Reaction in biphasic water/organic solvent system in the presence of surfactant: Inverse phase transfer catalysis versus interfacial catalysis

Article abstract of DOI:10.1016/S0040-4020(99)00994-1

Unexpected results obtained during the study of the influence of surfactant concentration on chalcone epoxidation by H₂O₂ in a water/heptane two-phase system in the presence of a surfactant (DTAB) led us to reconsider the catalytic mechanism of this reaction. Two stirring rates experiments, either low speed (i.e, 100 rpm) or high speed (i.e. 1200 rpm), lead us to discuss kinetic results on the basis of two competitive catalytic processes: an Inverse Phase Transfer Catalysis (IPTC) or an Interfacial Catalysis (IC).

Full text of DOI:10.1016/S0040-4020(99)00994-1

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 303–307
Reaction in Biphasic Water/Organic Solvent System in the
Presence of Surfactant: Inverse Phase Transfer Catalysis versus
Interfacial Catalysis
b
Bernard Boyer,a Araik Hambardzoumian,a Jean-Pierre Roquea, and Norair Beylerian
*
a
´
Laboratoire de Chimie Organique Physique, Universite Montpellier II, Place E. Bataillon, 34095 Montpellier, Cedex 5, France
bPhysical Chemistry Department, Yrevan State University, 1A. Manoukian Avenue, Yrevan, 375049 Armenia
Received 2 April 1999; accepted 3 November 1999
Abstract—Unexpected results obtained during the study of the influence of surfactant concentration on chalcone epoxidation by H2O2 in a
water/heptane two-phase system in the presence of a surfactant (DTAB) led us to reconsider the catalytic mechanism of this reaction. Two
stirring rates experiments, either low speed (i.e. 100 rpm) or high speed (i.e. 1200 rpm), lead us to discuss kinetic results on the basis of two
competitive catalytic processes: an Inverse Phase Transfer Catalysis (IPTC) or an Interfacial Catalysis (IC). ᭧ 1999 Elsevier Science Ltd. All
rights reserved.
Introduction
NPTC has been widely used to carry out a variety of reac-
tions such as substitution, C- and O-alkylation, oxidation
and dichloropropanation reactions,1,2 in simple and cheap
conditions.
The liquid–liquid phase transfer catalysis, discovered by
Jarousse1 in 1951, is a process favouring reactions between
a lipophilic substrate, dissolved in an organic phase, with a
hydrophilic reactant solubilized in water.
Another way to effect liquid–liquid phase transfer catalysis
consists of transferring the lipophilic substrate into water
where it reacts with the hydrophilic reactant. This more
recent approach has been called the ‘Inverse Phase Transfer
Catalysis’ (IPTC) by Mathias.3 Several methods have been
developed in order to transport the lipophilic substrate into
the water phase, such as:
The potential of this catalysis type applied to reactions
carried out in biphasic medium were really emphasized by
the work of Starks, Liotta, Dehmlow and Makosza, which
began in the 1970s.2 This catalytic procedure is still exten-
sively used in organic synthesis. Two types of liquid–liquid
phase transfer catalysis appear from a literature survey.
Indeed, the reaction can occur either in the organic phase
or in the aqueous one.
1. The temporary and reversible conversion of the lipophilic
substrate into a highly water-soluble ionic inter-
mediate.3–6
2. The complexation of a lipophilic substrate by chemically
We will name the first case ‘Normal’ Phase Transfer
Catalysis (NPTC); it is the most widely used. It consists
of transferring an anionic reactant from the water phase
via a lipophilic ion pair, to the organic phase where the
reaction with the lipophilic substrate takes place. The lipo-
philic ion pair generally corresponds to an anionic reactant
associated to either of the following:
modified cyclodextrins7 or by water-soluble calixarenes.8
Although these methods enhance the reactions rates in
biphasic medium, these present two major disadvantages,
i.e. the lack of general character and/or the use of costly and
not easily accessible catalysts.
