Directed dihydroxylation of cyclic allylic alcohols and trichloroacetamides using OsO4/TMEDA

Article abstract of DOI:10.1021/jo026161y

The oxidation of a range of cyclic allylic alcohols and amides with OsO₄/TMEDA is presented. Under these conditions, hydrogen bonding control leads to the (contrasteric) formation of the syn isomer in almost every example that was examined. Evidence for the bidentate binding of TMEDA to OsO₄ is presented and a plausible mechanism described.

Full text of DOI:10.1021/jo026161y

Dir ected Dih yd r oxyla tion of Cyclic Allylic Alcoh ols a n d
Tr ich lor oa ceta m id es Usin g OsO4/TMEDA
Timothy J . Donohoe,*,†,‡ Kevin Blades,‡ Peter R. Moore,‡ Michael J . Waring,‡
J on J . G. Winter,‡ Madeleine Helliwell,‡,§ Nicholas J . Newcombe,| and Geoffrey Stemp
Dyson Perrins Laboratory, South Parks Road, Oxford, OX1 3QY, UK, Department of Chemistry,
University of Manchester, Oxford Road, Manchester, M13 9PL, UK, AstraZeneca Pharmaceuticals,
Mereside, Alderley Park, Macclesfield, Cheshire, SK10 4TG, UK, and GlaxoSmithKline,
New Frontiers Science Park, Third Avenue, Harlow, Essex, CM19 5AW, UK
timothy.donohoe@chem.ox.ac.uk
Received J uly 9, 2002
The oxidation of a range of cyclic allylic alcohols and amides with OsO4/TMEDA is presented. Under
these conditions, hydrogen bonding control leads to the (contrasteric) formation of the syn isomer
in almost every example that was examined. Evidence for the bidentate binding of TMEDA to
OsO4 is presented and a plausible mechanism described.
In tr od u ction
Over the past few years, we have been engaged in a
program of research that aims to develop the directed
dihydroxylation of substituted alkenes. Perhaps the most
useful class of compounds that could be subjected to such
a protocol would be cyclic allylic alcohols because Kishi
demonstrated some years ago that these substrates are
dihydroxylated with a strong bias for the anti isomer;1
directed dihydroxylation would necessarily form the all-
syn isomer, which is difficult to access by other means,
Figure 1.
F IGURE 1. Syn versus anti selectivity for dihydroxylation.
Resu lts a n d Discu ssion
Oxid a tion of Cyclic Allylic Alcoh ols. We began our
studies with five-membered allylic alcohols (all made via
literature routes) and subjected four differently substi-
tuted compounds to oxidation with the OsO4/TMEDA
reagent, Scheme 1.3a Pleasingly, all of these compounds
gave the syn isomer as the major product, although the
level of stereoselectivity depended upon the substitution
pattern. In each example, the selectivity observed under
Upjohn conditions4 is also given so that a comparison can
be made between the result of a directed dihydroxylation
and that of a more standard dihydroxylation protocol.
One should be aware that with five-membered rings, both
allylic and homoallylic substituents can sometimes give
rise to syn products without recourse to hydrogen bond-
ing (allylic strain may be responsible, vide infra).5 Despite
this caveat, it can be seen that good levels of syn
selectivity are obtained with OsO4/TMEDA. In fact,
minimization of allylic strain between the hydroxyl and
the alkene methyl group may be responsible for the
increased preference for compound 3 to give syn products
relative to compound 1.
While our early attempts to design a protecting group
for allylic alcohols that would coordinate to osmium were
unsuccessful, we did discover that hydrogen bonding
between an allylic donor and OsO4 (as an acceptor) was
a useful method of forming the all-syn isomers.2 In
particular, we found that (in dichloromethane) the nature
of the amine additive used to speed up the reaction was
crucial to the degree of hydrogen bonding ability shown
by the oxidant. In fact, TMEDA additive was found to
produce a complex that was a very good hydrogen bond
acceptor and that was also capable of dihydroxylating
alkenes at -78 °C.3 We now wish to discuss in full our
results from oxidation of a wide variety of substituted
cyclic alkenes using the OsO4/TMEDA reagent.
* Corresponding author. Phone: (01865) 275649. Fax: (01865)
275674.
† Dyson Perrins Laboratory.
‡ University of Manchester.
§ To whom correspondence regarding the crystal structure should
be addressed.
(3) (a) Donohoe, T. J .; Moore, P. R.; Waring, M. J .; Newcombe, N. J .
Tetrahedron Lett. 1997, 38, 5027. (b) Donohoe, T. J .; Blades, K.; Moore,
P. R.; Winter, J . J . G.; Helliwell M.; Stemp, G. J . Org. Chem. 1999,
64, 2980. (c) Donohoe, T. J .; Waring M. J .; Newcombe, N. J . Tetrahe-
dron Lett. 1999, 40, 6881. (d) Donohoe, T. J .; Mitchell, L.; Waring, M.
J .; Bell, A.; Newcombe, N. J . Tetrahedron Lett. 2001, 42, 8951.
(4) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,
1973.
| AstraZeneca Pharmaceuticals.
GlaxoSmithKline.
(1) (a) Cha, J . K.; Christ W. J .; Kishi, Y. Tetrahedron Lett. 1983,
24, 3943. (b) Cha, J . K.; Christ, W. J .; Kishi, Y. Tetrahedron Lett. 1983,
24, 3947. (c) Cha, J . K.; Christ, W. J .; Kishi, Y. Tetrahedron 1984, 40,
2247. (d) Cha, J . K.; Kin, N.-S. Chem. Rev. 1995, 95, 1761.
(2) (a) Donohoe, T. J .; Garg, R.; Moore, P. R. Tetrahedron Lett. 1996,
37, 3407. (b) Donohoe, T. J .; Moore, P. R.; Beddoes, R. L. J . Chem.
Soc., Perkin Trans. 1 1997, 43.
(5) Poli, G. Tetrahedron Lett. 1989, 30, 7385. See also: Ward, S. E.;
Holmes, A. B.; McCague, R. Chem. Commun. 1997, 2085.
10.1021/jo026161y CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/25/2002
7946
J . Org. Chem. 2002, 67, 7946-7956

Directed Dihydroxylation of Allylic Alcohols and Amides
SCHEME 1a
SCHEME 2a
a
Reagents: (i) OsO4 (1 equiv), TMEDA (1 equiv), CH2Cl2, -78
°C, then NH2CH2CH2NH2; (ii) OsO4 (cat.), acetone, H2O, NMO,
rt; (iii) Ac2O, py.
Moreover, we presume that the lower level of selectiv-
ity displayed by 7 (relative to 1 and 5) is a consequence
of intramolecular hydrogen bonding within the substrate,
which reduces its ability to hydrogen bond to an incoming
reagent.
The ratios of stereosiomers were determined by 1H
NMR spectroscopy in conjunction with the crude reaction
mixtures. In each example shown in Scheme 1, the
relative stereochemistry was easy to assign because of
the different symmetry properties of the alcohol (or
peracetate) products; for example, the 13C NMR spectrum
of syn-2 consisted of three peaks.
One of the first observations to be made about the
OsO4/TMEDA oxidation is that it is not catalytic in
osmium; in fact, one forms an osmate ester by mixing
OsO4 (1 equiv), TMEDA (1 equiv), and alkene (1 equiv)
in CH2Cl2 at low temperature, vide infra. This osmate
ester must be cleaved during workup to liberate the diol.
Three methods have been used by the group: (1) Na2-
SO3 (aqueous) THF, ∆;6 (2) MeOH, HCl;3b and (3) H2-
NCH2CH2NH2 in CH2Cl2.3c Although all three methods
work well, generally, acidic methanol is the method of
choice even though silyl protecting groups do not often
survive this regime.7 Sodium sulfite is a reducing agent
and is presumed to reduce osmium(VI) and thereby
facilitate hydrolysis of the glycol ligand from the metal.
The other two sets of conditions are neither oxidizing nor
reducing and probably liberate the glycol ligand directly
a
Reagents: (i) OsO4 (1 equiv), TMEDA (1 equiv), CH2Cl2, -78
°C, then Na2SO3; (ii) OsO4 (cat.), acetone, H2O, NMO, rt; (iii) OsO4
(1 equiv), TMEDA (1 equiv), CH2Cl2, -78 °C, then HCl/MeOH;
(iv) Ac2O, py.
by displacement; our attempts to characterize the os-
mium-containing byproducts from these hydrolyses (es-
pecially with ethylenediamine) failed.
We then moved on to oxidize six-membered allylic
alcohols under directed dihydroxylation conditions, Scheme
2. The results showed that the directing effect was a
general one, and a variety of cyclic compounds were
oxidized with good levels of stereoselectivity. In fact,
oxidation of compound 13 represents one of the few cases
where our reagent is ineffective, forming syn- and anti-
14 with disappointing selectivity. This may be rational-
ized by considering the cleft into which the oxidant must
(6) See Schroeder, M. Chem. Rev. 1980, 80, 187 and ref 24.
(7) If retention of silyl groups is required, then an aqueous sodium
sulfite workup is recommended; see ref 25(b).
J . Org. Chem, Vol. 67, No. 23, 2002 7947

Donohoe et al.
SCHEME 3a
fit as it approaches the alkene: clearly, a pseudoaxial
hydroxyl group provides more hindrance to the osmium
reagent than a pseudoequatorial hydroxyl group. Also,
electronic deactivation of the alkene may be more preva-
lent with a pseudoaxial electronegative group. In support
of this hypothesis, an analogous difference in selectivity
has been observed for the directed epoxidation of these
two allylic alcohols.8
Examination of the sugar derived alkenes 17, 19, and
21 appears to show the opposite trend, as compound 19
contains a pseudoaxial hydroxyl group and is dihydroxy-
lated with higher diastereoselectivity than compound 17.
