Oxaziridinium salts as hydrophobic epoxidation reagents: Remarkable hydrophobically-directed selectivity in olefin epoxidation

Article abstract of DOI:10.1021/ja053660i

Selective epoxidation of cinnamates versus crotonate was used to detect hydrophobic binding of the cinnamates in the transition states with hydrophobic oxidizing agents in water solution. With peracids as oxidants, no such effect was seen, in accord with

Full text of DOI:10.1021/ja053660i

Published on Web 07/16/2005
Oxaziridinium Salts as Hydrophobic Epoxidation Reagents: Remarkable
Hydrophobically-Directed Selectivity in Olefin Epoxidation
Mark R. Biscoe and Ronald Breslow*
Department of Chemistry, Columbia UniVersity, New York, New York 10027
Received June 3, 2005; E-mail: rb33@columbia.edu
Recently we reported that use of the hydrophobic effect in the
reductions of ketones permits a dramatic increase in selectivity for
ketones located in a hydrophobic environment when a hydrophobic
borohydride is employed and the reaction medium is changed from
organic to aqueous.1 The face-face packing of the hydrophobic
units on the borohydride and the ketone in the transition state of
reduction minimizes the water-accessible hydrocarbon surface area,
leading to the observed relative rate acceleration.1b,2
Figure 1. Transition-state models of the epoxidation of styrene with (a)
perbenzoic acid and (b) the dioxirane derived from acetophenone.
We hoped that hydrophobic oxidizing reagents could also be
developed. Herein we report the ability of hydrophobic oxaziri-
dinium salts to effect relative rate increases in the selective
epoxidation of hydrophobic olefins that are an order of magnitude
greater than the largest increases observed in hydride reductions.1
Using substoichiometric hydrophobic iminium salts and oxone as
a co-oxidant, we further show that these selective epoxidations can
be made catalytic. In addition, we demonstrate the importance of
accurate transition-state models in the development of hydrophobic
reagents.
Table 1. Ratio of Epoxide Products (1:2) Formed in the
Competition Reactions of Crotonic Acid and Derivatives of
Cinnamic Acids with Peracids of Different Hydrophobicity (∆∆Gq
(kcal/mol) for Each Reaction Is in Parentheses)a,b
Ar
MMPP
PFPP
PAA
We considered perbenzoic acids, aryl ketone-derived dioxiranes,
and aryl oxaziridinium salts. Each class has a hydrophobic aryl
group theoretically capable of packing with a hydrophobic unit on
a substrate. However, computer models for perbenzoic acid
oxidation predict a spiro, highly synchronous transition state where
the hydrophobic phenyl ring assumes a position from which it is
incapable of packing on the hydrophobic unit of the olefin (Figure
1).3 By contrast, the calculations predict a synchronous transition
state for dioxirane or oxaziridinium cation epoxidation in which a
hydrophobic group would be located in a position conducive to
packing with the hydrophobic unit of the olefin during epoxidation
(Figure 1).3,4
On the basis of the model of the transition state for perbenzoic
acid epoxidation shown in Figure 1, no change in selectivity for
the epoxidation of the hydrophobic olefin should be observed when
the hydrophobicity of the peracid is varied. To test this, peracids
magnesium monoperoxyphthalate (MMPP), sodium perfluoromono-
peroxyphthalate (PFPP), and peracetic acid (PAA) were employed
Ph
44:56 (0.16)
16:84 (0.98)
47:53 (0.07)
93.3:6.7 (-1.56)
44:56 (0.16)
18:82 (0.88)
50:50 (0.00)
92.9:7.1 (-1.53)
75:25 (-0.65)
35:65 (0.37)
72:28 (-0.56)
93.2:6.8 (-1.55)
p-CF3Ph
2-naph
3
a Competition reactions were carried to ca. 5% epoxidation of the olefins
at room temperature and analyzed directly by HNMR (substrate concentra-
tions ranged from 5 to 300 mM; see Supporting Information for specific
concentrations). b Integer ratios are within an error (1%; other ratios are
within an error of (0.2% in at least duplicate runs.
