Molecule-induced homolysis of N-hydroxyphthalimide (NHPI) by peracids and dioxirane. A new, simple, selective aerobic radical epoxidation of alkenes

Article abstract of DOI:10.1016/j.tetlet.2005.12.089

Evidences are reported concerning the molecule-induced homolysis of NHPI by peracids and dioxirane; their combination can be utilized for the aerobic free-radical epoxidation of alkenes with selectivity quite different from the well-known epoxidation by peracids.

Full text of DOI:10.1016/j.tetlet.2005.12.089

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 1421–1424
Molecule-induced homolysis of N-hydroxyphthalimide (NHPI) by
peracids and dioxirane. A new, simple, selective aerobic
radical epoxidation of alkenes
Francesco Minisci,a,* Cristian Gambarotti,a Monica Pierini,a Ombretta Porta,a
Carlo Punta,a,* Francesco Recupero,a Marco Lucarinib and Veronica Mugnainib
aPolitecnico di Milano, Dipartimento di Chimica, Materiali e Ingegneria Chimica ‘G. Natta’, Via Mancinelli 7,
I-20131 Milano (MI), Italy
`
Universita di Bologna, Dipartimento di Chimica Organica ‘A. Mangini’, Via S. Giacomo, I-40126 Bologna (BO), Italy
b
Received 14 November 2005; revised 13 December 2005; accepted 16 December 2005
Abstract—Evidences are reported concerning the molecule-induced homolysis of NHPI by peracids and dioxirane; their combina-
tion can be utilized for the aerobic free-radical epoxidation of alkenes with selectivity quite different from the well-known epoxida-
tion by peracids.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Some years ago we reported evidences that the oxidation
of a variety of organic compounds (hydrocarbons,
alcohols, ethers, aldehydes, etc.) by dioxiranes1 and per-
acids2 can be explained by radical mechanism in clear
contrast with the widely accepted3,4 mechanism of ‘con-
certed oxenoid oxygen insertion’. Our interpretation1,2
involves a ‘molecule-induced homolysis’ in which the
transition states are related to the hydrogen abstractions
leading to radical pairs (Eqs. 1 and 2).
The controversial mechanisms, particularly with dioxir-
anes, are intriguing since the fast coupling of the radical
pair in the solvent cage leads to the same reaction prod-
ucts of the oxygen insertion mechanism and only few
radicals can escape from the cage giving typical free-
radical reactions. Very recently this controversy has
been resumed in the case of dioxiranes: one report5 cate-
gorically states that ‘a radical-type process has been
experimentally and theoretically rigorously discounted’,
whereas another research group reports6 that ‘theoreti-
cal calculations support both alternative reaction
mechanisms as feasible reaction pathways for the oxy-
genation of C–H bonds by dioxirane’ and also the experi-
mental results suggest that ‘the process splits into two
reaction pathways with different transition states leading
to the O–O homolysis and the O-atom insertion, respec-
tively, the incidence of each process depending on the
structure of the dioxirane’.6 What is not clear in this last
interpretation is the reason why the coupling of the radi-
cal pair (Eq. 2), leading to the same products of the
oxygen insertion mechanism, does not occur in the sol-
vent cage.
O
C
O
C
δ+
δ−
R
H + O
H
O
R'
R
H
O
H
O
R'
ð1Þ
ð2Þ
O
.
.
O
C R'
R
+ H
O
H +
O
O
CH3
CH3
O
CH3
CH3
δ+
+
R
H
R
H
O
δ−
.
O
O
CH3
CH3
.
R
+
H
We have ascribed the driving force for Eqs. 1 and 2 to the
high bond dissociation enthalpies (BDE) of the O–H
bonds formed in hydrogen abstraction, which are partic-
ularly effective with the weaker C–H bonds (tertiary
alkyl, benzyl, RCO–H, etc.). Recently we have evaluated7
Keywords: N-Hydroxyphthalimide; Epoxides; Alkenes; Peracids.
*
Corresponding authors. Tel.: +39 02 23993030; fax: +39 02 23993080
(F.M.); tel.: +39 02 23993026; fax: +39 02 23993080 (C.P.); e-mail
addresses: francesco.minisci@polimi.it; carlo.punta@polimi.it
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.12.089