However, the configuration at the anomeric center should
not be ignored here, as both alcohols 199 and 2110 have
anomeric configurations chosen to aid the directing effect.
Using this methodology, concise syntheses of usefully
protected forms of the sugars allose and tallose can now
be accomplished. Moreover, the selective, contrasteric,
oxidation of cyclohexadiene diol 15 represents a one-step
(and stereoselective) synthesis of conduritol D.11
a
Reagents: (i) OsO4 (1 equiv), TMEDA (1 equiv), CH2Cl2, -78
°C, then aqueous Na2SO3 ∆.
While most of the starting materials used in this
section were known in the literature, many of the
products (especially the syn isomers) are novel com-
pounds. The structures of the syn isomers of compounds
10, 12, and 14 were deduced from their symmetry when
compared to the corresponding anti isomers (syn-10 has
four resonances in its 13C NMR spectrum). The spectro-
scopic data from compound syn-16, produced from the
dihydroxylation of 15, were an exact match with those
of conduritol D. The anti isomer of 18 is a known
compound12 and could be identified as the minor compo-
nent of the TMEDA-mediated oxidation of dihdyropyran
17. Moreover, the structure of syn-20 was assigned by
1H NMR spectroscopy, which showed a W coupling
between the hydrogen atoms on C-2 and C-4, in agree-
ment with the literature precedent reported by Doherty.13
Finally, the stereochemistry of syn-22 was determined
by examination of the coupling constants on the tetrahy-
dropyran ring; for example, the coupling between H(C-
1) and H(C-2) is 10 Hz, clearly indicative of a trans
diaxial arrangement of these two hydrogen atoms.
From these results, one can conclude that as long as
the allylic alcohol is not in a particularly hindered
environment then the directed dihydroxylation reaction
is an efficient way of producing a useful range of syn
allylic alcohols in one step.
Moreover, the structure of the fully protected allose
monosaccharide syn-22 was assigned unambiguously by
X-ray crystallography; see Supporting Information.
Oxid a tion of P r otected Allylic Am in es. Our next
objective was to examine the directed dihydroxylation of
(protected) cyclic allylic amines, hoping for a similar
sense and level of stereoselectivity. Preliminary studies
concentrated on finding the optimum protecting group
that would encourage hydrogen bonding. Various deriva-
tives of both cyclopentenyl- and cyclohexenylamine were
prepared and then dihydroxylated under directing condi-
tions Scheme 3. As far as carbonyl-based protecting
groups are concerned, there was a clear correlation
between pKa of the CONH and the directing ability of a
particular protecting group; acetates and Boc groups were
reasonably good hydrogen bond donors, while the acidic
trifluoro- and trichloracetamides proved to be excellent
at promoting high syn selectivity (g25:1 selectivity means
that we could not detect the other diastereoisomer by
high-field NMR spectroscopy of the crude reaction mix-
ture and with an authentic sample of the anti isomer for
comparison).
(8) Chamberlain, P.; Roberts, M. L.; Whitham, G. H. J . Chem. Soc.
B 1970, 1374.
(9) Compound 19 was made via a Mitsunobu reaction on compound
23.
(10) Compound 21 (unstable) was prepared from 24 in two steps;
see: Valverde, S.; Garcia-Ochoa, S.; Martin-Lomas, M. J . Chem. Soc.,
Chem. Commun. 1987, 383.
In fact this correlation between pKa and syn selectivity
can be taken as evidence in favor of hydrogen bonding
control; the failure of the (more acidic) sulfonamides14 to
direct efficiently may be due to an increased steric
interaction between the SO2 unit and the oxidant, which
disfavors effective hydrogen bonding.
Of the two trihaloamide derivatives that showed
promise as protecting and activating groups, we decided
to investigate the trichloroacetamide group in detail
because allylic trichloroacetamides are easy to prepare
(11) Carless, H. A. J .; Busia, K.; Dove, Y.; Malik, S. S. J . Chem.
Soc., Perkin Trans. 1 1993, 2505. For a review of synthesis from
cyclohexadiene diols, see: Hudlicky, T.; Gonzalez, D.; Gibson, D. T.
Aldrichimica Acta 1999, 32, 35.
(12) Haque, M. B.; Roberts, B. P.; Tocher, D. A. J . Chem. Soc., Perkin
Trans. 1 1998, 2881.
(13) (a) Harris, J . M.; Keranen, M. D.; O’Doherty, G. A. J . Org.
Chem. 1999, 64, 2982. (b) Haukaas, M. H.; O’Doherty, G. A. Org. Lett.
2001, 3, 3899.
(14) (a) Xu, D.; Park, C. Y.; Sharpless, K. B. Tetrahedron Lett. 1994,
35, 2495. (b) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. Rev. 1994, 94, 2483.
7948 J . Org. Chem., Vol. 67, No. 23, 2002

Directed Dihydroxylation of Allylic Alcohols and Amides
SCHEME 4a
32 proved the relative stereochemistry in these series
unambiguously (see Supporting Information); the identity
of syn-26 was assigned by analogy to that of the corre-
sponding trifluoro analogue (see Figure 5).
As expected, oxidation of the six-membered cyclic
trichloroacetamides (also made via the Overman rear-
rangement)21 gave excellent levels of syn diastereoselec-
tivty, Scheme 5, while the Upjohn reaction lead to decent
levels of anti selectivity. Again, the pseudoaxially locked
amide 39 did not give high levels of stereoselectivity, in
line with that observed for the corresponding alcohol 13.
The directed dihydroxylation of dihydropyran template
43 proved to be an efficient way of preparing protected
talosamine syn-4422 in high yield. Pleasingly, the seven-
membered substrate 45 also responded well to directed
dihydroxylation, giving syn-46 as the sole product from
oxidation, although the nonselective dihydroxylation
under Upjohn conditions is surprising.
The stereochemistry of syn- and anti-36 was assigned
by 1H NMR spectroscopy on the respective TMEDA
1
osmate esters. In H NMR spectroscopy derived from the
syn isomer, the C-2 hydrogen appeared as a pseudo
triplet with J ) 3.8 Hz, while that from the anti isomer
showed C2(H) as a dd with J ) 3.9 and 9.5 Hz, clearly
indicating the stereochemical outcome of the reaction.
X-ray crystallographic analysis of the (syn) osmate ester
derived from 38 proved to be informative (see Figure 5)
and served to secure the stereochemistry of syn-38 itself.
The structures of syn-40 (isopropylidine acetal), syn-42,
and syn-46 were also determined by X-ray crystal-
lography so that the relative stereochemistry could be
assigned with confidence; see Supporting Information.
Stereochemistry in the talosamine series was assigned
by removal of the trichloroacetamide group from syn-44
(NaOH) and peracetylation (Ac2O) to form a tetraacetate
that had spectroscopic data identical to those reported
in the literature.22,23
a
Reagents: (i) OsO4 (1 equiv), TMEDA (1 equiv), CH2Cl2, -78
°C, then HCl/MeOH; (ii) OsO4 (cat.), acetone, H2O, NMO, rt; (iii)
dimethoxypropane, TFA; (iv) Ac2O, py.
via the Overman rearrangement,15 which would, there-
fore, provide a general and convenient route to the
oxidation precursors.
Pleasingly, the directed dihydroxylation of a series of
five-membered allylic trichloroacetamides was high-
yielding and extremely stereoselective for the syn isomer
Scheme 4. We presume that the enhanced acidity of the
trichloroacetamide (and also the trifluoro analogue) rela-
tive to that of the corresponding allylic alcohol (pKa
values are approximately 11.2 and 14.7, respectively)16
means that hydrogen bonding to the OsO4/TMEDA
reagent is more effective: this manifests itself as a more
syn-selective oxidation reaction. It is noteworthy that
under Upjohn conditions, none of the substrates shown
below give high levels of stereoselectivity; and this failure
to produce the anti isomer may be a manifestation of
allylic strain, as noted earlier by Poli. In fact, in this five-
membered ring series, it may be that access to the anti
stereoisomer is more problematic than access to the syn
diastereoisomer!
Mech an ism of th e Dir ected Dih ydr oxylation . With
a comprehensive set of results detailing the behavior of
the OsO4/TMEDA mixture toward alkenes, we wish to
comment on the mechanism of oxidation. Clearly, our
first question relates to obtaining proof that hydrogen
(17) Compound 25 was prepared via Overman rearrangement of the
corresponding alcohol (80%).
(18) Compound 27 was prepared from 33 (see ref 15a) via metath-
esis.
(19) Daluge, S.; Vince, R. Tetrahedron Lett. 1976, 17, 3005.
(20) Compound 31 was prepared from a mixture of monoacetates
34 (one shown), themselves formed from oxidation of cyclopentadiene
with Pb(OAc)4; see: Banwell, M. G.; Ebenbeck, W.; Edwards, A. J . J .
Chem. Soc., Perkin Trans. 1 2001, 114.
The relative stereochemistry of the diols shown in
Scheme 4 was more difficult to ascertain by NMR
spectroscopy, so we made extensive use of X-ray crystal-
lography. An X-ray structure of syn-28, anti-30, and anti-
(15) (a) Overman, L. E. J . Am. Chem. Soc. 1976, 98, 2901. (b)
Overman, L. E. Acc. Chem. Res. 1980, 13, 218. (c) Overman, L. E.;
Clizbe, L. A.; Freerks, R. L.; Marlowe, C. K. J . Am. Chem. Soc. 1981,
103, 2807. (d) Nishikawa, T.; Asai, M.; Ohyabu, N.; Isobe, M. J . Org.
Chem. 1998, 63, 188.
(16) These are calculated pKas (Advanced Chemistry Development,
pKa Predictor Programme version 5.12).
(21) Novel compounds 37 (76%), 39 (12%), 43 (94%), and 45 (59%)
were prepared via Overman rearrangement.