Contrary to the above examples, we recently explored a
general method dealing with the IPTC. It is based upon
the fact that any lipophilic substrate could be transferred
into the aqueous phase by means of hydrosoluble surfac-
tants. Indeed, it is well known that micelles formed in the
aqueous phase, above the cmc, are able to solubilize a frac-
tion of the substrate in equilibrium with the organic phase.
The reaction takes place at the surface of the micelles in the
water phase. The reaction product, normally lipophilic, is
transferred into the organic phase (Fig. 1).
1. A lipophilic quaternary onium (tetraalkyl ammonium ion,
tetraalkylphophonium) by exchanging its counterion.
2. Its own counterion which has been made lipophilic by
complexation with a crown ether.
Keywords: epoxidation; surfactant; phase-transfer catalysis; interface; stir-
ring rate.
* Corresponding author. Fax: ϩ33-4-67-14-38-88; e-mail: bboyer@ univ-
montp.fr
0040–4020/00/$ - see front matter ᭧ 1999 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(99)00994-1

304
B. Boyer et al. / Tetrahedron 56 (2000) 303–307
Figure 1. Inverse Phase Transfer Catalysis principle involving a surfactant as catalyst.
In fact, two catalytic processes are involved: (i) an inverse
phase transfer—the surfactant allows the transfer of the
lipophilic substrate into the water phase; (ii) a micellar cat-
alysis—the charged transition state formed at the micelle
organic phase: [chalcone]0ˆ0.1 M in heptane;
aqueous phase: [H2O2]0ˆ0.5 M; [NaOH]0ˆ0.5 M;
[DTAB] varying from 500 to 1 mM.
surface can be stabilized by the surfactant ionic head groups
At different times, an aliquot of the biphasic medium was
diluted 100 times in heptane. The absorbance of the solution
was measured at 300 nm, as chalcone was the only species
absorbing at this wavelength (1ˆ22400 MϪ1 cmϪ1 in hep-
tane).12 Absorbance variations for each DTAB concen-
tration, and without DTAB, are represented in Fig. 2.
only in the situation where the charges are opposite.9
We proved that this new type of IPTC is of interest
especially in the reduction of highly hydrophobic ketones
by sodium borohydride10 and in the epoxidation of
a,b-unsaturated ketones by hydrogen peroxide.11 Both reac-
tions were carried out in a water/heptane biphasic medium.
As expected, the decrease of the surfactant concentration
markedly reduces the reaction rate. Indeed, the lower its
concentration, the lower its solubilizing power as well as
its micellar catalysis efficiency. However, it was surprising
to observe catalytic effects with a surfactant concentration
as low as 1mM even though the presence of 0.5 M NaOH
lowers the cmc of DTAB (a cmc value of 0.64 mM in 0.5M
NaOH was measured, while it is equal to 14.4 mM in pure
water).10 Obviously, at 1 mM in DTAB, the micellar
concentration together with the resulting chalcone solubiliz-
ing capacity seemed too low to induce a significant IPTC
process. Under vigorous stirring an emulsion was observed;
we assumed that the large interface separating the two
phases would play an important role.
In order to optimize this last reaction we have examined the
influence of the surfactant concentration. Results observed
and presented herein, led us to re-examine the IPTC process
described above.
Results and Discussion
We have studied the influence of the dodecyltrimethyl-
ammonium bromide (DTAB) concentration on the rate
epoxidation of the chalcone, which was used as a model.
The reaction was carried out by vigorously stirring two
equal volumes of the following phases:
To check this point, we repeated the chalcone epoxidation
Figure 3. Inverse Phase Transfer Catalysis: absorbance variations
(lˆ300 nm) versus time for the epoxidation of chalcone at several
DTAB concentrations in biphasic media (heptane/water 1:1) under slow
stirring (100 rpm).
Figure 2. Interfacial Catalysis: absorbance variations (lˆ300 nm) versus
time for the epoxidation of chalcone at several DTAB concentrations in
biphasic medium (heptane/water 1/1) under vigorous stirring (1200 rpm).