1422
F. Minisci et al. / Tetrahedron Letters 47 (2006) 1421–1424
the BDE value of the O–H bond in NHPI (88.1 kcal/
mol); this value suggests to us that peracids and dioxir-
anes could give induced homolysis of NHPI (Eqs. 3
and 4) under mild conditions generating the phthalimido-
N-oxyl (PINO) radical which plays a key role in the
aerobic oxidations catalyzed by NHPI.7,8
solution at rt always leads to m-chlorobenzoic acid as
the main reaction product, whereas phenyl m-chloro-
benzoate and m-chlorobiphenyl are the by-products.
The only possible explanation for these last by-products
and for chlorobenzene is the formation of the acyloxyl
radical (Eq. 3), which reacts according to Eq. 7a–d.
O
C
O
C
O
H
O
R
N
O
H +
N
O
H
O
H
O
R
.
m-Cl C6H4
COO
C
NHPI
H
(
a
ð3Þ
)
O
C
ð7Þ
.
.
.
O
R
N
O
+
CO2 m-Cl C6H4
+
m-Cl C6H4
H O
2
+
COOC6H5
6
PINO
)
6
d
(
Cl
m-Cl C6H4
C6H5
C6H5
O
CH3
CH3
O
O
CH3
CH3
N
O H +
O
N
O
H
+
m-Chlorobenzoic acid is the main reaction product
because it is formed by hydrogen abstraction from
NHPI by the acyloxyl radical (Eq. 8).
ð4Þ
.
O
CH3
CH3
.
N
O
H
O
.
COO
COOH
The hypothetical coupling of the radical pairs generated
in Eqs. 3 and 4 is very likely reversible, so that the PINO
radical and the acyloxyl (Eq. 5) or the alkoxyl (Eq. 6)
radicals can escape from the solvent cage giving the typi-
cal free-radical reactions.
k8
.
+ H
O
N
k
8 > 109 M-1s-1
+
O N
Cl
Cl
ð8Þ
We expect Eq. 8 to be a very fast reaction on the basis
O
of the known rate constants for the hydrogen abstrac-
N
O
O
C R
tion by the peroxyl radical (7.2 · 103 Mꢁ1 sꢁ1
)
and
7
O
C
alkoxyl radical (109 Mꢁ1 sꢁ1);10 Eq. 8 is more exothermic
respectively of 22 and 6 kcal/mol than the correspond-
ing hydrogen abstractions by the peroxyl and alkoxyl
radicals.
ð5Þ
ð6Þ
O. .O
N
R
e
s
c
a
p
O
e
cage
.
.
O
C R
N
O
+
H
O
CH3
N
O
O
CH3
Our recent results7,8 have shown that the hydrogen
abstractions from C–H bonds by the PINO radical are
considerably faster than the corresponding hydrogen
abstractions by peroxyl radicals, in spite of the same en-
thalpy variation; a more marked polar effect in the reac-
tion by the PINO radical is, in our opinion,11 a factor
which contributes to this behavior. These results, com-
bined with the induced homolysis of NHPI by peracids
(Eq. 3) have suggested to us the possibility to utilize the
aerobic oxidation of aldehydes, catalyzed by NHPI, for
the epoxidations of alkenes by peracids generated
‘in situ’ under mild conditions (Eq. 10).
H
O
CH3
CH3
.
.
N
O
O
e
s
c
a
p
e
H
O
O
CH3
CH3
cage
.
.
O
N
+
Spectroscopic and chemical evidences, actually, support
this assumption. The EPR spectrum of the PINO radi-
cal7,9 was readily observed simply by adding at rt NHPI
to a solution of m-chloroperbenzoic acid in acetonitrile
(a(2H) = 0.46 G; a(N) = 4.77 G) or dimethyldioxirane
in acetone (a(2H) = 0.44 G; a(N) = 4.70 G), as it is
shown in Figure 1.
O
C
O
δ−
HPLC and GC analysis of the products arising from
m-chloroperbenzoic acid and NHPI in acetonitrile at rt
reveal the presence of m-chlorobenzoic acid as main
reaction product (ꢀ90%) and chlorobenzene (ꢀ10%)
as by-product. Moreover the same reaction in benzene
.
CH3
O
N
CH3
H
C
δ+
H
O
N
+
ð9Þ
O
.
CH3
C
HO N
+
O
C
NHPI
ð10Þ
C
C
CH CHO
O2
+
C
CH3COOH
+
+
3
Actually the aerobic oxidation of acetaldehyde in aceto-
nitrile solution at rt and atmospheric pressure of oxygen
in the presence of alkenes and catalytic amount of NHPI
leads to the corresponding epoxides. No oxidation oc-
curs under the same conditions in the absence of NHPI,
clearly indicating that Eq. 9 plays a key role in the aer-
obic epoxidation. However the selectivity observed with
several alkenes is quite different from that well-known12
with peracetic acid (Table 1).
Figure 1. EPR spectrum of PINO obtained by mixing NHPI (0.01 M)
with m-CPBA (0.01 M) in CH3CN at room temperature.