(22) Donohoe, T. J .; Blades, K.; Helliwell, M. Chem. Commun. 1999,
1733.
(23) Banaszek A.; Karpiesiuk, W. Carbohydr. Res. 1994, 251, 233.
J . Org. Chem, Vol. 67, No. 23, 2002 7949

Donohoe et al.
SCHEME 5a
SCHEME 6a
a
Reagents: (i) OsO4 (1 equiv), TMEDA (1 equiv), CH2Cl2, -78
°C, then HCl/MeOH; (ii) NaH, MeI.
F IGURE 2. Corey’s X-ray structure of an OsO4‚diamine
complex.
to the metal increases the electron density on osmium
and, in so doing, reduces back-bonding by the oxo ligands,
thus making these atoms more electron rich. Of course,
electron-rich oxo ligands are prone to efficient hydrogen
bonding.
A logical extension of this argument supposes that
TMEDA forms a bidentate complex with osmium tetrox-
ide, making both the metal and oxo groups even more
electron-rich again.
At the outset of this work, these (formally 20e)
complexes were unknown in the literature and were only
recently identified by Corey, who obtained a crystal
structure of a chiral 1,2-diamine bound to OsO4, Figure
2.24 In this complex, the bidentate nature of the chiral
1,2-diamine ligand can be seen clearly. In fact, there is
significant precedent for the reaction of a bidentate amine
with osmium tetroxide and an alkene.25 However, despite
their enhanced reactivity and potential for asymmetric
induction, the ability of these complexes to act as
hydrogen bond acceptors had not been uncovered.
Our own studies had concentrated on NMR spectra of
the OsO4/TMEDA reagent (which is only stable at tem-
peratures e-50 °C). Both 1H (300 MHz) and 13C NMR
(75 MHz) of a 1:1 mixture of OsO4 and TMEDA at -78
°C showed the presence of a single compound, with two
a
Reagents: (i) OsO4 (1 equiv), TMEDA (1 equiv), CH2Cl2, -78
°C, then HCl/MeOH; (ii) OsO4 (cat.), acetone, H2O, NMO, rt; (iii)
Ac2O, py.
bonding is responsible for the syn selectivity that we
observe. This was proven convincingly when the O-
methyl ether 47 and the N-methyl amide 49 were
prepared and then subjected to dihydroxylation with
OsO4/TMEDA, Scheme 6. Not only were both of these
oxidations slower than those of the parent alcohol or
amide, but each gave a single diastereoisomer that was
formed from oxidation from the least hindered face of the
alkene. Anti-48 is a known compound,2b and the stereo-
chemistry of anti-50 was correlated to a sample of anti-
36 that was N-methylated.
Being convinced that the allylic alcohols and amides
were acting as hydrogen bond donors, we next examined
the role of osmium tetroxide and its complexes as
hydrogen bond acceptors. In previous work, we had
shown that the complex of OsO4 with Me3N was a better
hydrogen bond acceptor than OsO4 alone. A simple
explanation for this is that coordination of a Lewis base
(24) Corey, E. J .; Sarshar, S.; Azimiora, M. D.; Newbold R. C.; Noe,
M. C. J . Am. Chem. Soc. 1996, 118, 7851.
(25) (a) Tomioka, K.; Nakajima, M.; Iitaka, Y.; Koga, K. Tetrahedron
Lett. 1988, 29, 573. (b) Corey, E. J .; DaSilva J ardine, P.; Virgil, S.;
Yuen, W.; Connell, R. D. J . Am. Chem. Soc. 1989, 111, 9243. (c)
Hanessian, S.; Meffre, P.; Girard, M.; Beaudoin, S.; Sance´au, J .-Y.;
Bennani, Y. J . Org. Chem. 1993, 58, 1991. (d) Oishi, T.; Iida, K. I.;
Hirama, M. Tetrahedron Lett. 1993, 34, 3573. (e) Vedejs, E.; Galante,
R. J .; Goekjian, P. G. J . Am. Chem. Soc. 1998, 120, 3613.
7950 J . Org. Chem., Vol. 67, No. 23, 2002

Directed Dihydroxylation of Allylic Alcohols and Amides
SCHEME 7
analogues of TMEDA. Oxidation using OsO4 and Me3N
is 1.2:1 syn selective, which is the selectivity expected
from a standard monodentate ligand that has no extra
possibilities for hydrogen bonding through the amine
ligand. The amines shown in Scheme 7 clearly fall into
two distinct categories: (i) those that can form a chelate
with osmium and (ii) those that cannot. The former give
high stereoselectivity, while the latter give lower levels
of control. If model B were correct, then amines with a
pendant methoxy or hydroxyl group should give levels
of syn selectivity that approach that of TMEDA; however,
they do not. The methylamine-derived compound (Me2-
NCH2NMe2) should also be able to participate in a
transition structure such as B but cannot form a four-
membered chelate with osmium. The result shown in
Scheme 7 shows that this ligand is not much better than
trimethylamine in promoting hydrogen bonding, so cre-
dence is given to model A.
F IGURE 3. Overlay of low-temperature NMR spectra, (CD3)2-
CO, -78 °C: (A) 13C NMR of TMEDA; (B) 13C NMR of OsO4/
TMEDA; (C) 1H NMR of TMEDA; (D) 1H NMR of OsO4/
TMEDA.
As mentioned earlier, the product of the reaction
between an alkene, OsO4, and TMEDA produces an
osmate ester, which must be hydrolyzed to produce the
diol product. Unusually, these osmate esters (18 elec-
trons) are quite stable and can be purified by chroma-
tography on silica. In fact, it is the stability of these
complexes that precludes the use of chelating diamines
with catalytic OsO4, which cannot be regenerated from
the osmate ester in situ by an oxidizing agent.26 Exami-
nation of the crystal structures of syn-osmate esters 51
and 52 derived from (TMEDA) OsO4-mediated dihy-
droxylation of the parent alkenes is informative and
reveals two interesting features. First, the chelated
nature of the TMEDA ligand is evident. Second, the
proximity of the amide N and glycolate oxygen attached
to osmium is noteworthy: these two atoms are closer
than the sum of Van der Waal’s radii, which is indicative
of a hydrogen bond, between them (Figure 5). Intramo-
lecular hydrogen bonding is also suggested by the NMR
spectra of these osmate esters, which show a particularly
low-field signal for the amide NH (δ 7-8 ppm) that is
not present in the corresponding anti osmate esters
(typically NH at δ 6-7 ppm). While these two features
of compounds 51 and 52 are clearly those of the product
of osmylation, they need not necessarily have been
conserved from the preceding transition states for osmy-
lation, so caution is appropriate when interpreting these
results.
F IGURE 4. Plausible interactions of TMEDA with OsO4.
peaks for the TMEDA ligand, and no uncomplexed
TMEDA was observed (Figure 3); these data are consis-
tent with a bidentate complex.
In addition, we also measured the low temperature
(-78 °C, CH2Cl2) IR spectra of OsO4 and OsO4/quinucli-
dine and OsO4/TMEDA in order to qualitatively assess
the OsdO bond order in this series. Under these condi-
tions, OsO4 absorbs at 953 cm-1, OsO4/quinuclidine at
910 cm-1, and OsO4/TMEDA at 880 cm-1. This trend
toward absorption at a lower wavenumber as the electron
density on the metal increases correlates well with our
observation that hydrogen bonding ability (and therefore
syn selectivity) increases in the order OsO4 < OsO4‚
monodentate amine < OsO4‚chelating diamine. Natu-
rally, caution is called for here because the three com-
plexes have different shapes, and therefore we cannot be
certain that we are observing the same type of absorption
in each case. Therefore, we present these data as
circumstantial evidence in favor of the electronic nature
of the OsO4/TMEDA complex.
If TMEDA forms (and then reacts via) a bidentate
complex, then this must hydrogen bond to the substrate
through an oxo ligand; see A, Figure 4. However, if the
reactive species contains TMEDA bound in a monoden-
tate fashion, then its extraordinary reactivity must be
explained by hydrogen bonding through the nonligated
amino group of TMEDA (B, Figure 4).
One aspect of the OsO4/TMEDA reagent not discussed
so far is the profound influence that TMEDA has on the
To investigate this second possibility, we performed the
oxidation of alcohol 11 using a series of bifunctionalized
(26) However, see: J onsson, S. Y.; Adolfsson, H.; Backva¨ll, J .-E. Org.
Lett. 2001, 3, 3463.
J . Org. Chem, Vol. 67, No. 23, 2002 7951

Donohoe et al.
F IGURE 5. N‚‚‚O distance in 51 is 2.67 Å and N‚‚‚O distance in 52 is 2.61 Å.
reactivity of OsO4 toward alkenes. Sharpless has mea-
sured the rate enhancement that quinuclidine imparts
onto the dihydroxylation reaction to be an approximately
100-fold increase (depending on the alkene substitution
pattern).27 Moreover, Corey states that bidentate (chiral)
diamine complexes are more reactive than complexes of
OsO4 with monodentate amines, again by a factor of
g100;24 clearly, this means that OsO4/TMEDA is ex-
pected to be 10 000 times more reactive than OsO4. This
estimate is borne by our observations that OsO4/TMEDA
accomplishes the osmylation of alkenes in minutes at -78
°C, under which conditions OsO4‚quinuclidine or OsO4
are inert: this fact alone demands an explanation involv-
ing a unique role for TMEDA. From our own studies, we
also know that the OsO4/TMEDA reagent does not show
the same levels of discrimination for electron-rich alkenes
over electron-poor alkenes as OsO4, which fits with it
being a relatively electron-rich oxidizing agent. So, how
can we explain the fact that TMEDA makes the oxidant
more electron rich and yet much more reactive, even
toward already electron-rich alkenes? Our observations
on this difficult question center around the role of the
metal during oxidation. Osmium tetroxide is a 16 electron
complex, whereas an osmate ester formed after osmyla-
tion is a 14 electron species and (inasmuch as these
compounds have a desire to attain an 18 electron con-
figuration) more electron deficient. For example, osmate
esters can readily bind two moles of a cinchona alkaloid
ligand, whereas coordination of only one such alkaloid
is observed for OsO4.28 If the metal becomes significantly
more electron deficient during osmylation, then the
presence of a chelating diamine, increasing electron
density on OsO4 beforehand, should help to assuage this
loss of electron density and promote the reaction.