B. Boyer et al. / Tetrahedron 56 (2000) 303–307
305
Table 1. Initial rates of the epoxidation reaction of chalcone versus DTAB concentration at two stirring rates (100 and 1200 rpm)
[DTAB] (mM)
500
250
100
50
25
10
r0 (fast)
1:5×10Ϫ1
2:0×10Ϫ2
1:1×10Ϫ1
8:0×10Ϫ2
6:2×10Ϫ2
2:2×10Ϫ2
5:2×10Ϫ4
rr0el (fast)
r0 (slow)
rr0el (slow)
288
212
154
119
42
1
5:0×10Ϫ3
6:0×10Ϫ4
2:0×10Ϫ4
6:1×10Ϫ5
1:3×10Ϫ5
1538
385
46
15
4.7
1
reaction under the same experimental conditions but
without vigorous stirring. We applied slow magnetic
stirring, just enough to avoid a concentration gradient in
each phase.
in heptane. That means that under the conditions used
dodecyltrimethylammonium is not able to transfer into
heptane with either bromine anion or hydroperoxide
anion.
2. Tetrabutylammonium hydrogenosulfate (TBAHS), well
known as an efficient normal phase transfer catalyst, was
tested to carry out chalcone epoxidation (4 ml of 0.1 M
chalcone in heptane and 4 ml of water containing 0.5 M
H2O2, 0.5 M NaOH and 0.1 M TBAHS). The percentages
of consumed chalcone at tˆ2 h were only 5 and 25%
under slow stirring (100 rpm) and vigorous stirring
(1200 rpm), respectively. In both cases, the reaction is
much slower than when DTAB is used at the same
concentration (see Figs. 1 and 2). Obviously, DTAB,
which is more efficient than TBAHS, acts in a different
way.
3. Sodium dodecylsulfate (SDS), an anionic surfactant, was
also evaluated in the same conditions as above using
TBAHS ([SDS]ˆ0.1 M). As already observed in the
reduction of hydrophobic ketones by sodium boro-
hydride,10 SDS catalysed the epoxidation reaction
under either slow or high stirring speed. The percentages
of epoxide formed at tˆ2 h were 17 and 31%, respec-
tively. It appears that SDS is a better catalyst than
TBAHS. However, it is clear that SDS cannot catalyse
the reaction by transferring HOOϪ either at the interface
or into heptane, as does TBAHS.
In such conditions an emulsion is not formed and the inter-
face area between the two phases is equal to the section area
of the used flask. Measuring the absorbance of diluted
aliquots of the upper heptane phase monitored the reaction.
Variations of the optical density versus time are plotted in
Fig. 3 at several DTAB concentrations.
It appears that the reaction is strongly accelerated when
DTAB concentration increases up to 500 mM. However,
the reaction rate is not significantly modified if the surfac-
tant concentration is ranging from 0 to 10 mM. We thought
that under slow stirring conditions the results fit in with
IPTC as presented above (Fig. 1). The higher the surfactant
concentration, the higher the solubilizing power and then
the more efficient the catalytic effect.
However, another way to explain the observed catalytic
effect could consist of the transfer of the HOOϪ reactive
anion in the organic phase via ion pairing with the dodecyl-
trimethylammonium cation, according to a normal PTC
process. To distinguish between these two possibilities we
completed the three following experiments:
All these results agree with the competition between two
catalytic processes, the predominant one depending upon
the stirring conditions.
1. Fifteen millilitres of heptane were equilibrated with
15 ml of water containing 0.5 M H2O2, 0.5 M NaOH
and 0.1 M DTAB. After 4 h the two phases were sepa-
rated and 10 ml of the heptane phase was dry evaporated
under vacuum. No detectable traces of solute were found
Finally, for each surfactant concentration, the initial rate (r0)
of the epoxidation reaction was determined under fast
(1200 rpm) and slow (100 rpm) stirring rates. Values are
reported in Table 1.
From this table, it is worthwhile pointing out the following
observations:
1. Whatever the stirring speed, increase of the surfactant
concentration enhances the rate of the epoxidation
reaction.
