Synthesis of Vicinal Amino Alcohols via a Tandem Acylnitrene Aziridination-Aziridine Ring Opening

Article abstract of DOI:10.1021/jo970473x

A facile synthesis of differently substituted vicinal amino alcohols is reported. We have demonstrated that these amino alcohols could be readily prepared from the oxazolidinones by treatment with aqueous base. We have synthesized a variety of substituted

Full text of DOI:10.1021/jo970473x

J . Org. Chem. 1997, 62, 4449-4456
4449
Syn th esis of Vicin a l Am in o Alcoh ols via a Ta n d em Acyln itr en e
Azir id in a tion -Azir id in e Rin g Op en in g
Stephen C. Bergmeier* and Dionne M. Stanchina
Division of Medicinal Chemistry, College of Pharmacy, The Ohio State University, Columbus, Ohio 43210
Received March 14, 1997X
A facile synthesis of differently substituted vicinal amino alcohols is reported. We have
demonstrated that these amino alcohols could be readily prepared from the oxazolidinones by
treatment with aqueous base. We have synthesized a variety of substituted bicyclic aziridine
precursors from the corresponding azidoformates using an intramolecular aziridination reaction.
The nucleophilic ring opening of these bicyclic aziridines using carbon, oxygen, nitrogen, sulfur,
and halogen-containing nucleophiles provided the oxazolidinones in good yield with high regiose-
lectivity. In all cases, nucleophilic attack occurred exclusively at the least substituted carbon of
the aziridine ring. Consequently, our approach allows for convenient and rapid access to the vicinal
amino alcohol functionality, an important structural component of many natural products.
Sch em e 1
In tr od u ction
The vicinal amino alcohol moiety is found in a wide
variety of biologically active molecules. Examples include
hydroxyamino acids such as bestatin,1 sphingolipids,2 and
balanol.3 A number of methods exist for the construction
of this functionality:4 hydride reduction of R-hydroxy
oximino ethers;4a reaction of aldehydes with chiral orga-
noboron reagents;4b cuprate, Michael, and cycloaddition
reactions of γ-aminoolefins;4c and a variety of other
methods4d have been used. In order to prepare a number
of amino alcohols with varying substitution at either end
of the molecule, one is often obliged to utilize a new
substrate for the synthesis of each different amino
alcohol. We had proposed that the formation of aziridines
such as 2 via the intramolecular addition of an azidofor-
mate to an olefin could allow one to prepare a single
reactive intermediate and readily convert it to a variety
of different amino alcohols.
mate gave a bicyclic aziridine in good yield. An intramo-
lecular dipolar cycloaddition of a carbamoyl azide has
recently been reported.7 This reaction did not yield an
aziridine; instead, nucleophilic addition to a betaine
intermediate provided an imidazolidinone. The thermal
or photochemical intermolecular reactions of azidofor-
mates with olefins to produce aziridines are well-known.8
These types of reactions are somewhat limited as cyclic
or strained olefins give the best yields. Furthermore,
high molar ratios of azidoformate to olefin are required.
Our plan was to cyclize an azidoformate (1) to the bicyclic
aziridine 2 and then open the aziridine with a desired
nucleophile (X) to produce 3 (Scheme 1).
Previously, only one intramolecular reaction between
an azidoformate and an olefin has been reported.5,6 In
this earlier report, the thermolysis of an aryl azidofor
Resu lts a n d Discu ssion
The desired azidoformate 6 was readily prepared by
reaction of cyclohexanecarboxaldehyde (4) with vinyl-
magnesium bromide to produce allylic alcohol 5. The
azidoformate can be synthesized by reaction of the alcohol
with CDI followed by NaN3.9 A solution of 6 in 1,1,2,2-
tetrachloroethane (TCE) was heated at reflux which
smoothly converted 6 to a mixture of chlorides 7a and
7b in very good yield.10 The ratio of 7a (trans) to 7b (cis)
is 6.7:1 (Scheme 2). The isomeric identity was deter-
* Author to whom correspondence should be addressed. Tele-
phone: 614-292-5499. E-mail: bergmeier.1@osu.edu.
X Abstract published in Advance ACS Abstracts, J une 1, 1997.
(1) (a) Umezawa, H.; Aoyagi, T.; Suda, H.; Hamada, M.; Takeuchi,
T. J . Antibiotics 1976, 29, 97. (b) Abe, F.; Shibuya, K.; Ashizawa, J .;
Takahashi, K.; Horinishi, H.; Matsuda, A.; Ishizuka, M.; Takeushi, T.;
Umezawa, H. J . Antibiot. 1985, 38, 411. For some representative
syntheses, see: (c) Fuji, K.; Kawabata, T.; Diryu, Y.; Sugiura, Y.
Heterocycles 1996, 42, 701. (d) Palomo, C.; Aizpurua, J . M.; Cuevas,
C. J . Chem. Soc., Chem. Commun. 1994, 1957. (e) Matsuda, F.;
Matsumoto, T.; Ohsaki, M.; Ito, Y.; Terashima, S. Chem. Lett. 1990,
723. (f) Pearson, W. H.; Hines, J . V. J . Org. Chem. 1989, 54, 4235.
(2) (a) Herold, P. Helv. Chim. Acta 1988, 71, 354-363 and references
therein. (b) Nimkar, S.; Menaldino, D.; Merrill, A. H.; Liotta, D.
Tetrahedron Lett. 1988, 29, 3037-3040.
(5) (a) Rhouati, S.; Bernou, A. J . Chem. Soc., Chem. Commun. 1989,
730-732.
(3) For information on isolation and structure determination of
balanol, see: (a) Kulanthaivel, P.; Hallock, Y. F.; Boros, C.; Hamilton,
S. M.; J anzen, W. P.; Ballas, L. M.; Loomis, C. R.; J iang, J . B.; Katz,
B.; Steiner, J . R.; Clardy, J . J . Am. Chem. Soc. 1993, 115, 6452-6453.
Synthesis: (b) Nicolaou, K. C.; Bunnage, M. E.; Koide, K. J . Am. Chem.
Soc. 1994, 116, 8402-8403. (c) Hughes, P. F.; Smith, S. H.; Olson, J .
T. J . Org. Chem. 1994, 59, 5799-5802. (d) Lampe, J . W.; Hughes, P.
F.; Biggers, C. K.; Smith, S. H.; Hu, H. J . Org. Chem. 1994, 59, 5147-
5148. (e) Adams, C. P.; Fairway, S. M.; Hardy, C. J .; Hibbs, D. E.;
Hursthouse, M. B.; Morley, A. D.; Sharp, B. W.; Vicker, N.; Warner, I.
J . Chem. Soc., Perkin Trans. 1 1995, 2355-2362.
(4) (a) Williams, D. R.; Osterhout, M. H.; Reddy, J . P. Tetrahedron
Lett. 1993, 34, 3271-3274. (b) Barrett, A. G. M.; Seefeld, M. A. J .
Chem. Soc., Chem. Commun. 1993, 339-341. (c) J urczak, J .; Golebio-
wski, A. Chem. Rev. 1989, 89, 149-164. (d) Reetz, M. T. Angew. Chem.
Int. Ed. Engl. 1991, 30 (12), 1531-1750.
(6) While the intramolecular cyclization of an azidoformate with an
olefin is rare, the intermolecular insertion reaction between the nitrene
and C-4 has been exploited to some advantage. See: (a) Lwowski, W.:
Carbonylnitrenes In Nitrenes; Lwowski, W., Ed.; Interscience: New
York, 1970; pp 185-224. (b) Edwards, O. E. Acylnitrene Cyclizations.
In Nitrenes; Lwowski, W., Ed.; Interscience: New York, 1970; pp 225-
304. (c) Lwowski, W.: Acyl Azides and Nitrene. In Azides and Nitrenes;
Scriven, E., Ed.; Academic: Orlando, 1984; pp 205-246. (d) Lwowski,
W. Nitrenes. React. Intermed. 1985, 3, 305-332.
(7) Deroose, F. D.; De Clercq, P. J . J . Org. Chem. 1995, 60, 321-
330.
(8) (a) Lwowski, W.; Mattingly, T. W., J r. J . Am. Chem. Soc. 1965,
87, 1947-1958. (b) Lwowski, W. Acyl Azides and Nitrene. In Azides
and Nitrenes; Scriven, E., Ed.; Academic: Orlando, 1984; pp 231-235.
(9) Yuan, P.; Plourde, R.; Shoemaker, M. R.; Moore, C. L.; Hansen,
D. E. J . Org. Chem. 1995, 60, 5360-5364.
S0022-3263(97)00473-8 CCC: $14.00 © 1997 American Chemical Society

4450 J . Org. Chem., Vol. 62, No. 13, 1997
Bergmeier and Stanchina
Sch em e 4
Sch em e 2
the reaction. This also had the benefit of improving the
ratio of trans:cis product from 6.7:1 to 10:1.13 If the
temperature was lowered even further, the reaction was
slowed such that starting material and aziridine were
the principle products observed.
