dichloroethane. 25 mmol of aqueous H2O2 was then added and the two-
type of Q cation had little effect on the epoxide yield and the rate
of reaction, except for (n-Bu)4N+, in which case the catalyst was
poorly soluble in the system. As H2O2 was used up in the
reaction, the catalyst precipitated as a white sediment. Similar
behaviour has been noted before.2h Addition of fresh H2O2
redissolved the catalyst and restarted the reaction. This could be
done several (3–4) times without breaking the ME. Hence, the
ME system allows reuse in situ by adding more hydrogen
peroxide, albeit with gradually reduced activity because of
water accumulation. Our attempts to reuse the catalyst by
filtering it off were unsuccessful—the isolated catalyst did not
dissolve in fresh ME. Therefore, for continuous processing,
water that forms in the reaction must be continuously removed
from the reaction system. This was attempted using ultra-
filtration.
The micellar enhanced ultrafiltration of the reaction mixture
was performed using a stirred batch filtration unit with Nadir®
membranes (4–10 kDa) under 5 bar nitrogen pressure. The
ultrafiltration retained ca. 90% of the catalyst behind the
membrane (from W analysis by ICP) and allowed removal of
by-product water from the reaction system. The epoxide
product could then be isolated from the filtrate, e.g. by
distillation or other methods discussed elsewhere.3 Hence, the
use of this ME system coupled with ultrafiltration shows the
potential for the development of continuous POM-catalysed
epoxidation of olefins with hydrogen peroxide.
phase mixture was heated to a certain temperature (50–70 °C). The reaction
was monitored by GC analysis performed on a Varian CP3800 chromato-
graph equipped with a 30 m ZB-WAX capillary column from Zebron. In
microemulsion, the epoxidation was carried out as follows. The catalyst
solution in chloroform was introduced into a round-bottomed flask, and the
solvent was removed under vacuum. The amount of catalyst was measured
by weighing the flask after solvent removal. 1.50 g Brij® 30 and 3.00 g (27
mmol) 1-octene were added to the catalyst. The flask was shaken until the
catalyst dissolved. The mixture was brought to a specified temperature
(room temperature or 50 °C), then 0.50 g (4.0 mmol) 27 wt% aqueous H2O2
was added. Samples were taken at appropriate intervals and analysed by GC
adding n-decane as a standard. Addition of n-decane to the reacting mixture
was undesirable to avoid breaking the microemulsion. Micellar enhanced
ultrafiltration of the reaction mixture was performed using a stirred batch
filtration unit with Nadir® membranes (4–10 kDa, 45 mm diameter) under
5 bar nitrogen pressure. The membranes were kindly provided by Nadir
Filtration. The phase diagram for the ternary system 1-octene–27% H2O2–
Brij 30 was obtained at 50 °C by adding 27% H2O2 aqueous solution
dropwise in 0.2 ml portions to 1-octene–Brij 30 mixtures of a 1+9 to 9+11
weight ratio, followed by visual inspection of the resulting system.
1 K. Weissermel and H.-J. Arpe, Industrial Organic Chemistry, 3rd edn.,
VCH, Weinheim, 1997.
2 (a) C. Venturello, R. D’Aloisio, J. C. J. Bart and M. Ricci, J. Mol. Catal.,
1985, 32, 107; (b) Y. Ishii and M. Ogawa, in Reviews on Heteroatom
Chemistry, Vol. 3, ed. A. Ohno and N. Furukawa, MY, Tokyo, 1990, p.
121; (c) L. Salles, C. Aubry, R. Thouvenot, F. Robert, C. Doremieux-
Morin, G. Chottard, H. Ledon, Y. Jeannin and J. M. Brégeault, Inorg.
Chem., 1994, 33, 871; (d) R. Neumann and A. M. Khenkin, J. Org.
Chem., 1994, 59, 7577; (e) D. C. Duncan, R. C. Chambers, E. Hecht and
C. L. Hill, J. Am. Chem. Soc., 1995, 117, 681; (f) N. M. Gresley, W. P.
Griffith, A. C. Laemmel, H. I. C. Nogueira and B. C. Parkin, J. Mol.
Catal. A, 1997, 117, 185; (g) I. V. Kozhevnikov, G. P. Mulder, M. C.
Steverink-de Zoete and M. G. Oostwal, J. Mol. Catal. A, 1998, 134, 223;
(h) Z. Xi, N. Zhou, Y. Sun and K. Li, Science, 2001, 292, 1139.
3 M.-J. Schwuger, K. Stickdorn and R. Schomäcker, Chem. Rev., 1995, 95,
849.
4 (a) P. Erra, C. Solans, N. Azemar, J. L. Parra, M. Clausse and D. Touraud,
Progr. Colloid Polym. Sci., 1987, 73, 150; (b) C. Larpent and H. Patin, J.
Mol. Catal., 1992, 72, 315; (c) J.-M. Aubry and S. Bouttemy, J. Am.
Chem. Soc., 1997, 119, 5286; (d) L. J. P. van den Broeke, V. G. de Bruijn,
J. H. M. Heinen and J. T. F. Keurentjes, Ind. Eng. Chem. Res., 2001, 40,
5240.
The authors thank the EPSRC, UK for support (grant GR/
N06762).
Notes and references
†
Experimental. All chemicals were purchased from Aldrich or Fluka and
used as received, except for 1-octene which was distilled prior to use. The
peroxo complexes Q3{PO4[WO(O2)2]4} were prepared according to the
literature method2a using Aliquat 336® (mainly CH3(n-C8H17)3NCl),
CH3(n-C8H17)3NCl or (n-C8H17)4NCl as a surfactant. The complexes were
stored as solutions in chloroform in a stoppered glass bottle at 4 °C. The
biphasic epoxidation of 1-octene was carried out as described by Venturello
et al.2a 2 ml of chloroform solution containing the catalyst, 50 mmol of
octene and 1 mmol n-decane (internal standard) were added to 10 ml
CHEM. COMMUN., 2003, 714–715
715