2. The initial rates ratio of this reaction realized in the same
surfactant concentrations but with different stirring speed
decreases when the DTAB concentrations increases. In
other words, initial rates performed by IPTC or IC get
closer as the surfactant concentration increases.
The relative increase of initial rates of the epoxidation
reaction performed with slow stirring (100 rpm) or with
fast stirring (1200 rpm) compared with the initial rate
measured at [DTAB]ˆ10 mM is represented in Fig. 4.
Figure 4. Relative increase of initial rates versus the surfactant concentra-
tion for the epoxidation reaction realized by IPTC or IC. [chalcone]ˆ0.1 M;
[NaOH]ˆ0.5 M; [H2O2]ˆ0.5 M; 25ЊC.
From the plots presented in Fig. 4, we note that the influence
of surfactant concentration on the reaction rate is more

306
B. Boyer et al. / Tetrahedron 56 (2000) 303–307
Scheme 1.
important in the case of IPTC than with IC. That agrees with
the difference of mechanism for these two processes. Thus,
in the case of the IPTC, the reaction rate is determined by
the micellar concentration in the aqueous phase; conse-
quently, the increase of the surfactant concentration strongly
enhances the reaction rate. In the case of IC the reaction
takes place at the interface. To cover this surface with a
monomolecular layer, only very weak quantities of surfac-
tant are needed.13 As soon as this surface is covered,
the reaction rate is no longer influenced by surfactant
concentration.
monophasic aqueous conditions.14 Surprisingly, when used
in
a biphasic medium (water/heptane), polymerized
micelles were less efficient10 than their non-polymerized
analogues in catalysing the reduction of hydrophobic
ketones by sodium borohydride under vigorous stirring.
At the same time, we observed that polymerized species
have no effect upon interfacial tension15 because they are
unable to form a monolayer at the interface. As a conse-
quence, even under vigorous stirring, they do not produce
emulsion.
So it appeared that under slow-speed stirring, catalysis
occurs mainly through an IPTC process. The surfactant
operates both in transferring the substrate in the aqueous
phase via micelles (provided micellar concentration is
high enough) and in stabilizing the transition state by elec-
trostatic interaction (micellar catalysis). On the contrary,
under vigorous stirring a large interfacial area results from
the formation of an emulsion. This allows a stronger
encounter probability of the lipophilic substrate with the
hydrophilic reactant. The reaction takes place mainly at
the interface. The surfactant reduces the interfacial tension,
increases the interfacial area and again stabilizes the
negative transition state formed, at the interface, close to
the ammonium groups of the surfactant. It is an Interfacial
Catalysis (IC) process. In this case, a small amount of
surfactant is enough to promote the catalytic pathway.
Under high-speed stirring, all these results lead to the
following conclusions:
• With polymerized surfactants, only the IPTC process is
involved. This explains their lower efficiency as catalysts
under biphasic conditions, despite their higher aptitude to
solubilize hydrophobic substrates in water and to
increase the micellar catalysis.
• On the contrary, ‘normal’ non-polymerized surfactants
operate via the more efficient IC process. The IPTC
process may also intervene, but only as a minor comple-
mentary pathway.
Further examination of the proposed Interfacial Catalytic
(IC) process is underway, along with its application to
additional reaction.
A question arises in the case where slow stirring is applied.
Actually, a mass transfer involved by IPTC could be the rate
limiting step, as the stirring speed is low. To answer this
question, we carried out the reaction in the following mono-
phasic conditions, in water:
Experimental
Dodecyltrimethylammonium bromide (DTAB), chalcone,
sodium hydroxide and 30% hydrogen peroxide were
obtained from commercial sources. UV studies were
performed on a SAFAS-170 spectrophotometer.
[H2O2]0ˆ0.5 M; [NaOH]0ˆ0.5 M; [chalcone]0ˆ0.01 M
and [DTAB]ˆ0.1 M.
General procedure for epoxidation reactions
The epoxidation is very quickly achieved (less than 1 min).
