Pseudophase Model for Reactivity in Reverse Micelles. The Case of the Water-Promoted Reaction between Two Lipophilic Reagents, Bromine and 1-Octene, in AOT Microemulsions

Article abstract of DOI:10.1021/jp960772a

Kinetic investigations of 1-octene bromination in AOT-isooctane-water microemulsions (13 / / tr, transfer coeffcients from isooctane to water, are 1.5E-5 and 8.8E-3 for alkene and bromine, respectively), bromination occurs in an aqueous microenvironment, as ilustrated by the high sensitivity of the bromination rate to the water content of the microemulsion.A kinetic pseudophase model describes the rate constant dependence on microemulsion composition satisfactorily by assuming competition between reactions at the interface and in the aqueous phase.Reasonable values for the coefficients of reagent partition between the interface and the two microphases and for the local bromination rate constants are obtained from the kinetic equations derived from the model.In particular, spectroscopically observed AOT-bromine complexation is in agreement with the high bromine concentration at the interface (K₂, bromine partition coefficient from isooctane to interface, = 6.8).The water-phase bromination rate constant, k_w = 1E8 M^-1 s^-1, is in the sanme range as that measured in bulk water.The lower limit for the interfacial rate constant, k_i, is 1E3 M^-1 s^-1, a value close to that observed in poorly aqueous methanol (MeOH/H2O = 95/5 v/v).These data are compared with those recently obtained in the same microemulsions for solvolysis, a reaction which, like bromination, is water-promoted but supposed to take place at the interface only.The results are discussed in terms of the chemical properties of the water molecules encased in the microemulsion droplets.

