Carbodiimide-Promoted Olefin Epoxidation with Aqueous Hydrogen Peroxide

Article abstract of DOI:10.1021/jo972026n

Commercially available carbodiimides in hydroxylic solvents containing hydrogen peroxide with mildly basic or acidic catalysts have been found to promote the epoxidation of olefins. A commercially available 30% aqueous solution of hydrogen peroxide serves as the oxidant for this process. The presumed reactive species is a peroxyisourea generated in situ by the addition of hydrogen peroxide to the carbodiimide.

Full text of DOI:10.1021/jo972026n

2564
J . Org. Chem. 1998, 63, 2564-2573
Ca r bod iim id e-P r om oted Olefin Ep oxid a tion w ith Aqu eou s
Hyd r ogen P er oxid e†,‡
George Majetich,* Rodgers Hicks, Guang-ri Sun, and Patrick McGill
Department of Chemistry, The University of Georgia, Athens, Georgia 30602
Received November 4, 1997
Commercially available carbodiimides in hydroxylic solvents containing hydrogen peroxide with
mildly basic or acidic catalysts have been found to promote the epoxidation of olefins.
A
commercially available 30% aqueous solution of hydrogen peroxide serves as the oxidant for this
process. The presumed reactive species is a peroxyisourea generated in situ by the addition of
hydrogen peroxide to the carbodiimide.
In tr od u ction a n d Ba ck gr ou n d
with formamide,12 nitriles (Payne oxidation),13 or Vils-
meier reagents14 to generate oxidants in situ.
The epoxidation of olefins is one of the most widely
studied reactions in organic chemistry.1 Peroxyacids, the
most common epoxidation reagents, are most often
prepared by treating the corresponding acid with anhy-
drous hydrogen peroxide.2 Unfortunately, anhydrous
hydrogen peroxide is not readily available except as part
of a stable complex.3 This, along with the diminishing
availability of high quality m-CPBA, has stimulated the
search for new epoxidizing agents.4
A dilute aqueous solution of H2O2 is the ideal oxygen
source for olefin epoxidation: it is safe, readily available,
and generates water as the sole byproduct.5 Only a few
electron-rich olefins undergo epoxidation with H2O2
alone; either the presence of a polarized multiple bond
or the addition of a promoter (such as a pertungstate salt6
or a metalloporphyrin7) is usually necessary. The epoxi-
dation of olefins using H2O2 with selenic acids,8 hexafluo-
roacetone,9 organosilane derivatives,10 and alkoxysul-
furans11 has been reported. Hydrogen peroxide reacts
Since the transfer of an oxygen atom from a molecule
of H2O2 to an olefin is a dehydration, numerous dehy-
drating agents have been used in epoxidation systems.15-19
In theory, the nucleophilic addition of hydroperoxide
anion to the electrophilic carbon of a carbodiimide, a
common dehydrating agent, would generate a peroxyi-
sourea that is isoelectronic with a peroxyacid; the trans-
fer of the electrophilic oxygen atom from this reactive
intermediate to an olefin would produce an epoxide and
a urea byproduct (Scheme 1).
Sch em e 1
† A patent application [U.S. 08/813039] has been filed for this
epoxidation procedure (March 6, 1997).
‡ Dedicated with appreciation and best wishes to Professors N. L.
Allinger and R. K. Hill on the occasion of their 70th birthdays.
(1) Rao, A. S. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Ley, S. V., Eds.; Pergamon: Oxford, 1991; Vol. 7, pp 357-
436.
(2) Fieser, M.; Fieser, L. F. Reagents for Organic Synthesis
Wiley-Interscience: New York, 1967; Vol. 1, p 135.
(3) (a) Heaney, H. Aldrichimica Acta 1993, 26, 35. (b) Heaney, H.
Topics Current Chem. 1993, 164, 1. (c) Cooper, M. S.; Heaney, H.;
Newbold, A.; Sanderson, W. Synlett 1990, 533 and references therein.
(d) Muzart, J . Synthesis 1995, 1325. (e) McKillop, A.; Sanderson, W.
R. Tetrahedron 1995, 51, 6145.
However, previous investigations found that carbodi-
imides are not effective as epoxidation promoters. Hamil-
ton and co-workers used nearly anhydrous H2O2 (98%),
acetic acid, and diisopropylcarbodiimide (DIC) in ethyl
acetate to epoxidize phenanthrene and pyrene with <3%
(4) For recent epoxidation procedures, see: (a) Murray, R.; Singh,
M.; Williams, B.; Moncrieff, H. J . Org. Chem. 1996, 61, 1830. (b) Yang,
D.; Yip, Y.; Tang, M.; Wong, M.; Zheng, J .; Cheung, K. J . Am. Chem.
Soc. 1996, 118, 491. (c) Zhang, W.; Loebach, J .; Wilson, S.; J acobsen,
E. J . Am. Chem. Soc. 1990, 112, 2801. (d) J acobsen, E.; Zhang, W.;
Muci, A.; Ecker, J .; Deng, L. J . Am. Chem. Soc. 1991, 113, 7063. (e)
Zhang, W.; J acobsen, E. J . Org. Chem. 1991, 56, 2296. (f) Sato, K.;
Aoki, M.; Oagawa, M.; Hashimoto, T.; Noyori, R. J . Org. Chem. 1996,
61, 8310. (g) Sharpless, K. B.; Rudollph, J .; Reddy, K. L.; Chiang, J . P.
J . Am. Chem. Soc. 1997, 119, 6189. (h) Wang, Z.-X.; Tu, Y.; Frohn,
M.; Zhang, J .-R.; Shi, Y. Ibid. 1997, 119, 11224.
(5) Strukul, G., Ed. Catalytic Oxidation with Hydrogen Peroxide as
Oxidant; Kluwer: Dordrecht, Netherlands, 1992.
(6) (a) Venturello, C.; Abneri, M.; Ricci, M. J . Org. Chem. 1983, 48,
3831. (b) Venturello, C.; D′Aloisio, R. J . Org. Chem. 1988, 53, 1553.
(7) Reynaud, J .; Battioni, P.; Bartoli, J .; Mansuy, D. J . Chem. Soc.,
Chem. Commun. 1985, 888.
(10) (a) Ho, T. L. Synth. Commun. 1979, 37. (b) Rebek, J .; McCready,
R. Tetrahedron Lett. 1979, 4337.
(11) Martin, L. D.; Martin, J . C. J . Am. Chem. Soc. 1977, 99, 3511.
(12) Chen, Y.; Reymond, J . Tetrahedron Lett. 1995, 36, 4015.
(13) (a) Payne, G.; Deming, P.; Williams, P. J . Org. Chem. 1961,
26, 659. (b) Payne, G. Tetrahedron 1962, 18, 763.
(14) (a) Dulcere, J .-P.; Rodriguez, J . Tetrahedron Lett. 1982, 23,
1887. (b) Rodriguez, J .; Dulcere, J .-P. J . Org. Chem. 1991, 56, 469.