Obviously, under biphasic conditions, mass transfer contri-
butes to limit the overall transformation rate. Nevertheless
the reaction takes place mainly within the aqueous phase as
an increase of the micellar concentration enhances the
reaction rate.
Reactions in two-phase media were studied according to the
following protocol. Four millilitres of the organic phase
(0.1 M chalcone in heptane) were added to 4 ml of the
aqueous phase. The aqueous phase was prepared just before
the beginning of the reaction by mixing together 2 ml of
1.0 M H2O2, 1 ml of 2.0 M NaOH and 1 ml of either
water or aqueous solution of DTAB at concentrations
ranging from 1 to 500 mM. The mixture is quickly stirred
at the appropriate speed (100 or 1200 rpm) in a jacketed
reactor thermostated at 25ЊC. At a given time, 20 ml of the
emulsion (1200 rpm) or 10 ml of the upper heptane phase
(100 rpm) were withdrawn and poured into a tube contain-
ing 2 ml of heptane. In the first case the tube was shaken and
allowed to stand for a few minutes in order to separate the
small amount of aqueous phase from the organic one. In
both cases the UV absorbance of the heptane solution was
It is worthwhile mentioning that a previously reported
behaviour of polymerized surfactants10 sustains our hypo-
thesis. These polymeric species were obtained by polymeri-
zation under g-irradiation of aqueous micellar solution of
v-unsaturated tensides (Scheme 1).
We observed that polymerization increased the solubilizing
power of a hydrophobic substrate in water.14 Moreover
polymerized micelles were more efficient than their non-
polymerized analogues in promoting micellar catalysis in

B. Boyer et al. / Tetrahedron 56 (2000) 303–307
307
B.; Hambardzoumian, A.; Roque, J. P.; Beylerian, N. Tetrahedron
1999, 55, 6147.
measured in a cuvette (0.1 cm) equilibrated cell holder
(25ЊC) of a SAFAS-170 spectrophotometer.
12. To confirm the relevance of spectrometric observations,
several experiments were achieved. In particular: (a) Epoxide
concentration can be monitored at 245 nm (eˆ16973 in heptane)
concurrently with the ketone concentration measured at 300 nm.
For a given kinetic, it is worthwhile to mention that half-life times
are very close for chalcone disappearance and epoxide formation,
excluding a significant accumulation of any intermediate. This
result is also confirmed by the observation of isosbestic points
when a series of spectra recorded versus time are superimposed.
(b) As chalcone could react with OHϪ giving a retrograde aldol
reaction,16 we tested the contribution of this side-reaction in the
absence of hydrogen peroxide under the following conditions:
[chalcone]0ˆ0.1 M; [DTAB]ˆ0.1 M; [NaOH]0ˆ0.03 or 0.50 M;
stirring speedˆ100 or 1200 rpm. All results reported in our
paper were obtained with 0.5 M NaOH, but because of the acidic
character of H2O2 the initial OHϪ concentration can be evaluated
as close to 0.03 M according the pKa of H2O2 (11.13).16 Whatever
the stirring speed, with [NaOH]0ˆ0.03 M, no significant spectral
change is detectable over a period of 4 h. Within the same period of
time and with [NaOH]0ˆ0.5 M, the percentage of chalcone trans-
formed is only 10 and 18% at 100 and 1200 rpm stirring speed,
respectively. It can be considered that the side-reaction of chalcone
with [OHϪ] is quite slow compared with the epoxidation rate as in
the presence of 0.5 M H2O2, [OHϪ] being always lower than 0.5 M.
13. At 1200 rpm, the surfactant quantity necessary to stabilize the
interface is very weak. A simple calculation shows that 4:3×1014
molecules are sufficient to cover 1 cm2 of the surface.17 For a
10 mM DTAB concentration, there are in 2 ml of the aqueous
phase 6:02×1021 molecules able to cover, in theory,
1:4×107 cm2: Clearly any surfactant increase cannot proportionally
increase the interface area. At this level of concentration, this latter
is certainly more influenced by the stirring rate than by the surfac-
tant concentration.