In summing up our data on the mechanism, we have
shown that the OsO4 TMEDA reagent is chelated, with
respect to the metal, both before and after the osmylation
event. The most logical extrapolation of these observa-
tions is to assume that the ligand remains bidentate
throughout the course of the oxidation: such a role for
the ligand also helps to explain the dramatically in-
creased reactivity of the complex relative to OsO4.quinucli-
dine. Of course, this particular mechanism requires that
the oxidant behave as a hydrogen bond acceptor through
its oxo ligands, a conclusion that is also supported by
experiments with a range of different analogues of
TMEDA. This reaction pathway leads directly to the
osmate ester as shown above with contrasteric (syn)
stereoselectivity.
Whatever the precise details of the osmylation reaction,
we have uncovered an interesting and useful variant of
the dihydroxylation reaction that shows excellent hydro-
gen bonding ability with a wide range of allylic alcohols
and amides giving access to all syn triol and amino-diol
motifs in five-, six-, and seven-membered rings. The
information gained about the effect of an amine additive
on the hydrogen bonding ability of the oxidant should
(28) J acobsen, E. N.; Marko, I.; France, M. B.; Svendsen, J . S.;
Sharpless, K. B. J . Am. Chem. Soc. 1989, 111, 737.
(27) Anderson, P. G.; Sharpless, K. B. J . Am. Chem. Soc. 1993, 115,
7047.
(29) For example, see: Blades, K.; Donohoe, T. J .; Winter, J . J . G.;
Stemp, G. Tetrahedron Lett. 2000, 41, 4701.
7952 J . Org. Chem., Vol. 67, No. 23, 2002

Directed Dihydroxylation of Allylic Alcohols and Amides
1
yield the crude product (as a single isomer by H NMR). The
prove to be useful in designing new dihydroxylating
systems, especially ones that are catalytic in transition
metal.29
crude product was purified by flash chromatography (2:1
dichloromethane-methanol) to yield the title compound as a
colorless oil (59 mg, 88%): 1H NMR (200 MHz; CD3OD) δ
3.80-3.72 (m, 2 H), 2.00-1.68 (m, 4 H), 1.22 (s, 3 H); 13C NMR
(75 MHz; CD3OD) δ 77.93, 77.62, 29.49, 23.25; IR (film) 3314
(br), 2968, 2944, 1058 cm-1; CIMS m/z (rel intensity) 133 (MH+,
30%,); C6H16NO3 requires 150.1130, found MH+ 150.1131.
Exp er im en ta l Section
Gen er a l Deta ils. All reactions were carried out under an
atmosphere of dry nitrogen. Proton nuclear magnetic reso-
nance spectra (NMR) were recorded at 300, 400, or 500 MHz.
13C NMR spectra were recorded at 75, 100, or 125 MHz.
Coupling constants (J ) are given in Hz. Infrared spectra (IR)
were recorded as evaporated films. Chemical ionization (CI)
was performed using NH3. Microanalysis was obtained from
the microanalytical section of the University of Manchester’s
Chemistry Department. All solvents and reagents requiring
purification were purified using standard laboratory tech-
niques according to methods published in Purification of
Laboratory Chemicals by Perrin, Armarego, and Perrin (Per-
gamon Press, 1966).
Gen er a l P r oced u r es for Dih yd r oxyla t ion : OsO4‚-
TMEDA. To a solution of substrate (0.01 M) and TMEDA (1.1
equiv) in dichloromethane precooled to -78 °C was added a
solution of OsO4 (1.05 equiv) in dichloromethane (∼1 mL). The
solution turned deep red and then brown-black. The solution
was stirred until complete (TLC analysis, ca. 1 h) before being
allowed to warm to room temperature.
Wor k u p A w ith Sod iu m Su lfite. The solvent was removed
in vacuo and replaced with THF (10 mL) and (aqueous) sodium
sulfite (10 mL). This mixture was heated at reflux for 3 h and
the product extracted as indicated.
Wor k u p B w ith Acid ic Meth a n ol. The solution was
concentrated under reduced pressure and the resulting residue
redissolved in methanol (10 mL). Hydrochloric acid (concen-
trated, ∼5 drops) was then added and the solution stirred for
2 h after which time the solvent was removed under reduced
pressure and the product isolated as indicated.
Wor k u p C w ith Eth ylen ed ia m in e. Ethylenediamine (5.0
equiv) was added to the crude reaction mixture at room
temperature and the resulting solution stirred for 48 h during
which time a brown precipitate formed. The solution was then
concentrated under reduced pressure and the product isolated
as indicated.
Up joh n Con d ition s. To a stirred solution of substrate (0.01
M) and N-methylmorpholine-N-oxide monohydrate (3.0 equiv)
in 4:1 acetone-water was added OsO4 (1 crystal, ∼1 mol %)
and the solution stirred until complete (TLC analysis, ca. 4
h). Sodium sulfite solution (saturated, 10 mL) was then added
and the solution stirred for an additional 30 min. The solvent
was then evaporated under reduced pressure and the product
isolated as outlined below.
syn -1,2,3,4-Tetr a a cetoxycyclop en ta n e Syn -6. cis-Cyclo-
pent-2-en-1,4-diol 5 (100 mg, 1.0 mmol) was oxidized with
OsO4/TMEDA using the ethylamine diamine workup. The
residue was redissolved in pyridine (10 mL) and acetic
anhydride (10 mL), and N,N-(dimethylamino)pyridine (5 mg,
catalytic) was added. The mixture was then heated at 80 °C
for 12 h. It was then allowed to cool and diluted with diethyl
ether (200 mL). The solution was then filtered through celite
and washed with dilute hydrochloric acid (100 mL, 2 M),
potassium carbonate solution (saturated, 100 mL), and brine
(100 mL). The ethereal layer was then dried (MgSO4) and
concentrated under reduced pressure to yield the crude
mixture of peracetylated products (>25:1 mixture by 1H NMR),
which were then purified by flash chromatography (10:1 light
petroleum-EtOAc) to yield the title compound as a colorless
oil (220 mg, 73%): 1H NMR (300 MHz, CDCl3, Me4Si) δ 5.21
(dd, J ) 4 and 2 Hz, 2 H), 5.08-5.16 (m, 2 H), 2.60 (dt, J ) 15
and 8 Hz, 1 H), 2.02 (s, 6 H), 2.00 (s, 6 H), 1.92-2.04 (m, 1 H);
13C NMR (75 MHz, CDCl3) δ 170.0, 169.6, 70.4, 69.1, 34.6, 20.6,
20.5; IR (film) 2994, 2941, 1744, 1435, 1372, 1231 cm-1; CIMS
m/z C13H22NO8 requires M 320.1345, found MNH4+, 320.1358.
Anal. Calcd for C13H18O8: C, 51.66; H, 6.00. Found: C, 51.89;
H, 6.29.
a n ti-1,2,3,4-Tetr a a cetoxycyclop en ta n e An ti-8. cis-1-Cy-
clopenten-3,4-diol 7 (100 mg, 1.0 mmol) was oxidized using
Upjohn conditions. The peracetylation and workup procedure,
which was identical to that for syn-6, gave the crude mixture
of isomers (10:1 mixture by 1H NMR). Purification by flash
chromatography (10:1 light petroleum-EtOAc) gave the title
compound as a colorless oil (182 mg, 62%): 1H NMR (300 MHz,
CDCl3) δ 5.43 (qd, J ) 4 and 1 Hz, 2 H), 5.34 (dd, J ) 4 and
2 Hz, 2H), 2.28 (t, J ) 5 Hz, 2H), 2.09 (s, 6H), 2.08 (s, 6H); 13
NMR (75 MHz, CDCl3) δ 169.9, 169.8, 73.5, 68.2, 34.8, 20.7,
20.5; IR (film) 2921, 2851, 1744, 1372, 1229 cm-1; CIMS m/z
(rel intensity) 320 (100%, MNH4+); C13H22NO8 requires
C
+
320.1345, found MNH4 320.1345.
syn -Cycloh exa n e-1,2,3-tr iol Syn -10. 2-Cyclohexene-1-ol
9 (50 mg, 0.51 mmol) was oxidized with OsO4/TMEDA using
the sodium sulfite workup. The crude reaction mixture was
then concentrated in vacuo to a gray powder. Ethanol (30 mL)
was added and the suspension stirred at room temperature
for 1 h. Filtration of this suspension through Celite and
concentration under reduced pressure gave a colorless solid
(80 mg). Flash chromatography (1:7 petrol-EtOAc) gave
inseparable syn- and anti-10 (66 mg, 98%) in a ratio of 9:1.
