Enecarbamates as Selective Substrates in Oxidations: Chiral-Auxiliary-Controlled Mode Selectivity and Diastereoselectivity in the [2+2] Cycloaddition and Ene Reaction of Singlet Oxygen and in the Epoxidation by DMD and mCPBA

Article abstract of DOI:10.1021/jo035745c

The stereochemical course of the oxidation of chiral oxazolidinone-substituted enecarbamates has been studied for singlet oxygen (¹O₂), dimethyldioxirane (DMD), and m-chloroperbenzoic acid (mCPBA) by examining of the special structural and stereoelectronic features of the enecarbamates. Valuable mechanistic insight into these selective oxidations is gained. Whereas the R¹ substituent on the chiral auxiliary is responsible for the steric shielding of the double bond and determines the sense of the π-facial diastereoselectivity, structural characteristic such as the Z/E configuration and the nature of the R ² group on the double bond are responsible for the extent of the diastereoselectivity. Stereoelectronic steering by the vinylic nitrogen functionality controls the mode selectivity (ene reaction vs [2+2] cycloaddition) in the case of ¹O₂.

Full text of DOI:10.1021/jo035745c

En eca r ba m a tes a s Selective Su bstr a tes in Oxid a tion s:
Ch ir a l-Au xilia r y-Con tr olled Mod e Selectivity a n d
Dia ster eoselectivity in th e [2+2] Cycloa d d ition a n d En e Rea ction
of Sin glet Oxygen a n d in th e Ep oxid a tion by DMD a n d m CP BA
Waldemar Adam,*,† Sara G. Bosio,† Nicholas J . Turro,‡ and Barbara T. Wolff†
Institut fu¨r Organische Chemie, Universita¨t Wu¨rzburg, Am Hubland, D-97074 Wu¨rzburg, Germany, and
Department of Chemistry, Columbia University, 3000 Broadway, New York, New York 10027
wadam@chemie.uni-wuerzburg.de
Received November 27, 2003
The stereochemical course of the oxidation of chiral oxazolidinone-substituted enecarbamates has
been studied for singlet oxygen (1O2), dimethyldioxirane (DMD), and m-chloroperbenzoic acid
(mCPBA) by examining of the special structural and stereoelectronic features of the enecarbamates.
Valuable mechanistic insight into these selective oxidations is gained. Whereas the R1 substituent
on the chiral auxiliary is responsible for the steric shielding of the double bond and determines the
sense of the π-facial diastereoselectivity, structural characteristic such as the Z/E configuration
and the nature of the R2 group on the double bond are responsible for the extent of the
diastereoselectivity. Stereoelectronic steering by the vinylic nitrogen functionality controls the mode
1
selectivity (ene reaction vs [2+2] cycloaddition) in the case of O2.
In tr od u ction
a urea functionality, have been steered through hydrogen
bonding.6 Most recently, we also showed that the dias-
tereoselectivity of the [2+2] cycloaddition of singlet
oxygen may be subject to control, when the oxazolidinone
chiral auxiliary is directly attached to the reactive double
bond in the form of an enecarbamate functionality.4
To date, the enecarbamates, readily prepared from the
corresponding aldehyde and an optically active oxazoli-
dinone auxiliary, have hardly7 been used to direct the
diastereoselectivity of reactions that involve their CC
double bond. In view of their special structural and
electronic features, these unusual substrates offer prom-
ising opportunities to explore their propensity in stereo-
selective oxidations. Variation of their structural fea-
tures, namely, the R/S configuration of the stereogenic
center, the size of the R1 substituent of the oxazolidinone
auxiliary, and the E/Z geometry of the CC double bond,
as well as the type of R2 substituent, enables a high
measure of steric manipulation in directing the attacking
oxidant.
Chiral auxiliaries have been successfully employed in
controlling the diastereoselectivity of numerous oxygen-
transfer reactions1 such as photooxygenations2 and ep-
oxidations.3 The control of the stereochemical course of
these reactions is of general synthetic interest, because
it gives access to enantiomerically pure oxygen-function-
alized building blocks.4
The best results so far in auxiliary-controlled singlet
oxygen reactions were obtained for substrates in which
the chiral auxiliary (oxazolidine or oxazolidinone) was
attached to a carbonyl functionality and not directly to
the double bond. Through the appropriate choice of the
chiral inductor, it is possible to control the diastereo-
1
selectivity of the [4+2] cycloaddition of O2 with sorbic
amide through steric interactions,5 whereas the ene
reaction with optically active oxazolidines, equipped with
† Universita¨t Wu¨rzburg.
‡ Columbia University.
(1) J ones, S. J . Chem. Soc., Perkin Trans. 1 2002, 95, 1-21. Bonini,
C.; Righi, G. Tetrahedron 2002, 58, 4981-5184.
(2) (a) Adam, W.; Griesbeck, A. Synthesis 1986, 1050-1052. (b)
Dussault, P. H.; Woller, K. R.; Hillier, M. C. Tetrahedron 1994, 50,
8929-8940. (c) Adam, W.; Wirth, T.; Pastor, A.; Peters, K. Eur. J . Org.
Chem. 1939, 4, 501-506. (d) Adam, W.; Gu¨thlein, M.; Peters, E.-M.;
Peters, K.; Wirth, T. J . Am. Chem. Soc. 1998, 120, 4091-4093. (e)
Adam, W.; Degen, H.-G.; Krebs, O.; Saha-Moller, C. R. J . Am. Chem.
Soc. 2002, 124, 12938-12939. (f) Adam, W.; Bosio, S. G.; Turro, N. J .
J . Am. Chem. Soc. 2002, 124, 8814-8815; (g) Adam, W.; Bosio, S. G.;
Turro, N. J . J . Am. Chem. Soc. 2002, 124, 14004-14005;
(3) (a) Adam, W.; Pastor, A.; Peters, K.; Peters, E.-M. Org. Lett. 2000,
2, 1019-1022. (b) Adam, W.; Schambony, S. B. Org. Lett. 2001, 3, 79-
82. (c) Adam, W.; Bosio, S. G.; Wolff, B. T. Org. Lett. 2003, 819-822.
(4) Adam, W.; Bosio, S. G.; Turro, N. J . J . Am. Chem. Soc. 2002,
124, 8814-8815.
Moreover, a definite advantage for stereochemical
control through such steric effects derives from the fact
that in these enecarbamates the distance between the
stereogenic center of the chiral auxiliary and the reactive
(5) Adam, W.; Gu¨thlein, M.; Peters, E.-M.; Peters, K.; Wirth, T. J .
Am. Chem. Soc. 1998, 120, 4091-4093.
(6) Adam, W.; Peters, K.; Peters, E.-M.; Schambony, S. B. J . Am.
Chem. Soc. 2000, 122, 7610-7611.
10.1021/jo035745c CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/06/2004
1704
J . Org. Chem. 2004, 69, 1704-1715