Full text of DOI:10.1021/jp960772a

12638
J. Phys. Chem. 1996, 100, 12638-12643
Pseudophase Model for Reactivity in Reverse Micelles. The Case of the Water-Promoted
Reaction between Two Lipophilic Reagents, Bromine and 1-Octene, in AOT Microemulsions
Iva B. Blagoeva,‡ Peter Gray,† and Marie-Franc¸oise Ruasse*,†
ITODYS, CNRS URA 34, UniVersite´ Paris 7-Denis Diderot, 1 rue Guy de la Brosse, 75005 Paris, France, and
Institute of Organic Chemistry, Bulgarian Academy of Sciences, ul. Academician G. BoncheV, bl. 9,
Sofia 1113, Bulgaria
ReceiVed: March 13, 1996X
Kinetic investigation of 1-octene bromination in AOT-isooctane-water microemulsions (13 e w ) [H2O]/
[AOT] e 24 and 6 e z ) [IO]/[AOT] e 57) shows that the reaction is first-order in alkene and first-order
in bromine, as usually found in protic media. Although both reagents are mainly located in the isooctane
phase (Ktr, transfer coefficients from isooctane to water, are 1.5 × 10-5 and 8.8 × 10-3 for alkene and bromine,
respectively), bromination occurs in an aqueous microenvironment, as illustrated by the high sensitivity of
the bromination rate to the water content of the microemulsion. A kinetic pseudophase model describes the
rate constant dependence on microemulsion composition satisfactorily by assuming competition between
reactions at the interface and in the aqueous phase. Reasonable values for the coefficients of reagent partition
between the interface and the two microphases and for the local bromination rate constants are obtained from
the kinetic equations derived from the model. In particular, spectroscopically observed AOT-bromine
complexation is in agreement with the high bromine concentration at the interface (K2, bromine partition
coefficient from isooctane to interface, ) 6.8). The water-phase bromination rate constant, kw ) 1 × 108
M-1 s-1, is in the same range as that measured in bulk water. The lower limit for the interfacial rate constant,
ki, is 103 M-1 s-1, a value close to that observed in poorly aqueous methanol (MeOH/H2O ) 95/5 v/v).
These data are compared with those recently obtained in the same microemulsions for solvolysis, a reaction
which, like bromination, is water-promoted but supposed to take place at the interface only. The results are
discussed in terms of the chemical properties of the water molecules encased in the microemulsion droplets.
Reverse micelles and w/o microemulsions1 are more and more
frequently used as microreactors2 for performing reactions in
constrained environments, taking advantage of reagent com-
partmentalization and of the particular properties of the encased
water.3-7 Nevertheless, the extension of these water-poor
microorganized media to a wide variety of reactions is still
limited by the paucity of relevant reactivity data. For example,
numerous organic reactions, the rate constants of which are
markedly higher in water than in less polar solvents,8,9 would
benefit from the hydrophilicity-hydrophobicity balance in
microemulsions. But the reactivity changes, resulting from the
difference3-7 in the structure of encased water compared to that
in network-organized bulk water, are far from being well-known
and still less well-controlled. In this respect, extensive kinetic
studies of reaction in reverse micelles and w/o microemulsions
are indispensable. However, these investigations can be inter-
preted in terms of reactivity, only if local reagent concentrations
and intrinsic rate constants in the various microphases of these
organized media can be obtained from the overall, apparent rate
data. Presently, the kinetic models10 that give access to these
intrinsic reactivity constants are still scarce and much less well-
hydrophobic reagent so that the reaction must necessarily occur
at the interphase.12,13 The local reagent concentrations and rate
constants are, therefore, readily obtained from a single-phase
model since the complete kinetic equation is greatly simplified.
More sophisticated reactions need to be investigated to validate
this complex kinetic model.
In this paper, we report results on the electrophilic bromi-
nation15 of 1-octene in AOT-isooctane-water microemul-
sions16 (AOT ) sodium bis(2-ethylhexyl)sulfosuccinate), which
involves reactions in at least two microphases. The two
reagents, bromine and the alkene, are highly hydrophobic and
are, therefore, mainly located in the oily microphase.17 How-
ever, this reaction needs water to occur,18 Scheme 1. The rate-
limiting ionization of the bromine-alkene charge transfer
complex must be electrophilically assisted by water molecules
in order to promote the departure of the leaving bromide. The
second, product-forming step is the nucleophilic trapping of the
bromonium ion either by its counter-bromide ion or by any
external nucleophile such as water. Consequently, it is expected
that bromination, if it occurs in AOT microemulsions, takes
place in an aqueous environment, i.e. at the interface and/or in
the water microphase. We show that the reaction does occur
in these w/o microemulsions, although the reagent concentra-
tions are very small in the aqueous microphases, and that the
pseudophase model describes the kinetic data fairly well when
two different reaction phases are taken into account.
documented than those for reactions in direct micelles.11
A
pseudophase model, analogous to that widely used in micellar
catalysis and which assumes competition between possible
reactions in three microphases (oil, water, and the interface),
has recently been proposed.12 However, its scope is presently
limited to reactions occurring a priori in one microphase
only.12,13 This is, in particular, the situation when the two
reagents are hydrophilic so that they can react together only in
the water phase14 or when a hydrophilic substrate reacts with a
Results
Table 1 summarizes the experimental second-order rate
constants, k2,app, of 1-octene bromination at 25 °C in AOT-
isooctane-water microemulsions of various compositions.
These data were obtained from the 1-octene concentration
dependence of the pseudo-first-order rate constants measured
by the stopped-flow technique, which ensured the rapid mixing
* To whom correspondence should be addressed.
† ITODYS, Paris, France.
‡ Bulgarian Academy of Sciences.
X Abstract published in AdVance ACS Abstracts, June 15, 1996.
S0022-3654(96)00772-1 CCC: $12.00 © 1996 American Chemical Society