Syn-10: 1H NMR (300 MHz, D2O) δ 3.81 (t, J ) 2.6 Hz, 1 H),
3.52 (ddd, J ) 10, 4.6 and 2.6 Hz, 2 H), 1.8-1.0 (m, 6 H); 13C
NMR (75 MHz, D2O) δ 72.55, 70.27, 26.29 and 18.77; IR (film)
3192 and 2927 cm-1; CIMS m/z (rel intensity) 150 (100%,
syn -Cyclop en ta n -1,2,3-tr iol Syn -2. 2-Cyclopentenol 1 (100
mg, 1.19 mmol) was oxidized with OsO4/TMEDA using the
ethylamine diamine workup. The residue was redissolved via
sonication in a mixture of ethanol (7.5 mL) and ethyl acetate
(40 mL) and the resulting solution filtered through celite and
concentrated under reduced pressure to yield the crude
1
mixture of syn and anti triols in a 7:1 ratio (by H NMR). The
MNH4+); C6H16NO3 requires M 150.1130, found MNH4
+
products were purified by flash chromatography (EtOAc) to
yield the title compound, as a mixture of isomers, as a clear
oil (107 mg, 76%). Pure syn-cyclopentane-1,2,3-triol 2 was then
separated by further chromatography: 1H NMR (300 MHz,
CD3OD) δ 4.02-4.09 (m, 2 H), 3.82 (t, J ) 5 Hz, 1 H), 1.76-
1.95 (m, 4 H); 13C NMR (75 MHz, CD3OD) δ 71.9, 69.9, 27.1;
IR (film) 3365 (br), 2962, 2926 cm-1; CIMS m/z (rel intensity)
154 (100%, M2NH4), 90 (40); C5H14NO3 requires M 136.0974,
150.1128.
syn -5-ter t-Bu tylcycloh exa n e-1,2,3-tr iol Syn -12. cis-5-
tert-Butylcyclohex-2-enol 11 (50 mg, 0.32 mmol) was oxidized
with OsO4/TMEDA using the sodium sulfite workup; the
isolation procedure, which was identical to that shown for syn-
10, gave a colorless solid. Chromatography through a small
plug of silica (eluting with 6:1 EtOAc/EtOH) gave syn-12 as a
colorless crystalline solid (55 mg, 91%) in a ratio of 24:1 (1H
NMR). The spectroscopic data for this compound matched that
in the literature.2b
Con d u r itol D Syn -16. cis-Cyclohexa-3,5-diene-1,2-diol 15
(50 mg, 0.45 mmol) was oxidized with OsO4/TMEDA using the
sodium sulfite workup; the isolation procedure, which was
identical to that shown for syn-10, gave a light brown oil (70
+
found MNH4 136.0979.
syn -2-Meth ylcyclop en ta n -1,2,3-tr iol Syn -4. 2-Methylcy-
clopent-2-en-1-ol 3 (50 mg, 0.51 mmol) was oxidized with OsO4/
TMEDA using the ethylamine diamine workup. The residue
was redissolved via sonication in a mixture of ethanol (7.5 mL)
and ethyl acetate (40 mL) and the resulting solution filtered
through celite and concentrated under reduced pressure to
J . Org. Chem, Vol. 67, No. 23, 2002 7953

Donohoe et al.
mg). Flash chromatography (9:1 EtOAc/MeOH) gave an in-
separable mixture of tetraols as a colorless oil (35 mg, 54%).
1H NMR spectroscopy (200 MHz/D2O) showed a mixture of
conduritol D and conduritol E in a ratio of 16:1.11
H), 5.61 (t, J ) 2.8 Hz, 1 H), 5.21 (dd, J ) 2.8 and 10.1 Hz, 1
H), 4.78 (dd, J ) 2.8 and 10.1 Hz, 1 H), 4.71 (d, J ) 10.0 Hz,
1 H), 4.19 (dd, J ) 1.9 and 12.3 Hz, 1 H), 4.08 (dd, J ) 4.7 and
12.3 Hz, 1 H), 4.03 (ddd, J ) 1.9, 4.7 and 10.0 Hz, 1 H), 2.17
(s, 3 H), 2.11 (s, 3 H), 1.98 (s, 3 H), 1.96 (s, 3 H); 13C NMR (75
MHz, CDCl3) δ 170.2, 169.5, 168.8, 168.7, 134.9, 134.4, 130.3,
128.8, 86.7, 72.6, 68.0, 65.4, 64.8, 61.4, 20.6, 20.5, 20.4; IR (thin
film) 1756, 1373 and 1227 cm-1; CIMS m/z (rel intensity) 490
(100%, MNH4+), 490 (100%, MNH4+); C20H28NO11S requires
M 490.1383, found MNH4+ 490.1383; mp 165 °C. Anal. Calcd
for C20H24O11S: C, 50.82; H, 5.08. Found: C, 50.87; H, 4.98.
2,3,4,6-Tetr a -O-a cetyl-1,5-a n h yd r o-D-a llitol Syn -18. Diol
17 (100 mg, 0.77 mmol) was oxidized with OsO4/TMEDA using
the sodium sulfite workup. The aqueous mixture was concen-
trated under reduced pressure and then dried in vacuo to a
gray solid. This solid was powdered, and pyridine (10 mL),
acetic anhydride (5 mL), and 4-(dimethylamino)pyridine (cata-
lytic) were added. The resulting black suspension was stirred
at room temperature under an atmosphere of nitrogen for 48
h. Diethyl ether (100 mL) was added and the mixture filtered
through Celite (washing with additional volumes of diethyl
ether). The combined filtrate was washed with 2 M HCl (100
mL), aqueous sodium hydrogen carbonate (100 mL), and brine
(100 mL). The organic extracts were dried (MgSO4) and
concentrated under reduced pressure to a light brown oil (201
mg). 1H NMR spectroscopy (300 MHz, CDCl3) showed a 1:6
mixture of tetraacetates anti and syn-18. Flash chromatogra-
phy (6:1 petrol-EtOAc) gave 2,3,4,6-tetra-O-acetyl-1,5-anhy-
dro-D-allitol syn-18 (161 mg, 63%) as a colorless oil: [R]27D +7.5
syn -N-(2,3-Dih yd r oxycyclop en tyl)-2′,2′,2′-tr ich lor oa c-
eta m id e Syn -26. Trichloroacetamide 25 (100 mg, 0.438 mmol)
was subjected to standard TMEDA dihydroxylation conditions
using the methanolic workup to produce a brown solid, which
was purified by column chromatography (60% ethyl acetate/
petrol) to produce syn-26 (92 mg, 80%) as a colorless solid: 1H
NMR (300 MHz, CDCl3) δ 7.55 (br s, 1 H), 4.20-4.27 (m, 2 H),
4.08 (dd, J ) 5.2 and 4.4 Hz, 1 H), 3.04 (br s, 1 H), 2.08-2.17
(m, 1 H) and 1.80-1.98 (m, 3 H); 13C NMR (75 MHz, (CD3)2-
CO) δ 160.6, 92.8, 72.8, 72.6, 52.3, 29.7 and 27.6; IR (film) 3305,
2925, 1693 and 1515 cm-1; CIMS m/z (rel intensity) 262 (100%,
MH+), 283 (28), 281 (100), 279 (93), 266 (21), 264 (73) and 262
(70); C7H11NO3Cl3 requires M 261.9804, found MH+ 261.9804;
mp 112-114 °C. Anal. Calcd for C7H10NO3Cl3: C, 32.0; H, 3.8;
N, 5.3. Found: C, 32.2; H, 4.0; N, 5.3.
1
(c 0.2, CHCl3); H NMR (300 MHz, CDCl3) δ 5.60 (t, J ) 2.6
Hz, 1 H), 4.95 (ddd, J ) 10, 5.5 and 2.6 Hz, 1 H), 4.84 (ddd, J
) 10 and 2.6 Hz, 1 H), 4.10-4.14 (m, 2 H), 3.78-3.88 (m, 2
H), 3.60 (t, J ) 10 Hz, 1 H), 2.10 (s, 3 H), 2.02 (s, 3 H), 1.94 (s,
3 H); 13C NMR (75 MHz, CDCl3) δ 170.69, 169.89, 169.30,
169.11, 71.71, 67.77, 66.40, 66.27, 63.45, 62.49, 20.70 (2) and
20.55 (2); IR (film) 2996 and 1747 cm-1; CIMS m/z (rel
intensity) 350 (10%, MNH4+) and 249 (100); C14H24NO9 re-
a n ti-N-(2,3-Dih yd r oxycyclop en tyl)-2′,2′,2′-tr ich lor oa c-
eta m id e An ti-26. Trichloroacetamide 25 (100 mg, 0.438
mmol) was subjected to standard Upjohn conditions, and the
products were purified by column chromatography (60% ethyl
acetate/petrol) to give two white solids; syn-26 (58 mg) and
anti-26 (32 mg) overall yield 78%. Anti-26: 1H NMR (300 MHz,
(CD3)2CO) δ 8.00 (br s, 1 H), 3.80-4.10 (m, 3 H), 3.64 (br s, 1
H), 1.80-1.97 (m, 2 H) and 1.38-1.63 (m, 2 H); 13C NMR (75
MHz, (CD3)2CO) δ 161.5, 77.1, 71.3, 57.3, 29.0 and 25.8; IR
(film) 3318, 2931, 1694 and 1517 cm-1; CIMS m/z (rel intensity)
283 (28%), 281 (100), 279 (95), 266 (20), 264 (60) and 262 (57);
C7H11NO3Cl3 requires M 261.9804, found MH+ 261.9809; mp
146-148 °C.
+
quires M 350.1451, found MNH4 350.1454.
2,3,4,6-Tetr a -O-a cetyl-1,5-a n h yd r o-D-m a n n itol An ti-18.
Diol 17 (50 mg, 0.39 mmol was oxidized using Upjohn
conditions. The peracetylation and workup procedure, which
was identical to that for syn-18, gave a crude mixture of
isomers. 1H NMR spectroscopy (300 MHz, CDCl3) of this oil
showed a mixture of the two tetraacetates anti-18 and syn-18
in a ratio of 16:1. Flash chromatography (4:1 petrol-EtOAc)
gave anti-18 as a colorless oil (89 mg, 70%). The spectroscopic
data for this compound matched that in the literature.12
2,3,4,6-Tetr a a cetoxy-1-ben zyloxy-ta lose Syn -20. Alcohol
19 (0.19 g, 0.54 mmol) was oxidized with OsO4/TMEDA using
the methanolic workup. The resulting crude tetraol was
dissolved in pyridine (20 mL), and to this was added acetic
anhydride (5 mL) and a catalytic amount of DMAP. The
reaction was allowed to stir overnight at room temperature.
