A simple, efficient alternative for highly stereoselective iodoacetoxylation of protected glycals

Article abstract of DOI:10.1016/j.tetlet.2004.10.166

Protected glycals are converted in high yields and selectivities in less than 2 h at low temperatures to 2-deoxy-2-iodoglycosyl acetates using the simple, inexpensive reagent mixture of ammonium iodide, hydrogen peroxide and acetic anhydride/acetic acid in acetonitrile. The corresponding 2-deoxy-2-bromoglycosyl acetates are obtained using ammonium bromide instead of the iodide, although longer reaction times are required and selectivities are inferior.

Full text of DOI:10.1016/j.tetlet.2004.10.166

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 9533–9536
A simple, efficient alternative for highly stereoselective
iodoacetoxylation of protected glycals
David W. Gammon,a,* Henok H. Kinfe,a Dirk E. De Vos,b
Pierre A. Jacobsb and Bert F. Selsb
aDepartment of Chemistry, University of Cape Town, Rondebosch 7701, South Africa
bCenter for Surface Chemistry and Catalysis, Katholieke Universiteit Leuven, Kasteelpark Arenberg 23,
B-3001 Leuven, Belgium
Received 29 September 2004; revised 22 October 2004; accepted 28 October 2004
Abstract—Protected glycals are converted in high yields and selectivities in less than 2h at low temperatures to 2-deoxy-2-iodogly-
cosylacetates using the simpel, inexpensive reagent mixture of ammonium iodide, hydrogen peroxide and acetic anhydride/acetic
acid in acetonitrile. The corresponding 2-deoxy-2-bromoglycosyl acetates are obtained using ammonium bromide instead of the
iodide, although longer reaction times are required and selectivities are inferior.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The preparation of 2-deoxy-2-iodoglycopyranosyl ace-
tates in the stereoselective synthesis of 2-deoxyglycosides
continues to attract attention.1–3 Glycosyl acetates are
easily activated as glycosyl donors, and the iodo substit-
uent provides stereo-directing anchimeric assistance in
the glycosylation step and is readily amenable to radi-
cal-induced reductive cleavage to give 2-deoxyglyco-
sides. As a radicalprecursor it aslo provides the basis
for introduction of alkenyl substituents, and it was our
work in this context4 together with our interest in devel-
oping environmentally benign synthetic protocols,5
which prompted an investigation of alternative methodo-
logies for the preparation of 2-deoxy-2-iodoglycopyran-
osylacetates from protected gyl casl.
RO
RO
RO
RO
I
O
RO
O
O
RO
RO
RO
RO
TfO
V
OAc
OAc
III
I
+
RO
RO
RO
RO
RO
OTf
O
RO
RO
RO
O
I
O
OAc
OAc
RO
I
OH
II
IV
VI
Scheme 1. Summary of the synthetic routes to 2-iodomannosyl- and
2-iodoglucosyl acetates.
followed by acetylation to give mixtures, which include
iodoacetates III and IV. In most cases this is limited
by the low diastereoselectivity of the iodohydrin forma-
tion, although in one instance7 the exclusive formation
of 3,4,6-tri-O-acetyl-2-deoxy-2-iodomannose has been
reported. The third category provides direct access to
the iodoacetates III and IV from glucals V using, in
most cases, a source of electrophilic iodine together with
acetic acid as solvent or co-solvent.2,3 Yields are gener-
ally good to excellent, and product selectivities range
from a modest ꢀ2:1 to the synthetically useful ꢀ11:1
in favour of the a-manno-configuration, achieved with
the CAN/NaI/AcOH procedure.2 Selectivity in favour
of the gluco-isomer results when tert-butyldimethylsilyl
protecting groups are used, and has been explained8
in terms of the preference in this glycal for the 5H4
The existing synthetic approaches to these 2-deoxy-2-
iodoglycopyranosyl acetates can be divided into three
categories, summarized for the gluco- and manno- cases
in Scheme 1. The first, involving displacement of 2-tri-
flates from otherwise protected glycosyl acetates (I and
II)6 provides access in good yields to the desired prod-
ucts III and IV, but relies on selective preparation and
careful handling of the moderately stable 2-triflates.
The second category (exemplified in Ref. 1) proceeds
via formation of iodohydrins VI from glucals V
Keywords: Glycals; 2-Izodoglycosides; Iodoacetates.