The importance of the solvent in this cyclization is not
yet clear. Carrying out the cyclization in TCE at 100 °C
with 10 mol % BHT provides both the aziridine and the
chloride. We speculate that upon heating TCE produces
a considerable quantity of either HCl or Cl radical.
The bicyclic aziridine 8 proved to be impossible to
purify. Attempts at purifying a relatively pure reaction
product via column chromatography on silica or alumina
did not give any products resembling the aziridine. We
have thus taken to quantifying the product via NMR
relative to an internal standard. This gives a yield for
8a /8b of 87%.
Sch em e 3
As with any azide, two modes of aziridine formation
are possible. Dipolar cyclization to form a triazoline
intermediate with rearrangement to form the aziridine
is well-known.14 Equally well-known is the formation of
the acylnitrene via thermolysis of azidoformates.5 In an
effort to determine which mode of aziridine formation
may be operative in our system, we carried out the
thermolysis of 6 in DMSO (Scheme 4). The only product
obtained from this reaction was the sulfoximine 9 in 83%
yield which is consistent with the product of a nitrene
and DMSO.15
As our goal was to be able to prepare vicinal amino
alcohols with a variety of groups on the alkyl chain, we
next investigated the reaction of bicyclic aziridine 8a with
a series of nucleophilic reagents (Scheme 5). In all but
one case, the opening of the aziridine 8a provided the
oxazolidinone 10 exclusively. The exceptional reactivity
of this aziridine makes possible very mild conditions for
nucleophilic ring opening. Oxygen nucleophiles such as
water,16 alcohols,17 esters,17d ethers,18 and carboxylic
acids16b,19 have been previously used to open aziridines.
When water was used to open the aziridine 8a to provide
mined via NMR experiments. For compound 7a , irradia-
tion of H3 produced a 3.3% NOE enhancement of the
signal for H1. Additionally the coupling constant J 2,3 for
7a (trans) was 3.23 Hz while the coupling constant J 2,3
for 7b (cis) was 7.55 Hz. This is consistent with the
observations that trans-oxazolidinones have smaller cou-
pling constants than the corresponding cis-oxazolidi-
nones.11
We speculated that the chloride might be coming from
HCl generated at high temperatures in the presence of
adventitious water. In an attempt to form only the
bicyclic aziridine, we lowered the temperature of the
reaction and included agents to remove water or acid
from the solution.12 These changes only lowered the
yields of 7 or decreased the rate of the reaction. Upon
changing the reaction solvent to methylene chloride (and
carrying out the reaction in a sealed tube), a dramatic
shift in product distribution was seen (Scheme 3). Now
(13) The predominance of the trans isomer can be rationalized by a
chair-like transition state as discussed in our earlier work. See ref 10.
(14) Padwa, A. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A.,
Ed.; Wiley: New York, 1984; Vol. 2, Chapter 12, pp 277-406.
(15) Although the products are not as well characterized, heating
in toluene or decane also gives only products consistent with nitrenes
reacting with the solvent. (a) Anderson, D. J .; Horwell, D. C.; Stanton,
E.; Gilchrist, T. L.; Rees, C.W. J . Chem. Soc., Perkin Trans. 1 1972,
1317-1320. (b) Colonna, S.; Stirling, C. J . J . Chem. Soc., Perkin Trans.
1 1974, 2120-2122. (c) Atkinson, R. S. Azides and Nitrenes Attached
to Elements Other than Carbon. In Azides and Nitrenes; Scriven, E.,
Ed.; Academic: Orlando, 1984; p 280. (d) Horner, L.; Christmann, A.
Chem. Ber. 1963, 388-398. (e) J ones, D. W.; Thornton-Pett, M. J .
Chem. Soc., Perkin Trans. 1 1995, 809-815.
1
the major products seen in the crude H NMR were the
bicyclic aziridines 8a /8b along with some polymeric
products. Lowering the temperature further (109 °C) and
including 10 mol % of 2,6-di-tert-butyl-4-methylphenol
(BHT) gave the bicyclic aziridine as the sole product of
(10) Bergmeier, S. C.; Stanchina, D. M. Tetrahedron Lett. 1995, 36,
4533-4536.
(11) Seebach, D.; Beck, A. K.; Mukhopadhyay, T.; Thomas, E. Helv.
Chim. Acta 1982, 65, 1101-1133.
(16) (a) Evans, D. A.; Faul, M. M.; Bilodeau, M. T.; Anderson, B. A.;
Barnes, D. M. J . Am. Chem. Soc. 1993, 115, 5328. (b) Davis, F. A.;
Zhou, P.; Reddy, G. V. J . Org. Chem. 1994, 59, 3243.
(12) Powdered dry 4 Å molecular sieves, CaSO4, MgSO4, and acid
scavengers such as poly(4-vinylpyridine) and 1,8-bis(dimethylamino)-
naphthalene (Proton Sponge) were used to preclude the formation of
HCl. We did not observe aziridine formation when any of these agents
were used. Only oxazolidinones were obtained in low yield with the
exception of the case of the Proton Sponge where only polymeric
material was formed.
(17) (a) Kogami, Y.; Okawa, K. Bull. Chem. Soc. J pn. 1987, 60, 2963.
(b) Stamm, H.; Speth, D. Arch. Pharm. (Weinheim, Ger.) 1989, 322,
277. (c) Pirrung, M. C.; Nunn, D. S. Bioorg. Med. Chem. Lett. 1992, 2,
1489. (d) Ho, M.; Chung, J . K. K.; Tang, N. Tetrahedron Lett. 1993,
34, 6513.
(18) Dehmlow, H.; Mulzer, J .; Seilz, C.; Strecker, A. R.; Kohlmann,
A. Tetrahedron Lett. 1992, 33, 3607.

Synthesis of Vicinal Amino Alcohols
J . Org. Chem., Vol. 62, No. 13, 1997 4451
Sch em e 5
Sch em e 6
with thiophenol under either acidic or basic conditions,
we obtained mixtures of products in low yields. Thiol-
acetic acid proved to be the reagent of choice as a clean
ring opening was achieved to provide 10d in 81% yield.
However, unlike the reaction with acetic acid where a
large excess of reagent was required, only 1 equiv of
thiolacetic acid was needed. No additional acid catalysis
was necessary.
Azides have also been used to open aziridine
rings.19d,21e,22,24a,c Our best results have been obtained
using TMSN3. We obtained an 8:1 mixture of oxazolidi-
nones 10e and 11, respectively, when TMSN3 was used
to open the aziridine. Reaction of 8a with sodium azide
under a variety of reaction conditions gave poor mixtures
of products in only moderate yields. We have also
attempted to introduce the nitrogen moiety by reaction
of 8a with benzylamine (Scheme 6). Although this was
unsuccessful, some interesting results were obtained.
Even though benzylamine appears to be a poor nucleo-
phile for this particular reaction, it is worth noting that
we do observe the opening of the oxazolidinone ring with
benzylamine to provide the aziridine 12, albeit in low
yield. In a similar system, Dauban and co-workers23 have
also observed this kind of intermolecular ring opening
when they attempted to open an aziridine using an
alcohol in the presence of a Lewis acid.
Halogens have also been used as nucleophiles to open
an aziridine ring.21e,22c,24 Our attempts to open the
aziridine 8a with HCl gave only low yields of the chloride
7. The use of TMSCl, on the other hand, readily provided
7a in very good yield. As previously discussed in this
report, 7 can also be formed directly from the azidofor-
mate by refluxing 6 in TCE. However, milder reaction
conditions and better selectivity result from first forming
the aziridine 8a and proceeding to open the ring with
TMSCl.
10a , a large excess of reagent as well as acid catalysis
proved to be critical to success of the reaction. A number
of acids were examined, and a catalytic amount of TFA
proved optimal. When the reaction was carried out with
1-2 equiv of water or with a 1:1 mixture of water/solvent,
a considerable quantity (30%) of another product which
was inseparable from 10a was obtained. On the basis
of 1H NMR decoupling experiments, it appeared to be the
product resulting from attack at the internal carbon of
the aziridine. A few examples of nucleophilic openings
of bicyclic aziridines have been reported.20 In several of
these cases, nucleophilic attack occurred exclusively at
the internal carbon of the bicyclic system.
Similarly, the reaction of 8a with methanol required
a large excess of reagent in addition to acid catalysis.
However, it proved necessary to change the acid catalyst
in order to achieve complete regioselectivity. Only
mixtures of products resulting from attack at both the
internal and terminal carbons of the aziridine were
obtained when TFA was employed while a catalytic
amount of sulfuric acid was required to provide 10b
exclusively.
We also wished to utilize organometallic reagents to
the open the aziridine ring as this is a well-known
Following a pattern analogous to the aforementioned
oxygen nucleophiles, we found that a large excess of
acetic acid was essential to the clean formation of the
acetate 10c. No additional reagents were required.
Reaction of 8a with nucleophiles such as NaOAc gave
only polar materials, none of which resembled the ox-
azolidinone 10c.
Ring-opening reactions of aziridines using sulfur-
containing nucleophiles such as thiophenol have been
well documented.21,22c,23 However, upon treatment of 8a
(21) (a) Lin, P.-Y.; Bellos, K.; Stamm, H.; Onistschemko, A. Tetra-
hedron, 1992, 48, 2359-2372. (b) Carreaux, F.; Dureault, A.; Depezey,
J . C. Synlett 1992, 527. (c) Voronkov, M. G.; Knutov, V. I.; Shevko, O.
