5-Hydroperoxycarbonylphthalimide: A new reagent for epoxidation

Article abstract of DOI:10.1039/a708179k

Peroxycarboxylic acids, widely used for epoxidation in industry and general research, have various drawbacks, such as difficulty of preparation in a pure state, cost and the possibly ring-opening of the product epoxides due to acid-catalysed reactions; a new reagent, 5-hydroperoxycarbonylphthalimide, overcomes these problems.

Full text of DOI:10.1039/a708179k

�
View Article Online / Journal Homepage / Table of Contents for this issue
5
-Hydroperoxycarbonylphthalimide: a new reagent for epoxidation
a
b
b
a
c
Alun P. James, Robert A. W. Johnstone,* † Moya McCarron, J. Phillip Sankey and Brian Trenbirth
a
Solvay Research and Development Laboratories, Widnes, UK WA8 0FE
Department of Chemistry, University of Liverpool, Liverpool, UK L69 3BX
Contract Chemicals, Knowsley Industrial Park South, Prescot, UK L34 9HY
b
c
Peroxycarboxylic acids, widely used for epoxidation in
industry and general research, have various drawbacks, such
as difficulty of preparation in a pure state, cost and the
possibly ring-opening of the product epoxides due to acid-
catalysed reactions; a new reagent, 5-hydroperoxycar-
bonylphthalimide, overcomes these problems.
O
HN
+
+
O
O
H
O
O
1
O
The formation of epoxides (oxiranes) from alkenes is a widely
used reaction, either for the product epoxides themselves or for
their preparation as intermediates in further synthesis. The
O
1
HN
O
H
manufacture of epoxides ranges in scale from the small
requirements of a research reaction to the huge tonnages of
industrial epoxides and bis-epoxides. Hydrogen peroxide, a
clean oxidant producing only water as a by-product, would be
ideal if it could be used to produce epoxides directly. Although
epoxidation can be effected with hydrogen peroxide through use
O
O
2
Scheme 1
2
3
of metal catalysts such as those of tungsten or molybdenum,
4
5
titanium silicalites and porphyrins, these catalysts are not
universally applicable for various reasons, including cost, lack
of specificity, inability to selectively epoxidise without forma-
tion of serious quantities of by-products and lack of reactivity
towards alkenes having electron-deficient double bonds. Such
catalysts have not found general favour, except in some
specialised instances. There is widespread use of hydrogen
peroxide by first converting it into a suitably reactive form, such
as tert-butyl peroxide and the popular peroxycarboxylic acids.
Of the latter, although many are known, only a few are in
Table 1 Alkenes epoxidised with 5-hydroperoxycarbonylphthalimide 1
5-PCP)a
(
Alkene
T/°Cb
Solvent
t/hc
2
,3-Dimethylbut-2-ene
20
20
20
CH
CH
CH
2
2
2
Cl
Cl
Cl
2
2
2
2
6
2
3
2
2
Cyclohexene
1
-Methylcyclohexene
2
0
Toluene
Cyclooctene
a-Pinene
20
20
20
20
35
35
35
60
35
60
0
CH
CH
CH
2
2
2
Cl
Cl
Cl
2
2
2
6
general use. Thus, peroxyformic, peroxyacetic, peroxyben-
b-Pinene
2.5
2 + 4
24
24
24
113
24
150
7
zoic, peroxyphthalic (also as its magnesium salt), peroxytri-
fluoroacetic and m-chloroperoxybenzoic acids are the ones
most frequently employed. Generally, the relatively cheap,
lower molecular mass, liquid peroxyformic and peroxyacetic
acids are used for large-scale industrial epoxidation and the