*
Corresponding author. Tel.: +27 21 650 2547; fax: +27 21 689
7499; e-mail: gammondw@science.uct.ac.za
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.10.166

9534
D. W. Gammon et al. / Tetrahedron Letters 45 (2004) 9533–9536
Table 1. Yields and selectivities in the iodoacetoxylation of protected
glucals (see Schemes 2 and 3)
conformation, which minimizes significant steric interac-
tions of the bulky silyl groups but also restricts access of
the electrophile to the b-face of the glucal.
Entry
Substrate
Products
Yield (%)
Ratioa
1
2
3
4
5
6
7
8
1
9a,b
85
100
100
95
83:17
91:9
93:7
93:7
17:83
3:2
In view of the limited number of efficient approaches to
these iodoacetates we felt there was scope for developing
alternatives, which exploit the simplicity and cost effec-
tiveness of the oxidation of iodide salts to provide iodo-
nium ion equivalents. The synthetic possibilities of
iodination of organic substrates with halide salts in the
presence of hydrogen peroxide and a catalyst have re-
cently been highlighted,1,9,10 and in our first attempts
at direct oxidative halogenation we treated tri-O-acet-
yl-D-glucal 1 and tri-O-benzyl-D-glucal 2 with NH4I
and 50% aqueous H2O2 in acetic acid, using conditions
reported to achieve easy iodination of phenol.11,12 The
desired 2-iodoacetates were formed but were accompa-
nied by significant amounts of the corresponding iodoh-
ydrins. However, addition of acetic anhydride to the
reaction mixture, ideally to intercept water present in
the reaction, ensured clean and selective conversion to
the iodoacetates, while the further modification of intro-
ducing acetonitrile as solvent with concomitant reduc-
tion of the amounts of Ac2O and AcOH, allowed for
lowering of the reaction temperature without the solu-
tion freezing. These modifications were in consideration
that low temperatures generally favour high stereoselec-
tivities, while in addition, the proposed method is optim-
ized with regard to the atom efficient use of the reagents,
with close to equivalent amounts of both iodide and
H2O2 required to realize full conversion in most
instances.
2
10a,b
11a,b
,b
5
612a
7
1
2
8
13a,b
14a,b
15a,b
16 –c
94
82
88
9a6
2:1
15:1:4
a Determined from 1H NMR of the reaction products before
separation.
detected in reactions of protected glucals, and in accord-
ance with previous findings the a-manno products pre-
dominated except in the case of the per-O-silylated
derivatives. The product distribution in the reaction of
benzylated galactal 8 was similar to previously reported
results with the talo-isomer 16a as the major product,
and a significant proportion of a-gluco-isomer 16c
resulting from 1,2-cis addition (Scheme 3).
The selectivities and yields compare very favourably
with reported methods, with those obtained with benzyl
or silyl protecting groups being the best yet reported,
and the reaction conditions are tolerant of a range of
protecting groups. The selectivity towards 2-deoxy-2-
iodomannopyranosylacetate increases noticeabyl from
entries 1–3, corresponding inter alia to changes in the
substituent at C-6 from acetylto benzylto tert-butyldi-
phenylsilyl. Interestingly, formation of 13b using NIS in
AcOH required heating at 100ꢁC for 10min8 whereas
the reaction proceeds at room temperature within 2h
using our method. In generalwe observed that seelcti-
vities were dependent on temperature, with lower tem-
peratures favouring a-manno selectivity, although
cooling below 0ꢁC led to unacceptably slow reactions.
Treatment of variously protected glucals and galactals
in this way13 (Scheme 2 and Table 1) provided efficient
and selective access to the desired 2-deoxy-2-iodoglyco-
sylacetates. In all cases complete conversion of the start-
ing glycal was achieved, and products were identified by
1H and 13C NMR spectroscopy after chromatographic
separation.14 Only the 1,2-trans addition products were
The possibility of using this methodology for bromo-
acetoxylation reactions was demonstrated by replace-
ment of NH4I with NH4Br in reactions of acetylated
and benzylated glucals 1 and 2 (Scheme 2 and Table 1,
entries 6 and 7), albeit that the bromoacetoxylation is
less stereoselective. NaI was also evaluated as an iodide
source instead of NH4I in the conversion of acetylated
glucal 1 to 9a and 9b. In this instance an improved
stereoselectivity of 7:1 in favour of 9a was observed
but the yield (64% overall) was inferior and the 1H
NMR spectrum of the reaction products provided
evidence for inseparable, hitherto unidentified products.