N. J . Gen. Chem. USSR 1992, 5, 704. (d) Hashimoto, H.; Shimada, K.;
Horito, S. Tetrahedron Lett. 1993, 34, 4953 -4956. (e) Osborn, H. M.
J .; Sweeney, J . B.; Howson, B. Synlett 1994, 145. (f) Hashimoto, M.;
Matsumoto, M.; Yamada, K.; Terashima, S. Tetrahedron Lett. 1994,
35, 2207-2210.
(22) (a) Gutherie, R. D.; Williams, G. J . J . Chem. Soc., Perkin Trans.
1 1976, 801-804. (b) For a preparation of TMSN3, see: Brandstrom,
A.; Lamm, B.; Palmerlz, I. Acta Chem. Scand. B 1974, 28, 699-701.
(c) Burgaud, B. G. M.; Horwell, D. C.; Padova, A.; Pritchard, M. C.
Tetrahedron 1996, 52, 13035-13050. (d) Leung, W.-H.; Yu, M.-T.; Wu,
M.-C.; Yeung, L.-L. Tetrahedron Lett. 1996, 37, 891-892.
(23) Dauban, P.; Chiaroni, A.; Riche, C.; Dodd, R. H. J . Org. Chem.
1996, 61, 2488-2496.
(24) For examples of opening activated aziridines with iodide, see:
(a) Tanner, D.; Birgersson, C.; Dhaliwal, H. K. Tetrahedron Lett. 1990,
31, 1903-1906. (b) Nadir, U. K.; Basu, N. Tetrahedron Lett. 1992, 33,
7903-7906. For examples of opening of activated aziridines with
bromide, see: (c) Dureault, A.; Tranchepain, I.; Depezay, J . C. J . Org.
Chem. 1989, 54, 5324-5330. For examples of opening of aziridines with
chloride, see: (d) Kyburz, E.; Els, H.; Majnoni, S.; Englert, G.; von
Planta, C.; Furst, A.; Plattner, P. A. Helv. Chim. Acta 1966, 49, 359.
(e) Gurjar, M. K.; Patil, V. J .; Yadav, J . S.; Rama Rao, A. V. Carbohydr.
Res. 1984, 129, 267.
(19) (a) Takeuchi, H.; Koyama, K. J . Chem. Soc., Perkin Trans. 2
1981, 121. (b) J ones, D. S.; Srinivasan, A.; Kasina, S.; Fritzberg, A.
R.; Wilkening, D. W. J . Org. Chem. 1989, 54, 1940. (c) Kim, N.-S.;
Kang, C. H.; Cha, J . K. Tetrahedron Lett. 1994, 35, 3489. (d) Legters,
J .; Willems, J . G. H.; Thijs, L.; Zwanenburg, B. Recl. Trav. Chim. Pays-
Bas 1992, 111, 59.
(20) (a) De Kimpe, N.; Boelens, M.; Contreras, J . Tetrahedron Lett.
1996, 37, 3171-3174. (b) Kim, D.-K.; Kim, G.; Kim, Y. W. J . Chem.
Soc., Perkin Trans. 1 1996, 803-808. (c) Armstrong, R. W.; Moran, E.
J . J . Am. Chem. Soc. 1992, 371-372. (d) Hashimoto, M.; Matsumoto,
M.; Yamada, K.; Terashima, S. Tetrahedron Lett. 1994, 35, 2207-2210.

4452 J . Org. Chem., Vol. 62, No. 13, 1997
Bergmeier and Stanchina
Sch em e 8
Sch em e 7
reaction of aziridines.25 However, the reaction of 8a with
organolithium or Grignard reagents gave a poor mixture
of products. A clean ring opening was effected by treating
the intermediate aziridine with a cuprate. Both alkyl
(n-Bu2CuLi) and aryl (PhLi/CuI) cuprates were employed.
While we did not encounter any problems with the more
reactive alkyl cuprate, reaction of 8a with Ph2CuLi led
to products resulting from opening of the aziridine with
iodide. In order to circumvent this problem, it was
necessary to use only a catalytic amount of copper iodide
to provide 10f exclusively.
Oxazolidinones are commonly used as protecting groups
for amino alcohols.26 They can be generated from the
cyclocarbamation of 1,2- and 1,3-amino alcohols27 and can
be cleaved using a variety of methods.26,28 In general,
we have not converted the oxazolidinones to the free
amino alcohols as these compounds are difficult to handle
and purify. However, this transformation can be readily
accomplished with aqueous base. For example, the
conversion of 10f to 13 proceeds in 76% yield with LiOH
(Scheme 7).29 The acetate derivative 10c and the chloride
derivative 7a are incompatible with the basic conditions
required to cleave these oxazolidinones and therefore
cannot be converted to the free amino alcohol using these
reaction conditions.
Aldehydes other than cyclohexanecarboxaldehyde (4)
were utilized to prepare azidoformates that were subse-
quently cyclized to the corresponding aziridines. The
allylic alcohols were prepared by addition of vinylmag-
nesium bromide to the corresponding aldehyde. Conver-
sion of the allylic alcohol to the azidoformate utilizes the
sequential reaction of the alcohol with CDI, followed by
NaN3 (Scheme 8). The azidoformates were then cyclized
to the aziridines 16a -d in generally good yield. For
characterization, the aziridines were directly converted
to the alkyl derivatives 17a -d by reaction with n-butyl
cuprate. Pivaldehyde was selected because we believed
that the bulky tert-butyl moiety would enhance the
stereoselectivity of the aziridination reaction. This was
indeed the case as cyclization of the azidoformate 15a
gave the trans isomer 16a exclusively.
basic groups were included in our selected set of alde-
hydes. 2-(Benzyloxy)acetaldehyde was chosen in order
to determine whether the Lewis basic oxygen moiety
would react with the nitrene. This did not appear to be
the case as side products which could be formed from
1
such a reaction were not evident in the crude H NMR
spectrum. In order to ascertain whether the presence of
an aromatic ring would lead to products derived from
nitrene insertion, hydrocinnamaldehyde was included as
one of our chosen aldehydes. We did not observe the
presence of any such products in the crude 1H NMR
spectrum and were able to obtain the aziridine 16c in
67% yield. Similarly, 1,2,3,6-tetrahydrobenzaldehyde
was selected because it contains an olefin which could
potentially react with the nitrene. However, the presence
of the olefin was of no consequence since the olefinic
protons of the six-membered ring were still present in
the crude 1H NMR spectrum of the aziridine 16d .
Clearly, olefins or aromatic rings immediately adjacent
to the azidoformate would produce mixtures of products.
Unlike the more sterically hindered azidoformate 15a ,
cyclization of azidoformates 15b-d gave mixtures of both
the trans and cis aziridines 16b-d . In all of these cases,
the trans isomer was the major isomer. Upon treatment
of 16b-d with n-butyl cuprate, followed by purification,
only the trans oxazolidinones 17b-d were isolated.
Further investigation into the scope and selectivity of this
reaction is currently underway in our laboratories.
We were concerned that other sites for nitrene reaction
could be problematic. Thus, aromatic rings and Lewis
We have also examined the use of nonterminal olefins.
Starting with trans-2-hexen-1-ol, the azidoformate was
prepared and cyclized to give a 43:1 mixture of bicyclic
aziridines 20a and 20b (Scheme 9). We had anticipated
that only the trans aziridine 20a would be formed.
Apparently, some fraction of the acylnitrene that is
formed is in the triplet state which then adds nonste-
reospecifically in a stepwise fashion as described by
Skell30 allowing for the formation of 20b. It is believed
that for an intermolecular reaction of a thermolytically
generated nitrene with an olefin, the amount of triplet
nitrene which is produced is often on the order of 10%
(25) (a) Dureault, A.; Tranchepain, I.; Depezay, J -C. J . Org. Chem.
1989, 54, 5324-5330. (b) Baldwin, J . E.; Adlington, R. M.; O’Neil, I.
A.; Schofield, C.; Spivey, A. C.; Sweeney, J . B. J . Chem. Soc., Chem.
Commun. 1989, 1852-1854. (c) Baldwin, J . E.; Farthing, C. N.; Russell,
A. T.; Schofield, C. J .; Spivey, A. C. Tetrahedron Lett. 1996, 37, 3761-
3764. (d) Aggarwal, V. K.; Vennall, G. P. Tetrahedron Lett. 1996, 37,
3745-3746. (e) Osowska-Pacewicka, K.; Zwierzak, A. Synthesis, 1996,
333-335. (f) Chakraborty, T. K.; Gangakhedkar, K. Tetrahedron Lett.
1991, 32, 1897-1898.
(26) Ishizuka, T.; Kunieda, T. Tetrahedron Lett. 1987, 28, 4185-
4188.
(27) (a) Yi, F.; Wang, T.; Izawa, S.; Kobayashi, S.; Ohno, M. J . Am.