(15) (a) Rebek, J .; McCready, R.; Wolf, S.; Mossman, A. J . Org.
Chem. 1979, 44, 1485. (b) Rebek, J .; Wolf, S.; Mossman, A. J . Org.
Chem. 1978, 43, 180.
(16) R. McCready, Ph.D. Dissertation, University of Pittsburgh,
1980.
(17) S. Wolf, Ph.D. Dissertation, University of California, Los
Angeles, 1977.
(18) Krishnan, S.; Kuhn, D.; Hamilton, G. Tetrahedron Lett. 1977,
1369.
(19) Rebek, J .; Wolf, S.; Mossman, A. J . Chem. Soc., Chem. Commun.
1974, 711.
(8) Grieco, P. A.; Yokoyama, Y.; Gilman, S.; Nishizawa, M. J . Org.
Chem. 1977, 42, 2034.
(9) Heggs, R. P.; Ganem, B. J . Am. Chem. Soc. 1979, 101, 2484.
S0022-3263(97)02026-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/01/1998

Carbodiimide-Promoted Olefin Epoxidation
J . Org. Chem., Vol. 63, No. 8, 1998 2565
Sch em e 2
H2O2 prior to workup resulted in the rapid decomposition
of the epoxide. On the other hand, sodium sulfite, which
is less acidic than sodium bisulfite, consumed any
remaining H2O2 without causing problems. Neverthe-
less, this neutralization step is unnecessary for most
oxidations. In a few cases the polarity of the epoxide and
the unreacted DCC are similar, which makes chromato-
graphic separation tedious. Instead, stirring an ethereal
solution of the crude reaction mixture with 5% aqueous
acetic acid for 20 min completely hydrolyzed any unre-
acted DCC without compromising the oxidation yield.
1,3-Cyclooctadiene was used to test the suitability of
procedure A for the selective epoxidation of conjugated
double bonds. Using these conditions, the mono- and bis-
epoxides shown in eq 3 were isolated in 39% and 24%
yield, respectively. To suppress the formation of the bis-
conversion (Scheme 2).18 Rebek and co-workers reported
that the epoxidation of olefins using anhydrous H2O2 and
a carbodiimide failed.19,20 The authors stated that “com-
mercially available dicyclohexylcarbodiimide (DCC) failed
to epoxidize cyclododecene in THF containing H2O2 even
when acidic (HCl or HBF4) or basic (NaHCO3) catalysts
were present”.15a
We discovered that treating styrene with a 10-fold
molar excess of 30% aqueous H2O2 in methanol and 1
equiv of DCC and KHCO3 at room temperature produced
styrene oxide in 75% yield within 2 h (eq 1).21 When the
reaction was run without the DCC, epoxidation occurred
in 54% yield, although complete reaction required eigh-
teen hours.
epoxide, 5 equiv of the diene relative to the carbodiimide
were used. This ratio, referred to here as procedure B,
produced the mono-epoxide in 65% yield as the only
product (eq 4). This alternative procedure, wherein the
carbodiimide is the limiting reagent, is preferred over
procedure A for the epoxidation of alkenes that are
inexpensive relative to the cost of the carbodiimide or for
the monoepoxidation of conjugated dienes.
To rule out direct epoxidation by H2O2, the less
electron-rich substrate methyl 10-undecylenoate was
treated under the same conditions used with styrene (eq
2). Although the epoxidation was incomplete even after
24 h, methyl oxiranenonanoate was isolated in 64% yield.
Repeating this reaction without DCC gave only unreacted
olefin after 24 h.
Other weak inorganic bases such as sodium bicarbon-
ate, sodium carbonate, sodium phosphate (mono- or di-
or tribasic), or their potassium equivalents can be
substituted for potassium bicarbonate in procedures A
or B without any loss of efficiency. Although epoxidation
proceeds using only 0.1 mol equiv of the weak inorganic
base versus carbodiimide, the rate and yield of the
reaction are greatly improved if stoichiometric amounts
of base and carbodiimide are used. Strong organic bases
such as sodium ethoxide react with the nascent epoxide,
thereby destroying it. Triethylamine, a weak organic
base, does not promote epoxidation, because the amine
undergoes rapid oxidation. In contrast, pyridine is an
effective catalyst.
Op tim iza tion of Th e Aqu eou s
H2O2-Ca r bod iim id e Ep oxid a tion P r ocess
Having found conditions that allowed carbodiimides to
function as epoxidation promoters, we sought to optimize
this new process and establish its scope and limitations.
R-Pinene was used as a common substrate for these
studies.
(b) Solven t Stu d ies. Several reaction media were
compared using procedure A and R-pinene. In general,
the solvent system was chosen to dissolve, or at least
partially dissolve, the reactant olefin and not to react
with the carbodiimide, the H2O2, or the product. While
simple alcohols such as methanol or ethanol are the
preferred solvents, 2-propanol or 1-butanol were also
acceptable. Epoxidation occurred in aromatic alcohols,
such as benzyl alcohol or phenol, but the lower volatility
and lower water solubility of these alcohols made the
isolation of the product difficult.
(a ) P r oced u r es A a n d B. Throughout this manu-
script procedure A involves the use of 2 mol equiv of DCC
and KHCO3 relative to the olefin (typically 0.10 to 1.00
mmol) and 1 mL of a 30% aqueous solution of H2O2 (at
least a 5-fold molar excess) in 5 mL of methanol or
ethanol. The dicyclohexyl urea (DCU) byproduct is
sparingly soluble in simple alcohols, making it easy to
remove by filtration. The epoxide is isolated by ethereal
workup. Adding sodium bisulfite to destroy the excess
(20) Rebek, J .; Wolf, S.; Mossman, A. Abstracts of the 172nd
National Meeting of the American Chemical Society, ORGN 109, 1976.
(21) Majetich, G.; Hicks, R. Synlett 1996, 649.
The choice of reaction solvent was always a compro-
mise between the requisite protic character and the

2566 J . Org. Chem., Vol. 63, No. 8, 1998
Majetich et al.
Ta ble 1. Th e Ep oxid a tion of r-P in en e w ith P r oced u r e A
likely. These considerations prompted us to substitute
diisopropylcarbodiimide (DIC), an easily handled liquid
that is less allergenic than DCC, or ethyl(3-(dimethy-
lamino)propyl)carbodiimide (EDC), a water-soluble car-
bodiimide, in the epoxidation. Both DIC and EDC were
as effective dehydrating agents as DCC. However, DIC
is easier to work with which makes it the carbodiimide
of choice for large-scale epoxidations.
in Differ en t Solven t System s
solvent
reaction time % yield epoxide
2-propanol
1-butanol
C6H5CH2OH/CH3OH(1:1)
C6H5OH/CH3OH(1:1)
acetone
DMF
1,4-dioxane
THF
24 h
24 h
2 days
2 days
24 h
24 h
24 h
4 days
24 h
48 h
40%
40%
none
20%
trace
trace
trace
trace
trace
28%
35%
The epoxidation of oleic acid and vinylacetic acid, two
water-soluble substrates, was attempted using the water-
soluble carbodiimide EDC and water as the solvent. In
both cases an insoluble gummy material formed, and
workup of these reactions provided only the unreacted
alkenes.