R1O
R2O
R2O
R1O
R1O
X
R2O
R2O
O
O
O
X
+
R2O
R2O
OAc
OAc
R1
R1
R2
R2
X
1
2
5
6
7
Ac
Bn
Ac
Bn
9a,b
10a,b
11a,b
12a,b
13a,b
14a,b
15a,b
Ac
Bn
Ac
Bn
I
I
TBDPS Ac
TBDPS Ac
I
TBDPS Bn
TBDPS TBDPS
TBDPS Bn
I
I
Br
Br
TBDPS TBDPS
Ac
Bn
Ac
Bn
Concerning the proposed mechanism of the reaction, the
appearance of a brown/yellow colour in the reaction
Scheme 2. Haloacetoxylation of protected glucals.
OBn
BnO
OBn
BnO
BnO
OBn
O
OBn
O
BnO
BnO
BnO
BnO
I
O
O
+
+
OAc
BnO
I
I
OAc
OAc
8
16a
16b
16c
Scheme 3. Iodoacetoxylation of protected galactals.

D. W. Gammon et al. / Tetrahedron Letters 45 (2004) 9533–9536
9535
was added NH4I (64mg, 0.44mmol), and Ac2O (0.5mL)
and the resulting mixture cooled to 0ꢁC. H2O2 (25lL of a
50% aqueous solution in water, 0.44mmol) was added and
the solution stirred for 1h at 0ꢁC, when TLC showed the
reaction was complete. A 0.1M solution of sodium
thiosulfate was then added until the brownish colour
disappeared, and the solution was cooled in an ice-water
bath before adding cold 10% aq NaOH until the solution
became slightly basic. The resultant mixture was extracted
with ethylacetate, and the combined organic phases
washed once with brine, then dried (MgSO4) and concen-
trated. Separation of the product mixture by chromato-
graphy on silica gel yielded the pure isomers 9a (144mg,
70%) and 9b (30mg, 15%).
solutions upon addition of H2O2 to the other reactants
suggests the presence of molecular I2, formed either by
reaction of IÀ with H2O2 under acidic conditions17,18
or by reaction of IÀ with peracetic acid, generated upon
addition of H2O2 to acetic anhydride in the presence of
acetic acid.19 It is possible that an initial and rapid for-
mation of a p-complex between the olefin and I2 is fol-
lowed by a rate-determining abstraction of IÀ by the
peracetic acid.19 The high degree of stereoselectivity in
solvolysis of the resulting iodonium species and the fact
that in the absence of acetic anhydride the iodoacetates
predominate over iodohydrins suggests that acetic acid
attacks the iodonium species directly. The presence of
an excess of acetic anhydride ensures that the concentra-
tion of water in the reaction mixture is minimized,
allowing for successful addition of the acetate or acetic
acid to the cyclic iodonium species.
14. Analytical data for known compounds 9a, 9b, 10a, 10b,
13a, 13b, 14a, 14b, 15a, 15b, 16a, 16b and 16c were
consistent with published data (Refs. 5, 7, 14, 15 and 16);
data for new compounds is as follows. 11a: mp 153–
155ꢁC. 1H NMR (300MHz, CDCl3): d 7.77–7.34 (m, 10H,
2 · Ph), 6.45 (d, 1H, J = 1.5Hz, H-1), 5.63 (t, 1H,
J = 9.6Hz, H-4), 4.60 (dd, 1H, J = 4.2, 9.6Hz, H-3), 4.52
(dd, 1H, J = 1.5, 4.2Hz, H-2), 3.98–3.92 (m, 1H, H-5),
3.72 (d, 2H, J = 2.7Hz, H-6a and H-6b), 2.12, 2.09, 1.93
(3s, 9H, 3 · CH3CO2), 1.08 (s, 9H, Me3C–Si). 13C NMR
(100MHz, CDCl3): d 170.0, 169.0, 168.2 (3 · CH3CO2),
135.8, 135.7, 133.2, 133.1, 129.7, 129.6, 127.6 (2 · Ph), 95.0
(C-1), 74.0 (C-5), 69.2 (C-3), 67.1 (C-4), 62.0 (C-6), 27.3
(C-2), 26.7 (Me3C–Si), 20.9, 20.8, 20.5 (3 · CH3CO2), 19.2
(Me3C–Si). IR (cmÀ1): 3004, 2933, 2859, 2346, 1751, 1473,
1429, 1371, 1299, 1240, 1220, 1142, 1112, 1062, 1008, 977,
942. Anal. Calcd for C28H35IO8Si: C, 51.38; H, 5.39.
Found: C, 51.42; H, 5.31.