Chem. Soc. 1982, 104, 6465-6466. (b) Hirama, M.; Nishizaki, I.;
Shigemoto, T.; Ito, S. J . Chem. Soc., Chem. Commun. 1986, 393-394.
(c) Bartlett, P. A.; Tauzella, D. J .; Barstow, J . F. Tetrahedron Lett.
1982, 23, 619-622.
(28) Dyen, M. E.; Swern, D. Chem. Rev. 1967, 67, 197.
(29) Kemp, S. J .; J ianming, B.; Pedersen, S. F. J . Org. Chem. 1996,
61, 7162-7167.
(30) (a) Skell, P. S.; Woodworth, R. C. J . Am. Chem. Soc. 1956, 78,
4496. (b) Woodworth, R. C.; Skell, P. S. J . Am. Chem. Soc. 1959, 81,
3384.

Synthesis of Vicinal Amino Alcohols
J . Org. Chem., Vol. 62, No. 13, 1997 4453
Sch em e 9
these aziridines can be purified, although the yield does
drop upon chromatography.
In conclusion, we have developed a method for the
preparation of differently substituted vicinal amino al-
cohols. This method utilizes a stereoselective intramo-
lecular aziridination to form a bicyclic aziridine which is
then reacted with the desired nucleophile to produce
vicinal amino alcohols.
Exp er im en ta l Section
Gen er a l. Melting points were taken using a capillary
melting point apparatus and are uncorrected. Thin-layer
chromatography (TLC) was performed on precoated silica gel
F254 aluminum foils. 1H NMR spectra referenced to TMS were
recorded at 250 MHz. Nuclear Overhauser enhancement
(NOE) experiments were carried out at both 270 and 500 MHz.
Mass spectra were recorded in the electron impact (EI) mode
unless otherwise indicated. All reactions were carried out
under
a nitrogen atmosphere. Pyridine and NEt3 were
distilled from calcium hydride and stored over potassium
hydroxide. Anhydrous DMS and anhydrous DMSO were used
as received from Aldrich Chemical Co. DMF was distilled over
barium oxide and stored over 4 Å molecular sieves. Acetic
anhydride and oxalyl chloride were distilled under nitrogen
immediately prior to use. Benzyl bromide was distilled at
reduced pressure before use. THF was distilled from sodium/
benzophenone ketyl prior to use. Benzene, CH2Cl2, and
chlorotrimethylsilane were distilled from calcium hydride prior
to use. 1,1′-Carbonyldiimidazole was recrystallized from dry
THF immediately before use. Copper(I) iodide was purified
prior to use.32 All other reagents were used as purchased.
Chromatography refers to flash chromatography on silica gel
according to the method of Still.33 CAUTION: Azid es a r e
p oten tia lly exp losive, esp ecia lly u p on h ea tin g. Wh ile
w e h a ve obser ved n o p r oblem s w ith sta bility or for m a -
tion of th e a zid ofor m a tes, ca r e sh ou ld be exer cised .
Cycloh exa n ea llyl Alcoh ol (5). To a solution of vinylmag-
nesium bromide (60 mL of a 1.0 M solution in THF, 60 mmol)
in THF (100 mL) at 0 °C was added cyclohexanecarboxalde-
hyde (5.60 g, 50 mmol). The reaction was stirred at 0 °C until
it was complete by TLC (4 h). The reaction was quenched by
the addition of saturated aqueous NH4Cl and extracted with
EtOAc (3 × 60 mL), dried (MgSO4), filtered, concentrated, and
chromatographed to afford 6.50 g of 5 (92%) as a colorless oil:
1H NMR (CDCl3) δ 5.80 (m, 1H), 5.15 (m, 2H), 3.80 (m, 1H),
1.90-1.55 (m, 6H), 1.55-0.80 (m, 5H); 13C NMR (CDCl3) δ
139.8, 115.2, 77.6, 77.5, 76.5, 43.5, 28.7, 28.3, 26.1, 26.1; HRMS
calcd for C9H16O 140.1202, found 140.1193.
1-[(Azid oca r bon yl)oxy]-1-cycloh exyl-2-p r op en e (6). To
a solution of 5 (3.50 g, 25.0 mmol) in benzene (80 mL) and
pyridine (5.93 g, 75.0 mmol) was added 1,1’-carbonyldiimida-
zole (8.11 g, 50.0 mmol). The reaction mixture was stirred at
rt for 3.5 h. It was then diluted with EtOAc (100 mL), washed
with brine (2 × 100 mL), dried (MgSO4), filtered, and concen-
trated. To this crude material was added DMF (117 mL)
followed by NaN3 (8.12 g, 125.0 mmol). The reaction mixture
was then acidified to ca. pH 4 with concentrated HCl and
stirred at rt for 18 h. It was then diluted with EtOAc (100
mL) and washed with brine (3 × 100 mL). The aqueous layers
were then extracted with EtOAc (2 × 100 mL). The combined
organic layers were dried (MgSO4), filtered, concentrated, and
purified by chromatography to afford 4.61g of 6 (88%) as a
colorless oil: 1H NMR (CDCl3) δ 5.70 (m, 1H), 5.25 (m, 2H),
4.90 (t, 1H), 1.85 -1.50 (m, 6H), 1.50-0.70 (m, 5H); 13C NMR
(CDCl3) δ 156.9, 133.8, 118.8, 84.0, 41.4, 28.4, 28.3, 28.2, 25.8,
25.7; IR (neat) 2100, 1750 cm-1; HRMS calcd for C10H15N3O2
209.1165, found 209.1173.
and is dependent upon the concentration of the olefin.8b
In general, as the concentration of olefin increases, the
amount of nitrene in the triplet state decreases.31 Since
our system involves the intramolecular reaction of a
nitrene and an olefin, we expected that the amount of
20b that is formed should be significantly less than 10%
due to the fact that the concentration of olefin is at its
highest possible level. This appears to be the case since
a 43:1 mixture of 20a and 20b is obtained. We antici-
pated then that the azidoformate resulting from cis-2-
hexen-1-ol would cyclize to give primarily the cis product.
Unfortunately, this was not the case and a 2:1 mixture
of the cis:trans aziridines 20a and 20b was formed. The
fact that the amount of trans aziridine formed was
significantly greater than expected seems to indicate
either that the conversion from the singlet state to the
triplet state is faster than previously reported31a,b or that
the lifetime for the nitrene is longer than might be
anticipated.
The aziridine 20a can also be readily opened to provide
alcohol 21 in high yield. The alcohol 21 was subsequently
acetylated in order to ascertain (by homonuclear proton
decoupling experiments) whether nucleophilic ring open-
ing had occurred at the internal carbon of the aziridine
20a to form a six-membered ring or whether it had
occurred at the external carbon to provide the five-
membered oxazolidinone 21. On the basis of these
experiments, we found the latter to be the case since
decoupling of the signal at 1.50 ppm corresponding to the
protons on the alkyl chain resulted in the collapse of the
signal for the proton on the carbon of the aceylated
alcohol at 4.95 ppm. It is interesting to note that 20a
and 20b can also be formed in refluxing TCE. Appar-
ently, ring-opening reactions of these aziridines are not
as facile as those of terminal aziridines. Furthermore,
(4S*,5R*)-4-(Ch lor om eth yl)-5-cycloh exyl-1,3-oxazolidin -
2-on e (7a ) a n d (4R*,5R*)-4-(Ch lor om eth yl)-5-cycloh exyl-
1,3-oxa zolid in -2-on e (7b). A solution of 6 (2.10 g, 10 mmol)
in TCE (200 mL) was heated to reflux for 4 h. The solvent
(31) (a) McConaghy, J . S., J r.; Lwowski, W. J . Am. Chem. Soc. 1967,
89, 2357. (b) McConaghy, J . S., J r.; Lwowski, W. J . Am. Chem. Soc.
1967, 89, 4450. (c) Mishra, A.; Rice, S. N.; Lwowski, W. J . Org. Chem.
1968, 33, 481.
(32) Kauffman, G. B.; Teter, L. A. Inorg. Synth. 1961, 7, 9-12.
(33) Still, C. W.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 50, 2923.