CH3OH/EtOAc (1:1)
CH3CH2OH/anhydrous H2O2
CH3OH/H2O (1:1)
24 h
lipophilic character needed for substrate solubility. This
led us to examine the feasibility of using water-miscible
cosolvents for the oxidation. Unfortunately, the epoxi-
dations of R-pinene in acetone, DMF, 1,4-dioxane, and
THF were all unsatisfactory (Table 1).
(e) P r oced u r es D a n d E. Seeking an acid-catalyzed
procedure to complement the base-initiated epoxidations,
several acids were examined. Strong acids gave poor
results. For example, using p-TsOH, the DCC-based
epoxidation of 1-dodecene gave 1,2-epoxydodecane in 19%
yield, while HCl (introduced as a 0.12 M solution in
methanol) gave only a 7% conversion. Ion-exchange
resins, such as Amberlite IR-120, also afforded low yields
of 1,2-epoxydodecane (16%). We discovered that Amber-
lite IRC-50, a polymer-bound sulfonic acid, was a practi-
cal catalyst for this process. The use of 2 mol equiv of
DCC relative to the olefin (typically 0.10 to 1.00 mmol)
and 1 mL of a 30% aqueous solution of H2O2 in 5 mL of
methanol and 10 mg of IRC-50 beads is called procedure
D. A second set of acid-catalyzed conditions was estab-
lished in which the carbodiimide is the limiting reagent
and the alkene is in excess (procedure E). As one would
expect, strongly acidic ion-exchange resins promote both
epoxidation and the consumption of the oxidized product.
Since carbodiimide-promoted olefin epoxidations re-
quire solvents having a hydroxyl group, we were curious
whether the presence of water is essential for epoxida-
tion. To answer this question an anhydrous ethereal
solution of H2O2 was prepared22 and combined with
absolute ethanol to give an anhydrous ethanolic solution
2.5 M in H2O2.23 The epoxidation of R-pinene occurred
in 28% yield using this solution as both the solvent and
the H2O2 source. Comparing this result with the 68%
yield obtained for this reaction in wet methanol and the
83% yield obtained in wet ethanol using procedure B,
indicates that while water is not essential for epoxidation,
its presence does increase the yield.
Solubility problems are often encountered when the
water exceeds 50 volume % of the solvent. This was first
observed when R-pinene was epoxidized in a 1:1 mixture
of methanol and water (supplemented by the water
contributed by the 30% aqueous H2O2). Under these
conditions, the reaction mixture formed two immiscible
phases and the yield decreased significantly. In light of
this observation, the best solvent system should be
homogeneous.
Transition metal and lanthanide-based Lewis acids
were unsuitable as catalysts because they accelerated the
decomposition of H2O2 instead of its addition to the
carbodiimide. Other Lewis acids, such as triethylalumi-
num, zinc bromide, cerium chloride, and boron trifluoride
etherate were also unsuitable as catalysts because they
facilitated the rapid conversion of DCC to DCU.
(c) P r oced u r e C. The use of sodium phosphate to
facilitate the addition of the hydroperoxide anion to the
carbodiimide permitted the efficient epoxidation of
R-pinene in a 1:1 mixture of methanol and ethyl acetate;
no epoxidation was observed using this solvent system
with KHCO3. This result was unexpected since the
tribasic phosphate initially formed an insoluble complex
upon addition of aqueous H2O2. Nevertheless, after
stirring the reaction mixture for 12 h at room tempera-
ture, workup afforded an 83% yield of epoxide. Further
studies established that while acetone, 1,4-dioxane, and
THF were useful cosolvents for this oxidation, no oxida-
tion was observed when DMF was used. Moreover,
sodium phosphate (mono- or dibasic), sodium carbonate,
potassium phosphate, or potassium carbonate were ac-
ceptable catalysts. This modification of procedure A is
designated as procedure C.
(f) Stoich iom etr y of th e Oxid a n t. More than a
5-fold excess of H2O2 (relative to the limiting reagent)
was used routinely throughout these studies. The ep-
oxidation of R-pinene was carried out under conditions
A at four different oxidant stoichiometries to determine
the minimum amount of oxidant required. While an
excess of H2O2 is clearly needed, the yield of epoxide only
slightly improved when more than 3 equiv of H2O2 were
used (cf. Chart 1). Even a 3% aqueous solution of H2O2
was effective as the source of oxidantsprovided that at
least 2 equiv of H2O2 were used and that the longer
reaction times required were acceptable.
(g) Alter n a te H2O2 Sou r ces. Stable adducts of H2O2,
such as sodium percarbonate [2Na2CO3-3H2O2],3d,e readily
release H2O2 when exposed to water and can result in a
violent exothermic reaction when used in the presence
of an oxidizable substrate. We speculated that the
decomposition of the adduct would be slower in methanol,
thereby permitting it to serve as both the source of
oxidant and as the base for the carbodiimide-promoted
epoxidation process.24 This conjecture proved to be valid
as summarized in Scheme 3.
(d ) Ch oice of th e Ca r bod iim id e. The waxy solid
DCC is a potent allergen that crumbles and scatters
easily during handling, making inadvertent skin contact
(22) Hudlicky, M. Oxidations in Organic Chemistry; ACS Monograph
186; American Chemical Society; Washington, DC, 1990; p 7. To
prevent the accumulation of peroxides these solutions were prepared
fresh for each use.
(23) (a) Kingzett, C. J . Chem. Soc. 1880, 37, 792. (b) Schimpf, H. A
Manual of Volumetric Analysis, Wiley: New York, 1909, p 214.
(24) Sodium percarbonate in acetic acid is not useful for epoxida-
tions, see: Ohta, A.; Ohta, M. Synthesis 1985, 216.

Carbodiimide-Promoted Olefin Epoxidation
J . Org. Chem., Vol. 63, No. 8, 1998 2567
Ch a r t 1. Ep oxid e Yield ver su s Stoich iom etr y of
H2O2
comparable yields under all the conditions employed; (3)
monosubstituted olefins typically epoxidize in lower yield
with both the H2O2-carbodiimide procedure and the
Payne system than with the more robust oxidant m-
CPBA; and (4) steric congestion near the double bond
retards epoxidation with the carbodiimide-promoted
procedures.
Tables 4 and 5 summarize the epoxidation of three
rigid bicyclic alkenes and four conformationally flexible
olefins, respectively. The carbodiimide-promoted epoxi-
dations and the Payne oxidation display similar selectiv-
ity, and both processes were as facially selective as
m-CPBA. Limonene, which has two double bonds with
different degrees of substitution, gave a 1:1 mixture of
isomeric epoxides 14a and 14b and trace amounts of the
bis-epoxide with the Payne system, whereas the carbo-
diimide-promoted process displayed a modest preference
for the oxidation of the trisubstituted endocyclic olefin
(a 3:2 mixture of 14a and 14b and <2% of the bis-
epoxide).
