi and
C. Hoornaert, Tetrahedron Lett., 1982, 23, 2502; D. L. J. Clive, P. L.
Beaulieu and L. Set, J. Org. Chem., 1984, 49, 1313; J. K. Crandell and
W. I. Michaley, J. Org. Chem., 1984, 49, 4244; G. Just and G.
Sacripante, Can. J. Chem., 1987, 65, 104, A. V. Rama Rao, J. S. Yadav,
C. S. Rao and S. Chandrasekhar, J. Chem. Soc., Perkin Trans. 1, 1990,
1211; C. K. Sha, T. S. Jean and D. C. Wang, Tetrahedron Lett., 1990, 31,
3745; D. L. Boger and R. J. Mathvink, J. Am. Chem. Soc., 1990, 112,
4003; T. Honda, M. Satoh and Y. Kobayashi, J. Chem. Soc., Perkin
Trans. 1, 1992, 1557; J. Marco-Contelles, Synth Commun., 1994, 24,
1293; J. Marco-Contelles, M. Bernabe, D. Ayala and B. Sanchez,
J. Org. Chem., 1994, 59, 1234; this details a 6-endo-dig cyclisation;
R. E. McDevitt and B. Fraser-Reid, J. Org. Chem., 1994, 59, 3250.
6 A. L. J. Beckwith, Tetrahedron, 1981, 37, 3073; A. L. J. Beckwith and
C. J. Easton, J. Am. Chem. Soc., 1981, 103, 615; V. Malatesta and K. U.
Ingold, J. Am. Chem. Soc., 1981, 103, 609, A. L. J. Beckwith and C. H.
Schiesser, Tetrahedron, 1985, 41, 3925; P. R. Jenkins, M. C. R.
Symons, S. E. Booth and C. J. Swain, Tetrahedron Lett., 1992, 33,
3545.
7 B. Mekki, G. Singh and R. H. Wightman, Tetrahedron Lett., 1991, 32,
5143 and references cited therein.
8 G. Stork and T. Takahashi, J. Am. Chem. Soc., 1977, 99, 1275.
9 S. Jiang, G. Singh and R. H. Wightman, Chem. Lett., 1996, 67.
10 P. J. Garegg and B. Samuelsson, J. Chem. Soc., Perkin Trans. 1, 1980,
2866.
11 J.-C. Malanda and A. Doutheau, J. Carbohydr. Chem., 1993, 12, 999.
12 W. T. Borden and A. Nicolaides, J. Am. Chem. Soc., 1991, 113, 6750
and refs. therein.
13 L. Pingli and M. Vandewalle, Tetrahedron, 1994, 50, 7061; L. Pingli
and M. Vandewalle, Synlett., 1994, 228.
The exo-methylenecyclohexane 13 (Scheme 3) was hydro-
borated and gave after oxidation the primary alcohol 15 in 94%
yield. Removal of the protecting groups of 15 with 6 m HCl gave
carba-a-l-gulopyranose 16 whose spectral properties were in
accord with those reported in the literature.13 Additional
structural proof was obtained by acetylation of 16 with excess
acetic anhydride and pyridine which afforded the pentaacetate
17 in 100% yield. The cyclohexene 14 underwent hydro-
boration/oxidation (Scheme 4) with BH3–Me2S and afforded
the protected carba-b-d-rhamnose derivative 18 in 77% yield.
The stereochemistry of the newly formed chiral centres was anti
with the C-4 hydroxy group b as a result of hydroboration
occurring from the opposite face from the O-isopropylidene
group, and this was confirmed by NOE experiments. Removal
of the MOM and isopropylidene protection proceeded unevent-
fully with 6 m HCl and afforded the fully deprotected carba-b-d-
Received in Liverpool, UK, 20th March 1998; 8/02228C
1506
Chem. Commun., 1998