Enecarbamates as Selective Substrates in Oxidations
TABLE 1. Syn th esis of th e Dia ster eom er ic
En eca r ba m a tes 1
dra
Z:E
entry
enecarbamate
R1
R2
1
2
3
4
5
6
7
8
9
1a
H
Ph(Me)CH
Ph(Me)CH
Ph(Me)CH
Ph(Me)CH
Ph(Me)CH
Ph(Me)CH
iPr
Me
Me
Me
Me
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
53:47
73:27
71:29
50:50
83:17
(4R)-1b
(4R)-1c
(4S)-1d
(4S)-1e
(4S)-1f
(4R)-1g
1h
(4R)-1i
(4R)-1j
(4S)-1k
(4S)-1l
R-Me
R-iPr
S-iPr
S-Ph
S-tBu
R-iPr
H
R-Me
R-iPr
S-Ph
S-tBu
10
11
12
Me
a
Yields >95% in all cases; the diastereomeric ratios were
determined from the areas under the characteristic signals in the
1H NMR spectra directly on the crude reaction mixture (error (5%
of the stated value); for ratios >95:5, the minor diastereomer was
not detected; material balances >95%.
CC double bond functionality is significantly less com-
pared to the usual Evans’ oxazolidinone auxiliary,8 linked
by means of an amide functionality.9
As oxidants we have chosen singlet oxygen, dimethyl-
dioxirane, and m-chloroperbenzoic acid to examine the
stereoselectivity of the photooxygenation and epoxidation
as a function of the size and type of R1 and R2 substitu-
ents and the E/Z geometry of the enamine-type double
bond. For the photooxygenation it was also of mechanistic
interest to assess the competition between [2+2] cycload-
dition and the ene reaction, when the R2 substituent
carries allylic hydrogen atoms. Analogous to the stereo-
electronic directing effect observed in vinyl ethers,10 it
was expected that such mode selectivity should also be
expressed by the enamine-type functionality, dependent
on its E/Z configuration.
Herein we present the results of our extensive study,
which provides valuable mechanistic insight into the
selective oxidations by singlet oxygen, DMD, and mCP-
BA, as well as promising applications.
FIGURE 1. Crystal structures of the enecarbamates (Z,4R,3′S)-
1c (ref 2f) and (Z,4S)-1k (see the Supporting Information), the
hydroperoxide (4R,1′S)-3j (ref 2g), and the diol unlike-8 (ref
2f).
Resu lts
obtained in high yield (>95%) by refluxing the chiral
oxazolidinones9 and the various aldehydes in toluene,
catalyzed by p-toluenesulfonic acid. In the case of en-
ecarbamates 1a -f, the only double bond diastereomer
obtained from the condensation of 2,3-diphenylbutanal
with the corresponding oxazolidinone was the Z one
(Table 1, entries 1-6), as was proved by NOESY mea-
surements (see the Supporting Information). A high
selectivity was also observed when 2,3-diphenylbutanal
was replaced by 3-methyl-2-phenylbutanal (1g, entry 7).
By condensing the different oxazolidinones with the
2-phenylpropanal, the enecarbamates 1h -l were ob-
tained (entries 8-12). In these cases both double bond
Syn th esis a n d Con figu r a tion a l Assign m en t of th e
En eca r ba m a tes 1. The enecarbamates 1 (Table 1) were
(7) J aney, J . M.; Iwama, T.; Kozmin, S. A.; Rawal, V. H. J . Org.
Chem. 2000, 65, 9059-9068. Xiong, H.; Hsung, R. P.; Shen, L.; Hahn
J . M. Tetrahedron Lett. 2002, 4449-4453. Xiong, H.; Hsung, R. P.;
Berry, C. R.; Rameshkumar, C. J . Am. Chem. Soc. 2001, 123, 7174-
7175. Rameshkumar, C.; Xiong, H.; Tracey, M. R.; Berry, C. R.; Yao,
L. J .; Hsung, R. P. J . Org. Chem. 2002, 67, 1339-1345.
(8) (a) Evans, D. A.; Chapman, K. T.; Bisaha, J . J . Am. Chem. Soc.
1988, 110, 1238-1256. (b) Evans, D. A.; Urpi, F.; Somers, T. C.; Clark,
J . S.; Bilodeau, M. T. J . Am. Chem. Soc. 1990, 112, 8215-8216. (c)
Evans, D. A.; Nelson, S. G. J . Am. Chem. Soc. 1988, 110, 1238-1256.
(d) Ager, D. J .; Prakash I.; Schaad, D. R. Aldrichim. Acta 1997, 30,
3-12.
(9) (a) Evans, D. A.; Chapman, K. T.; Bisaha, J . J . Am. Chem. Soc.
1988, 110, 1238-1256. (b) Evans, D. A.; Urpi, F.; Somers, T. C.; Clark,
J . S.; Bilodeau, M. T. J . Am. Chem. Soc. 1990, 112, 8215-8216. (c)
Evans, D. A.; Nelson, S. G. J . Am. Chem. Soc. 1997, 119, 9309-9310.
(d) Ager, D. J .; Prakash I.; Schaad, D. R. Aldrichim. Acta 1997, 30,
3-12.
(10) (a) Rousseau, G.; Le Perchec, P.; Conia, J . M. Tetrahedron Lett.
1977, 2517-2520. (b) Lerdal, D.; Foote, C. S. Tetrahedron Lett. 1978,
3227-3230. (c) Inagaki, S.; Fujimoto, H.; Fukui, K. Chem. Lett. 1976,
749-752.
J . Org. Chem, Vol. 69, No. 5, 2004 1705

Adam et al.
TABLE 2. Mod e a n d Dia ster eoch em ica l Selectivities in th e P h otooxygen a tion of th e En eca r ba m a tes 1a
selectivityb
mode
diastereo
[2+2]:ene
2:3c
ene
ul:lk
[2+2]
entry
substrate
1a
R1
R2
R3
(1S,2S):(1R,2R)
1
2
3
4
5
H
Ph(Me)CH
(S)-Ph(Me)CH
(R)-Ph(Me)CH
(R)-Ph(Me)CH
(S)-Ph(Me)CH
(S)-Ph(Me)CH
(R)-Ph(Me)CH
(R)-Ph(Me)CH
(R)-Ph(Me)CH
(S)-Ph(Me)CH
(R)-Ph(Me)CH
(S)-Ph(Me)CH
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Ph
Me
Me
Me
Me
Ph
Ph
Ph
Ph
>95
50:50
>95:5
>95:5
>95:5
>95:5
<5:95
>95:5
>95:5
<5:95
<5:95
<5:95
<5:95
>95:5
(Z,4R,3′S)-1b
(Z,4R,3′R)-1b
(Z,4R,3′R)-1c
(Z,4R,3′S)-1c
(Z,4S,3′S)-1d
(Z,4R,3′R)-1c
(Z,4R,3′R)-1c
(Z,4S,3′R)-1e
(Z,4S,3′S)-1e
(Z,4S,3′R)-1f
(Z,4S,3′S)-1f
(Z,4R)-1g
E-1h
Z-1h
(E,4R)-1i
(E,4R)-1j
(E,4S)-1k
(E,4S)-1l
(Z,4R)-1i
(Z,4R)-1j
R-Me
R-Me
R-iPr
R-iPr
S-iPr
R-iPr
R-iPr
S-Ph
S-Ph
S-tBu
S-tBu
R-iPr
H
>95
>95
>95
>95
6
>95
7d
8e
9
>95
>95
>95
10
11
12
13
14
15
16
17
18
19
20
21
22
23
>95
>95
>95
iPr
Ph
Me
Ph
Ph
Ph
Ph
Me
Me
Me
Me
>95
15:85
80:20
16:84
36:64
8:92
H
R-Me
R-iPr
S-Ph
S-tBu
R-Me
R-iPr
S-Ph
S-tBu
88:12
83:17
71:29
91:9
53:47
56:44
85:15
>95:5
n.d.
n.d.
n.d.
23:77
80:20
75:25
87:13
60:19f
n.d.
>95:5
>95:5
<5:95
<5:95
(Z,4S)-1k
(Z,4S)-1l
a
The photooxygenations were run in CDCl3 at -32 °C with TPFPP (5,10,15,20-tetrakis(pentafluorophenyl)porphine) as sensitizer until
complete conversion of the enecarbamate. Determined from the areas under the characteristic signals in the 1H NMR spectrum directly
b
on the photooxygenate (error (5% of the stated value); material balance >95% for all reactions. c In entries 1-13, the ene product
corresponding to 3 was not observed. CD3OD:CDCl3 (4.1:1). e CD3COCD3:CDCl3 (2.4:1). f Also 21% of the endoperoxide 6l was obtained.
d
isomers were formed, and the E/Z ratio depended on the
individual substituent in the oxazolidinone ring. After
chromatographic separation of the E/Z isomers on silica
gel, the double bond configuration was proved by NOESY
measurement; in the case of (Z,4S)-1k , an X-ray analysis
of one of the diastereomers confirmed the proposed
structure (Figure 1).
The unsubstituted enecarbamate 1h and the phenyl
derivative 1k were formed nondiastereoselectively (ca.
50:50 mixture of the two double bond isomers), whereas
the methyl and the isopropyl substituents gave ap-
preciable amounts of the E diastereomers. The tert-butyl-
substitueted enecarbamate (4S)-1l displayed the highest
E selectivity; evidently, the steric repulsion of the de-
manding tert-butyl substituent with the phenyl group on
the double bond is the most effective.
ecarbamates, only [2+2] cycloaddition was observed. The
dioxetanes 2 could not be isolated, because they readily
decomposed at room temperature (ca. 20 °C) to the
expected carbonyl products, especially on attempted silica
gel chromatography.
The photooxygenation of the unsubstituted enecar-
bamate 1a displayed no diastereoselectivity (Table 2,
entry 1), whereas already the methyl derivative (Z,4R)-
1b (Table 2, entries 2 and 3), and certainly the isopropyl
derivative (Z,4R)-1c (entries 4 and 5), afforded the
diastereomerically pure dioxetanes (4R,3′S,4′S)-2b and
(4R,3′S,4′S)-2c. The other possible diastereomers could
1
not be observed even in the 600-MHz H NMR spectra
of the photooxygenate.
The change of the configuration for the R1 substituent
in the oxazolidinone from R to S has as a consequence
the inversion of the configuration at the newly formed
stereogenic centers in the dioxetane 2c, i.e., 4R,3′S,4′S
versus 4S,3′R,4′R (entries 5 and 6). Also in the polar
solvent mixtures CD3OD:CDCl3 (4.1:1, entry 7) and CD3-
COCD3:CDCl3 (2.4:1, entry 8), the diastereomerically
pure dioxetane (4R,3′S,4′S,1′′R)-2c was formed in the
photooxygenation of enecarbamate (Z,4R,3′R)-1c.
The photooxygenation of (Z,4S,3′R)-1e and (Z,4S,3′S)-
1e (R1 ) Ph, entries 9 and 10) and (Z,4S,3′R)-1f and
(Z,4S,3′S)-1f (R1 ) tBu, entries 11 and 12) showed also a
diastereoselectivity of >95:5 in the formation of the
dioxetanes (4S,3′R,4′R)-2e and (4S,3′R,4′R)-2f. The pho-
tooxygenation of enecarbamate (Z,4R)-1g, in which the
P h otooxygen a tion s. The photooxygenation of the
enecarbamates
1 was performed at -35 °C with
5,10,15,20-tetrakis(pentafluorophenyl)porphine (TPFPP)
as the sensitizer and a 800-W sodium lamp as the light
source. In the enecarbamates 1a -f, the 1-phenylethyl
substituent at the C2′ position of the double bond was
chosen to minimize the ene reaction, since the required
coplanar alignment of the only allylic hydrogen atom11
is conformationally encumbered. Indeed, for these en-
(11) Gollnick, K.; Hartmann, H.; Paur, H. In Oxygen and Oxy-
Radicals in Chemistry and Biology; Rodgers, M. A. J ., Powers, E. L.,
Eds.; Academic Press: New York, 1981; pp 379-395.
1706 J . Org. Chem., Vol. 69, No. 5, 2004