In summary, we have shown that the simple, cost effec-
tive and environmentally benign combination of NH4I
(or NH4Br), 50% aq H2O2 and Ac2O/AcOH in CH3CN
at low temperatures achieves efficient and highly stereo-
selective haloacetoxylation of protected glycals. The
application of this methodology to a wider range of ole-
fins is currently being investigated, together with its
compatibility with a wider array of protecting groups,
and will be reported later in full.
1
11b: H NMR (300MHz, CDCl3): d 7.67–7.33 (m, 10H,
Acknowledgements
2 · Ph), 5.87 (d, 1H, J = 9.3Hz, H-1), 5.30 (dd, 1H,
J = 9.3, 11.2Hz, H-4), 5.12 (t, 1H, J = 9.3Hz, H-3), 3.98
(dd, 1H, J = 9.3, 11Hz, H-2), 3.80–3.65 (m, 3H, H-5, H-6a
and H-6b), 2.18, 2.09, 1.89 (3s, 9H, 3 · CH3CO2), 1.04 (s,
9H, Me3C–Si). 13C NMR (100MHz, CDCl3): d 169.6,
169.3, 168.4 (3 · CH3CO2), 136.0, 135.6, 133.1, 133.0,
129.7, 129.6, 127.7, 127.6, 127.5 (2 · Ph), 94.0 (C-1), 75.7,
75.4 (C-3 and C-5), 68.8 (C-4), 62.2 (C-6), 29.7 (C-2), 26.7
(Me3C–Si), 20.7, 20.6, 20.5 (3 · CH3CO2), 19.2 (Me3C–
Si). IR (cmÀ1): 3075, 2995, 2930, 2857, 2346, 1759, 1464,
1428, 1366, 1271, 1267, 1263, 1260, 1238, 1217, 1113, 1058,
1007, 904, 889. Anal. Calcd for C28H35IO8Si: C, 51.38; H,
5.39. Found: C, 51.50; H, 5.55.
We thank the NRF (South Africa) and the Ministry of
the Flemish Community (Flanders, Belgium) for finan-
cialsupport for this research under the RSA-Felmish
Community bilateral agreement.
References and notes
´
1. Barluenga, J.; Marco-Arias, M.; Gonzales-Bobes, F.;
Ballesteros, A. Chem. Eur. J. 2004, 10, 1677–1682.
2. Roush, W. R.; Narayan, S. Org. Lett. 1999, 1, 899–902.
3. Durham, T. B.; Roush, W. R. Org. Lett. 2003, 5, 1871–
1874.
4. Gammon, D. W.; Hunter, R.; Steenkamp, D. J.; Mud-
zunga, T. T. Bioorg. Med. Chem. Lett. 2003, 13, 2045–
2049.
5. Sels, B. F.; De Vos, D. E.; Jacobs, P. A. J. Am. Chem. Soc.
2001, 123, 8350–8359.
6. Binkley, R. W.; Ambrose, M. G.; Hehemann, D. G. J.
Carbohydr. Chem. 1987, 6, 203–219.
7. Djurendic, E.; Vukojevic, N.; Dramicanin, T.; Canadi, J.;
Miljkovic, D. J. Serb. Chem. Soc. 1998, 63, 685–688.
8. Chong, P. Y.; Roush, W. R. Org. Lett. 2002, 4, 4523–4526.
9. Higgs, D. E.; Neen, M. I.; Detty, M. R. Org. Lett. 2000, 3,
349–352.
10. Sels, B. F.; Brosius, R.; De Vos, D. E.; Jacobs, P. A.;
Gammon, D. W.; Kinfe, H. H. Adv. Synth. Catal., in
press.
11. Lubbecke, H.; Boldt, P. Tetrahedron 1978, 34, 1577–1579.
12. Sempere C. J.; Nomen Ribe, R.; Serra Hosta, E.; Lopez
Belmonte, L., ES Patent number WO 2004/014829 A1,
2004.
