Stereochemical Features of the Physical and Chemical Interactions of Singlet Oxygen with Enecarbamates

Article abstract of DOI:10.1021/ol035856y

(Equation Presented) Oxazolidinone-substituted enecarbamates represent a mechanistically rich system for the study of stereoelectronic, steric, and conformational effects on stereoselectivity and mode selectivity in ¹O₂ [2 + 2] and ene reactions. Photooxygenation of these enecarbamates with ¹O₂ leads to diastereomerically pure dioxetanes that decompose to yield an oxazolidinone carbaldehyde and one of the two enantiomers of methyldesoxybenzoin in enantiomeric excess. Stereoselectivity originates at the allylic stereocenter, a result supported by quenching studies, computational analysis, and deuterium solvent isotope effects.

Full text of DOI:10.1021/ol035856y

ORGANIC
LETTERS
2003
Vol. 5, No. 26
4951-4953
Stereochemical Features of the Physical
and Chemical Interactions of Singlet
Oxygen with Enecarbamates
,‡
Thomas Poon,† Nicholas J. Turro,* Jessica Chapman,† P. Lakshminarasimhan,‡
Xuegong Lei,‡ Steffen Jockusch,‡ Roberto Franz,‡ Ilyas Washington,‡
Waldemar Adam,§ and Sara G. Bosio§
Department of Chemistry and Department of Chemical Engineering,
Columbia UniVersity, 3000 Broadway, Mail Code 3119, New York, New York 10027,
Joint Science Department, W.M. Keck Science Center, 925 N. Mills AVe.,
Claremont McKenna, Pitzer, and Scripps Colleges, Claremont, California 91711, and
Institut fu¨r Organische Chemie, UniVersita¨t Wu¨rzburg, 97074 Wu¨rzburg, Germany
turro@chem.columbia.edu
Received September 24, 2003
ABSTRACT
Oxazolidinone-substituted enecarbamates represent a mechanistically rich system for the study of stereoelectronic, steric, and conformational
effects on stereoselectivity and mode selectivity in 1O [2 + 2] and ene reactions. Photooxygenation of these enecarbamates with 1O leads
2
2
to diastereomerically pure dioxetanes that decompose to yield an oxazolidinone carbaldehyde and one of the two enantiomers of
methyldesoxybenzoin in enantiomeric excess. Stereoselectivity originates at the allylic stereocenter, a result supported by quenching studies,
computational analysis, and deuterium solvent isotope effects.
1
Reaction1 of the enecarbamate 1 with singlet oxygen (1O2)
results in the formation of dioxetane 2 with >95% dia-
stereoselectivity (Scheme 1). The dioxetane 2 is derived from
attack of 1O2 from the face anti to the isopropyl substituent
at the oxazolidinone stereocenter-1 (C-1). This diastereo-
selectivity is consistent with classical steric π-facial shielding
of attack of 1O2 by the isopropyl group at C-1. Enecarbamate
1 also possesses a second stereogenic center at C-5. Since
C-1 and C-5 are both in proximity of the reactive CdC
1 with O2 established that an alkyl substitutent at C-1
produces a high diastereoselectivity of dioxetane formation.
However, because the reactions were run to 100% comple-
tion, the results were silent as to the role of the influence of
the stereochemistry at C-5 on the stereoselectivity of di-
oxetane formation and in particular the possible enantio-
selectiVity of formation of MDB at partial conVersion
(Scheme 1).
1
double bond being attacked by O2, both C-1 and C-5 are
expected to influence the stereochemical outcome of di-
oxetane formation. The results of the reported reactions1 of
Scheme 1
* Phone: (+1) 212-854-2175. Fax: (+1) 212-932-1289.
† Claremont McKenna, Pitzer, and Scripps Colleges.
‡ Columbia University.
§ Universita¨t Wu¨rzburg.
(1) (a) Adam, W.; Bosio, S. G.; Turro, N. J. J. Am. Chem. Soc. 2002,
124, 8814. (b) Adam, W.; Bosio, S. G.; Turro, N. J. J. Am. Chem. Soc.
2002, 124, 14004.
10.1021/ol035856y CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/22/2003