Enecarbamates as Selective Substrates in Oxidations
TABLE 3. Dia ster eoselectivities of th e Ep oxid a tion of En eca r ba m a tes 1 by DMD a n d m CP BA
diastereoselectivitya
entry
substrate
Z-1h
reagent
R1
R2
R3
2′S:2′R
1
2
3a
3b
4
5
6
7
8
DMD
mCPBA
DMD
DMD
mCPBA
DMD
mCPBA
DMD
mCPBA
DMD
H
H
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
50:50
50:50
60:40
63:37
70:30
Z-1h
(Z,4R)-1j
(Z,4R)-1j
(Z,4R)-1j
(Z,4S)-1k
(Z,4S)-1k
(Z,4S)-1l
(Z,4S)-1l
E-1h
R-iPr
R-iPr
R-iPr
S-Ph
S-Ph
S-tBu
S-tBu
H
8 (10):92 (90)
8 (7):92 (93)
7 (6):93 (94)
7:93
9
50:50
50:50
53:47
50:50
10
11
12
13
14
15
16
17
18
19
E-1h
mCPBA
DMD
mCPBA
DMD
mCPBA
DMD
mCPBA
DMD
H
(E,4R)-1j
(E,4R)-1j
(E,4S)-1k
(E,4S)-1k
(E,4S)-1l
(E,4S)-1l
1a
R-iPr
R-iPr
S-Ph
S-Ph
S-tBu
S-tBu
H
40:60
48:52b
25 (25):75 (75)
21:79
rac-Ph(Me)CH
(R)-Ph(Me)CH
(S)-Ph(Me)CH
<5:95
(Z,4R,3′R)-1c
(Z,4R,3′S)-1c
DMD
DMD
R-iPr
R-iPr
<5:95
34:66
a
1
Diastereoselectivities were determined by H NMR spectroscopy (error (5% of the stated value) directly on the reaction mixture and
by HPLC analysis (error (2% of the stated value); the results of the diol 10 prepared from epoxide 4 (DMD reactions) and from the ester
5 (mCPBA reactions) are given in parentheses. Also 50% epoxide (2′S,3′R)-4d and 2% ester (2′S,3′R)-5d were obtained. In Table 3, only
b
the configurations at the C2′ position are given, namely the stereochemical descriptors 2′S and 2′R, because this position does not depend
on the initial double bond configuration of the enecarbamates Z-1 and E-1. For clarity and convenience, the 2′S and 2′R descriptors are
marked in bold face in the text.
phenylethyl substituent on the double bond is replaced
by an isopropyl substituent, gave exclusively the di-
oxetane (4R,3′S,4′S)-2g (entry 13).
diastereoselectivity. In the case of the enecarbamates E-1,
the diastereoselectivity could not be determined, because
the dioxetanes were too labile and only the decomposition
products were detected and no conclusion could be
reached on the diastereoselectivity of the Z versus E
enecarbamates.
The enecarbamates 1h -l all possess ene-active allylic
hydrogen and, thus, the [2+2] cycloaddition is competed
for by the ene reaction. The Z- and E-configured enecar-
bamates displayed a substantial difference in the mode
selectivity: Whereas the photooxygenation of the enecar-
bamates E-1 proceeded preferably along the ene mode
to the hydroperoxides 3 (entries 14 and 16-19), the Z-1
diastereomers yielded mainly the [2+2]-cycloadducts 2
(entries 15 and 20-23).
Ep oxid a tion w ith DMD a n d m CP BA. The enecar-
bamates 1h and 1j-l were epoxidized by dimethyldi-
oxirane (DMD) in acetone and m-chloroperbenzoic acid
(mCPBA) in chloroform (Table 3) within a few hours at
20 °C. The epoxidation of the achiral substrates Z-1h (R1
) H, entry 1) and E-1h (entry 9) by DMD gave, as
expected, a 50:50 mixture of the diastereomeric epoxides.
The diastereoselectivity of the ene reaction with the
ene carbamates E-1 ranges from 71:29 for enecarbamate
(E,4S)-1k (entry 18) to 91:9 for (E,4S)-1l (entry 19); no
diastereomers are possible for the enecarbamate E-1h
(R1 ) H, entry 14). In contrast, the ene reaction of the
Z-1 isomers covered a wide spread in the diastereomeric
ratios, which extended from essentially unselective for
i
The epoxidation of the chiral (Z,4R)-1j (R1 ) Pr) at 20
°C gave a mixture of the diastereomeric epoxides 4j in a
2′S:2′R diastereomeric ratio (dr) of only 60:40 (entry 3a);1
no significant increase in the dr value was observed at
-25 °C (entry 3b).
An increase in the size of the substituent R1 on the
oxazolidinone ring caused a significant increase in the
diastereoselectivity of the epoxidation by DMD: For the
enecarbamate (Z,4S)-1k (R1 ) Ph, entry 5), the major
epoxide isomer (2′R,3′R)-4k was obtained in a high
diastereomeric ratio (92:8); this was also the case with
i
(Z,4R)-1i (R1 ) Me, entry 20) and (Z,4R)-1j (R1 ) Pr,
entry 21) to highly selective for (Z,4S)-1k (R1 ) Ph, entry
22) and expectedly (Z,4S)-1l (R1 ) tBu, entry 23). For both
sets of ene products, the major diastereomer has the
unlike configuration and the extent in the facial dif-
ferentiation follows the steric demand of the R1 substitu-
ent in the oxazolidinone ring. Furthermore, the diaste-
reofacial differentiation is more pronounced for the Z
(entries 20-23) than the E isomers (entries 16-19).
The diastereoselectivity of the [2+2] cycloaddition is
essentially perfect in the case of the enecarbamates Z-1
(entries 20-23), except for Z-1h (R1) H, entry 15), which
has no stereogenic centers and cannot display any
(Z,4S)-1l (R1
)
tBu, entry 7), for which the major
diastereomer was (2′R,3′R)-4l. The epoxidation of the E-1
isomers by DMD showed the same trend, in that for the
achiral E-1h (R1 ) H) no selectivity was observed (entry
9), whereas the 2′S:2′R ratios range from 53:47 for (E,4R)-
i
1j (R1 ) Pr, entry 11), to 40:60 for (E,4S)-1k (R1 ) Ph,
entry 13), to 25:75 for (E,4S)-1l (R1 ) tBu, entry 15), that
J . Org. Chem, Vol. 69, No. 5, 2004 1707