We have found that carbodiimide-promoted olefin
epoxidations represent a practical way to prepare acid-
and heat-sensitive epoxides. â-Pinene oxide or indene
oxide are too acid-sensitive to be prepared by m-CPBA,
but are easily prepared by using the H2O2-carbodiimide
procedures (cf. substrates 8 and 10, Tables 3 and 4,
respectively).
Sch em e 3
The preparative capabilities of this protocol were also
explored. In large-scale reactions rapidly mixing the
reagents together was exothermic; hence, the slow ad-
dition of H2O2 with efficient cooling was required. Cy-
clooctene oxide and allylbenzene oxide were synthesized
on a 20-g scale in 58% and 61% yield, respectively, using
procedure A. The use of procedure B achieved the large-
scale preparation of cyclooctene oxide and â-pinene oxide
in 57% and 60% yield, respectively.
A neighboring group, or Henbest,25 directing effect is
possible when a coordinating group is present, as in the
epoxidation of allylic alcohols. Thus five allylic or ho-
moallylic alcohols were treated with H2O2 and DCC to
look for a Henbest directing effect. These results, along
with those obtained by using either Payne oxidation or
metal-catalyzed epoxidation,26 are presented in Table 6.
For each substrate, each of the H2O2-carbodiimide
procedures gave the same product distribution. The most
pronounced directing effect was observed with the ho-
moallylic alcohol terpinen-4-ol (16), which gave the syn
epoxide independent of the choice of oxidant.27 In
contrast, the oxidation of homoallylic alcohol 18 gave a
chromatographically inseparable mixture of 18a and 18b,
with only a modest preference for the face with the
hydroxyl group.28 Significant Henbest directing effects
were observed in the epoxidation of allylic alcohols 19
and 20.
(h ) Effect of Hea tin g Usin g P r oced u r e A. The
epoxidation of R-pinene in a 1:1 mixture of methanol and
ethyl acetate was sluggish at room temperature; reflux-
ing this reaction mixture for 3 h produced R-pinene
epoxide in 71% yield. Unfortunately, this result is
misleading because R-pinene epoxide is stable under
these conditions. Further work showed that heating the
reaction mixtures of more reactive epoxides gave low
yields. Control experiments verified that the decomposi-
tion of the peroxyisourea to DCU and the rate of epoxide
consumption by either the solvent (methanol) or the
alkoxide are accelerated at higher temperatures. When
styrene was reacted using procedure A under reflux only
2-methoxy-2-phenylethanol and 2-methoxy-1-phenyle-
thanol were isolated. Thus, while some carbodiimide-
promoted epoxidations may benefit from heating, in
general it should be avoided.
(i) P r oced u r es F a n d G. Standard m-CPBA and
Payne oxidations are included for comparative purposes
as procedures F and G, respectively. The seven sets of
experimental conditions are summarized in Table 2.
We foresaw that the epoxidation of R,â-unsaturated
enones using base-catalyzed procedures would involve
nucleophilic addition of a hydroperoxide anion29 instead
of the involvement of the peroxyisourea intermediate.
Resu lts
(25) Henbest, H.; Wilson, R. J . Chem. Soc. 1957, 1958.
(26) (a) Itoh, T.; J itsukawa, K.; Kaneda, K.; Teranishi, S. J . Am.
Chem. Soc. 1979, 101, 159. (b) Finn, M.; Sharpless, K. B. In Asymmetric
Synthesis; Morrison, J ., Ed.; Academic Press: New York, 1985, Vol. 5,
pp 247-308.
(27) Mitchell, P. Org. Prep. Proceed. Int. 1990, 22(4), 534.
(28) Barrelle, M.; Apparu, M. Bull. Soc. Chim. Fr. 1972, 2016.
(29) House, H. O. Modern Synthetic Reactions, 2nd ed.; Benjamin:
Reading, MA, 1972 and references therein.
A variety of unactivated olefins (internal or terminal,
acyclic or cyclic) were epoxidized using the five optimized
H2O2-carbodiimide procedures (Tables 3, 4, and 5).
Scrutiny of Table 3 reveals four trends: (1) olefins having
at least two alkyl substituents give good yields of the
epoxides; (2) tri- and tetrasubstituted allylic alcohols give

2568 J . Org. Chem., Vol. 63, No. 8, 1998
Ta ble 2. Sta n d a r d Con d ition s for Ep oxid a tion s
Majetich et al.
reaction time
method
olefin equiv
dehydrating agent (equiv)
oxygen source
promoting agent (equiv)
A
B
C
D
E
F
1
5
1
1
5
1
1
DCC (2)
DCC (1)
DCC (2)
DCC (2)
DCC (2)
none
H2O2
H2O2
H2O2
H2O2
H2O2
m-CPBA
H2O2
KHCO3 (2)
KHCO3 (2)
Na3PO4‚12H2O (2)
catalytic Amberlite IRC-50
catalytic Amberlite IRC-50
none
14-24 h
24 h to 72 h
24 h
24 h
24 h
30 min to 12 h
48 h
G
C6H5CN (1.5)
KHCO3 (1)
Indeed, cyclohexenone was completely converted to the
corresponding epoxy ketone in only 20 min under proce-
dure A without consumption of the DCC; under acidic
conditions C, epoxidation of cyclohexenone was not
observed.
Ketones can undergo acid-catalyzed Baeyer-Villiger
oxidation with peroxyacids to generate esters, an undes-
ired side-reaction in the epoxidation of nonconjugated
enones. While control experiments showed that the
peroxyisourea did not react with ketones, we found that
treatment of 2-allylcyclohexanone or 2-prenylcyclohex-
anone for 48 h with H2O2 and DCC under either basic or
acidic conditions produced complex mixtures. These
results contrast with the selectivity possible using a
peroxycarboximidic acid intermediate (Payne oxidation)
or a peracid.
Ta ble 3. Ep oxid a tion of a Va r iety of Olefin s
Although the carbodiimide-based process takes place
under rather mild conditions, limitations do exist. For
example, the reaction of homoallylic alcohol 22 with
procedure A gave a complex mixture of products (Figure
1). However, if the hydroxyl group was protected (cf.
acetate 13, Table 5), an epoxide could be isolated. Allylic
alcohols 23 and 24 also gave unknown products with the
H2O2-carbodiimide procedure.
Ta ble 5. Ep oxid a tion of Con for m a tion a lly F lexible
Cyclic Olefin s
Ta ble 4. Ep oxid a tion of Br id ged Bicyclic Olefin s

Carbodiimide-Promoted Olefin Epoxidation
J . Org. Chem., Vol. 63, No. 8, 1998 2569
F igu r e 1.