higher molecular mass, solid m-chloroperoxybenzoic acid is
widely employed for research or very small scale purposes.
There are serious disadvantages to the use of peroxyformic and
peroxyacetic acids, including their difficulty of synthesis in
high concentration, their lack of stability, the need for buffering
of a reaction medium to lessen acid-induced side reactions and
the difficulty of recycling the parent acid. m-Chloroperoxy-
benzoic acid is not available commercially in large quantities or
in pure form because of its easy thermal decomposition and
sensitivity to mechanical shock;7 the peroxyacid is also
expensive because of the cost of the starting acid and the need
to use methanesulfonic acid to catalyse its formation from
hydrogen peroxide. Nevertheless, m-chloroperoxybenzoic acid
does have desirable properties, in being capable of effecting
many kinds of epoxidation and in giving a product acid that is
relatively insoluble in non-polar solvents. The present research
has shown that 5-hydroperoxycarbonylphthalimide 1 (5-PCP;
Limonene
CH Cl2
d
2
Oct-1-ene
CH
CH
CH
ClCH
CH Cl
ClCH
CH Cl
2
2
2
Cl
Cl
Cl
2
cis-Stilbene
trans-Stilbene
Allyl phenyl ether
Styrene
2
2
2
CH
2
Cl
Cl
2
2
Diallyl maleate
6
1
2
CH
2
-Methylhept-5-en-2-one
,2-Diallyloxyethane
2
2
55
EtOAc
70
a
To conc. H SO4 (29.4 g; 98% w/w) was added water (1.7 g;
demineralised), with stirring and cooling, followed by hydrogen peroxide
(3.66 g; 85% w/w; 0.0915 mol), maintaining the reaction temperature below
5 °C. After addition had been completed, the mixture was warmed to
0 °C, at which stage 1,3-dihydro-1,3-dioxoisoindole-5-carboxylic acid
ref. 10) (5-carboxyphthalimide 2; 5 g) was added. The mixture was stirred
at 40 °C for 45 min, after which it was cooled and quenched by addition of
ice–water. 5-Hydroperoxycarbonylphthalimide 1 (5-PCP) crystallised out.
The product was filtered off, washed normally with water to pH 3 and air-
dried to give the pure compound in > 90% yield, mp 175 °C. b In a typical
2
1
4
(
reaction (Scheme 1), 1-methylcyclohexene (48 mg; 0.5 mmol) in CH Cl (2
2
2
ml) was stirred with 5-PCP (0.12 g; 0.6 mmol) at room temperature for 2 h.
The course of the reaction was monitored by GC and by GC–MS. At the end
of reaction, the precipitated 5-carboxyphthalimide 2 was filtered off, the
solvent was evaporated and the product was weighed and examined by H
NMR spectroscopy. All the epoxides are known and authentic materials
were used for comparison with the epoxides produced with 5-PCP. Yields
were between 98–100%. Approximate time to complete reaction, as
revealed by the disappearance of the starting alkene. d With 1 equiv. of
-PCP, only the 1,2-mono(epoxide) was isolated after 2 h. Addition of a
1
1
,3-dioxo-1,3-dihydroisoindole-5-peroxycarboxylic
acid;
8
Scheme 1) is an excellent, cheaper reagent for effecting
epoxidation, ameliorating or overcoming many of the disadvan-
tages listed above. Its preparation is described in a footnote to
Table 1.
c
5
The alkenes shown in Table 1 were epoxidised at tempera-
tures between 0 and 60 °C in various solvents (Scheme 1). The
further 1 equiv. of 5-PCP at this stage gave the 1,2:9,10 bis(epoxide) after
a further 4 h.
Chem. Commun., 1998
429