Adam et al.
TABLE 4. Tr a n sfor m a tion of th e Dioxeta n es 2a ,c,g,k to th e Diols 8-10 for th e Ch em ica l Cor r ela tion of Th eir
Con figu r a tion s
entry
substrate
R1
R2
yield (%)
product
2S:2Ra
1
2
3
4
5
(4R,3′S,4′S,1′′R)-2c
(4R,3′S,4′S,1′′S)-2c
2a
(4R,3′S,4′S)-2g
(4S,3′R,4′R)-2k
R-iPr
R-iPr
H
(R)-Ph(Me)CH
(S)-Ph(Me)CH
rac-Ph(Me)CH
(CH3)2CH
30
35
25
41
30
unlike-8
like-8
8
9
>95:5
>95:5
50:50
>95:5
<5:95
R-iPr
S-Ph
CH3
10
a
Diastereoselectivity determined by HPLC analysis (2% error of the stated value) of the diols 8-10; the HPLC analyses were carried
out on chiral columns (Daicel CHIRALCEL OD-H) on a Kontron instrument, furnished with a UVIKON 720 spectrophotometer and a
CHIRALYSER 1.6 from IBZ Massetechnik (Hannover, Germany).
SCHEME 1. Ep oxid a tion of th e Dia ster eom er ic
(Z,4R,3′R)-1c a n d (Z,4R,3′S)-1c En eca r ba m a tes w ith
DMD
is, the diastereoselectivity increases with the size of the
R1 substituent, but only moderately.
In the epoxidation with mCPBA, only a low conversion
(ca. 10%) of the enecarbamates 1h and 1j-l was observed
directly in the 1H NMR spectrum of the reaction mixture.
The major epoxide diastereomer was in all cases the same
as that in the DMD reaction. Unfortunately, under these
reaction conditions, the epoxide was opened exclusively
at the C2′ position to the ester 5 by the acid generated
from mCPBA, which could not be avoided by using
aqueous NaHCO3 as buffer (not shown in Table 3).
As expected, the achiral substrate Z-1h (R1 ) H, entry
SCHEME 2. In d ep en d en t Syn th esis of Au th en tic
Diol u n like-8
2) was epoxidized unselectively by mCPBA. The diaste-
reoselectivity in the epoxidation of the chiral enecarbam-
ates Z-1j-l increases as the size of the R1 substituent
on the oxazolidinone ring is increased. Thus, the dr
values range from 70:30 (2′S:2′R) for the isopropyl-
substituted enecarbamate (Z,4R)-1j (entry 4), to 8:92 for
the phenyl-substituted derivative (Z,4S)-1k (entry 6), to
1c was also fully epoxidized after 3 d (entry 19); however,
7:93 for the tert-butyl one (Z,4S)-1l (entry 8). The
the epoxides (2′R,3′R,1′′S)-4c and (2′S,3′S,1′′S)-4c were
formed in a diastereomeric ratio of only 67:33. Clearly,
E-configured enecarbamates (entries 12, 14, and 16) were
epoxidized by mCPBA in a lower diastereoselectivity than
this low π-facial differentiation for the (Z,4R,3′S)-1c
enecarbamate compared to the high one for the (Z,4R,3′R)-
1c diastereomer discloses a pronounced substrate inher-
ent match/ mismatch situation.
the Z-configured counterparts (entries 4, 6, and 8), as was
the case for DMD (entries 3, 5, and 7 versus 11, 13, and
15). Only for the substrate with the large tert-butyl
substituent (E,4S)-1l (entry 16) was an appreciable dr
value of 21:79 observed in favor of the (2′S,3′S)-4l isomer.
The mCPBA epoxidation of the (E,4S)-1k (R1 ) Ph)
enecarbamate is unusual in that for this substrate the
(2′R,3′S)-4k diastereomer is ring-opened to the ester
(2′S,3′S)-5k , whereas the (2′S,3′R)-4k epoxide persists
esterification. Presumably, the nucleophilic attack of the
m-chlorobenzoic acid on the C2′ position of the latter
epoxide is sterically hindered.
To assess the influence of an additional stereogenic
center within the enecarbamate structure and to uncover
a possible match/mismatch effect, the racemic enecar-
bamate 1a and the optically active derivatives (Z,4R,3′R)-
1c and (Z,4R,3′S)-1c were epoxidized. Thus, the enecar-
bamate 1a was converted in high diastereoselectivity
(>95:5, entry 17) to the epoxide (2′S*,3′S*,1′′S*)-4a . In
contrast, the epoxidation of the enecarbamate (Z,4R,3′R)-
1c with DMD (Scheme 1) proceeded very slowly and gave
after 3 d at ca. 20 °C the epoxide (2′R,3′R,1′′R)-4c in high
diastereoselectivity (>95:5, entry 18) and 88% yield. The
corresponding 3′S-diastereomeric enecarbamate (Z,4R,3′S)-
Con figu r a tion a l Assign m en ts. The absolute config-
uration of the dioxetanes 2a ,c,g,k was established by
reduction with L-methionine12 (Table 4), followed by
removal of the chiral auxiliary through treatment with
NaBH4/DBU13 to afford the diols 8-10. In the case of
dioxetane (4R,3′S,4′S,1′′R)-2c (entry 1), the R configu-
ration at the C1′′ position is known from an X-ray
diffraction analysis of the S-configured enecarbamate
diastereomer (Z,4R,3′S)-1c (Figure 1). The independent
synthesis of diol 8 from desoxybenzoin (Scheme 2) gave
a diastereomeric mixture (92:8) of the unlike and like
products. The diastereomers were separated by silica gel
chromatography and the NMR spectra of the unlike
diastereomer [the relative configuration was determined
by X-ray analysis (Figure 1)] were identical with those
of the diol 8 obtained from the dioxetane (3′S,4′S,1′′R)-
(12) Adam, W.; Hadjiarapoglou, L.; Mosandl, T.; Saha-Mo¨ller, C. R.
Angew. Chem., Int. Ed. Engl. 1991, 30, 200-202.
(13) Gaul, C.; Scha¨rer, K.; Seebach, D. J . Org. Chem. 2001, 66,
3059-3073.
1708 J . Org. Chem., Vol. 69, No. 5, 2004