The pyridine was removed in vacuo until a dry black solid was
obtained. The product was purified by column chromatography
(40% ethyl acetate in petroleum ether 40-60) to afford syn-
20 as a clear oil (0.22 g, 92%): 1H NMR (300 MHz, CDCl3) δ
7.40-7.27 (m, 5 H), 5.34-5.28 (m, 2 H), 5.13 (dt, J ) 3.8 and
1.2 Hz, 1 H), 4.95 (d, J ) 1.2 Hz, 1 H), 4.71 (d, J ) 11.8 Hz, 1
H), 4.54 (d, J ) 11.8 Hz, 1 H), 4.27 (dt, J ) 1.2 and 7.2 Hz, 1
H), 4.19-4.15 (m, 2 H), 2.12 (s, 3 H), 2.12 (s, 3 H), 2.10 (s, 3
H), 1.97 (s, 3 H); 13C NMR (75 MHz, CDCl3) δ 170.3, 170.1,
169.9, 169.6, 136.1, 128.5, 128.2, 128.1, 97.3, 69.6, 67.5, 66.8,
65.9, 65.6, 62.0, 20.8, 20.7, 20.6, 20.5; IR (thin film) 3032, 1769,
1372, 1225, 1069 cm-1; CIMS m/z (rel intensity) 456 (100%,
MNH4+), 331 (70); C21H30NO10 requires M 456.1870, found
syn -N -(2,3-Dih yd r oxy-1-m et h yl-cyclop en t yl)-2′,2′,2′-
tr ich lor oa ceta m id e Syn -28. Trichloroacetamide 27 (70 mg,
0.29 mmol) was subjected to standard TMEDA dihydroxylation
conditions using the methanolic workup and the resulting
brown oil purified by column chromatography (80% diethyl
ether/petrol) to give syn-28 (65 mg, 81%) as a colorless solid:
1H NMR (300 MHz, CDCl3) δ 7.61 (br s, 1 H), 4.28 (br q, J )
6 Hz, 1 H), 3.70 (d, J ) 6 Hz, 1 H), 3.20 (br s, 2 H), 2.34-2.49
(m, 1 H), 1.88-2.02 (m, 1 H), 1.60-1.76 (m, 2 H), 1.43 (s, 3
H); 13C NMR (75 Mz, CDCl3) δ 160.9, 93.5, 78.8, 71.6, 61.3,
33.0, 30.2 and 23.1; IR (film) 3368, 2972, 2940, 1702 and 1505
cm-1; CIMS m/z (rel intensity) 297 (35%), 295 (95), 293 (100,
MH+), 280 (20), 278 (50) and 276 (60); C8H13NO3Cl3 requires
M 293.0226, found MH+ 293.0230; mp 62-64 °C.
2,2-Dim eth yl-6-(2′,2′,2′-tr ich lor oa cetyla m in o)-tetr a h y-
d r o-cyclop en ta [1,3]d ioxole-4-ca r boxylic Acid Meth yl Es-
ter Syn -30. Trichloroacetamide 29 (100 mg, 0.349 mmol) was
subjected to standard TMEDA dihydroxylation conditions with
methanolic workup, and the resulting brown oil was purified
by column chromatography (60% ethyl acetate/petrol) to give
a 5:1 mixture of inseparable syn- and anti-diols (98 mg, 88%)
as colorless oils. A small sample of these diols was converted
into its acetonide, see below, for separation: 1H NMR (300
MHz, CDCl3) δ 7.24 (br d, J ) 7.6 Hz, 1 H), 4.90 (t, J ) 6.0
Hz, 1 H), 4.63 (t, J ) 6.0 Hz, 1 H), 4.00-4.19 (m, 1 H), 3.76 (s,
3 H), 2.80 (dt, J ) 12.5 and 6.0 Hz, 1 H), 2.28 (dt, J ) 12.5
and 6.0 Hz, 1 H), 2.05 (q, J ) 12.5 Hz, 1 H), 1.49 (s, 3 H), 1.37
(s, 3 H); 13C NMR (75 Mz, CDCl3) δ 170.0, 161.4, 111.5, 92.3,
79.5, 78.2, 52.0, 51.9, 45.7, 29.2, 25.5, 24.1; IR (film) 3383, 2921,
1737, 1692 and 1524 cm-1; CIMS m/z (rel intensity) 381 (30%),
379 (90) and 377 (100, MNH4+); C12H17NO5Cl3 requires M
+
MNH4 456.1866.
2,3,4,6-Tetr a a cetoxy-6-ben zen esu lfon yl a llose Syn -22.
â-Sulfone 21 (0.10 g, 0.26 mmol) was oxidized with OsO4/
TMEDA using the methanolic workup. The resulting crude
tetraol was dissolved in pyridine (20 mL), and to this was
added acetic anhydride (5 mL) and a catalytic amount of
DMAP. The reaction was allowed to stir overnight at room
temperature. The pyridine was removed in vacuo until a dry
black solid was obtained. The product was purified by column
chromatography (60% diethyl ether in petroleum ether 40-
60) to afford the tetraacetate syn-22 as a colorless crystalline
solid (0.10 g, 82%): 1H NMR (500 MHz, CDCl3) δ 7.90 (d, J )
7.5 Hz, 2 H), 7.71 (t, J ) 7.4 Hz, 1 H), 7.57 (t, J ) 7.5 Hz, 2
7954 J . Org. Chem., Vol. 67, No. 23, 2002

Directed Dihydroxylation of Allylic Alcohols and Amides
360.0172, found MH+ 360.0167; mp 186-187 °C. Anal. Calcd
for C12H16NO5Cl3: C, 40.0; H, 4.5; N, 3.9. Found: C, 39.9; H,
4.5; N, 3.8.
(m, 3 H), 3.00-3.70 (br m, 2 H), 1.58-1.84 (m, 5 H), 1.32-
1.46 (m, 1 H);
70.3, 52.1, 28.4, 26.1, 18.2, IR (film) 3407, 2942, 1698, 1515
cm-1; CIMS m/z (rel intensity) 295 (91%) and 293 (100,
MNH4+); C8H13NO3Cl3 requires M 275.9961, found MH+
275.9966.
13C NMR (75 MHz, CDCl3) δ 161.8, 92.6, 70.8,
2,2-Dim eth yl-6-(2′,2′,2′-tr ich lor oa cetyla m in o)-tetr a h y-
d r o-cyclop en ta [1,3]d ioxole-4-ca r boxylic Acid Meth yl Es-
ter An ti-30. Trichloroacetamide 29 (100 mg, 0.35 mmol) was
subjected to standard Upjohn conditions; the crude product
was redissolved in acetone (20 mL), and then 2,2-dimethox-
ypropane (182 mg, 5 equiv) and trifluoroacetic acid (3 drops)
were added. The mixture was stirred for 1 h, concentrated in
vacuo, and purified by column chromatography (20% ethyl
acetate/petrol) to separate syn-30 (58 mg, 46%) and anti-30
(64 mg, 51%) acetonides as two colorless solids; overall, a 1:1.1
syn:anti ratio was observed (122 mg, 97%): 1H NMR (300
MHz, CDCl3) δ 8.16 (br d, J ) 7.1 Hz, 1 H), 4.77 (d, J ) 5.4
Hz, 1 H), 4.47 (d, J ) 5.4 Hz, 1 H), 4.33 (t, J ) 7.1 Hz, 1 H),
3.69 (s, 3 H), 3.08 (d, J ) 8.3 Hz, 1 H), 2.46 (ddd, J ) 7.1, 8.3
and 14.8 Hz, 1 H), 1.96 (d, J ) 14.8 Hz, 1 H), 1.40 (s, 3 H),
1.22 (s, 3 H); 13C NMR (75 Mz, CDCl3) δ 176.5, 161.5, 111.1,
92.4, 85.8, 83.5, 57.6, 52.8, 51.0, 30.8, 26.4, 24.0; IR (film) 3383,
2992, 1737, 1692, 1524 cm-1; CIMS m/z (rel intensity) 381
(25%), 379 (82), 377 (100, MNH4+), 364 (22), 362 (71) and 360
a n ti-N-(2,3-Dih yd r oxycycloh exyl)-2′,2′,2′-tr ich lor oa c-
eta m id e An ti-36. Trichloroacetamide 35 (100 mg, 0.412
mmol) was subjected to the standard Upjohn reaction condi-
tions, and the resulting brown oil was purified by column
chromatography (60% ethyl acetate/petrol) to produce a color-
less oil (35 mg, 31%) and a white solid (77 mg, 68%). 1H NMR
showed the oil to be a mixture of syn:anti diols and the solid
to be anti-36: 1H NMR (300 MHz, (CD3)2CO) δ 7.90 (br s, 1
H), 4.00-4.13 (m, 2 H), 3.79 (s, 2 H), 3.63-3.71 (br m, 1 H),
1.72-2.02 (m, 3 H), 1.40-1.60 (m, 3 H); 13C NMR (75 MHz,
CDCl3) δ (CCl3 not observed) 162.9, 75.2, 69.6, 52.8, 30.3, 30.1,
18.2; IR (film) 3407, 2943, 1699, 1514 cm-1; CIMS m/z (rel
intensity) 295 (28%), 293 (30, MNH4+), 280 (32), 278 (88) and
276 (100); C8H13NO3Cl3 requires M 275.9961, found MH+,
275.9956.; mp 115-117 °C.