View Article Online
alkenes were chosen to explore the general suitability of 5-PCP
for epoxidation. Thus, the reactivity of the alkenes towards
electrophilic oxygen ranged from the electron-rich 2,3-di-
methylbut-2-ene (containing a tetraalkyl-substituted double
bond) to the electron-deficient allyl phenyl ether (containing a
monosubstituted, oxygen-deactivated double bond). Other al-
kenes were chosen to demonstrate the stability of the resulting
epoxide under the reaction conditions (styrene, a-pinene,
means that the hydrogen ion concentration in solution is
extremely low. In turn, this means that acid-catalysed changes
in the epoxide, when formed, are reduced to a very low level,
enabling the isolation of even acid-sensitive epoxides in high
yield and purity. 5-PCP itself is partly soluble in the solvents
described here. This implies it has all of the kinetic advantages
of a homogeneous reaction but, being a very weak acid, does not
produce a significant hydrogen ion concentration in the reaction
medium. Thus, there is little or no effect from acid-catalysed
decomposition of the product epoxide. In the most sensitive
epoxides prepared in this work (e.g. styrene oxide), reaction
proceeded to virtually 100% yield with no evidence for acid-
catalysed rearrangement products (phenylacetaldehyde in the
case of styrene oxide). Until full hazard information is
available, 5-PCP should be treated with caution, as with any
peroxy compound.
1,2-diallyloxyethane), the stereo integrity of reaction (cis- and
trans-stilbene) or the selectivity of reaction (limonene, having
two different double bonds).
In each case, epoxidation proceeded to give virtually a 100%
yield of very pure product epoxide. Reaction rates were as
expected from the known mechanism of epoxidation with
peroxyacids.1 Thus, at room temperature, reaction of 2,3-di-
methylbut-2-ene, 1-methylcyclohexene, cyclohexene, cyclo-
,6
octene and a- and b-pinene with 5-PCP was fast in CH
the less reactive alkenes, such as oct-1-ene, allyl phenyl ether,
,2-diallyloxyethane and diallyl maleate, higher temperatures
were needed to achieve reaction in a reasonable time. Such
reactions could be carried out in CH Cl under reflux (35 °C),
,2-dichloroethane (at 50 °C), EtOAc (at 55 °C) and toluene (at
0 °C). When the epoxidation was carried out at higher
2 2
Cl . For
Notes and References
1
†
E-mail: rj05@liv.ac.uk
2
2
1
5
1 See for example, M. Hudlicky, Oxidations in Organic Chemistry, ACS
Monograph 186, American Chemical Society, Washington DC, 1990,
pp. 60–65; J. March, Advanced Organic Chemistry, Wiley, New York,
temperatures than these, reaction was faster and a clean product
was still obtained, but there was some small loss of active
oxygen over long periods of time. In these cases, somewhat
more than a 1 equiv. of peroxyacid to each double bond was
needed. Thus, after a period of about 100 h at 60 °C with 2
equiv. of 5-PCP, epoxidation of the two allylic bonds in
diallylmaleate ceased at about 60–70% because of depletion of
the peroxyacid; addition of a further 1 equiv. (0.5 equiv. for
each bond) caused the reaction to proceed to completion and to
give a very pure product.
1
985, pp. 733–735.
C. Venturello, M. Abneri and M. Ricci, J. Org. Chem., 1983, 48,
831.
2
3
3
4
R. Landau, G. Sullivan and D. Brown, Chem. Technol., 1979, 9, 602.
A. Tuel and Y. B. Taarit, J. Chem. Soc., Chem. Commun., 1994,
1
667.
5
A. M. d’A Rocha Gonsalves, R. A. W. Johnstone, M. M. Pereira and
J. Shaw, J. Chem. Soc., Perkin Trans. 1, 1991, 645.
6 D. Swern, in Organic Peroxides, ed. D. Swern, Wiley-Interscience,
New York, 1970, vol. 1, pp. 313–374 and see ref. 1 above.
7
M. Hirano, S. Yakabe, A. Satoh, J. H. Clark and T. Morimoto, Synth.
Commun., 1996, 26, 4591; T. Mino, S. Masuda, M. Nishio and
M. Yamashita, J. Org. Chem., 1997, 62, 2633.
In all of the reactions, the product acid (5-carboxyphthali-
mide 2) was almost totally insoluble in the solvent. This
behaviour is highly advantageous in two ways. First, the
product acid can be filtered off easily from the reaction medium
and re-used; this potentially low-cost acid can be recycled with
hydrogen peroxide to give little or no environmentally dis-
advantageous by-product. Second, the insolubility of the acid,
due to extensive strong intermolecular hydrogen-bonding,9
8
9
J. P. Sankey and A. P. James, (Interox Chemicals), PCT Int. Appl., WO
9
1 09,843, 11th July, 1991; Chem. Abstr., 1991, 115, 282510h.
N. Feeder and W. Jones, Acta Crystallogr., Sect. C, 1994, 50, 824.
1
0 W. H. Perkin and J. F. S. Stone, J. Chem. Soc., 1925, 127, 2295.
Received in Cambridge, UK, 13th November 1997; 7/08179K
430
Chem. Commun., 1998