When non-hydrophobic PAA is used, selectivity for all hydro-
phobic olefins except fused 3 actually increases, which is contrary
to expectations if hydrophobic direction were involved. This
increase in selectivity results from the low steric bulk of PAA. When
MMPP and PFPP are employed, repulsion likely exists between
the rotating hydrophobic unit of the olefin and the aryl ring of the
peracid. This repulsion is removed when nonbulky PAA is used or
when rotation of the hydrophobic unit of the olefin is prevented
through a ring fusion as with 3. Thus, in agreement with models,
there is no facilitating hydrophobic interaction in the transition states
of these peracid epoxidations.
The results were very different in epoxidations with isolable
oxaziridinium reagents,6 selected for evaluation as likely hydro-
phobically accelerated epoxidation reagents. Previously reported
oxaziridiniums 6 and 7 were employed in these experiments. The
in competition reactions of derivatives of cinnamic acid with
crotonic acid in NaHCO3/D2O. MMPP contains a hydrophobic
phenyl ring, PFPP contains a significantly more hydrophobic5
perfluorinated phenyl ring, and PAA contains no hydrophobic
substituent. The results in Table 1 show that there is no preference
for the hydrophobic olefin to form product 1 when MMPP or PFPP
are used relative to simple peracetic acid, PAA.
rigid hydrophobic backbone structure and the orientation of the
oxygen atom in a pseudoaxial position6c suggested that a significant
9
10812
J. AM. CHEM. SOC. 2005, 127, 10812-10813
10.1021/ja053660i CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Ratio of Epoxide Products (4:5) Formed in the
Competition Reactions of Crotonic Acid and Derivatives of
Cinnamic Acids with Dioxirane and Oxaziridinium Epoxidizing
Reagents under Different Conditions (∆∆Gq (kcal/mol) for Each
Reaction Is in Parentheses)a,b
Ar
solvent
DMDO
oxaz. 6
oxaz. 7
Ph
Ph
D2O
61:39
(-0.27)
59:41
(-0.22)
22:78
(0.75)
22:78
(0.75)
68:32
(-0.45)
66:32
(-0.39)
90.5:9.5
(-1.34)
90:10
96.5:3.5
(-1.97)
88:12
(-1.18)
84:16
98.1:1.9
(-2.34)
91.8:8.2
(-1.43)
92.7:7.3
(-1.51)
69:31
(-0.47)
99.8:0.2
(-3.68)
98.1:1.9
(-2.34)
99.9+:0.1
(-4.10)
99.4:0.6
(-3.03)
1:1 (v:v) iPrOD:
D2O
D2O
p-CF3Ph
p-CF3Ph
2-naph
2-naph
3
(-0.98)
54:46
Figure 2. Catalytic cycle using hydrophobic iminium salts.
1:1(v:v) iPrOD:
D2O
D2O
(-0.10)
98.7:1.3
(-2.57)
95.9:4.1
(-1.87)
99.7:0.3
(-3.44)
98.9:1.1
(-2.67)
The oxaziridinium epoxidations have been made catalytic through
the use of the associated iminium salts with oxone as a stoichio-
metric co-oxidant (Figure 2).8 These iminium-catalyzed epoxida-
tions successfully reproduce the selectivities observed in the
noncatalytic epoxidations of Table 2. Five to ten turnovers of the
iminium catalyst are consistently achieved in these reactions. The
turnover number is dependent upon the pH at which the reactions
are conducted, as decomposition of the oxaziridium to an isoquino-
linium species occurs more rapidly in more alkaline conditions (see
Supporting Information). While oxone can act as an epoxidizing
agent when not in the presence of a catalyst, this background process
is very slow compared to the rate of hydrophobic epoxidation.
We have shown previously that hydrophobically induced sub-
strate selectivity in borohydride reductions can be translated into
regioselective reactions in a diketone.1 Thus, we expect that our
selective oxygen-atom transfers will also be regioselective, and
perhaps stereoselective, in polyenes.
1:1(v:v) iPrOD:
D2O
D2O
3
1:1 (v:v) iPrOD:
D2O
(-1.30)
a Competition reactions were carried to ca. 5% epoxidation of the olefins
at room temperature in the solvent with NaHCO3 and analyzed directly by
HNMR (substrate concentrations ranged from 5 to 300 mM; see Supporting
Information for specific concentrations). b Integer ratios are within an error
(1%; other ratios are within an error of (0.2% in at least duplicate runs.
hydrophobic binding interaction was likely in the transition state
of epoxidation. Because dioxirane epoxidation exhibits a nearly
identical transition-state geometry to that for oxaziridinium epoxi-
dation,3,4 we considered dimethyl dioxirane (DMDO) a reasonable
non-hydrophobic control for the hydrophobic oxaziridiniums.