syn -N-(5-ter t-Bu t yl-2,3-d ih yd r oxycycloh exyl)-2′,2′,2′-
tr ich lor oa ceta m id e Syn -38. Trichloroacetamide 37 (50 mg,
0.17 mmol) was subjected to standard TMEDA dihydroxylation
conditions with methanolic workup and then purified by
column chromatography (40% ethyl acetate/petrol) to give syn-
38 (55 mg, 96%) as a colorless solid: 1H NMR (300 MHz,
CDCl3) δ 7.19 (br d, J ) 7.7 Hz, 1 H), 3.90 (s, 1 H), 3.78 (ddd,
J ) 3.9, 7.7 and 12 Hz, 1 H), 3.64 (br d, J ) 12 Hz, 1 H), 2.90
(br s, 1 H), 2.42 (br s, 1 H), 1.66 (br d, J ) 12 Hz, 2 H), 1.31
(qt, J ) 12 Hz, 2 H), 1.27 (br q, J ) 12 Hz, 1 H), 0.91 (s, 9 H);
13C NMR (75 MHz, CDCl3) δ 161.2, 92.6, 71.3, 70.1, 51.9, 42.9,
+
(73); C12H20N2O5Cl3 requires M 377.0438, found MNH4
377.0430; mp 92-94 °C. Anal. Calcd for C12H16NO5Cl3 C, 40.0;
H, 4.5; N, 3.9. Found: C, 40.3; H, 4.5; N, 3.8.
syn -2,3,4-Tr ia cet oxy-(2,2,2-t r ich lor oa cet yla m in o)-cy-
clop en ta n e Syn -32. Trichloroacetamide 31 (120 mg, 0.49
mmol) was subjected to standard TMEDA conditions, with the
methanolic workup. The resulting brown oil was dissolved in
pyridine (20 mL), and to this were added acetic anhydride (5
mL) and a catalytic amount of DMAP. The reaction was
allowed to stir overnight at room temperature. The pyridine
was removed in vacuo until a dry black solid was obtained.
The residue was purified by column chromatography (80%
diethyl ether in petroleum ether 40-60) to yield syn-32 as a
colorless solid (165 mg, 83%): 1H NMR (300 MHz, CDCl3) δ
7.24 (brd, J ) 8.4 Hz, 1 H), 5.48 (t, J ) 4.4 Hz, 1 H), 5.30-
5.22 (m, 2 H), 4.68-4.56 (m, 1 H), 2.74 (ddd, J ) 8.4, 8.4 Hz,
and 15.1 Hz, 1 H), 2.14 (s, 3 H), 2.09 (s, 3 H), 2.07 (s, 3 H),
1.93 (ddd, J ) 4.1, 6.0 and 15.1 Hz, 1 H); 13C NMR (75 MHz,
CDCl3) δ 169.6, 169.1, 168.8, 161.1, 92.3, 71.8, 70.1, 69.5, 49.7,
34.8, 20.5, 20.5, 20.3; IR (film) 3422, 3348, 2941, 1723, 1519,
1224, 822 cm-1; CIMS m/z (rel intensity) 421 (90%, MNH4+),
404 (5, MH+); C13H20N2O7Cl3 requires M 421.0336, found
32.2, 28.6, 27.4, 26.0; IR (film) 3409, 2962, 1704, 1512 cm-1
;
CIMS m/z (rel intensity) 351 (93%), 349 (100, MNH4+), 336
(22), 334 (63) and 332 (69); C12H24N2O3Cl3 requires M 349.0852,
found MNH4 349.0853; mp 112-113 °C. Anal. Calcd for C12H20
-
NO3Cl3: C, 43.3; H, 6.1; N, 4.2. Found: C, 43.2; H, 6.1; N, 4.1.
a n ti-N-(5-ter t-Bu tyl-2,3-d ih yd r oxycycloh exyl)-2′,2′,2′-
tr ich lor oa ceta m id e An ti-38. Trichloroacetamide 37 (50 mg,
0.17 mmol) was subjected to standard Upjohn conditions and
the resulting black solid purified by column chromatography
(40% ethyl acetate/petrol) to give syn-38 and anti-38 diols (53
mg, 95%) as white solids: 1H NMR (300 MHz, (CDCl3) δ 6.70
(br d, J ) 7.0 Hz, 1 H), 4.02-4.18 (m, 2 H), 3.43-3.52 (m, 1
H), 3.22 (br d, J ) 6.2 Hz, 1 H), 2.63 (br s, 1 H), 2.07 (ddd, J
) 3.0, 6.9 and 12.3 Hz, 1 H), 2.02 (ddd, J ) 2.8, 6.5 and 12.3
Hz, 1 H), 1.73 (tt, J ) 2.8 and 12.3 Hz, 1 H), 1.25 (dt, J ) 2.8
+
MNH4 421.0332; mp 88 °C.
a n ti-2,3,4-Tr ia cetoxy-(2,2,2-tr ich lor oa cetyla m in o)-cy-
clop en ta n e An ti-32. Trichloroacetamide 31 (150 mg, 0.50
mmol) was subjected to standard Upjohn conditions, and the
resulting brown oil was dissolved in pyridine (20 mL); to this
were added acetic anhydride (5 mL) and a catalytic amount
of DMAP. The reaction was allowed to stir overnight at room
temperature. The pyridine was removed in vacuo until a dry
black solid was obtained. The products were purified by column
chromatography (80% diethyl ether in petroleum ether 40-
60) to yield the title compounds as a colorless oil (126 mg, 63%).
Separation of pure anti-32 was possible by chromatography:
1H NMR (300 MHz, CDCl3) δ 7.30 (brd, J ) 7.7 Hz, 1 H), 5.36-
5.28 (m, 2 H), 5.07 (ddd, J ) 2.1, 4.5 and 7.8 Hz, 1 H), 4.34
(qt, J ) 7.8 Hz, 1 H), 2.98 (ddd, J ) 7.8, 7.8 and 14.8 Hz, 1 H),
2.12 (s, 3 H), 2.09 (s, 3 H), 2.07 (s, 3 H), 1.93 (ddd, J ) 4.5, 7.7
and 14.8 Hz, 1 H); 13C NMR (75 MHz, CDCl3) δ 170.9, 169.5,
169.3, 162.0, 92.0, 74.4, 74.0, 73.8, 53.9, 34.0, 20.8, 20.6, 20.5;
IR (film) 3420, 3020, 1746, 1714, 1371, 1224, 822 cm-1; CIMS
m/z (rel intensity) 421 (90%, MNH4+), 404 (5, MH+); C13H20N2O7-
Cl3 requires M 421.0336, found MNH4+ 421.0338; mp 117 °C.
and 12.3 Hz, 1 H), 1.09 (q, J ) 12.3 Hz, 1 H), 0.89 (s, 9 H); 13
NMR (75 MHz, CDCl3) δ 163.0, 92.4, 75.6, 69.6, 53.0, 39.2,
31.8, 31.7, 31.6, 27.4; IR (film) 3304, 2962, 1686 and 1529 cm-1
CIMS m/z (rel intensity) 351 (82) and 349 (100, MNH4+);
12H24N2O3Cl3 requires M 349.0852, found MNH4 349.0858;
C
;
C
mp 181-182 °C. Anal. Calcd for C12H20NO3Cl3: C, 43.3; H,
6.1; N, 4.2. Found: C, 43.2; H, 6.0; N, 4.1.
syn -N-(5-ter t-Bu t yl-2,3-d ih yd r oxy-cycloh exyl)-2′,2′,2′-
tr ich lor oa ceta m id e Syn -40 a n d a n ti-N-(5-ter t-Bu tyl-2,3-
d ih yd r oxy-cycloh exyl)-2′,2′,2′-tr ich lor oa ceta m id e An ti-
40. Trichloroacetamide 39 (100 mg, 0.335 mmol) was subjected
to standard TMEDA dihydroxylation conditions with metha-
nolic workup, and the resulting brown solid was purified by
column chromatography (60% ethyl acetate/petrol) to yield a
1.5:1 mixture of syn-40 and anti-40 isomers (107 mg, 97%) as
a colorless solid. The diol diastereoisomers were inseparable
and characterized together: 1H NMR (300 MHz, CDCl3) δ 8.75
(br d, J ) 8.2 Hz, 1 H), 6.71 (br d, J ) 7.6 Hz, 1 H) 3.66-4.40
(m, 2 × 4 H), 3.60 (br s, 1 H), 3.05 (br s, 1 H), 1.05-2.05 (m,
2 × 5 H), 0.89 (s, 9H) and 0.92 (s, 9 H); 13C NMR (75 MHz;
CDCl3) δ 162.5, 161.5, 92.9, 92.1, 70.9, 69.6, 69.3, 68.5, 52.6,
52.3, 41.5, 34.6, 32.1, 31.9, 31.5, 29.9, 29.2, 27.3 (2), and 24.4;
IR film 3454-3320 (br), 2961, 1702 and 1515 cm-1; CIMS m/z
(rel intensity) 351 (45%), 332 (100, MH+); C12H21NO3Cl3
syn -N-(2,3-Dih yd r oxycycloh exyl)-2′,2′,2′-tr ich lor oa ce-
ta m id e Syn -36. Trichloroacetamide 35 (0.100 g, 0.412 mmol)
was subjected to the standard TMEDA dihydroxylation condi-
tions with methanolic workup, and the resulting orange
mixture was purified by column chromatography (60% ethyl
acetate/petrol) to yield syn-36 (111 mg, 99%) as a colorless
oil: 1H NMR (300 MHz, CDCl3) δ 7.86 (br s, 1 H), 3.86-4.06
J . Org. Chem, Vol. 67, No. 23, 2002 7955

Donohoe et al.
requires M 332.0587, found MH+ 332.0580. Anal. Calcd for
12H20NO3Cl3: C, 43.3; H, 6.1; N, 4.2. Found: C, 43.5; H, 6.3;
N, 4.1.