Discu ssion a n d Mech a n istic Con sid er a tion s
Hydroxylic solvents permitted epoxidation, whereas no
oxidation was observed in aprotic solvents or under
anhydrous conditions. Thus, hydrogen bonding must
stabilize both the peroxyisourea intermediate and the
transition state leading to its formation. The decline in
yield observed when too much water is present in the
reaction medium can be attributed to the lower solubili-
ties of DCC and the reactant olefin in the medium: too
much water would also affect the solvation of the per-
oxyisourea. We believe that an equilibrium exists be-
tween two intramolecularly associated forms of the
peroxyisourea which lead to epoxidation (cf. i and ii, eq
5) and an open, solvated form (cf. iii) whereby the
electrophilic oxygen atom is shielded. This solvation
prevents this conformer from reacting with the olefin and
leads to the unwanted consumption of the peroxyisourea.
This equilibrium is highly influenced by the composition
of the solvent. Once the concentration of water in the
reaction media exceeds 50%, the equilibrium favors the
open form of the peroxyisourea (cf. iii).
A similar, albeit less dramatic, rate reduction is
observed for peroxyacid epoxidations in solvents that can
function as Lewis bases.30 For example, the rate constant
for the epoxidation of cyclohexene with m-CPBA at 20
°C drops from 2.25 × 10-2 M-1 s-1 in dichloromethane30a
to 2.5 × 10-4 M-1 s-1 in tert-butyl alcohol.30b
Presumably, oleic and vinylacetic acids fail to react in
water, despite the use of water-soluble carbodiimides,
because the hydrocarbon chains of these substrates
cluster together to form micelles, leaving the alkene
moiety inaccessible for oxidation.
The structure of the intramolecular hydrogen-bonded
peroxyisourea reagent dictates that this reactive inter-
mediate cannot exhibit a steric preference (cf. Figure 2).
The cyclic nature of this form of the peroxyisourea orients
the alkyl substituent attached to the imine nitrogen away
from the electrophilic oxygen atom. To avoid eclipsing
the C-O and CdN bonds, the remaining alkyl group
attached to the amide nitrogen is directed out of the plane
of the hydrogen-bonded chelate. Thus, both alkyl sub-
stituents are poorly situated to interact sterically with
an olefin approaching the electrophilic oxygen. Similar
arguments hold for oxidations using peroxyacids and
peroxycarboximidic acids. Thus, when facial specificity
is observed, as with the bridged bicyclic olefins 9, 10, and
11 (Table 4), it is due to the steric interactions between
the olefin and the electrophilic oxygen atoms of the
reagents rather than the steric influences of their sub-
stituents.
Ta ble 6. Hen best Dir ectin g Effect Stu d ies
F igu r e 2.
A basic premise throughout this study is that the
oxidant is a peroxyisourea. While this is a reasonable
assumption to make, we have been unable to verify its
formation experimentally. This prompted us to evaluate
the possibility of other oxidizing species.
Only two alternative oxidants can be postulated for the
base-promoted epoxidations. First, hydroperoxide ion
can add to the carbonyl group of the dicyclohexyl urea,
which becomes part of the reaction mixture once either
epoxidation or singlet oxygen generation begins.17 This
adduct (iv) can form either a peroxyisourea (v) or a
(30) (a) Paquette, L.; Barrett, J . Org. Synth. 1969, 49, 62. (b)
Renolen, P.; Ugelstad, J . J . Chim. Phys. 1960, 57, 634. (c) Curci, R.;
DiPrete, R.; Edwards, J .; Modena, G. J . Org. Chem. 1970, 35, 740. (d)
Kavcic, R.; Plesnicar, B. J . Org. Chem. 1970, 35, 2033.

2570 J . Org. Chem., Vol. 63, No. 8, 1998
Majetich et al.
Sch em e 4
Sch em e 6
Sch em e 5
which can then react with H2O2 (Scheme 6). The transfer
of the electrophilic oxygen from the peroxyacid to an
olefin would achieve epoxidation and make both pro-
cesses catalytic.
Ta ble 7. Ep oxid a tion of 25
promoting % axial
reagent
solvent
CH3OH
agent
attack
polymer-bound peroxyacid
H2O2-DCC (procedure A)
H2O2-DCC (procedure B)
H2O2-DCC (procedure C)
H2O2-DCC (procedure D)
H2O2-DCC (procedure E)
m-CPBA (procedure F)
m-CPBA
none
68
29
32
27
28
32
71
78
30
CH3OH
KHCO3
CH3CH2OH KHCO3
THF/EtOAc Na3PO4
peroxyacid (vi) as shown in Scheme 4; either of which
could act as the oxygen transfer agent. To test this,
R-pinene was treated using procedure A with DCU as
the dehydrating agent instead of DCC. Only a trace of
R-pinene oxide was detected after 48 h. This result is
not surprising since the carbonyl of a urea is less
electrophilic than that of formamide. Alternatively, the
formation of an azaoxirane (vii) can be postulated via
the pathways shown in Scheme 5. This oxidant can be
discounted because such a highly strained species could
not survive in the protic reaction medium and because
the byproduct of the epoxidation would be a carbodiimide,
not the urea observed.
The acid-catalyzed procedures C and D require a
weakly acidic ion-exchange resin to assist the nucleophilic
addition of H2O2 to the carbodiimide. If a peroxyisourea
is not involved in these oxidations, then the catalyst itself
must play an active role. Toward this end we can
speculate that a resin-bound oxidant is involved. More
specifically, if the DCC reacts first with the polymer-
bound carboxylic acid units, and then with the H2O2
present, a peroxyacid would be produced (eq 6). Alter-
natively, adjacent carboxylic acid moieties can be dehy-
drated by the DCC to form a polymer-bound anhydride31
CH3OH
IRC-50
CH3CH2OH IRC-50
CH2Cl2
CH3OH
KHCO3
none
H2O2-benzonitrile (procedure G) CH3OH
KHCO3
A series of experiments were carried out to test if resin-
bound peroxyacids were actually involved in the epoxi-
dation. 1-tert-Butyl-4-methylenecyclohexane (25) is an
useful substrate for testing the selectivity of peroxyi-
soureas and peroxyacids because it gives widely varying
ratios of O-axial and O-equatorial epoxides depending on
the oxidant used.32 When olefin 25 was treated with the
polymer-bound peroxyacid (produced by treating IRC-50
with methanesulfonic acid and H2O233) in methanol, the
percent of axial attack was consistent with that produced
by peroxyacids in methanol. In contrast, the product
distributions obtained with the five H2O2-carbodiimide
procedures are similar to those obtained using an H2O2-
benzonitrile system (Table 7). These observations argue
against a polymer-bound peroxyacid as the oxidizing
species and dictate the involvement of a common inter-
mediate in both the acid and base-catalyzed H2O2-
carbodiimide epoxidations. Given all of these consider-
ations, the oxygen transfer agent is most likely a
peroxyisourea.