The concentration of chalcone was thus monitored as a
function of time. Each absorbance value reported is the
mean of at least three independent readings differing by
less than 5%.
References
1. Jarousse, M. J. C. R. Acad. Sci. Paris, Ser. C 1951, 232, 1424.
2. Starks, C. M.; Liotta, C. L. Phase Transfer Catalysis: Principle
and Techniques; Academic: New York, 1978; Brandstrom, A. Adv.
Phys. Org. Chem. 1977, 15, 267; Dehmlow, E. V.; Dehmlow, S. S.
Phase Transfer Catalysis, 2nd ed., Verlag Chemie: Weinheim,
1983; Freedman, H. H. Pure Appl. Chem. 1986, 58, 857; Makosza,
M.; Fedorynski, M. Adv. Catal. 1985, 35, 375; Harada, A. J. Synth.
Org. Chem. Jpn. 1990, 48, 517; Goldberg, Y. Phase Transfer
Catalysis. Selected Problems and Applications; Gordon: Berk-
shire, 1992, pp. 359–366; Starks, C. M.; Liotta, C. L.; Hacpern,
M. Phase Transfer Catalysis: Fundamentals, Applications, and
Industrial Perspectives; Chapman: London, 1994, pp. 179–183.
3. Mathias, L. J.; Waidya, R. A. J. Am. Chem. Soc. 1986, 108,
1093.
4. Fife, W. K.; Xin, Y. J. Am. Chem. Soc. 1987, 109, 1278.
5. Kuo, C. S.; Jwo, J. J. J. Org. Chem. 1992, 57, 1991.
6. Wamser, C. C.; Yates, J. A. J. Org. Chem. 1989, 54, 150.
7. Trifonov, A. Z.; Nikiforov, T. T. J. Mol. Catal. 1984, 24, 15;
Tanaka, N.; Yamaguchi, A.; Araki, Y.; Araki, M. Chem. Lett.
1987, 715; Deratani, A.; Lelievre, G.; Maraldo, T.; Sebille, B.
Carbohydr. Res. 1989, 192, 215.
8. For the use of calixarenes as normal phase transfer-catalysts,
see Taniguchi, H.; Otsuji, Y.; Nomura, N. E. Bull. Chem. Soc. Jpn.
1995, 68, 3563; Nomura, E.; Taniguchi, H.; Kawaguchi, K.; Otsuji,
Y. J. Org. Chem. 1993, 58, 4709; Araki, K.; Yanagi, A.; Shinkai, S.
Tetrahedron 1993, 49, 6763; Shimizu, S.; Kito, K.; Sasaki, Y.;
Hirai, C. Chem. Commun. 1997, 1629.
14. Boyer, B.; Lamaty, G.; Leydet, A.; Roque, J. P.; Sama, P. New
J. Chem. 1992, 16, 883 see also p. 887.
15. Boyer, B.; Durand, S.; Lamaty, G.; Moussamou-Missima,
9. Al-Lohedan, H.; Bunton, C. A.; Mhala, M. M. J. Am. Chem.
Soc. 1982, 104, 6654.
`
J. M.; Pavia, A. A.; Pucci, B.; Roque, J. P.; Rouviere, J. J. Chem.
Soc. Perkin Trans. II 1991, 1311.
10. Boyer, B.; Betzer, J. F.; Lamaty, G.; Leydet, A.; Roque, J. P.
New J. Chem. 1995, 19, 807.
11. (a) Boyer, B.; Hambardzoumian, A.; Lamaty, G.; Leydet, A.;
Roque, J. P.; Bouchet, P. New J. Chem. 1996, 20, 985. (b) Boyer,
16. Temple, R. D. J. Org. Chem. 1970, 35, 1275.
17. Bruice, T. C. J. Am. Chem. Soc. 1968, 90, 1333; Nogami, H.;
Hasegawa, J.; Iwatsuru, M. Chem. Pharm. Bull. 1970, 18, 2297.