The results of the competition reactions with oxaziridium salts
6 and 7 and DMDO (Table 2) show a very large increase in
selectivity for the hydrophobic olefin when 6 or 7 is used in place
of DMDO in an aqueous medium. This selectivity increases directly
with the hydrophobicity of the olefin. In the competition reaction
of a 2-naphthyl olefin and a methyl olefin, a change in ∆∆Gq of
3.23 kcal favoring epoxidation of the 2-naphthyl olefin was
observed (but minimal chirality) when oxaziridinium 7 was
employed in place of DMDO. This corresponds to a 240-fold
relative rate increase for epoxidation of the 2-naphthyl olefin, the
largest change in selectivity we have observed in a hydrophobically
accelerated atom-transfer reaction.
When a 1:1 ratio of D2O:iPrOD was used as the solvent for the
competition reactions, little or no change is observed for reactions
involving DMDO. By contrast, with hydrophobic oxaziridinium
reagents, ca. 1 kcal of binding energy is consistently lost (Table 2)
as a result of the alcohol-induced damping of the hydrophobic effect,
and the rates of reactions are much slower. Even so, significant
selectivity is still present, indicating that a completely aqueous
solution is not required. Thus for water insoluble substrates our
demonstrated selectivities in mixed solvents, reflecting more than
just hydrophobic binding forces, could be useful.
Acknowledgment. We thank the NSF and NIH for financial
support of this work.
Supporting Information Available: NMR and mass spectra,
procedures for the synthesis of previously unreported compounds, and
procedures for competition experiments. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Biscoe, M. R.; Breslow, R. J. Am. Chem. Soc. 2003, 125, 12718-
12719. (b) Biscoe, M. R.; Uyeda, C.; Breslow, R. Org. Lett. 2004, 6,
4331-4334.
(2) (a) Breslow, R. Acc. Chem. Res. 1991, 24, 159-164. (b) Blokzijl, W,;
Engberts, J. Angew. Chem., Int. Ed. Engl. 1993, 32, 1545-1579. (c) Otto,
S.; Engberts, J. Org. Biomol. Chem. 2003, 1, 2809-2820.
(3) Houk, K. N.; Liu, J.; DeMello, N. C.; Condroski, K. R. J. Am. Chem.
Soc. 1997, 119, 10147-10152.
(4) Washington, I.; Houk, K. N. J. Am. Chem. Soc. 2000, 122, 2948-2949.
(5) Sheppard, W. A.; Sheetz, C. M. Organic Fluorine Chemistry; W. A.
Benjamin, Inc.: New York, 1969.
(6) (a) Milliet, P.; Picot, A.; Lusinchi, X. Tetrahedron Lett. 1976, 19, 1573-
1576. (b) Milliet, P.; Picot, A.; Lusinchi, X. Tetrahedron 1981, 37, 4201-
4208. (c) Bohe, L.; Hanquet, G.; Lusinchi, M.; Lusinchi, X. Tetrahedron
Lett. 1993, 34, 7271-7274.
(7) Hanquet, G.; Lusinchi, X.; Milliet, P. Tetrahedron 1993, 49, 423-438.
(8) (a) Aggarwal, V. K.; Wang, M. F. Chem. Commun. 1996, 191-192. (b)
Page, P. C. B.; Rassias, G. A.; Bethell, D.; Schilling, M. B. J. Org. Chem.
1998, 63, 2774-2777. (c) Minakata, S.; Takemiya, A.; Nakamura, K.;
Ryu, I.; Komatsu, M. Synlett 2000, 12, 1810-1812.
Extensive oxidation of substrate carboxylate groups (but not un-
ionized carboxylic groups) compete with olefin epoxidation in the
mixed solvent, while very little is observed in D2O.7
JA053660I
9
J. AM. CHEM. SOC. VOL. 127, NO. 31, 2005 10813