Su m m a r y
We have developed a practical method for the epoxi-
dation of olefins that uses solutions of 30% aqueous H2O2
as the oxidant. The best solvent system consists of a
(31) For the preparation of polymer-bound anhydrides, see: Scott,
L.; Rebek, J .; Oysyanko, L.; Sims, C. J . Am. Chem. Soc. 1974, 96, 1606.
(32) (a) Carlson, R.; Behn, N. J . Org. Chem. 1967, 32, 1363. (b)
Vedejs, E.; Dent, W.; Kendall, J .; Oliver, P. J . Am. Chem. Soc. 1996,
118, 3556.

Carbodiimide-Promoted Olefin Epoxidation
J . Org. Chem., Vol. 63, No. 8, 1998 2571
P r oced u r e B: Ep oxid a tion w ith a Lim ited Qu a n tity
of DCC. Solid KHCO3 (73 mg, 0.73 mmol) was suspended in
a solution of the olefin (5.00 mol equiv relative to DCC) and
DCC (150 mg, 0.73 mmol) in 5 mL of ethanol. Hydrogen
peroxide (1 mL of a 30% solution, 8.80 mmol) was added, and
the reaction mixture was stirred at room temperature for 24
h. During the workup, the solids were removed by filtration,
and the filtrate was subjected to ethereal workup.
single phase with a simple alcohol and less than 50
volume percent of water. The yields of epoxide obtained
with this method are comparable to those produced by
the Payne oxidation and are only slightly lower than
those available by using high quality m-CPBA. Finally,
we postulate that the most likely reactive species is a
peroxyisourea.
P r oced u r e C: Ep oxid a tion w ith Sod iu m P h osp h a te in
Meth a n ol/Eth yl Aceta te. Solid Na3PO4‚12H2O(2.00 mol
equiv relative to the olefin) was suspended in a solution of the
olefin (100 mg) and DCC (2.00 mol equiv relative to the olefin)
in 2.5 mL of methanol and 2.5 mL of ethyl acetate. Hydrogen
peroxide (1 mL of a 30% solution, 8.80 mmol) was added, and
the reaction mixture was stirred at room temperature for 24
h. During workup, the solids were removed by filtration, and
the filtrate was subjected to ethereal workup.
P r oced u r e D: Acid -Ca ta lyzed Ep oxid a tion w ith Ex-
cess DCC. Amberlite IRC-50 (10 mg) was suspended in a
solution of the olefin (100 mg) and DCC (2.00 mol equiv
relative to the olefin) in 5 mL of methanol. Hydrogen peroxide
(1 mL of a 30% solution, 8.80 mmol) was added, and the
mixture was stirred at room temperature for 24 h. During
the workup, the solids were removed by filtration, and the
filtrate was subjected to ethereal workup.
Exp er im en ta l Section
Gen er a l. All reactions were carried out on a 0.10 to 1.00
mmol scale under an atmosphere of nitrogen and monitored
by TLC analysis until the starting material was completely
consumed. Unless otherwise indicated, ethereal workup con-
sisted of the following procedure: the reaction was quenched
at room temperature with saturated aqueous ammonium
chloride. The organic solvent was removed under reduced
pressure on a rotary evaporator, and the residue was taken
up in ether, washed with brine and dried over anhydrous
MgSO4. Filtration, followed by concentration at reduced
pressure on a rotary evaporator and at 1 Torr to a constant
weight, afforded a crude residue which was purified by flash
chromatography using NM silica gel 60 (230-400 mesh ASTM)
and distilled reagent grade solvents. Microanalysis was
performed by Atlantic Microlab, Inc., Atlanta, GA. 1H NMR
spectra were obtained in CDCl3 and were calibrated using
trace CHCl3 present (δ 7.26) as an internal reference.
1-Cyclohexene-1-methanol (2), 2-methyl-1-cyclohexene-1-
methanol (3), and 3-cyclohexene-1-methanol (cf. 13) were
obtained from stocks maintained in our inventory. Other
olefinic substrates were obtained from commercial sources. The
A.C.S. reagent grade H2O2 used was obtained from J . T. Baker
as a 30% w/v aqueous solution, signifying that 100 mL of
solution contains 30 g of H2O2. Since commerical 30% aqueous
solutions of H2O2 lose oxygen over time, even recently pur-
chased bottles of this reagent may contain varying amounts
of oxidants. Because of this, the H2O2 concentration of each
bottle was determined via titration.23
Complete chromatographic separation of unreacted DCC
from epoxides containing no other polar functional groups is
often tedious. Except in cases of extremely acid-sensitive
products such as â-pinene oxide, the remaining DCC can be
hydrolyzed by stirring an ethereal solution of the crude epoxide
with 5% aqueous acetic acid for 20 min. The ethereal phase
is then washed with sat. NaHCO3, dried over anhydrous
MgSO4, and concentrated. Complete chromatographic separa-
tion of the epoxide from DCU is straightforward.
On occasion, sufficient solid sodium sulfite was added in
small portions to neutralize (based on iodide starch paper) any
excess H2O2 prior to standard ethereal workup.
The authors have provided Chemical Abstracts Service
(CAS) registry numbers for known reaction products. Spectral
data are provided for new compounds. The following ab-
breviations are used throughout this section: hexanes (H) and
diethyl ether (E).
P r oced u r e E: Acid -Ca ta lyzed Ep oxid a tion w ith
a
Lim ited Qu a n tity of DCC. Amberlite IRC-50 (10 mg) was
suspended in a solution of the olefin (5.00 mol equiv relative
to DCC) and DCC (150 mg, 0.73 mmol) in 5 mL of ethanol.
Hydrogen peroxide (1 mL of a 30% solution, 8.80 mmol) was
added, and the mixture was stirred at room temperature for
24 h. During the workup, the solids were removed by
filtration, and the filtrate was subjected to ethereal workup.
P r oced u r e F : Ep oxid a tion w ith m -CP BA. To a 0 °C
solution of 100 mg of the olefin in CH2Cl2 was added a solution
of m-CPBA (1.30 mol equiv relative to the olefin) in CH2Cl2.
The mixture was slowly warmed to room temperature, stirred,
and monitored by TLC analysis prior to ethereal workup.
P r oced u r e G: A P a yn e Ep oxid a tion . Solid KHCO3 (1.00
mol equiv relative to the olefin) was suspended in a solution
of the olefin (100 mg) and benzonitrile (1.50 mol equiv relative
to the olefin) in 1 mL of methanol. Hydrogen peroxide (a 30%
solution, 1.5 mol equiv relative to the olefin) was added, and
the mixture was stirred at room temperature for 48 h. The
solvent was removed at reduced pressure, and the resulting
residue was purified by chromatography.
(b) Ep oxid a tion of Su bstr a tes: Ep oxid a tion of Meth -
ylen ecycloh exen e (1). Methylenecyclohexene (100 mg, 1.04
mmol) was converted to 6-oxaspiro[2.5]octane [75934-94-0] in
74% yield using procedure A, in 89% yield using procedure B,
in 85% using procedure C, in 70% yield using procedure D, in
72% yield using procedure E, in 75% yield using procedure F,
and in 47% yield using procedure G.