13. A representative procedure is as follows: To a solution of
glucal 1 (100mg, 0.37mmol) in AcOH/CH3CN (1:1, 2mL)
12a:1H NMR (400MHz, CDCl3): d 7.77–7.18 (m, 20H,
4 · Ph), 6.45 (d, 1H, J = 1.6Hz, H-1), 4.97 (d, 1H,
J = 10.4Hz, CH2Ph), 4.74 (d, 1H, J = 11.6Hz, CH2Ph),
4.65 (d, 1H, J = 10.4Hz, CH2Ph), 4.57 (d, 1H, J = 11.6Hz,
CH2Ph), 4.48 (dd, 1H, J = 1.6, 4.4Hz, H-2), 4.26 (t, 1H,
J = 9.4Hz, H-4), 4.00 (dd, 1H, J = 3.2, 11.6Hz, H-6a),
3.88–3.82 (m, 2H, H-5 and H-6b), 3.25 (m, 1H, H-3), 2.01
(s, 3H, CH3CO2), 1.11 (s, 9H, Me3C–Si). 13C NMR
(100MHz, CDCl3): d 168.5 (CH3CO2), 135.9, 135.7, 129.6,
129.5, 128.5, 128.4, 128.1, 128.0, 127.9, 127.7, 127.6, 127.5
(4 · Ph), 95.8 (C-1), 75.5, 75.5, 75.1 (C-3, C-4, C-5,
CH2Ph), 71.2 (CH2Ph), 62.2 (C-6), 31.1 (C-2), 26.9
(Me3C–Si), 20.8 (CH3CO2), 19.3 (Me3C–Si). IR (cmÀ1):
3071, 3033, 3000, 2933, 2860, 2585, 2346, 1750, 1589, 1497,
1473, 1455, 1428, 1371, 1295, 1272, 1269, 1265, 1261, 1259,
1219, 1143, 1112, 1028, 1000, 937, 868. Anal. Calcd for
C38H43IO6Si: C, 60.79; H, 5.77. Found: C, 61.14; H, 5.94.
1
12b: H NMR (400MHz, CDCl3): d 7.70–7.21 (m, 20H,
4 · Ph), 5,84 (d, 1H, J = 9.6Hz, H-1), 4.98 (d, 1H,
J = 10Hz, CH2Ph), 4.91 (d, 1H, J = 4Hz, CH2Ph), 4.89
(d, 1H, J = 4.4Hz, CH2Ph), 4.77 (d, 1H, J = 10.8Hz,
CH2Ph), 4.00 (dd, 1H, J = 9.6, 10.4Hz, H-2), 3.95 (m, 2H,
H-6a and H-6b), 3.87 (t, 1H, J = 9Hz, H-4), 3.78 (dd, 1H,

9536
D. W. Gammon et al. / Tetrahedron Letters 45 (2004) 9533–9536
J = 9, 10.4Hz, H-3), 3.53–3.49 (m, 1H, H-5), 2.19 (s, 3H, for C38H43IO6Si: C, 60.79; H, 5.77. Found: C, 61.18; H,
CH3CO2), 1.05 (s, 9H, Me3C–Si). 13C NMR (100MHz,
CDCl3): d 168.8 (CH3CO2), 135.9, 135.5, 129.6, 126.5,
128.5, 128.4, 128.2, 128.0, 127.9, 127.7, 127.6, 127.5
(4 · Ph), 94.4 (C-1), 85.4 (C-3), 78.9 (C-4), 76.6 (C-5),
75.8 (C2Ph), 75.1 (CH2Ph), 62.0 (C-6), 30.5 (C-2), 26.8
(Me3C–Si), 20.7 (CH3CO2), 19.3 (Me3C–Si). IR (cmÀ1):
3073, 3033, 2999, 2931, 2858, 2670, 2349, 1760, 1589,
1497, 1472, 1455, 1428, 1362, 1310, 1271, 1267, 1263, 1260,
1222, 1152, 1112, 1068, 1048, 1029, 1013, 889. Anal. Calcd
5.75.
15. Miljkovic, D.; Djurendic, E.; Vukojevic, N.; Gasi, K.;
Csanadi, J. Carbohydr. Res. 1992, 233, 251–253.
16. Marzabadi, C. H.; Spilling, C. D. J. Org. Chem. 1993, 58,
3761–3766.
17. Copper, C. L.; Koubek, E. J. Chem. Ed. 1998, 75, 87–90.
18. Liebhafsky, H. A.; Mohammad, A. J. Am. Chem. Soc.
1933, 55, 3977–3988.
19. Ogata, Y.; Aoki, K. J. Org. Chem. 1966, 31, 1625–1629.