Enecarbamates as Selective Substrates in Oxidations
SCHEME 3. Assign m en t of th e Absolu te
Con figu r a tion of th e Dioxeta n e 2c by Ch em ica l
Cor r ela tion w ith Diol 8
SCHEME 5. Assign m en t of th e Rela tive
Con figu r a tion of th e Ep oxid e 4a by Ch em ica l
Cor r ela tion w ith Diol 8
10 gave the same enantiomeric ratio (10:90) as the
diastereomeric ratios (8:92) of the epoxide 4k and of the
ester 5k (within the experimental error). The absolute
configuration of the major enantiomer was found to be
R by comparison with an authentic reference sample of
enantiomerically enriched diol 9, prepared independently
according to the literature14 (see the Supporting Informa-
tion).
SCHEME 4. Tr a n sfor m a tion of th e Ep oxid e 4k a n d
th e Ester 5k to th e Com m on Diol 10a
With the configuration of the major enantiomer of diol
10 known to be R, the configuration at the C2 position
in the major isomer of the epoxide 4k and the ester 5k ,
both derived from enecarbamate (Z,4S)-1k , may also be
assigned R, provided that the C2 site is not involved in
the ring-opening hydrolysis of 4k and the esterification
of 4k to 5k . That this is so could be unequivocally
established for the latter process, since NMR analysis of
the reaction mixture revealed that exclusively the re-
gioisomer with nucleophilic attack on the C2′ position had
been formed. Consequently, the absolute configuration
of the ester 5k is 1S,2R and, hence, that of the epoxide
4k is 2′R,3′R.
The configurational assignment of the epoxides 4a and
4c derived from the DMD epoxidation of the enecarbam-
ates 1a , (Z,4R,3′S)-1c, and (Z,4R,3′R)-1c was achieved
in analogy with that of epoxides 4j,k . In the case of the
diastereomerically pure racemic epoxide 4a , hydrolysis
afforded the like diastereomer of the diol 8 (Scheme 5).
From the relative configuration of the diol like-8, the
relative configuration of the epoxide 4a was assigned to
be (2′S*,3′S*,1′′S*)-4a .
According to the same procedure, the diastereomeric
epoxide mixture of (2′R,3′R,1′′S)-4c and (2′S,3′S,1′′S)-4c
was transformed to the diol 8, whose RS:SS-diastereo-
meric ratio was determined to be 67:33 by chiral HPLC
analysis. The same method was applied for the assign-
ment of the diastereomerically pure epoxide 4c derived
from (Z,4R,3′R)-1c. The resulting diol like-8 indicates the
2′R,3′R,1′′R configuration for the newly formed stereo-
genic centers in the epoxide 4c.
a
The configuration refers to the major stereoisomer.
2c; hence, this diol is also unlike configured. In the 2,3-
diphenyl-1,2-butanediol (8) derived from the dioxetane
(3′S,4′S,1′′R)-2c, the C2 stereogenic center (C1′′ in the
dioxetane) is already known to have the R configuration
(Scheme 3), because the dioxetane was formed from the
diastereomerically pure enecarbamate (Z,4R,3′R)-1c; thus,
it follows that the C2 site in diol 8 possesses the S
configuration, and the dioxetane 2c, derived from
(Z,4R,3′R)-1c, has the 4R,3′S,4′S,1′′R configuration.
The absolute configuration of the dioxetane 2c, ob-
tained from (Z,4R,3′S)-1c (entry 2), was determined in
the same way and the configuration was shown to be
4R,3′S,4′S,1′′S. In the case of the dioxetane 2a that
pertains to enecarbamate 1a (R′ ) H), the transformation
to the diol 8 gave a 50:50 diastereomeric mixture (entry
3).
By following the same procedure, the dioxetane
(4R,3′S,4′S)-2g was derivatized to the known diol 9
(Table 4, entry 4) and characterized by chiral HPLC
analysis. For this purpose, a racemic sample of the diol
9 (see the Supporting Information), whose HPLC data
are known, was employed as reference.
The absolute configuration of the dioxetane (4S,3′R,4′R)-
2k was determined by conversion to the corresponding
diol 10 (Table 4, entry 5) and chiral HPLC analysis, with
the enantiomerically pure diol as reference. The absolute
configuration for the major diastereomer of the hydro-
peroxide 3j was determined to be 4R,1′S (unlike) by
means of X-ray analysis (Figure 1).
Discu ssion
The results in Table 2 (photooxygenations) and Table
3 (epoxidations) show that the enecarbamates 1 are
highly effective in controlling the diastereoselectivity of
oxyfunctionalization by a number of electrophilic oxi-
dants, which include singlet oxygen, DMD, and mCPBA.
The characteristic structural features of the substrate,
both of the oxazolidinone chiral auxiliary and of the CC
double bond functionality, dictate the diastereoselective
course of these oxidations through conformational effects.
These derive mainly from steric interactions conditioned
The absolute configurations of the epoxides 4 and
esters 5 were assessed by their transformation to the
common diol 10 (Scheme 4, shown exemplarily for the
epoxide 4k and the ester 5k ). The hydrolysis of the
epoxide 4k produced from the enecarbamate (Z,4S)-1k
was catalyzed by p-toluenesulfonic acid (p-TsOH) in an
acetone/water mixture, followed by reductive cleavage of
the chiral auxiliary with NaBH4/DBU. Similarly, the
ester 5k obtained in the mCPBA epoxidation of the
enecarbamate (Z,4S)-1k was readily reduced to the diol
10 by NaBH4. Chiral HPLC analysis of the resulting diol
(14) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J .; J oeng, K. S.; Kwong, H. L.; Morika, A. K.; Wang, Z. M.;
Xu, D. Q.; Zhang, X. L. J . Org. Chem. 1992, 57, 2768-2771.
J . Org. Chem, Vol. 69, No. 5, 2004 1709

Adam et al.
F IGURE 2. Possible conformations A and B of the enecarbamate (Z,4R)-1c.
F IGURE 4. Steric interactions in the matched and mis-
matched cases of the enecarbamates 1.
F IGURE 3. Distinct steric shielding for the enecarbamates
E-1 und Z-1.
on conformational effects, as well as from the stereoelec-
tronic control in the attack of the oxidant (singlet oxygen).
For convenience, we shall first enumerate these substrate-
specific aspects and subsequently elaborate how they
account for the observed selectivity.
Str u ctu r a l a n d Ster eoelectr on ic Ch a r a cter istics
of th e En eca r ba m a tes. The preferred conformation of
enecarbamate 1 plays an important role in the diastere-
ochemical differentiation observed in this work. The
shielding of only one side of the CC double bond is caused
by the relative position of the chiral oxazolidinone with
respect to the olefinic functionality. The X-ray structure
of the enecarbamate (Z,4R,3′S)-1c (Figure 1) shows that
a relatively small dihedral angle (ca. 38°) is favored
between the carbonyl functionality and the CC double
bond. For this reason, the R1 substituent of the oxazoli-
dinone occupies preferentially one of the half-spaces of
the double bond and is thereby in an ideal position to
interact sterically with the incoming oxidant. Of the two
possible conformations A and B for the Z-configured
enecarbamate 1c, NOESY measurements reveal that in
solution the conformer A dominates, in which the R1
substituent and not the carbonyl group is directed toward
the phenyl group at the double bond to minimize steric
repulsions within the substrate molecule (Figure 2).
A relevant structural feature for stereochemical control
is the distinct conformations of the double bond isomers
E-1 and Z-1. From DFT calculations (see the Supporting
Information) it is apparent that the R1 substituent (tBu,
Ph) covers up the CC double bond better in the Z isomers
than in the E case (Figure 3), which results in more
effective shielding of the CC double bond; thus, a higher
stereochemical control is expected for the Z-configured
than for the E-configured enecarbamates. For the Z
isomer, already the sterically less demanding methyl and
isopropyl substituents promote a better diastereodiffer-
entiation than for the corresponding E isomer (Figure 3).
An interesting structural feature offers the presence
of an additional stereogenic carbon atom in the olefinic
functionality of the enecarbamates 1: The stereochemical
course of the reaction may be influenced through the
“match-mismatch” effect for the two possible diastere-
omers (Figure 4). As the structures reveal, the steric
interactions between the two stereocenters in the R,R
F IGURE 5. HOMO of the (Z,4S)-1k enecarbamate (calculated
by the B3LYP method) and its orbital-directing effect of the
singlet-oxygen attack.
diastereomer are less severe compared with those in the
R,S diastereomer, such that distinct conformations are
expected and, consequently, differences in the stereo-
chemical control occur.
An important stereoelectronic feature of the enecar-
bamates stems from the presence of the vinylic nitrogen
functionality. In the photooxygenation of the enecarbam-
ates 1, this lone-pair-bearing substituent plays the same
role that is taken by the methoxy group in the photooxy-
genation of methoxy styrenes.10 Analogous to this well-
documented “methoxy effect”,15 which depends on the
stereoelectronic interaction between the lone pair of the
methoxy group and the incoming singlet oxygen, also in
the case of the enecarbamates 1, the lone pair of the
vinylic nitrogen steers the attack of the singlet oxygen
preferentially to the side of the double bond that bears
the oxazolidinone chiral auxiliary (Figure 5).
On the basis of these structural and stereoelectronic
characteristics of the enecarbamates 1, we shall now
rationalize the selectivities observed in the oxidation of
these mechanistically informative substrates. First we
shall discuss separately the mechanistic aspects of the
photooxygenation and epoxidation and subsequently
point out conclusive common traits.
(15) Inagaki, S.; Fujimoto, H.; Fukui, K. Chem. Lett. 1976, 749-
752.
1710 J . Org. Chem., Vol. 69, No. 5, 2004