F. Minisci et al. / Tetrahedron Letters 47 (2006) 1421–1424
1423
Table 1. Epoxidation of olefins by aerobic oxidation of acetaldehyde, catalyzed by NHPIa
Olefin
Catalyst (%)
Reaction time (h)
Yield of epoxide (%)
1-Hexene
1-Hexene
1-Octene
1-Decene
1-Decene
1-Decene
1-Dodecene
Methyl oleate
cis-2-Hexene
Cyclooctene
Cyclooctene
NHPI (10)
NHPI (10)
NHPI (10)
NHPI (10)
NHPI (10)
—
NHPI (10)
NHPI (10)
NHPI (10)
NHPI (10)
—
24
48
24
24
48
24
23
24
24
27
24
61
70
80
81
94
—
81
—
—
96
—
NHPI (10)
NHPI (10)
14
24
O
R(+)-limonene
2-Methyl-2-butene
cis (67), trans (33)
—
a General procedure: a solution of 5 mmol of olefins, 15 mmol of acetaldehyde and 0.5 mmol of NHPI in 10 mL of acetonitrile was stirred at rt in
atmospheric pressure of O2 for the time reported in the table; the known epoxides were isolated by flash chromatography and characterized by
NMR and MS spectra (identical with authentical samples).
Thus a-olefins and cyclic olefins give good yields of
O
.
ð13Þ
.
CH3
N
O
H
O
H
O
C
CH3
H O
+ +
2
CO
+
epoxides, whereas internal acyclic olefins are unreactive
under the same conditions; the internal acyclic olefins
are, on the contrary, more reactive than a-olefins12 with
peracetic acid and also the stereochemistry for the epoxi-
dation of limonene is different: in the aerobic epoxida-
tion, 67% of the cis isomer and 33% of the trans
isomer are formed, whereas 41% of the cis isomer and
59% of the trans isomer are the reaction products with
peracid; moreover 2-methyl-2-butene is 300 times more
reactive than propene with peracetic acid, whereas it is
unreactive towards the aerobic epoxidation.
N O
2
An analogous transition state should not occur with per-
benzoic acid, as shown by Eq. 7a, due to the stronger
Ph–C bond (that is reflected in the quite different rates
of decarboxylation of PhCOOÅ and CH3COOÅ, respec-
tively, ꢀ106 and ꢀ109 sꢁ1). A transition state (Eq. 14)
analogous to Eq. 13 could be involved for Eq. 12.
R
CH
R
CH
O
C
.
CH3
CO
+
2
+
O
ð14Þ
H3C
O
O
CH2
CH2
This selectivity clearly suggests that the epoxidation by
aerobic cooxidation of aldehyde/olefin previously re-
ported13 occurs by a different mechanism since internal
olefins are readily epoxidated.
The fact that the internal acyclic olefins are unreactive
towards the aerobic epoxidation could suggest that per-
acetic acid is not formed according to Eq. 15.
O
O
k15
+
k15 > 104 M-1s-1
.
O
N
Peracetic acid, therefore, does not appear to be respon-
sible for the aerobic epoxidation of the alkenes, cata-
lyzed by NHPI. Our interpretation involves the
reaction of the acyl radical with O2 (Eq. 11) followed
by the addition of the acylperoxyl radical to the olefin
and decomposition of the radical adduct (Eq. 12).
HO
+
N
R C
R
C
.
OO
OO
H
ð15Þ
We have not evaluated the rateꢁ1coꢁn1stant k15, but we
expect a value higher than 104 M
because we have
s
determined a value of 7.2 · 103 Mꢁ1 sꢁ1 for the hydro-
gen abstraction from NHPI by t-BuOOÅ radical and
Eq. 15 is about 5 kcal/mol more exothermic (the BDE
values of the O–H bonds are, respectively, 88 and
93 kcal/mol for t-BuOO–H and RCOOO–H). The high
rate constant of k15 suggests that peracetic acid is actu-
ally formed according to Eq. 15 in competition to the
addition of the olefin12 (Eq. 12), but its reaction with
NHPI (Eqs. 3 and 13) is faster than the epoxidation of
alkenes.
.
.
O
O
ð11Þ
CH3
CH3
O2
C
O
CH3
C
+
O
.
O
O
O
O
O
O
.
H2C CHR
CH3
C
C
CH2 CHR
+
ð12Þ
O
.
CH R
H2C
CH3 CO
+
2
+
Acetaldehyde revealed to be much more effective than
benzaldehyde for the epoxidation of alkenes; a possible
explanation is that the CH3–C bond could be involved
in the transition state of the induced homolysis of NHPI
by the CH3COOOH (Eq. 13) contributing to lower the
activation energy.
The aerobic cooxidation of acetaldehyde and benzyl
derivatives catalyzed by NHPI has also been reported,14
but the authors did not realize that peracids give the in-
duced homolysis of NHPI generating the PINO radical
(Eq. 3), which plays the key role in the aerobic
oxidation.