4454 J . Org. Chem., Vol. 62, No. 13, 1997
Bergmeier and Stanchina
was then removed under vacuum. The crude oil was chro-
reaction stirred at rt for 21.5 h. The reaction mixture was
then quenched with saturated NaHCO3, dried (MgSO4), fil-
tered, concentrated, and chromatographed to afford 100 mg
of 10b (77%) as a colorless oil: 1H NMR (CDCl3) δ 3.95 (dd,
1H, J ) 5.84 Hz (decoupled signal at 3.70 ppm)), 3.70 (m, 1H,
J ) 4.86 Hz (decoupled signal at 3.30 ppm)), 3.30 (m, 5H),
1.90-1.50 (m, 6H), 1.30-1.10 (m, 5H); 13C NMR (CDCl3) δ
159.1, 82.9, 75.2, 70.6, 59.2, 54.6, 41.9, 27.7, 27.4, 26.6, 26.2,
25.7, 25.5. Anal. Calcd for C11H19NO3: C, 61.95; H, 8.98; N,
6.57. Found: C, 62.28; H, 8.74; N, 6.66.
(4R*,5R*)-5-Cycloh exyl-4-(a cetoxym eth yl)-1,3-oxa zoli-
d in -2-on e (10c). To a solution of glacial acetic acid (2 mL)
cooled to 10 °C was added a solution of 8 (200 mg, 1.10 mmol)
in 1,4-dioxane (2 mL). The reaction mixture was stirred at
10 °C for 5 h and then allowed to warm to rt where it stirred
for 20 h. The reaction mixture was then concentrated and
matographed to give 1.90 g (87%) of 7a /7b as an off-white solid
1
in a 6.7:1 mixture by H NMR. Recrystallization from Et2O/
hexane gave 1.51 g of 7a (70%) as a colorless solid: mp 93-
96 °C; 1H NMR (CDCl3) δ 4.10 (dd, J ) 4.4, 6.3 Hz, 1H), 3.83,
(m, 1H), 3.51 (d, J ) 5.52 Hz, 2H), 1.89-1.57 (m, 6H), 1.29-
1.04 (m, 5H); 13C NMR (CDCl3) δ 158.9, 83.9, 62.1, 46.2, 43.7,
41.9, 27.7, 27.2, 26.1, 25.5, 25.4; NOE experiment, irradiation
of the multiplet at 4.10 ppm showed a 3.3% enhancement for
the signal at 3.51 ppm; HRMS calcd for C10H16ClNO2 217.0871,
found 217.0848.
Alter n ative P r epar ation of (4S*,5R*)-4-(Ch lor om eth yl)-
5-cycloh exyl-1,3-oxa zolid in -2-on e (7a ). To a solution of
freshly distilled chlorotrimethylsilane (72 mg, 0.66 mmol) in
THF (1 mL) cooled to -30 °C was added a solution of 8 (100
mg, 0.55 mmol) in THF (1 mL). The reaction was stirred at
-30 °C for 4 h and was then warmed to rt where stirring was
continued for 19 h. The reaction mixture was then diluted
with EtOAc (20 mL), washed with water (2 × 35 mL) and brine
(1 × 40 mL), dried (MgSO4), filtered, concentrated, and
chromatographed (3.5:1 hexanes/EtOAc) to afford 95 mg of 7a
(80%) as a colorless solid identical to that prepared above.
(4R*,5R*)-5-Cycloh exyl-1-oxa-3-azabicyclo[3.1.0]h exan -
2-on e (8a) an d (4S*,5R*)-5-Cycloh exyl-1-oxa-3-azabicyclo-
[3.1.0]h exa n -2-on e (8b). A solution of 6 (400 mg, 1.91 mmol)
and BHT (42 mg, 0.19 mmol) in CH2Cl2 (30 mL) was placed in
an ACE Glass Model 8648B 100 mL capacity pressure tube.
CAUTION: Rea ction s con d u cted in p r essu r e tu bes a r e
p oten tia lly exp losive a n d sh ou ld be ca r r ied ou t beh in d
a p r otective sh ield . Wh ile w e h a ve obser ved n o p r ob-
lem s, ca r e sh ou ld be exer cised . The reaction vessel was
cooled to -78 °C, evacuated, and sealed. The reaction mixture
was then warmed to 109 °C in an oil bath for 13 h. It was
then cooled to room temperature and concentrated to afford
300 mg (87%) of a mixture of 8a and 8b in 10:1 ratio,
chromatographed (95:5 Et2
O/MeOH) to afford 185 mg of 10c
(70%) as a colorless solid: mp 99-102 °C; 1H NMR (CDCl3) δ
3.95 (m, 3H, J ) 6.94 Hz (decoupled signal at 3.75 Hz)), 3.75
(m, 1H, J ) 5.02 Hz (decoupled signal at 3.95 Hz)), 2.00 (s,
3H), 1.65 (m, 6H), 1.10 (m, 5H); 13C NMR (CDCl3) δ 170.5,
159.1, 82.6, 65.5, 53.7, 41.8, 27.3, 27.2, 25.9, 25.4, 25.3, 20.4.
Anal. Calcd for C12H19NO4: C, 59.73; H, 7.93; N, 5.80.
1
Found: C, 59.67; H, 7.81; N, 5.79. H-1H COSY experiment
(270 MHz, CDCl3) showed a cross-peak at 6.85 ppm (NH)
coupled with methine at 3.75 ppm.
(4S*,5R*)-4-[(Acet ylt h io)m et h yl]-5-cycloh exyl-1,3-ox-
a zolid in -2-on e (10d ). To a solution of thiolacetic acid (0.097
g, 1.23 mmol) in THF (2 mL) cooled to -30 °C was added a
solution of 8 (180 mg, 1.03 mmol) in THF (2 mL). The reaction
mixture was stirred at -30 °C for 4 h and was warmed to rt
where stirring continued overnight. The reaction mixture was
then concentrated and chromatographed to afford 210 mg of
10d (81%) as a colorless solid: mp 80-82 °C; 1H NMR (CDCl3)
δ 3.90 (dd, 1H), 3.70 (m, 1H, J ) 4.34 Hz (decoupled signal at
3.05 and 2.90 ppm); decoupled signal at 3.90 ppm lead to
collapse of signal into a triplet), 3.05 (dd, 1H, (decoupled signal
at 3.70 ppm lead to collapse of signal at 3.05 ppm)), 2.90 (dd,
1H, (decoupled signal at 3.70 ppm lead to collapse of signal at
2.90 ppm)), 2.30 (s, 3H), 1.70 (m, 6H), 1.15 (m, 5H); 13C NMR
(CDCl3) δ 194.8, 158.8, 85.1, 54.6, 41.8, 34.2, 30.5, 27.6, 27.3,
26.1, 25.6, 25.4. Anal. Calcd for C12H19NO3S: C, 56.01; H,
7.44; N, 5.44. Found: C, 56.35; H, 7.49; N, 5.42.
1
respectively. An H NMR analysis of the crude material using
the signal for the aromatic protons (6.85 ppm) of BHT as an
internal standard was used to determine the yield of the
reaction: 1H NMR (CDCl3) δ 4.30 (d, 1.1H), *3.05 (m, 0.1H),
2.95 (m, 0.9H), 2.50 (d, 0.9H), *2.45 (d, 0.1H), *2.20 (d, 0.1H),
2.10 (m, 1H), 1.75 (m, 6H), 1.20 (m, 5H) (* indicates minor
isomer); NOE experiment (CDCl3, 500 MHz), irradiation of
peak at 4.30 ppm shows a 1.34% enhancement of the signal
at 2.95 ppm and a 2.68% enhancement of the signal at 2.10
ppm; irradiation of peak at 2.95 ppm shows a 3.75% enhance-
ment of the signal at 2.50 ppm.
S,S-Dim et h yl-N-[((1-cycloh exyl-2-p r op en yl)oxy]ca r -
bon yl]su lfoxim in e (9). A solution of 6 (200 mg, 0.96 mmol)
in DMSO (30 mL) along with BHT (21 mg, 0.096 mmol) was
placed in an Ace Glass Model 8648B 100 mL capacity pressure
tube. The reaction vessel was cooled to -78 °C, evacuated,
and sealed. The reaction mixture was then warmed to 100
°C in an oil bath for 13 h. It was then cooled to room
temperature, concentrated, and chromatographed to give 103
mg of 9 (83%): 1H NMR (CDCl3) δ 5.70 (m, 1H), 5.15 (t, 2H),
4.85 (t, 1H), 3.25 (s, 6H), 1.65 (m, 6H), 1.05 (m, 5H); 13C NMR
(CDCl3) 158.7, 135.3, 117.3, 80.9, 41.8, 41.7, 41.5, 28.5, 28,4,
26.2, 25.8, 25.8; IR (neat) 1758, 1663, 1020 cm-1; HRMS calcd
for C12H21NO3S 259.1243, found 259.1231.
(4R*,5R*)-4-(Azid om eth yl)-5-cycloh exyl-1,3-oxa zolid in -
2-on e (10e) a n d (4R*,5S*)-4-(Azid om eth yl)-5-cycloh exyl-
1,3-oxa zolid in -2-on e (11). To a stirred suspension of NaN3
(72 mg, 1.10 mmol) in DMF (1 mL) cooled to 0 °C was added
chlorotrimethylsilane (130 mg, 1.20 mmol) dropwise over 5
min.34 The reaction mixture was stirred for 30 min at 0 °C
and 30 min at rt. A solution of 8 (180 mg, 1.00 mmol) in DMF
(0.50 mL) was added at rt and stirring continued at rt for 5 h.
The reaction mixture was diluted with EtOAc (20 mL) and
washed with water (3 × 50 mL), dried (MgSO4), filtered,
concentrated, and chromatographed to afford 160 mg of 2
(73%) as a colorless solid as an 8:1 mixture of 10e (trans):11
1
(cis): mp 89-92 °C; H NMR (CDCl3) δ *4.20 (m, 0.12H, J )
10.17 Hz (decoupled signal at *3.75 Hz)), 4.00 (m, 0.87H, J )
6.00 Hz (decoupled signal at 3.65 Hz)), *3.75 (m, 0.12H), 3.65
(m, 0.87H), 3.35 (m, 2H), 1.65 (m, 6H), 1.10 (m, 5H) (* indicates
minor cis isomer); 13C NMR (CDCl3) 158.9, 83.5, 54.8, 54.4,
41.8, 27.6, 27.3, 26.1, 25.6, 25.4; IR (CCl4) 2107, 1749 cm.-1
HRMS calcd for C10H16N4O2 + H (M + H) 225.1275, found
225.1348.