Enecarbamates as Selective Substrates in Oxidations
SCHEME 6. Z- a n d E-Con figu r ed En eca r ba m a tes 1k for th e Atta ck of Sin glet Oxygen fr om below th e P la n e
of th e P a p er
SCHEME 7. P r efer r ed π-F a cia l Atta ck of Sin glet
Oxygen , Con tr olled by Ster ic Sh ield in g of th e R1
Su bstitu en t in th e En eca r ba m a te 1
Mod e Selectivity in th e P h otooxygen a tion . From
the results in Table 2 it is clear that the mode selectivity
of the substrates may be divided into two groups: The E
isomers react preferentially according to ene mode,
whereas for the Z isomers the [2+2] cycloaddition is
favored. This diastereomer-dependent dichotomy may be
understood in terms of the orbital-directing effect of the
vinylic nitrogen functionality. The orbital interaction
between the HOMO (Figure 5, for (Z,4S)-1k ) of the
enecarbamate and the LUMO of the incoming singlet
oxygen directs the attack onto the side that bears the
nitrogen atom (Scheme 6). Since for the enecarbamates
Z-1 no allylic hydrogen atom is present on this side of
the double bond, [2+2] cycloaddition takes place prefer-
ably (entries 15 and 20-23). The [4+2] cycloaddition may
compete when the double bond and the phenyl substitu-
ent are essentially coplanar, as is presumably the case
for substrate Z-1l (Table 2, entry 23).
In contrast to Z-1, in the enecarbamates E-1 the
directing nitrogen atom of the enamine functionality and
the methyl group are located on the same side of the
double bond. Thus, the incoming singlet oxygen abstracts
an allylic hydrogen atom from the methyl group and a
high mode selectivity in favor of the ene reaction is
expressed (entries 14 and 16-19). In the case of the
substrates 1a -g, not even traces of ene reaction are
observed (Table 2, entries 1-13). The reason for this is
that not only does the vinylic nitrogen atom favor [2+2]
cycloaddition, but also 1,2-allylic strain disfavors the
competing ene reaction. As the first conformer of the
enecarbamates 1a -g in Figure 6 conveys, the allylic
hydrogen atom to be abstracted cannot assume a copla-
nar alignment with the π orbital of the CC double bond
because of steric repulsion between the methyl group at
the chirality center and the vinylic phenyl substituent
(Figure 6).
already a methyl group in the oxazolidinone ring suffices
to obtain perfect π-facial control (Table 2, entries 2 and
3), a fact that illustrates the efficacious steric shielding
inherent in these enecarbamates. This is accentuated by
the finding that no diastereoselectivity is observed in the
absence of chirality in the oxazolidinone (R1 ) H, entry
1).
From the results in Table 2 it becomes evident that
the R-configured R1 substituents induce two new S-
configured stereogenic centers (entries 2-5, 7-8, and 13),
as shown in Scheme 7. In contrast, the S-configured R1
substituents afford the inverse diastereoselectivity (en-
tries 6 and 9-12). Furthermore, the configuration at the
C3′ position of the enecarbamate (Scheme 7) does not
influence the diastereoselectivity, as is demonstrated by
the fact that the attack of singlet oxygen takes place from
the same π face, irrespective of whether the 3′R- or 3′S-
configured enecarbamate (Z,4R,3′R)-1c or (Z,4R,3′S)-1c
(entries 4 and 5, Table 2) is employed. Furthermore, the
complete diastereoselectivity observed in the case of the
Dia ster eoselectivity in th e [2+2] Cycloa d d ition
of Sin glet Oxygen . The high diastereoselectivity in the
[2+2] cycloaddition of singlet oxygen is demonstrated by
all chiral substrates 1b-g (Table 2, entries 2-13). Thus,
i
enecarbamate (Z,4R)-1g (R2 ) Pr, entry 13), in which
the phenylethyl substituent is replaced by an isopropyl
group, confirms that only the stereogenic center on the
oxazolidinone dictates the π-facial selectivity. This effect
is in part due to the steering nature of the vinylic
nitrogen functionality, which promotes syn attack of
singlet oxygen (Scheme 7).
Furthermore, the diastereoselectivity of the photooxy-
genation of enecarbamate (Z,4R,3′R)-1c is independent
of the solvent polarity (entries 7 and 8, Table 2), which
suggests that the carbonyl group plays no significant role
in the stereochemical outcome of the [2+2] cycloaddition.
This is in good agreement with the X-ray structure of
(Z,4R,3′S)-1c (Figure 1), which shows that the carbonyl
group points away from the double bond functionality and
F IGURE 6. Sterically hindered coplanar and preferred in-
plane conformations of the abstractable allylic hydrogen atom
in the enecarbamates 1a -g.
J . Org. Chem, Vol. 69, No. 5, 2004 1711