Scheme 2
To elucidate the mechanistic details of the reaction of the
1
enecarbamate 1 and O2 and to determine the role of the
1
stereocenter at C-5 on the stereoselectivity of attack of O2
on the double bond of 1, we have investigated (1) the rate
1
constant for quenching of O2 by two diastereomers of 1
(1a and 1b), by the oxazolidinone 3, and by the enecarbamate
1
4; (2) the efficiency of chemical reaction of O2 with the
two diastereomers of 1 (1a and 1b); and (3) the enantiomeric
excess (ee) of the ketone MDB formed from decomposition
of the dioxetane after reaction of 1 with 1O2 at low conversion
(Schemes 2 and 3).
Figure 1. Stern-Volmer plot to determine the total quenching
rate constants, k(q+r), where I0 is the intensity of the chemilumi-
nescence of 1O2 at 1270 nm in the absence of quencher and I is the
intensity in the presence of quencher 1, 3, 4, or 8. For experimental
details, see ref 3b.
Due to the longer lifetime of 1O2 in CDCl3 (7 ms)5 compared
to CHCl3 (0.23 ms),5 the conversions are expected, and
found, to be higher in CDCl3 (30%) relative to CHCl3 (15%),
for the same amount of endoperoxide decomposition. Also
as expected, the increase in conversion was accompanied
by a lower ee of MDB (30% ee in CHCl3 compared to 10%
Scheme 3
1
ee in CDCl3). Similar results were found when O2 was
generated with 7 as a photosensitizer. The oxazolidinone 3,
which does not possess a reactive CdC bond, was found to
1a and 1b react with 1O2 to form the dioxetane 2, arising
from stereoselective attack anti to the isopropyl group at C-1.1
Subsequent decomposition of the dioxetanes at complete
conversion results in the formation of nearly equal yields of
(R)-MDB and (S)-MDB.2 We report that after low (<30%)
conversions of 1a and 1b, followed by decomposition of
the dioxetane, 2, the MDB produced is significantly en-
riched (ee ca. 33%) in (S)-MDB or (R)-MDB, respec-
1
be a poorer quencher (Figure 1) of O2 (0.34 × 105 M-1
s-1), and the unsubstituted enecarbamate 4 was found to be
a better quencher (1.7 × 105 M-1 s-1) than the isopropyl-
substituted enecarbamate, 1 (1.0 × 105 M-1 s-1).
1
The quenching of O2 by any substrate can proceed via
two pathways: (1) a chemically productive “reactive”
pathway with a rate constant, kr, and (2) a chemically
unproductive, net physical “quenching” pathway with a rate
constant, kq. The relative efficiency of the reaction of 1 with
1O2 was determined by comparison of the chemical reaction
1
tively (Scheme 2). At low conversion, the reaction of O2
with the enecarbamate 4 (which possesses no stereocenter
at position 1) results in only a slight ee (ca 4%) in the MDB
produced by decomposition of the precursor dioxetanes
(Scheme 3).
The rates of quenching of 1O2 by 1a and 1b were
determined by monitoring the quenching of 1O2 phos-
phorescence at 1270 nm.3 1O2 was produced by both the
thermal decomposition of 1,4-dimethylnaphthalene endo-
peroxide (6)4 or by photosensitization with 5,10,15,20-
tetraphenyl-21H,23H-porphine (7).5 Stern-Volmer analysis
(Figure 1) of the data yielded an indistinguishable quenching
rate constant of 1.0 ( 0.2 × 105 M-1 s-1 for 1a and 1b.
1
of O2 with 1 and with 1-methylcyclohexene (8)6 as a
standard. The rate of disappearance of 8 in CCl4 was ca. 10
times faster than the rate of disappearance of 1 under
comparable conditions (Supporting Information). Since the
1
maximum efficiency of reaction of 8 with O2 is 1.0, the
1
maximum efficiency of reaction of 1 with O2 is 0.1. This
result demonstrates clearly that the major path of interaction
1
of 1 with O2 is a quenching mode rather than a reactiVe
mode.
(3) (a) Khan, A. U.; Kasha, M. Proc. Natl. Acad. Sci. U.S.A. 1979, 76,
6047. (b) Ben-Shabat, S.; Itagaki, Y.; Jockusch, S.; Sparrow, J. R.;
Nakanishi, K.; Turro, N. J. Angew Chem., Int. Ed. 2002, 41, 814.
(4) (a) Wasserman, H. H.; Larese, D. L. Chem. Commun. 1972, 253. (b)
Adam, W.; Prein, M. Acc. Chem. Res. 1996, 29, 275.
(5) Schmidt, R.; Afshari, E. Ber. Bunsen-Ges. Phys. Chem. 1992, 96,
788.
(2) (a) That the relative yields are not exactly 50:50 is attributed to the
observed small excess of one of the diastereomers of 1 after purification
by column chromatography. (b) An authentic sample of (S)-MDB was
synthesized using a literature procedure (McKenzie, A.; Roger, R.; Wills,
G. O. J. Chem. Soc. 1926, 779). Thus, the stereochemical assignments in
Scheme 2 are absolute.
(6) Monroe, B. M. J. Am. Chem. Soc. 1981, 103, 7253.
4952
Org. Lett., Vol. 5, No. 26, 2003