(4R*,5R*)-5-Cycloh exyl-4-(h yd r oxym eth yl)-1,3-oxa zo-
lid in -2-on e (10a ). To a solution of 8 (260 mg, 1.43 mmol) in
water (3.1 mL) cooled to 5 °C was added trifluoroacetic acid
(8.1 mg, 0.071 mmol). The ice bath was removed, and the
reaction mixture was stirred at rt for 21 h. The reaction
mixture was then diluted with EtOAc and separated. The
organic layer was dried (MgSO4), filtered, concentrated, and
chromatographed (6:1 EtOAc/hexanes) to afford 200 mg of 10a
(4R*,5R*)-4-Ben zyl-5-cycloh exyl-1,3-oxa zolid in -2-on e
(10f). To a suspension of CuI (98 mg, 0.51 mmol) in THF (1.6
mL) cooled to -78 °C was added PhLi (2.56 mL of a 2.0 M
solution in cyclohexane/ether, 5.1 mmol). The reaction mixture
was then warmed to -30 °C and stirred for 45 min. The
mixture was then cooled to -78 °C, and a solution of 8 (93
mg, 0.51 mmol) in THF (1 mL) was added. The reaction
stirred at -78 °C for 40 min and was subsequently warmed
to rt by removing the ice bath. Stirring continued at rt for 28
h. After being quenched with saturated aqueous NH4Cl, the
1
(71%) as a colorless solid: mp 92-94 °C; H NMR (CDCl3) δ
4.10 (m, 1H), 3.65 (m, 2H), 3.50 (m, 1H), 1.65 (m, 6H), 1.15
(m, 5H); 13C NMR (CDCl3) δ 160.2, 82.8, 64.1, 56.9, 41.7, 27.5,
27.3, 26.1, 25.5, 25.4. Anal. Calcd for C10H17NO3: C, 60.28;
H, 8.59; N, 7.03. Found: C, 60.36; H, 8.37; N, 7.01.
(4R*,5R*)-5-Cycloh exyl-4-(m eth oxym eth yl)-1,3-oxa zo-
lid in -2-on e (10b). To a solution of 8 (111 mg, 0.63 mmol) in
MeOH (2.4 mL) cooled to 0 °C was added concentrated H2SO4
(6 mg, 0.063 mmol). The ice bath was removed and the
(34) Vorbruggen, H.; Krolikiewicz, K. Collect. Czech. Chem. Com-
mun. 1979, 35.

Synthesis of Vicinal Amino Alcohols
J . Org. Chem., Vol. 62, No. 13, 1997 4455
reaction mixture was extracted with EtOAc (2 × 30 mL). The
organic layers were washed with water (2 × 35 mL) and brine
(1 × 35 mL), dried (MgSO4), filtered, concentrated, and
chromatographed to afford 100 mg of 10f (75%) as a yellow
oil: 1H NMR (CDCl3) δ 7.20 (m, 5H), 4.05 (dd, 1H, J ) 5.88
Hz (decoupled signal at 3.75 ppm)), 3.75 (m, 1H, J ) 4.65 Hz
(decoupled signal at 2.80 ppm)), 2.80 (d, 2H), 1.80-1.40 (m,
6H), 1.10 (m, 5H); 13C NMR (CDCl3) δ 158.7, 136.2, 129.1,
128.8, 127.1, 85.5, 56.4, 42.5, 41.8, 27.7, 27.4, 26.1, 25.6, 25.5;
HRMS calcd for C16H21NO2 259.1573, found 259.1616.
(4R*,5R*)-5-Cycloh exyl-4-p en t yl-1,3-oxa zolid in -2-on e
(10g). To a suspension of CuI (192 mg, 1.01 mmol) in THF
(3.5 mL) cooled to -78 °C was added n-BuLi (0.88 mL of a 2.3
M solution in hexanes, 2.02 mmol). The reaction mixture was
then warmed to rt and stirred for 45 min. After the solution
was cooled to -78 °C, 8 (175 mg, 0.965 mmol) in THF (1.3
mL) was added and stirring continued at -78 °C for 20 min.
The reaction mixture was then warmed to rt by removing the
ice bath and stirring was continued for 4 h. The reaction was
quenched at this time with saturated aqueous NH4Cl and
extracted with EtOAc (2 × 35 mL). The organic layers were
washed with water (1 × 35 mL) and brine (1 × 35 mL), dried
(MgSO4), filtered through Celite, concentrated, and chromato-
graphed (5:1 hexanes/EtOAc) to afford 160 mg of 10g (73%)
as a colorless oil: 1H NMR (CDCl3) δ 3.85 (dd, 1H, J ) 5.1 Hz
(decoupled signal at 1.50 ppm)), 3.50 (m, 1H, J ) 4.99 Hz
(decoupled signal at 1.50 ppm)), 1.90-0.70 (m, 22H); 13C NMR
(CDCl3) δ 159.7, 86.3, 55.2, 41.9, 36.7, 36.2, 31.5, 31.4, 29.2,
29.1, 29.1, 27.9, 27.4, 26.2, 25.9, 25.5, 25.2, 24.7, 22.4, 22.3,
22.2, 13.7. Anal. Calcd for C14H25NO2: C, 70.25; H, 10.53;
N, 5.85. Found: C, 70.29; H, 10.60; N, 5.66.
for the preparation of 6 on a 50 mmol scale in 75% yield as a
colorless oil from 14a : 1H NMR (CDCl3) δ 5.80 (m, 1H), 5.30
(m, 2H), 4.85 (d, 1H), 0.90 (m, 9H); 13C NMR (CDCl3) δ 156.9,
132.3, 119.7, 87.0, 34.3, 26.1, 25.5; IR (neat) 3025, 2138, 1735
cm-1; HRMS calcd for C8H13N3O2 183.1117, found 183.0984.
1-(Ben zyloxy)-2-[(a zid oca r bon yl)oxy]-3-bu ten e (15b).
The azidoformate 15b was prepared by the same method used
for the preparation of 6 on a 4.9 mmol scale in 71% yield as a
colorless oil from 14b: 1H NMR (CDCl3) δ 7.35 (m, 5H), 5.85
(m, 1H), 5.40 (m, 3H), 4.55 (m, 2H), 3.60 (m, 2H); 13C NMR
(CDCl3) δ 156.7, 137.6, 131.9, 128.3, 127.6, 127.5, 119.2, 77.9,
73.2, 70.8; IR (neat) 3020, 2187, 1731 cm-1; HRMS calcd for
C12H13N3O3 247.0958, found 247.0951.
1-P h en yl-3-[(a zid oca r bon yl)oxy]-4-p en ten e (15c). The
azidoformate 15c was prepared by the same method used for
the preparation of 6 on a 6.2 mmol scale in 79% yield from
14c: 1H NMR (CDCl3) δ 7.20 (m, 5H), 5.80 (m, 1H), 5.25 (m,
3H), 2.65, (m, 2H), 2.00 (m, 2H); 13C NMR (CDCl3) δ 156.6,
140.6, 134.9, 128.4, 128.2, 126.4, 126.0, 118.2, 79.0, 35.4, 31.1;
IR (neat) 3086, 2134, 1736 cm-1; HRMS calcd for C12H13N3O2
231.1008, found, 231.1009.
1-[(Azid oca r bon yl)oxy]-1-(3-cycloh exen yl)-2-p r op en e
(15d ). The azidoformate 15d was prepared by the same
method used for the preparation of 6 on a 9.4 mmol scale in
94% yield from 14d as a 1:1 mixture of diastereomers: 1H
NMR (CDCl3) δ 5.65 (m, 3H), 5.30 (m, 2H), 5.00 (m, 1H), 1.90
(m, 6H), 1.25 (m, 1H); 13C NMR (CDCl3) δ 156.8, *156.7, 133.6,
*133.5, 126.9, *126.7, *125.3, 125.1, *119.2, 118.9, *83.3, 83.0,
*77.5, 77.0, 37.5, *37.2, 27.0, *26.8, 24.6, *24.2, 24.1 (* indi-
cates other isomer); IR (neat) 2188, 2135, 1754, 1731 cm-1
HRMS calcd for C10H13N3O2 207.1009, found 207.1002.