Adam et al.
polarization of the CC double bond. Electron-rich sub-
stituents in the R region increase the electron density in
the â region through conjugation (+M effect) and the
attack will be preferred there. This implies an asym-
metric perepoxy-like structure for the transition state,
with the O-1 atom closer to the double bond in the â
region. For the ene reaction, the steric interaction with
the substituent R1 is less severe, because the terminal
O-2 atom is closer to the methyl group for the hydrogen
abstraction, whereas in the [2+2] cycloaddition, the
terminal O-2 atom must come into the R region to cyclize
the dioxetane ring and the interaction with R1 is much
more intensive.
Dia ster eoselectivity in th e DMD a n d m CP BA
Ep oxid a tion s. The epoxidation of the enecarbamate
1j-l showed also a high degree of diastereoselectivity
(Table 3), although not as high as for the photooxygen-
ations (Table 2). The diastereoselectivity of the epoxida-
tion is again controlled by the R1 substituent of the
oxazolidinone ring, since the epoxidation of the parent
achiral enecarbamate 1h (R1 ) H) is unselective (Table
3, entries 1 and 2). Moreover, the configuration of
stereogenic center in the oxazolidinone ring determines
the absolute configuration of the epoxide product, that
is, the R configuration in the enecarbamate Z-1j (entries
3 and 4, Table 3) leads mainly to the (4R,2′S,3′S)-4j
product, whereas the S configuration in the substrates
Z-1k (entries 5 and 6) and Z-1l (entries 7 und 8) affords
mainly the corresponding 2′R,3′R epoxides.
To account for the fact that the size of the R1 substitu-
ent dictates the extent of the diastereoselectivity of the
epoxidation, we propose the mechanism in Figure 8 for
the oxygen transfer, in which for convenience of com-
parison the attacks from above and below for both the E
and the Z diastereomers are given. When X ) H, the
steric hindrance for the 1′Re,2′Re attack is more or less
the same as that for the 1′Si,2′Si attack, since a favorable
conformation may be selected in which the methyl groups
are out of the way of the incoming oxidant. When,
however, R1 is a tert-butyl group (X ) CH3 in Figure 9)
as in the enecarbamate Z-1l (entries 7 and 8, Table 3),
the 1′Re,2′Re attack is severely obstructed and the
highest 1′Si,2′Si diastereoselectivity (dr ) 7:93) is ob-
tained; the (2′R,3′R)-4l epoxide is formed almost exclu-
sively for both oxidants DMD and mCPBA.
The preferably attacked π face remains the same for
the Z- or E-configured enecarbamates, i.e., the 1′Si,2′Si
face for the Z and the 1′Si,2′Re face for the E isomer in
Figure 8; however, the extent of the diastereofacial
differentiation depends decisively on the double bond
configuration, as manifested by the fact that the epoxi-
dation of the Z isomers is always more selective than that
of the E isomers. This difference is due to the conforma-
tional effects in the Z and E isomers, which express the
shielding efficacy of the double bond in the enecarbamate.
As may be seen from the lowest energy conformations of
the E-1l and Z-1l isomers calculated by the DFT method
(see the Supporting Information), the shielding of the
upper face of the double bond, i.e., 1′Re,2′Re for Z-1l and
the 1′Re,2′Si for E-1l (Figure 8), by the tBu group is more
effective in the case of the Z than the E diastereomer.
Expectedly, the 1′Si,2′Si attack should be facilitated for
the Z case, as is documented by the higher dr values for
the Z-1l substrate (Table 3, entries 7 and 8) versus those
F IGURE 7. Fukui’s polarity model of singlet oxygen applied
to the photooxygenation of the enecarbamate substrate.
is too remote to interact sterically and/or electronically
with the attacking 1O2. Thus, the high diastereoselectivity
in the dioxetane formation from the Z-1i-l enecarbam-
ates (Table 2, entries 20-23) may be explained in terms
of the synergistic interplay between the directing vinylic
nitrogen effect and the effective shielding of one of the π
faces of the CC double bond by the R1 substituent
(Scheme 7). Unfortunately, the diastereoselectivity of the
[2+2] cycloaddition for the enecarbamates E-1i-l could
not be assessed (Table 2, entries 16-19), because these
dioxetanes decompose during the photooxygenation.
Dia ster eoselectivity in th e En e Rea ction of Sin -
glet Oxygen . The diastereoselectivity of the ene reaction,
like the mode selectivity, depends strongly on the double
bond configuration of the enecarbamate (Figure 3). In the
case of the Z-1i-l derivatives (entries 20-23, Table 2),
relatively little (13-25%) ene product is formed and the
diastereoselectivity depends strongly on the steric hin-
drance of the R1 substituent, which follows the order Me
t
iPr < Ph , Bu. In fact, to obtain the ene product, an
anti attack of singlet oxygen must occur with respect to
the R1(Ph) substituent (Scheme 6), because the methyl
group with the abstractable hydrogen atoms is on the
opposite side and less likely accessible. In the case of the
methyl and isopropyl R1 substituents (Table 2, entries
20 and 21), the steric shielding is not sufficiently effective
to obtain appreciable diastereofacial differentiation in the
anti attack (entries 20 and 21, Table 2). On the contrary,
the phenyl and tert-butyl R1 substituents (entries 22 and
23) are sufficiently spacious to cover up one π face of the
double bond and block hydrogen abstraction from the anti
direction.
In the case of the E-1 substrates, the oxazolidinone ring
is on the same side of the double bond as the methyl
group (Scheme 6). Thus, through the steering effect of
the vinylic nitrogen, the singlet oxygen enters the double
bond from the side of the oxazolidinone ring (syn attack)
and encounters more severe steric interactions with the
R1 substituent. Accordingly, the diastereoselectivity in
the ene reaction of the E-1 substrates is generally higher
(entries 16-19, Table 2) than that for the corresponding
Z isomer (entries 20-23) and already a methyl substitu-
ent gives about the same diastereoselectivity (88:12) as
a tert-butyl substituent (91:9).
The generally lower diastereoselectivity of the ene
reaction with respect to the [2+2] cycloaddition depends
on the distinct steric interactions between the incoming
singlet oxygen and the R1 substituent for these two
reaction modes. According to Fukui’s theoretical analy-
sis,15 the alignment of the approaching O-1 atom (Figure
7) determines whether the R or â region of the double
bond will be attacked, which in turn depends on the
1712 J . Org. Chem., Vol. 69, No. 5, 2004

Enecarbamates as Selective Substrates in Oxidations
F IGURE 8. Preferred π-facial attack of the DMD epoxidation as a function of the size of the X substituent in the enecarbamates
E-1j,l and Z-1j,l (X ) H, CH3).
F IGURE 9. Preferred configurations for the match/mismatch effect in the Z,4R,3′R and Z,4R,3′S diastereomers of the enecarbamate
1c versus the Z,3′R enantiomer of 1a and the Z,4R enantiomer of 1j.
for E-1l (entries 15 and 16). Thus, only the large steric
demand of a tert-butyl group suffices to cover up the
double bond sufficiently in the E-configured enecarbam-
ate 1l to induce an appreciable diastereoselectivity.
Th e Ma tch /Mism a tch Effect in th e Ep oxid a tion .
The match/ mismatch effect in the epoxidation is docu-
mented for the set of enecarbamates 1a (only the phen-
ylethyl stereogenic center), the diastereomeric pair
(Z,4R,3′R)-1c and (Z,4R,3′S)-1c (both stereogenic cen-
ters), and (Z,4R)-1j (only the oxazolidinone stereogenic
center), whose stereoprojections are given in Figure 9.
In the enecarbamate 1a , the phenylethyl substituent
fully controls the attack of the DMD on the double bond
(Figure 9), since complete diastereoselectivity is observed
(Table 3, entry 17). Within the phenylethyl substituent,
the phenyl group is sterically more demanding and, thus,
points away from the remaining substrate (Figure 9).
Since the hydrogen atom at the chirality center is
sterically less demanding than the methyl group, it is
oriented toward the double bond, while the methyl group
shields the π face above the double bond. Consequently,
the DMD attack takes place preferably from the bottom
π face and leads to the (2′R*,3′R*,1′′R*)-4a epoxide.
In the epoxidation of the diastereomeric enecarbamates
(Z,4R,3′R)-1c and (Z,4R,3′S)-1c, the complete diastereo-
selectivity for (Z,4R,3′R)-1c indicates that this constitutes
the matched case (Table 3, entry 18), whereas the low
diastereoselectivity of 67:33 (entry 19) for the epoxidation
of (Z,4R,3′S)-1c suggests that this is the mismatched
situation. From the steric interactions displayed in
Figure 9, it becomes evident that the phenyl group of the
phenylethyl substituent in (Z,4R,3′R)-1c points away
from the rest of the substrate to minimize 1,2-allylic
strain. In this preferred conformation, the methyl group
of the phenylethyl substituent hinders the attack from
the upper side of the double bond and overcompensates
for the influence of the isopropyl group on the oxazoli-
dinone ring. This is substantiated by the fact that little
control in the diastereoselectivity is observed for the
epoxidation of the enecarbamate (Z,4R)-1j, which pos-
sesses only chirality in the oxazolidinone ring (Figure 9).
In the (Z,4R,3′S)-1c diastereomer, the R1 substituent
of the oxazolidinone sterically interacts with the phenyl-
ethyl substituent on the CC double bond (Figure 9) in
competition with 1,2-allylic strain. Consequently, the
steric repulsion between the methyl and the isopropyl
groups populates a minimum-energy conformation in
which the methyl substituent obstructs the attack of the
oxidant DMD on the upper π face, whereas the phenyl
group of the phenylethyl substituent shields the lower π
face. Thus, the (Z,4R,3′S)-1c diastereomer constitutes the
mismatch case, for which a much lower (67:33) diaste-
reoselectivity (entry 17) is observed.
That in the epoxidation the phenylethyl substituent
controls the diastereoselectivity better than the oxazoli-
dinone auxiliary contrasts the results seen in the pho-
tooxygenation, for which the reverse situation applies.
This divergence may be explained in terms of the
following two factors: On one hand, DMD is larger than
singlet oxygen and, thus, the steric interactions of the
phenylethyl substituent at the double bond are more
intensive; on the other hand, the steering effect of the
vinylic nitrogen atom in singlet-oxygen attack accentu-
ates the steric interactions with the R1 substituent of the
oxazolidinone auxiliary.
Syn th etic Ap p lica tion . The highly diastereoselective
photooxygenation of the enecarbamate 1 offers the op-
portunity to exploit these advantageous substrates for
the synthesis of valuable, optically active building blocks.16
For example, the reduction of the dioxetanes 2 to optically
active diols constitutes such a utilization of this unprec-
(16) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, B. K. Chem.
Rev. 1994, 94, 2483-2547.
J . Org. Chem, Vol. 69, No. 5, 2004 1713