The substituent at C-1 of 1b controls1 the facial direction
of reactiVe O2 attack, which occurs exclusively from the
The experimental observation (Scheme 2) is that (1R,5R)-
1b reacts with O2 faster than (1R,5S)-1b. Thus, we can
1
1
face opposite to the isopropyl group. Thus, the mechanistic
analysis needs to consider only the influence of C-5 on the
trajectories of 1O2 toward the double bond from the top face.
In Figure 2, the lowest energy conformations at C-5 for (1R,
5R)- and (1R, 5S)-1b can be used as a working structural
basis to analyze the enantiomeric selectivity of reaction with
1O2.
conclude that although E and F have nearly identical energies
computationally, to be consistent with the experimental
results, F probably exists as the major conformer at equi-
librium.
In conclusion, the reaction of 1O2 with enecarbamates has
proven to be rich in testable aspects of stereoselectivity at
the two stereogenic centers of these molecules. The results
1
are remarkable in that O2, a sterically undemanding and
simple diatomic molecule, is still subject to significant
stereochemical effects on its chemical reactivity and quench-
ing, as well as its diastereomeric and enantiomeric selectivity.
Our results demonstrate that the isopropyl group at C-1
completely controls the diastereofacial attack on the double
bond of 1, and the absolute configuration at C-5 exerts a
significant control on the enantioselectivity of attack from
the face opposite to the isopropyl group. It has been well
1
established that O2 is quenched by X-H vibronic interac-
tions.9 It is possible that the remarkable stereoselectivity is
due to vibronic physical quenching of the sterically blocking
group rather than, or in addition to, classic steric effects.
This possibility can be tested by replacing the hydrogens of
the isopropyl group at C-1 and the hydrogens of the methyl
group at C-5 with deuteria.
Figure 2. Isopropyl group shields the bottom face of the alkene
in 1b. A-C are the lowest energy conformations for the (1R,5R)-
isomer. D-F are the lowest energy conformations for the (1R,5S)-
isomer. Energies were computed at the PM3 level and are in kcal/
mol.
Acknowledgment. The authors at Columbia thank the
NSF (Grant CHE-01-10655) for generous support of this
research. W.A. and S.G.B. acknowledge the generous
financial support by the Deutsche Forschungsgemeinschaft
(Schwerpunktprogramm “Peroxidchemie: Mechanistische
und pra¨parative Aspekte des Sauerstofftransfers”). T.P.
thanks The W. M. Keck Foundation and Claremont
McKenna, Pitzer, and Scripps Colleges for research funding.
J.C. thanks the NSF and Columbia University for REU
support.
The energies of the three lowest energy conformations
shown in Figure 2 at C-5 for 1a and 1b were calculated at
the PM3 level of theory.7 For the (1R,5R)-1b conformations,
the lowest energy conformation is C where the hydrogen
eclipses the double bond. This conformation places the
methyl group over the top face of the alkene. For the (1R,5R)-
1b conformations, the two lowest energy conformations are
(F) with the hydrogen and (E) with the methyl eclipsed to
the double bond. Both E and F were found to be of the same
energy by both PM3 and MM28 (not shown) computations.
In F, the phenyl group is positioned over the top face of the
Supporting Information Available: Experimental details
and a comparison of the efficiency of reaction of 1 and 8
1
with O2. This material is available free of charge via the
Internet at http://pubs.acs.org.
1
alkene, clearly providing a steric barrier to attack by O2.
OL035856Y
(7) As implemented in SPARTAN version 5.0 (Wavefunction, Inc.: Irvine,
CA).
(8) As implemented in Macromodel: Mohamadi, F.; Richards, N. G. J.;
Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.;
Hendrickson, T.; Still, W. C. J. Comput. Chem. 1990, 11, 440.
(9) (a) Merkel, P. B.; Nilsson, R.; Kearns, D. R. J. Am. Chem. Soc. 1972,
94, 1030. (b) Hurst, J. R.; Schuster, G. B. J. Am. Chem. Soc. 1983, 105,
5756. (c) Salokhiddinov, K. I.; Byteva, I. M. Opt. Spectrosc. (USSR) 1983,
55, 282. (d) Rodgers, M. A. J. J. Am. Chem. Soc. 1983, 105, 6201. (e)
Ogilby, P. R.; Foote, C. S. J. Am. Chem. Soc. 1983, 105, 3423.
Org. Lett., Vol. 5, No. 26, 2003
4953