syn -N-(2,3-Dih ydr oxy-1,5,5-tr im eth ylcycloh exyl)-2′,2′,2′-
mmol) was subjected to standard Upjohn conditions, and the
resulting brown oil was dissolved in pyridine (20 mL); to this
were added acetic anhydride (5 mL) and a catalytic amount
of DMAP. The reaction was allowed to stir overnight at room
temperature. The pyridine was removed in vacuo until a dry
black solid was obtained. The products were purified by column
chromatography (40% diethyl ether in petroleum ether 40-
60) to yield the title compounds as a colorless solid (124 mg,
85%): 1H NMR (300 MHz, CDCl3) δ 7.02 (br d, J ) 8.2 Hz, 1
H), 5.22 (br d, J ) 9.2 Hz, 1 H), 5.08 (dd, J ) 2.1 and 9.2 Hz,
1 H), 4.32-4.20 (m, 1 H), 2.12 (s, 3 H), 2.04 (s, 3 H), 2.08-
1.60 (m, 8 H); 13C NMR (75 MHz, CDCl3) δ 171.1, 170.2, 161.3,
92.5, 76.0, 72.4, 53.3, 30.8, 27.7, 23.1, 22.7, 21.0, 20.8; IR (film)
3339, 2939, 1713, 1523, 1240 cm-1; CIMS m/z (rel intensity)
391 (100%, MNH4+), 374 (5, MH+); C13H22N2O5Cl3 requires M
C
tr ich lor oa ceta m id e Syn -42. Trichloroacetamide 41 (100 mg,
0.351 mmol) was subjected to standard TMEDA dihydroxyla-
tion conditions with methanolic workup to a produce a brown
solid, which was purified by column chromatography (60%
ethyl acetate/petrol) to yield syn-42 (97 mg, 87%) as a colorless
crystalline solid: 1H NMR (300 MHz, CD3OD) δ 3.42 (d, J )
3 Hz, 1 H) 3.38-3.32 (m, 1 H), 2.35 (dd, J ) 14 and 2 Hz, 1
H), 1.86-1.78 (m, 1 H), 1.58 (s, 3 H), 1.58-1.48 (m, 1 H), 1.39
(d, J ) 14 Hz, 1 H), 1.20 (s, 3H) and 0.82 (s, 3 H); 13C NMR
(75 MHz; CDCl3) δ 161.9, 94.6, 76.1, 71.3, 59.7, 45.2, 43.8, 34.1,
31.0, 29.7 and 24.3; IR (film) 3424 (br), 2925) and 1704 cm-1
;
CIMS m/z (rel intensity) 318 (100%, MH+); C11H19NO3Cl3
requires M 318.0519, found MH+ 318.0514; mp 167-169 °C.
Anal. Calcd for C11H18NO3Cl3 C, 41.5; H, 5.7; N, 4.4. Found:
C, 41.4; H, 5.8; N, 4.4.
+
391.0594, found MNH4 391.0594; mp 70 °C.
a n t i-N -(2,3-Dih yd r oxycycloh e xyl)-N -m e t h yl-2′,2′,2′-
tr ich lor oa ceta m id e An ti-50. Trichloroacetamide 49 (100 mg,
0.39 mmol) was subjected to standard TMEDA dihydroxylation
conditions with methanolic workup, and the resulting orange
solid was purified by column chromatography (60% ethyl
acetate/petrol) to give the title compound (88 mg, 78%) as a
colorless oil: 1H NMR (300 MHz, CDCl3) δ 4.47 (br s, 1 H),
3.86 (dd, J ) 2.0 and 11.7 Hz, 1 H), 3.68 (dt, J ) 3.4 and 11.7
Hz, 1 H), 2.85 (s, 3 H), 1.96-2.18 (m, 3 H), 1.16-1.81 (m, 5
H); IR (film) 3433, 2941, 1742 cm-1; CIMS m/z (rel intensity)
311 (33%), 309 (96) and 307 (100, MNH4+); C9H14NO3Cl3
requires M 289.0039, found M+ 289.0037.
(1R,2S,4R,5S,6S)-3,4,6-Tr iacetoxy-1-(ben zyloxy)-2-(2,2,2-
tr ich lor oa cetyla m in o)-tetr a h yd r op yr a n Syn -44. Trichlo-
roacetamide 43 (179 mg, 0.362 mmol) was subjected to
standard TMEDA dihydroxylation conditions with methanolic
workup to a produce a brown solid. The resulting crude tetraol
was dissolved in pyridine (20 mL), and to this were added
acetic anhydride (5 mL) and a catalytic amount of DMAP. The
reaction was allowed to stir overnight at room temperature.
The pyridine was removed in vacuo until a dry black solid was
obtained. The product was purified by column chromatography
(10% ethyl acetate in petroleum ether 40-60) to afford syn-
44 as a colorless oil (0.156 mg, 82%): 1H NMR (500 MHz,
CDCl3) δ 7.60 (d, J ) 6 Hz, 1 H), 7.40-7.25 (m, 5 H), 5.43 (s,
1 H), 5.41 (t, J ) 4 Hz, 1 H), 5.00 (s, 1 H), 4.66 (AB, J ) 12
Hz, 2 H), 4.44 (dd, J ) 5 and 10 Hz, 1 H), 4.32 (td, J ) 7 and
1 Hz, 1 H), 4.21-4.08 (m, 2 H), 2.20 (s, 3 H), 2.08 (s, 3 H),
2.00 (s, 3 H); 13C NMR (75 MHz; CDCl3) δ 170.2, 169.2, 169.0,
161.7, 135.9, 128.6, 128.3, 128.1, 98.22, 92.41, 70.0, 67.26,
66.62, 64.72, 61.57. 50.47, 20.70, 20.59, 20.38; CIMS m/z (rel
intensity) 559 (100%), 557 (100, MNH4+); C21H28N2O9Cl3
syn -N-(5-ter t-Bu t yl-2,3-osm a d ioxycycloh exyl)-2′,2′,2′-
tr ich lor oa ceta m id e]-[N′,N′,N′′,N′′-tetr a m eth yleth ylen e-
d ia m in e] Osm a te Diester Syn -51. cis-Trichloroacetamide 37
(50 mg, 0.17 mmol) was subjected to standard TMEDA
osmylation conditions to produce a brown oil, which was
purified by column chromatography (10% methanol/dichlo-
romethane) to give the title compound (106 mg, 96%) as a
brown crystalline solid: 1H NMR (300 MHz, CDCl3) δ 7.95 (br
d, J ) 7.3 Hz, 1 H), 4.30-4.48 (m, 2 H), 3.75-3.86 (m, 1 H),
3.13 (s, 4 H), 2.91 (s, 3 H), 2.89 (s, 3 H), 2.85 (s, 3 H), 2.84 (s,
3 H), 1.88-1.98 (m, 1 H), 1.70-1.80 (m, 1 H), 1.49-1.63 (m, 1
H), 1.57-1.27 (m, 2 H), 0.90 (s, 9 H); 13C NMR (75 MHz, CDCl3)
δ 160.8, 93.3, 87.7, 86.5, 64.6, 64.1, 52.1, 51.7 (2), 51.6, 51.5,
42.7, 32.4, 28.2, 27.9, 27.6; IR (film) 3393, 2958, 2866, 1710,
1498 cm-1; CIMS m/z (rel intensity) 672 (38%), 670 (48, MH+)
+
requires M 557.0860, found MNH4 557.0868.
syn -N-(2,3-Dia cetoxyoxycycloh ep tyl)-2′,2′,2′-tr ich lor o-
a ceta m id e Syn -46. Trichloroacetamide 45 (100 mg, 0.39
mmol) was subjected to standard TMEDA conditions, with
methanolic workup, and the resulting brown oil was dissolved
in pyridine (20 mL); to this were added acetic anhydride (5
mL) and a catalytic amount of DMAP. The reaction was
allowed to stir overnight at room temperature. The pyridine
was removed in vacuo until a dry black solid was obtained.
The products were purified by column chromatography (40%
diethyl ether in petroleum ether 40-60) to yield the title
compound as a colorless solid (115 mg, 78%): 1H NMR (300
MHz, CDCl3) δ 7.72 (brd, J ) 7.2 Hz, 1 H), 5.20-5.10 (m, 2
H), 4.38-4.23 (m, 1 H), 2.17 (s, 3 H), 2.08 (s, 3 H), 2.06-1.60
(m, 8 H); 13C NMR (75 MHz, CDCl3) δ 171.1, 169.5, 161.0, 92.6,
76.2, 74.4, 51.8, 28.5, 28.1, 24.5, 21.7, 21.1, 20.8; IR (film) 3341,
2942, 1710, 1513, 1243 cm-1; CIMS m/z (rel intensity) 391
(100%, MNH4+), 374 (5, MH+); C13H22N2O5Cl3 requires M
+
and 668 (MNH4 30%); C18H38N4O5Cl3Os requires 670.1257,
found MH+ 670.1264.
Ack n ow led gm en t. We thank the EPSRC, the Le-
verhulme Trust, AstraZeneca, and GlaxoSmithKline for
supporting this work.
Su p p or tin g In for m a tion Ava ila ble: Detailed procedures
and spectroscopic data for compounds 19, 21, 27, 31, 37, 39,
43, and 45 and X-ray pictures and CIF data for syn-22, syn-
28, anti-30, anti-32, syn-40, syn-42, syn-46, 51, and 52. This
material is available free of charge via the Internet at
http://pubs.acs.org.
+
391.0594, found MNH4 391.0596; mp 90 °C.
a n ti-N-(2,3-Dia cetoxyoxycycloh exyl)-2′,2′,2′-tr ich lor o-
a ceta m id e a n ti-46. Trichloroacetamide 45 (100 mg, 0.39
J O026161Y
7956 J . Org. Chem., Vol. 67, No. 23, 2002