Adam et al.
selectivity of 1O2 is caused by the favorable HOMO-
LUMO interaction with the vinylic-nitrogen functionality,
which steers the enophile toward the syn side of the
double bond with respect to the oxazolidinone. When this
side bears a methyl group, allylic hydrogen abstraction
takes place and the ene product is formed; when, how-
ever, the phenyl group is located cis to the oxazolidinone
substituent, [2+2] cycloaddition occurs. For the latter
reaction mode, the synergistic interplay between the
vinylic nitrogen functionality and the shielding by the
R1 substituent cause the diastereoselectivity in the [2+2]
cycloaddition to be essentially perfect (dr > 95:5) already
for the relatively small methyl group. The bulky phenyl-
ethyl substituent on the CC double bond plays no
significant role, because the singlet oxygen attacks the
Z diastereomer anti to the phenylethyl group due to the
steering by the vinylic nitrogen functionality. In contrast,
in regard to diastereoselectivity, the vinylic nitrogen
functionality is inconsequential for the epoxidation.
Although it would be expected that by increasing the size
SCHEME 8. Syn th esis of th e Op tica lly Active
En eca r ba m a tes 1, Th eir Ster eoselective Oxid a tion
to th e Resp ective Dioxeta n es 2 a n d Ep oxid es 4,
Th eir Tr a n sfor m a tion to th e En a n tiom er ica lly
En r ich ed Diols 8-10, a n d th e Recycliza tion of th e
Ch ir a l Oxa zolid in on e Au xilia r y
1
of the oxidant (DMD versus O2) the diastereoselectivity
should increase, the highest dr value for the epoxidation
is at best 93:7 versus >95:5 for the photooxygenation.
This is the consequence of the lack of stereoelectronic
steering by the vinylic nitrogen functionality for DMD
1
versus 1O2. Contrary to O2, DMD attacks anti to the
edented oxidation chemistry (Scheme 8). The chiral
oxazolidinone auxiliary may be recovered after the reduc-
tive cleavage and recycled. Clearly, this represents an
attractive feature of our new asymmetric oxidation
methodology.
Similarly, the epoxidation by DMD and mCPBA
(Scheme 8) entails an alternative highly diastereoselec-
tive route to the same enantiomerically enriched oxy-
functionalized products. All the advantages of the sub-
strate-controlled selectivities for the photooxygenation
are inherent in the epoxidation process, which should
establish itself as an attractive tool for the synthesis of
asymmetric oxyfunctionalized compounds.
oxazolidinone substituent to avoid the steric hindrance
by the R1 substituent.
Th e Ster ic Effects of th e R1 Su bstitu en t on th e
Oxa zolid in on e Au xilia r y. The favored conformation of
the enecarbamates 1 in solution positions the R1 sub-
stituent of the chiral auxiliary to one π face of the double
bond, thereby it sterically shields the incoming oxidant
(1O2, DMD, and mCPBA) and causes a highly diastereo-
selective attack from the opposite side. The size of R1
strongly influences the extent of the shielding and
thereby the diastereoselectivity of the ene reaction.
Although for both the [2+2] cycloaddition and the ene
reaction a perepoxy-like transition state is traversed, the
allylic hydrogen abstraction in the ene reaction occurs
far away from the R1 substituent and the steric interac-
tions are not as effective as in the case of the [2+2]
cycloaddition, in which the tailing oxygen atom must
come near the R1 substituent to close the dioxetane ring.
Thus, the control of the diastereoselectivity in the singlet-
oxygen ene reaction is lower than that of the [2+2]
cycloaddition. The size of the R1 substituent also influ-
ences the diastereoselectivity in the epoxidation reac-
tions, in which a sterically demanding substituent such
as the bulky tert-butyl group may completely cover one
face of the double bond and encumber the attack of the
oxidant from that direction.
Mech a n istic Syn op sis
The present mechanistic work on the diastereoselec-
tivity and mode selectivity in the photooxygenation (1O2)
and epoxidation (DMD, mCPBA) of the enecarbamates
1 has revealed that the characteristic structural and
electronic features of these substrates are decisively
responsible in steering the attack of the oxidant. The
structural characteristics encompass the R1 substituent
in the oxazolidinone chiral auxiliary, the E,Z configura-
tion of the double bond, and the stereogenic substituent
on that functionality, while the electronic factor relates
to the enamine-type double bond. The structural aspects
provide the necessary steric interactions toward the
incoming oxidant through proper conformational align-
ment, and the electronic feature is expressed in terms of
the stereoelectronic control through the enamine-type
nitrogen functionality. The salient consequences of these
factors on the selectivities (mode and diastereo) of the
photooxygenation (1O2) and epoxidation (DMD and mCP-
BA) will now be highlighted, first for the stereoelectronic
and subsequently for the steric effects.
Th e Ster ic Effects of th e P h en yleth yl Su bstitu en t
on th e CC Dou ble Bon d . This bulky, chiral phenylethyl
substituent as the only stereogenic center in the enecar-
bamate may control completely the diastereoselectivity
of the DMD epoxidation, whereas it exercises no influence
in the singlet-oxygen [2+2] cycloaddition and ene reac-
tion. Through the vinylic-nitrogen steering effect, the
phenylethyl group is not on the trajectory of the incoming
1O2. DMD is sterically significantly more demanding than
1O2 and encounters severe hindrance by the phenylethyl
group such that substantial diastereoselectivity is ob-
tained during the epoxidation.
Th e Ster eoelectr on ic Effect of th e Vin ylic Nitr o-
gen F u n ction a lity. The observed E,Z-dependent mode
1714 J . Org. Chem., Vol. 69, No. 5, 2004

Enecarbamates as Selective Substrates in Oxidations
Th e Com p osite Ster ic Effects of th e R1 a n d th e
P h en ylet h yl Su b st it u en t s. The steric interactions
between two stereogenic centers, namely the phenylethyl
substituent at the CC double bond and the R1 group of
the oxazolidinone auxiliary, may influence the diaste-
reoselectivity of the oxidation, a phenomenon known as
the “match-mismatch” effect. For the epoxidation, the
R1 group (e.g. isopropyl) alone causes little diastereofacial
differentiation, but a high diastereoselectivity is imposed
by the phenylethyl group alone; for the photooxygenation,
the opposite applies. On one hand, are both stereogenic
centers present, when they occupy different half-spaces
above and below the CC double bond, they do not
sterically interact with one another and the phenylethyl
group controls completely the diastereoselectivity. On the
other hand, when the two substituents occupy the same
half-space, conformational changes are imposed by their
mutual steric interactions and the diastereoselective
control is severely reduced. These effects play only a role
in the epoxidation and none in the photooxygenation
E-configured isomers because of conformational align-
ment of the double bond substituents, which is reflected
in the higher diastereoselectivity of the Z versus the E
1
isomers with DMD and mCPBA, but not for O2. For the
latter, the stereoelectronic steering by the vinylic nitro-
gen functionality overcomes the influence of the E/Z
configurations in the case of the ene reaction. For the
[2+2] cycloaddition of 1O2, no conclusions may be drawn
because the dioxetanes could only be detected for the Z
diastereomers (diastereoselectivity >95:5).
Ack n ow led gm en t. This work was generously sup-
ported by the Deutsche Forschungsgemeinschaft
(Schwerpunktprogramm “Peroxidchemie”) and the Fonds
der Chemischen Industrie. We thank M. Schwarm
(Degussa Hu¨ls AG, Hanau) for a generous gift of the
optically active oxazolidinones and alaninol, J . Bialas
for the synthesis of DMD, and H. Ihmels for helpful
discussion, and K. Wissel for valuable assistance.
1
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails and X-ray crystallographic data for (Z,4S)-1k . This
material is available free of charge via the Internet at
http://pubs.acs.org.
because, as already stated, O2 is not subject to diaste-
reoselective control by the phenylethyl group.
Th e Con for m a tion a l Effects of th e E/Z Con figu -
r a tion s. For the Z-configured enecarbamates the shield-
ing is substantially better than that for the corresponding
J O035745C
J . Org. Chem, Vol. 69, No. 5, 2004 1715