Ep oxid a tion of 1-Cycloh exen ylm eth a n ol (2). 1-Cyclo-
hexenylmethanol (100 mg, 0.89 mmol) was converted to
7-oxabicyclo[4.1.0]heptane-1-methanol [17550-61-7] in 65%
yield using procedure A, in 68% yield using procedure B, in
64% using procedure C, in 57% yield using procedure D, in
60% yield using procedure E, in 68% yield using procedure F,
and in 67% yield using procedure G.
This section has been organized as follows: (a) General
Epoxidation Procedures; (b) Epoxidation of Various Substrates;
and (c) Large-scale Epoxidations. In addition to these experi-
mentals, detailed experimental descriptions of the various
carbodiimide-based epoxidations we discussed within the text
can be found in the Supporting Information section of this
article.
(a ) Gen er a l Ep oxid a tion P r oced u r es. P r oced u r e A:
Ep oxid a tion w ith Excess DCC. Solid KHCO3 (2.00 mol
equiv relative to the olefin) was suspended in a solution of the
olefin (100 mg) and DCC (2.00 mol equiv relative to the olefin)
in 5 mL of methanol. Hydrogen peroxide (1 mL of a 30%
solution, 8.80 mmol) was added, and the reaction mixture was
stirred at room temperature for 24 h. During workup, the
solids were removed by filtration and the filtrate was subjected
to ethereal workup.
Epoxidation of (2-Meth ylcycloh ex-1-en yl)m eth an ol (3).
(2-Methylcyclo-hex-1-enyl)methanol (100 mg, 0.79 mmol) was
converted to 2,3-oxa-3-methylbicyclo[4.1.0]heptane-1-methanol
in 67% yield using procedure A, in 70% yield using procedure
B, in 77% using procedure C, in 63% yield using procedure D,
in 65% yield using procedure E, in 70% yield using procedure
F, and in 64% yield using procedure G. This epoxide has the
following spectral data/physical properties: 1H NMR (250
MHz) 1.15-1.59 (m, 4 H), 1.38 (s, 3 H), 1.61-2.00 (m, 4 H),
3.60-3.79 (m, 2 H); 13C (62.7 MHz) 64.8 (t), 64.5 (s), 62.8 (s),
31.7 (t), 26.6 (t), 20.7 (t), 20.1 (t), 19.9 (q) ppm; IR (film) 3400
br., 2934, 1434, 1031, 833 cm-1. Anal. for C8H14O2. Calcd: C,
67.56%; H, 9.93%. Found: C, 67.75%; H, 9.68%.
Ep oxid a tion of P r op -2-en ylben zen e (4). Prop-2-enyl-
benzene (100 mg, 1.69 mmol) was converted to phenylmethy-
(33) Harrison, C.; Hodge, P. J . Chem. Soc., Perkin Trans. 1 1976,
605.

2572 J . Org. Chem., Vol. 63, No. 8, 1998
Majetich et al.
cyclohexen-1-yl)oxirane [31684-93-2] (14b) in 35% yield.
loxirane [4436-24-1] in 59% yield using procedure A, in 61%
yield using procedure B, in 39% using procedure C, in 42%
yield using procedure D, in 54% yield using procedure E, in
84% yield using procedure F, and in 41% yield using procedure
G.
Ep oxid a tion of 1-Dod ecen e (5). 1-Dodecene (100 mg,
0.59 mmol) was converted to 1,2-dodecyloxirane [2855-19-8]
in 40% yield using procedure A, in 28% yield using procedure
B, in 52% using procedure C, in 27% yield using procedure D,
in 32% yield using procedure E, in 91% yield using procedure
F, and in 36% yield using procedure G.
Ep oxid a tion of Styr en e (6). Styrene (100 mg, 0.96 mmol)
was converted to styrene oxide [96-09-03] in 75% yield using
procedure A (cf. initial observations), in 80% yield using
procedure B, in 75% using procedure C, in 73% yield using
procedure D, in 39% yield using procedure E, in 88% yield
using procedure F, and in 68% yield using general procedure
G.
A
10% yield of 1-methyl-4-(2-methoxiranyl)oxabicyclo[4.1.0]-
heptane (14c) [96-08-2] was also isolated. An inseparable
mixture (1:1) of epoxides 14a and 14b was obtained in 85%
yield using procedure B. A 2% yield of bis-epoxide 14c was
also isolated. Using procedure C an inseparable mixture (1:
1) of epoxides 14a and 14b was obtained in 39% yield. Using
procedure D an inseparable mixture (1:1) of epoxides 14a and
14b was obtained in 49% yield. A 15% yield of bis-epoxide
14c was also isolated. Using procedure E an inseparable
mixture (1:1) of epoxides 14a and 14b was obtained in 85%
yield. A 1% yield of bis-epoxide 14c was also isolated. Using
procedure F an inseparable mixture (1:1) of epoxides 14a and
14b was obtained in 75% yield. An 11% yield of bis-epoxide
14c was also isolated. Using procedure G an inseparable
mixture (1:1) of epoxides 14a and 14b was obtained in 70%
yield. A 16% yield of bis-epoxide 14c was also isolated.
Ep oxid a tion of (E)-2,2-Dim eth ylh ep t-3-en e (7). (E)-2,2-
Dimethylhept-3-ene (100 mg, 0.79 mmol) was converted to (E)-
3,4-oxa-2,2-dimethylheptane in 48% yield using procedure A,
in 36% yield using procedure B, in 0% using procedure C, in
14% yield using procedure D, in 19% yield using procedure E,
in 59% yield using procedure F, and in 15% yield using
procedure G. This epoxide has the following spectral data/
physical properties: 1H NMR (250 MHz) 0.91 (s, 9 H), 0.96 (t,
3 H, J ) 7.1 Hz), 1.33-1.65 (m, 4 H), 2.46 (d, 1 H, J ) 2.4
Hz), 2.75-2.85 (m, 1H); 13C (62.7 MHz) 66.9 (d), 55.3 (d), 34.3
(t), 30.6 (s), 25.8 (q), 19.4 (t), 13.9 (q) ppm; IR (film) 2958, 1363,
Ep oxid a tion of 1,3-Cycloocta d ien e (15). 1,3-Cycloocta-
diene (100 mg, 0.92 mmol) was converted to 9-oxabicyclo[6.1.0]-
non-2-ene (15a) [6690-12-6] in 39% yield and 3,10-dioxatricyclo-
[7.1.0.02,4]decane-(1R, 2R, 4R, 9R) (15b) [97373-34-7] in 24%
yield. 1,3-Cyclooctadiene gave only epoxide 15a in 65% yield
using procedure B. In contrast, procedure C produced a 45%
yield of epoxide 15a and a 14% yield of 15b. Using procedure
D, diene 15 produced a 18% yield of monoepoxide 15a and a
17% yield of bis-epoxide 15b. 1,3-Cyclooctadiene gave only
epoxide 15a in 12% yield using procedure E and in 80% yield
using procedure F. Using procedure G, diene 15 produced a
46% yield of monoepoxide 15a and a 13% yield of bis-epoxide
15b.