1424
F. Minisci et al. / Tetrahedron Letters 47 (2006) 1421–1424
G.; Surya Prakasch, G. K. J. Org. Chem. 2005, 70, 7919,
References and notes
and references cited therein.
7. Amorati, R.; Lucarini, M.; Mugnaini, V.; Pedulli, G. F.;
Minisci, F.; Recupero, F.; Fontana, F.; Astolfi, P.; Greci,
L. J. Org. Chem. 2003, 68, 1747.
1. Minisci, F.; Zhao, L.; Fontana, F.; Bravo, A. Tetrahedron
Lett. 1995, 36, 1697; Bravo, A.; Fontana, F.; Fronza, G.;
Minisci, F.; Serri, A. Tetrahedron Lett. 1995, 36, 6945;
Bravo, A.; Fontana, F.; Fronza, G.; Minisci, F.; Zhao, L.
J. Org. Chem. 1998, 63, 254.
2. Bravo, A.; Bjorsvik, H. R.; Fontana, F.; Minisci, F.; Serri,
A. J. Org. Chem. 1996, 61, 9409.
3. Asensio, G.; Mello, R.; Gonzalez-Nunez, M. E.; Boix, C.;
˜
Royo, J. Tetrahedron Lett. 1997, 38, 2373; Curci, R.;
Dinoi, A.; Fusco, C.; Lillo, M. A. Tetrahedron Lett. 1996,
37, 249; Simakov, P. A.; Choi, S. Y.; Newcomb, M.
Tetrahedron Lett. 1998, 39, 8187; Adam, W.; Curci, R.;
D’Accolti, L.; Dinai, A.; Fusco, C.; Gasparrini, F.; Kluge,
R.; Paredes, R.; Shulz, M.; Smers, A. K.; Veloza, L. A.;
Weinkatz, S.; Winde, R. Chem. Eur. J. 1997, 3, 105.
4. Schneider, H. J.; Mueller, W. J. Org. Chem. 1985, 50,
4609; Tari, M.; Matsuda, R.; Asakawa, Y. Tetrahedron
Lett. 1985, 26, 227.
8. (a) Sakaguchi, S.; Nashiwaki, Y.; Kitamura, T.; Ishii, Y.
Angew. Chem., Int. Ed. 2001, 40, 222; (b) A review: Ishii,
Y.; Iwahama, S.; Sakaguchi, S. Adv. Synth. Catal. 2001,
343, 393; (c) A review: Minisci, F.; Recupero, F.; Pedulli,
G. F.; Lucarini, M. J. Mol. Catal. 2003, 204–205, 63–69.
9. (a) Lemaire, H.; Rassat, A. Tetrahedron Lett. 1964, 2245;
(b) Mackor, A.; Wajer, Th. A. J. W.; De Boer, Th.
Tetrahedron 1968, 24, 4542.
10. Coseri, S.; Mendenhall, G. D.; Ingold, K. U. J. Org.
Chem. 2005, 70, 4629.
11. Minisci, F.; Recupero, F.; Cecchetto, A.; Gambarotti, C.;
Punta, C.; Faletti, R.; Paganelli, R.; Pedulli, G. F. Eur. J.
Org. Chem. 2004, 2004, 109–119.
12. Banillon, G.; Lick, C.; Schank, K. The Chemistry of
Peroxides. In The Chemistry of Functional Groups; Patai,
S., Ed.; John Wiley: New York, 1983; p 289.
13. Lassila, K. R.; Waller, F. J.; Werkheiser, S. E.; Wressell,
A. L. Tetrahedron Lett. 1994, 35, 8077.
5. Adam, W.; Zhao, C. G. In Handbook of C–H Transfor-
mation; Dyker, G., Ed.; Wiley-VCH, 2005; Vol. 2, p 507,
and references cited therein.
14. Einhorn, C.; Einhorn, J.; Marcadal, C.; Pierre, J.-L. Chem.
Commun. 1997, 447.
6. Gonzalez-Nunez, M. E.; Royo, J.; Mello, R.; Baguena,
˜
M.; Martinez Ferrer, J.; Ramirez de Arellano, C.; Asensio,