;
(1R*,2R*)-1-Cycloh exyl-2-a m in o-3-p h en yl-1-p r op a n ol
(13). Oxazolidinone 10f (100 mg, 0.39 mmol), LiOH (490 mg,
11.6 mmol), EtOH (7 mL), and water (3 mL) were heated at
reflux for 17 h.29 The reaction mixture was concentrated to 2
mL and extracted with EtOAc (3 × 30 mL). The combined
organic extracts were dried (MgSO4), filtered, concentrated,
and chromatographed to give 68 mg of 13 (76%) as a colorless
(4R*,5R*)-5-ter t-Bu t yl-4-p en t yl-1,3-oxa zolid in -2-on e
(17a ). The aziridine 16a was prepared by the same method
used for the preparation of 8a and 8b on a 3.2 mmol scale in
84% yield from 15a and was used immediately in the next
reaction: 1H NMR (CDCl3) δ 4.10 (m, 1H), 2.90 (m, 1H,
decoupled signal at 4.10 ppm leads to the collapse of the signal
at 2.90 ppm), 2.40 (d, 1H), 2.00 (d, 1H), 0.90 (m, 9H). To a
suspension of copper(I) iodide (500 mg, 2.64 mmol) in THF
(9.6 mL) cooled to -78 °C was added n-BuLi (2.40 mL of a 2.3
M solution in hexanes, 5.53 mmol). The reaction mixture was
warmed to -40 °C and stirred for 40 min. The solution was
then cooled to -78 °C, and 16a (410 mg, 2.64 mmol) in THF
(3.42 mL) was added and stirring continued at -78 °C for 20
min. The reaction mixture was slowly warmed to rt by
removing the ice bath. After 4 h, the reaction was quenched
with a saturated aqueous solution of NH4Cl, extracted with
EtOAc (3 × 50 mL), washed with brine (2 × 50 mL), dried
(MgSO4), filtered, concentrated, and chromatographed to afford
370 mg of 17a (66%) as a yellow oil: 1H NMR (CDCl3) δ 3.80
(d, 1H, J ) 4.67 Hz (decoupled signal at 3.50 ppm leads to the
collapse of the signal at 3.80 ppm)), 3.50 (m, 1H (decoupled
signal at 3.80 ppm leads to the collapse of the signal at 3.50
ppm)), 1.40 (m, 8H), 0.90 (m, 12H); 13C NMR (CDCl3) δ 159.9,
89.6, 53.3, 37.1, 34.1, 31.4, 24.8, 24.4, 22.3, 13.9. Anal. Calcd
for C12H23NO2: C, 67.57; H, 10.87; N, 6.56. Found: C, 67.72;
H, 11.15; N, 6.54. No NOE enhancements were observed.
(4R *,5R *)-5-(Be n zyloxy)-4-p e n t yl-1,3-oxa zolid in -2-
on e (17b). The aziridine 16b was prepared by the same
method used for the preparation of 8a and 8b on a 3.8 mmol
scale in 70% yield as a 3:1 mixture of trans:cis isomers from
15b and was used immediately in the next reaction: 1H NMR
(CDCl3) δ 7.30 (m, 5H), *4.85 (m, 0.33 H, J ) 5.10 Hz
(decoupled signal at 3.65 ppm; decoupled signal at 3.10 ppm
leads to the collapse of the signal at *4.85 ppm)), 4.65 (m,
0.67H (decoupled signal at 3.65 and 3.10 ppm leads to the
collapse of the signal at 4.65 ppm)), 4.50 (m, 2H), 3.65 (m, 2H
(decoupled signal at 4.65 ppm leads to the collapse of the signal
at 3.65 ppm), 3.10 (m, 1H, J ) 3.41 Hz (decoupled signal at
2.50 ppm; decoupled signal at *4.85, 2.50, *2.40, 2.10 ppm
leads to the collapse of the signal at 3.10 ppm)), 2.50 (d, 0.67
H) (decoupled signal at 3.10 ppm leads to the collapse of the
signal at 2.50 ppm), *2.40 (d, 0.33H (decoupled signal at 3.10
ppm leads to the collapse of the signal at *2.40 ppm)), *2.35
(d, 0.33H (decoupled signal at 3.10 ppm leads to the collapse
of the signal at *2.35 ppm)), 2.10 (d, 0.67H (decoupled signal
1
solid: mp 81-83 °C; H NMR (CDCl3 + D2O) δ 7.20 (m, 5H),
4.70 (m, 2H), 3.00 (m, 1H), 2.55 (m, 1H), 1.75 (m, 6H), 1.25
(m, 5H); 13C NMR (CDCl3) δ 139.3, 129.2, 128.6, 126.4, 29.9,
28.1, 26.5, 26.4, 26.2; HRMS calcd for C15H23NO 233.1780,
found 233.1743.
4,4-Dim eth yl-1-p en ten -3-ol (14a ). The allylic alcohol 14a
was prepared by the same method as 5 and was immediately
used for the preparation of 15a . Analytical data matched that
reported for the known compound.35
1-(Ben zyloxy)-3-bu ten -2-ol (14b). The allylic alcohol 14b
was prepared by the same method used for the preparation of
5 on an 8 mmol scale in 61% yield. Analytical data matched
that reported for the known compound.36
1-P h en yl-4-p en ten -3-ol (14c). The allylic alcohol 14c was
prepared by the same method used for the preparation of 5
on a 1.86 mmol scale in 73% yield from commercially available
hydrocinnamaldehyde: 1H NMR (CDCl3) δ 7.30 (m, 5H), 5.95
(m, 1H), 5.25 (m, 2H), 4.15 (m, 1H), 2.75 (m, 2H), 1.90 (m,
2H); 13C NMR (CDCl3) δ 141.7, 140.9, 128.2, 128.1, 125.6,
114.4, 72.0, 38.3, 31.4; HRMS calcd for C11H14O 162.1045,
found 162.1046.
3-Cycloh exen yla llyl Alcoh ol (14d ). The allylic alcohol
14d was prepared by the same method used for the prepara-
tion of 5 on a 18.2 mmol scale in 87% yield from commercially
available 1,2,3,6-tetrahydrobenzaldehyde as a 1:1 mixture of
diastereomers: 1H NMR (CDCl3) δ 5.85 (m, 1H), 5.65 (m, 2H),
5.20 (m, 2H), 3.90 (m, 1H), 1.85 (m, 6H), 1.25 (m, 1H); 13C NMR
(CDCl3) δ *139.7, 139.6, 127.1, *126.9, 126.2, *126.1, 115.7,
*115.5, 77.5, *77.1, 39.4, 27.6, *26.9, *25.0, 24.7, *24.2 (* in-
dicates other isomer); HRMS calcd for C9H14O 138.1045, found
138.1044.
4,4-Dim eth yl-3-[(a zid oca r bon yl)oxy]-1-p en ten e (15a ).
The azidoformate 15a was prepared by the same method used
(35) Vittorelli, P.; Peter-Katalinic, J .; Mukherjee-Muller, G.; Hansen,
H.-J .; Schmid, H. Helv. Chim. Acta 1975, 58, 1379-1425.
(36) Takano, S.; Nishizawa, S.; Akiyama, M.; Ogasawara, K. Het-
erocycles 1984, 22, 1779-1788.

4456 J . Org. Chem., Vol. 62, No. 13, 1997
Bergmeier and Stanchina
at 3.10 ppm leads to the collapse of the signal at 2.10 ppm))
(* indicates minor isomer). The oxazolidinone 17b was pre-
pared by the same method used for the preparation of 17a on
a 0.26 mmol scale in 82% yield as a 3:1 mixture of trans/cis
isomers from 16b: 1H NMR (CDCl3) δ 7.30 (m, 5H), 4.55 (s,
2H), 4.30 (m, 1H, J ) 2.89 Hz (decoupled signal at 3.65 ppm)),
3.65 (m, 3H (decoupled signal at 4.30 ppm leads to the collapse
of the signal at 3.65 ppm)), 1.55 (m, 2H), 1.25 (m, 6H), 0.85 (t,
3H); 13C NMR (CDCl3, trans isomer) δ 158.5, 137.6, 128.5,
127.9, 127.8, 127.7, 80.8, 73.7, 70.3, 54.6, 35.5, 31.5, 28.6, 24.9,
22.4, 13.8; HRMS calcd for C16H23NO3 277.1679, found 277.1689.
(4R*,5R*)-4-P en t yl-5-p h en et h yl-1,3-oxa zolid in -2-on e
(17c). The aziridine 16c was prepared by the same method
used for the preparation of 8a and 8b on a 1.3 mmol scale in
67% yield as a 3:1 mixture of trans:cis isomers, respectively,
from 15c and was used immediately in the next reaction: 1H
NMR (CDCl3) δ 7.25 (m, 5H), *4.65 (m, 0.33H, J ) 4.81 Hz
(decoupled signal at 2.00 ppm; decoupled signal at *3.05 ppm
leads to collapse of signal at *4.65 ppm)), 4.55 (m, 0.67H
(decoupled signal at 2.00 ppm leads to collapse of signal at
4.55 ppm)), *3.05 (m, 0.33H (decoupled signal at *4.65 ppm
leads to collapse of signal at *3.05 ppm)), 2.75 (m, 2.67H), 2.50
(d, 1H), 2.40 (d, 1H), 2.00 (m, 2H) (* indicates minor isomer);
NOE experiment (CDCl3, 270 MHz), irradiation of the peak
at *4.65 ppm shows a 5.18% enhancement of the signal at
*3.05 ppm; irradiation of the peak at *3.05 ppm shows a 7.23%
enhancement of the signal at *4.65 ppm. The oxazolidinone
17c was prepared by the same method used for the preparation
of 17a on a 0.53 mmol scale in 93% yield as a 3:1 mixture of
trans:cis isomers from 16c: 1H NMR (CDCl3) δ 7.20 (m, 5H),
*4.55 (m, 0.33H), 4.15 (m, 0.67H), *3.70 (m, 0.33H, J ) 7.48
Hz (decoupled signal at 1.35 ppm)), 3.40 (m, 0.67H, J ) 5.33Hz
(decoupled signal at 1.35 ppm)), 2.75 (m, 2H), 1.95 (m, 2H),
1.35 (m, 8H), 0.80 (t, 3H) (* indicates minor cis isomer); 13C
NMR (CDCl3, trans isomer) δ 158.8, 128.6, 128.4, 126.2, 81.7,
77.5, 57.9, 36.6, 35.2, 31.6, 31.5, 31.2, 25.0, 22.7, 22.6, 22.4,
14.0, 13.8; HRMS calcd for C16H23NO2 261.1730, found 261.1713.