909 cm-1
. Anal. for C9H18O. Calcd: C, 75.98%; H, 12.76%.
Found: C, 75.85%; H, 12.71%.
Ep oxid a tion of In d en e (8). Indene (100 mg, 0.86 mmol)
was converted to indene oxide in 69% yield using procedure
A, in 59% yield using procedure B, in 47% using procedure C,
in 73% yield using procedure D, in 98% yield using procedure
E, in 0% yield using procedure F, and in 42% yield using
procedure G.
Ep oxid a tion of r-P in en e (9). R-Pinene (100 mg, 0.73
mmol) was converted to R-pinene oxide [1686-14-2] in 68%
yield using procedure A, in 83% yield using procedure B, in
30% using procedure C, in 73% yield using procedure D, in
68% yield using procedure E, in 74% yield using procedure F,
and in 33% yield using procedure G.
Ep oxid a tion of â-P in en e (10). â-Pinene (100 mg, 0.73
mmol) was converted to â-pinene oxide [6931-54-01] in 91%
yield using procedure A, in 60% yield using procedure B, in
49% using procedure C, in 48% yield using procedure D, in
65% yield using procedure E, in 0% yield using procedure F,
and in 27% yield using procedure G.
Ep oxid a tion of Bicyclo[2.2.1]h ep t-2-en e (11). Olefin 11
(100 mg, 1.06 mmol) was converted to 3-oxatricyclo[3.2.1.02,4]-
octane [278-74-0] in 64% yield using procedure A, in 61% yield
using procedure B, in 50% using procedure C, in 68% yield
using procedure D, in 57% yield using procedure E, in 79%
yield using procedure F, and in 48% yield using procedure G.
Ep oxid a tion of 1-Meth yl-4-isop r op ylcycloh exen e (12).
Olefin 12 (100 mg, 0.72 mmol) was converted to 1-methyl-4-
(1-methylethyl)-7-oxabicyclo[4.1.0]heptane [3626-19-5] as a 1:1
mixture of diastereomers in 83% yield using procedure A, in
61% yield using procedure B, in 74% using procedure C, in
38% yield using procedure D, in 44% yield using procedure E,
in 73% yield using procedure F, and in 63% yield using
procedure G.
(c) La r ge-Sca le Ep oxid a tion s: Ep oxid a tion of Cy-
cloocten e w ith P r oced u r e A. Solid NaHCO3 (30.00 g, 0.36
mol) was suspended in a solution of cyclooctene (20.00 g, 0.18
mol) and DCC (73.00 g, 0.36 mol) in 420 mL of methanol.
Hydrogen peroxide (150 mL of a 30% solution, 1.32 mol) was
added dropwise over a 1-h period. An ice-water bath was
used to keep the reaction temperature below 25 °C. The
mixture was stirred at room temperature for an additional 48
h. The solids were removed by filtration, and the filtrate was
partitioned between ether and water. Standard ethereal
workup, followed by distillation (bp 54-56°/5 mm), gave 13.1
g (58%) of cyclooctene oxide (9-oxabicyclo(6.1.0)nonane) [286-
62-4].
Ep oxid a tion of P r op -2-en ylben zen e w ith P r oced u r e
A. Solid Na2CO3 (36.00 g, 0.34 mol) was suspended in a
solution of prop-2-enylbenzene (20.00 g, 0.17 mol), and DCC
(70.00 g, 0.34 mole) in 500 mL of ethanol. Hydrogen peroxide
(160 mL of a 30% solution, 1.40 mol) was added dropwise over
a 1-h period. An ice-water bath was used to keep the reaction
temperature below 25 °C. The mixture was stirred at room
temperature for an additional 48 h. The solids were removed
by filtration, and the filtrate was partitioned between ether
and water. Standard ethereal workup, followed by vacuum
distillation (58-65 °C/1 mm), gave 14.2 g (61%) of epoxide
[4436-24-1].
Ep oxid a tion of Cycloh exen e w ith P r oced u r e B. Solid
KHCO3 (19.40 g, 194 mmol) was suspended in a solution of
cyclohexene (65.60 g, 0.79 mol) and DIC (20.00 g, 159 mmol)
in 400 mL of ethanol. Hydrogen peroxide (100 mL of a 30%
solution, 0.88 mol) was added dropwise over a 3-h period. An
ice-water bath was used to keep the reaction temperature
below 25 °C. The mixture was stirred at room temperature
for an additional 18 h. Approximately two-thirds of the solvent
was carefully removed at room temperature in vacuo. Solids
were removed by filtration, and the filtrate was partitioned
between ether and water. Standard ethereal workup, followed
by vacuum distillation, gave 8.90 g (57%) of cyclooctene oxide.
Ep oxid a tion of Cycloh exen e-4-ca r bin ol Aceta te (13).
Olefin 13 (100 mg, 0.65 mmol) was converted to 7-oxabicyclo-
[4.1.0]heptane-3-methanol acetate in 62% yield as a 1:1
mixture of diasteroemers, (1R, 3R, 6R) [81370-41-4] and (1R,
3â, 6R) [81370-42-5], using procedure A, in 68% yield using
procedure B, in a 86% yield using procedure C, in 29% yield
using procedure D, in 62% yield using procedure E, in 75%
yield using procedure F, and in 52% yield using procedure G.
Ep oxid a tion of Lim on en e (14). Using procedure A
limonene (100 mg, 0.73 mmol) was converted to an inseparable
mixture (1:1) of 1-methyl-4-(1-methylethenyl)-7-oxabicyclo-
[4.1.0]heptane (14a ) [1195-92-2] and 2-methyl-2-(4-methyl-3-
Ep oxid a tion of (-)-â-P in en e w ith P r oced u r e B. Solid
KHCO3 (19.4 g, 194 mmol) was suspended in a solution of (-)-

Carbodiimide-Promoted Olefin Epoxidation
J . Org. Chem., Vol. 63, No. 8, 1998 2573
â-pinene (66.10 g, 480 mmol) and DCC (20.00 g, 96.9 mmol)
in 300 mL of ethanol. Hydrogen peroxide (100 mL of a 30%
solution, 0.88 mol) was added dropwise over a 6-h period. An
ice-water bath was used to keep the reaction temperature
below 25 °C. The mixture was stirred at room temperature
for an additional 18 h. Approximately two-thirds of the solvent
was carefully removed at room temperature on a rotary
evaporator. Solids were removed by filtration, and the filtrate
was partitioned between ether and water. Standard ethereal
workup, followed by vacuum distillation (bp 98-100°/27 mm),
gave 8.85 g (60%) of (-)-â-pinene oxide.
Su p p or tin g In for m a tion Ava ila ble: Experimental details
(20 pages). This material is contained in libraries on micro-
fiche, 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 O972026N