(4R*,5R*)-5-(3-Cycloh exen yl)-4-p en tyl-1,3-oxa zolid in -
2-on e (17d ). The aziridine 16d was prepared by the same
method used for the preparation of 8a and 8b on a 3.2 mmol
scale in 75% yield as an 11:1 mixture of trans:cis isomers,
respectively, from 15d and was used immediately in the next
reaction: 1H NMR (CDCl3) δ 5.60 (m, 2H), 4.40 (m, 1.09H),
*3.05 (m, 0.09H), 2.90 (m, 0.91H), 2.45 (d, 1H), 1.90 (m, 7H)
(* indicates minor cis isomer); NOE experiment (CDCl3, 270
MHz), irradiation of the peak at 4.40 ppm shows a 7.21%
enhancement of the signal at *3.05 ppm. The oxazolidinone
17d was prepared by the same method used for the prepara-
tion of 17a on a 2.3 mmol scale in 71% yield as an 11:1 mixture
of trans:cis isomers from 16d : 1H NMR (CDCl3, trans isomer)
δ 5.65 (m, 2H), 4.00 (m, 1H, J ) 5.80 Hz (decoupled signal at
3.50 ppm)), 3.50 (m, 1H (decoupled signal at 4.00 ppm leads
to the collapse of the signal at 3.50 ppm)), 2.00 (m, 6H), 1.30
(m, 9H), 0.85 (t, 3H); 13C NMR (CDCl3) δ 159.4, 127.5, 128.7,
125.4, 124.8, 85.9, 85.3, 55.5, 55.4, 38.0, 36.4, 36.2, 31.5, 29.3,
26.7, 26.4, 24.9, 24.7, 24.6, 24.4, 23.9, 23.4, 22.4, 13.8. Anal.
Calcd for C14H23NO2: C, 70.85; H, 9.77; N, 5.90. Found: C,
70.96; H, 9.66; N, 5.60.
dines 20a and 20b were prepared by the same method used
for the preparation of 8a and 8b on a 2.54 mmol scale in 78%
yield from 19 as a 43:1 mixture of 20a and 20b, respectively:
1H NMR (CDCl3, 500 MHz) δ 4.35 (m, 2H), *3.15 (m, 0.02H),
2.95 (m, 1H), *2.65 (m, 0.02H), 2.35 (m, 1H), 1.45 (m, 4H),
0.90 (m, 3H) (* indicates minor isomer); NOE experiment
(CDCl3, 500 MHz), irradiation of peak at *2.65 ppm shows
3.28% enhancement of signal at 3.15 ppm; irradiation of peak
at *3.15 ppm shows a 3.89% enhancement of signal at *2.65
ppm and a 4.41% enhancement of signal at 4.35 ppm; irradia-
tion of peak at 2.95 ppm shows 1.36% enhancement of signal
at 2.35 ppm and a 3.88% enhancement of signal at 4.35 ppm;
irradiation of peak at 2.35 ppm shows a 1.29% enhancement
of signal at 2.95 ppm.
(4S*)-4-[(1R*)-1-Hydr oxybu tyl]-1,3-oxazolidin -2-on e (21).
To a solution of the aziridine 20 (120 mg, 0.83 mmol) in a 1:1
mixture of water/dioxane (0.5 mL:0.5 mL) cooled to 0 °C was
added trifluoroacetic acid (9 mg, 0.083 mmol). The reaction
mixture was stirred at 0 °C for 4.5 h. It was then concentrated
and chromatographed (3:1 EtOAc/hexanes) to afford 120 mg
of 21 (91%), a colorless oil, as a separable mixture of diaster-
eomers in 43:1 trans/cis ratio: 1H NMR (CDCl3 + D2O, trans
isomer) δ 4.35 (m, 2H), 3.80 (m, 1H), 3.65 (m, 1H), 1.35 (m,
4H), 0.85 (t, 3H); 13C NMR (CDCl3) 160.9, 70.9, 65.8, 60.2, 56.9,
34.2, 18.7, 14.0, 13.7; FAB-MS m/ z 160 (M + H). Anal. Calcd
for C7H13NO3: C, 52.82; H, 8.23; N, 8.80. Found: C, 52.88;
H, 8.08; N, 8.71.
(2Z)-1-[(Azid oca r bon yl)oxy]h exen e (23). The azidofor-
mate 23 was prepared by the same method used for the
preparation of 6 on a 6 mmol scale in 93% yield as a colorless
oil from commercially available cis-2-hexen-1-ol (22): 1H NMR
(CDCl3) δ 5.55 (m, 1H), 5.50 (m,1H), 4.65 (d, 2H), 2.00 (m, 2H),
1.30 (m, 2H), 0.80 (t, 3H); 13C NMR (CDCl3) δ 157.0, 136.2,
121.9, 63.7, 29.2, 22.2, 13.1; IR (neat) 3029, 2167, 1731 cm-1
HRMS calcd for C7H11N3O2 169.0852, found 169.0861.
;
P r ep a r a tion of (4R*,6S*)-6-P r op yl-1-oxa -3-a za bicyclo-
[3.1.0]h exa n -2-on e (20a ) a n d (4R*,6R*)-6-P r op yl-1-oxa -3-
a za bicyclo[3.1.0]h exa n -2-on e (20b) fr om 23. The aziri-
dines 20a and 20b were prepared by the same method used
for the preparation of 8a and 8b on a 2.54 mmol scale in 78%
yield from 23 as a 1:2 mixture of 20a and 20b, respectively:
1H NMR (CDCl3, 500 MHz) δ 4.35 (m, 2H), 3.15 (m, 1H), *2.95
(m, 0.5H), 2.65 (m, 1H, J ) 4.81 Hz (decoupled signal at 1.45
ppm)), *2.35 (m, 0.5H, J ) 3.68 Hz (decoupled signal at 1.45
ppm)), 1.45 (m, 4H), 0.90 (m, 3H) (* indicates minor isomer);
NOE experiment (CDCl3, 500 MHz), irradiation of peak at 3.15
ppm shows a 7.45% enhancement of signal at 2.65 ppm and a
6.60% enhancement of signal at 4.35 ppm; irradiation of peak
at *2.95 ppm shows a 1.47% enhancement of signal at *2.35
ppm and a 2.91% enhancement of signal at 4.35 ppm; irradia-
tion of peak at 2.65 ppm shows a 7.34% enhancement of signal
at 3.15 ppm; irradiation of peak at *2.35 ppm shows a 0.97%
enhancement of signal at *2.95 ppm.
Ack n ow led gm en t. This project was funded in part
by the American Association of Colleges of Pharmacy
Grant Program for New Investigators with funds from
The Burroughs Wellcome Fund and the American
Foundation for Pharmaceutical Education.
(2E)-1-[(Azid oca r bon yl)oxy]h exen e (19). The azidofor-
mate 19 was prepared by the same method used for the
preparation of 6 on a 6 mmol scale in 93% yield as a colorless
oil from commercially available trans-2-hexen-1-ol (18); 1H
NMR (CDCl3) δ 5.70 (m, 1H), 5.45 (m, 1H), 4.45 (d, 2H), 1.90
(m, 2H), 1.25 (m, 2H), 0.75 (t, 3H); 13C NMR (CDCl3) δ 156.8,
137.5, 122.5, 68.6, 33.9, 21.6, 13.0; IR (neat) 2932, 2186, 1731
cm-1; HRMS calcd for C7H11N3O2 169.0852, found 169.0854.
P r ep a r a tion of (4R*,6S*)-6-P r op yl-1-oxa -3-a za bicyclo-
[3.1.0]h exa n -2-on e (20a ) a n d (4R*,6R*)-6-p r op yl-1-oxa -3-
a za bicyclo[3.1.0]h exa n -2-on e (20b) fr om 19. The aziri-
Su p p or tin g In for m a tion Ava ila ble: 1H and 13C NMR
spectra of 5-9, 10e, 10f, 11, 13, 14c, 14d , 15a -d , 16a -d ,
17b, 17c, 19, 20a , 20b, and 23 (41 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O970473X