Pseudophase Model for Reactivity in Reverse Micelles
J. Phys. Chem., Vol. 100, No. 30, 1996 12639
SCHEME 1
TABLE 1: Experimentala and Calculatedb Second-Order
Rate Constants for 1-Octene Bromination in AOT-
Isooctane-Waterc Microemulsions of Various Compositionsd
exp
2,app
calc
no.
wd
zd
[AOT]e
[H2O]e
k
k
2,app
1
2
13.3
13.3
13
57.8
27.7
18
0.099
0.19
0.28
0.38
0.588
0.19
0.27
0.37
0.372
0.56
0.19
0.27
0.36
0.54
0.35
0.52
1.32
2.53
3.64
5.05
7.82
3.02
4.24
5.99
5.91
8.90
3.78
5.32
7.16
10.80
8.40
12.48
38.8
80.6
40.01
81.45
116.4
170.2
279.3
92.82
138.5
199.3
198.1
328.3
110.3
164.9
243.3
410.7
294.9
498.3
3
112
4
13.3
13.3
15.9
15.7
16.2
15.9
15.9
19.9
19.7
19.9
20
12.1
166
282
94.9
5
6.68
27.7
18
12.2
12.1
6.76
27.7
18
12.1
6.71
12
6.67
6
7
140
207
216
328
109
174
256
388
276
502
8
9
Figure 1. Second-order plots for 1-octene bromination in AOT-
isooctane-water microemulsions: (O) [AOT] ) 0.38 M, [IO] ) 4.66
M, and [H2O] ) 4.22 M; (4) [AOT] ) 0.37 M, [IO] ) 4.50 M, and
[H2O] ) 5.98 M. For nonzero intercept, see text.
10
11
12
13
14
15
16
24
24
(correlation coefficient, r > 0.99). Therefore, 1-octene bromi-
nation is first-order in alkene. The slopes of these plots gave
the k2,app values collected in Table 1. It must be mentioned
that their intercepts are not zero, as would be expected for usual
bimolecular reactions. This comes from the fact that the
microemulsions contained bromine-consuming impurities. There-
fore, before the addition of the reagents, they were presaturated
with bromine. Consequently, a small portion of 1-octene, used
for the preparation of the alkene-containing microemulsion at
several concentrations, was consumed by the bromine excess
before the kinetic run itself.
Bromine-containing microemulsions were not stable over long
periods (1-2 h) due to a slow reaction of bromine with the
solvent. This reaction led to complex kinetic curves, typical
of radical chain reactions22 (incubation period, changes of kinetic
order, autocatalytic accelerations, etc.). However, this uncon-
trolled and slow bromine consumption did not interfere with
the much faster 1-octene bromination and did not perturb the
monoexponential signals described above. In agreement with
this finding, the addition of isoamyl nitrile, a radical scavenger,23
to the microemulsion did not modify the values of the
experimental rate constants.
a k2,app in M-1 s-1 (molar units) measured with a reproducibility on
4-6 runs better than (2%. b In M-1 s-1, calculated by eq 14. c Water
contained 0.1 M NaBr and 10-4 M HClO4. d w ) [H2O]/[AOT] and z
) [IO]/[AOT]. e In M in the total solution volume.
of the two microemulsions, each containing one of the reagents.
Bromine disappearance was monitored spectroscopically at 290
nm (ꢀ in the 104 range). Alkene concentrations were in the
range of 5 × 10-3 to 5 × 10-2 M in the total volume of the
microemulsion, while bromine was about 10-4 M. The kinetic
signals were always rigorously monoexponential; that is, the
reaction was clearly first-order in bromine. In particular, there
was no induction period on the reaction time scale (0.1-2 s),
which would suggest a slow mixing of the initial microemulsions
containing either the bromine or the alkene. As expected,
reagent exchange between the microdroplets (ke ) 109 M-1 s-1
)
was fast compared to the reaction time.19
The water phase of the microemulsions contained 0.1 M
sodium bromide so that the reaction could be monitored at 290
nm, where the tribromide ion, obtained from bromine by the
very fast equilibrium (1), absorbs strongly.20 With this proce-
The microemulsion composition, defined as usual by w )
[H2O]/[AOT] and z ) [IO]/[AOT] (IO, isooctane), was varied
from w ) 13 to 24, while z was between 6 and 57. Since the
aim of this work was to confirm experimentally the ability of
a pseudophase model to describe quantitatively the reactivity
in these media, w ) 13 was chosen as the lower limit of the
investigated microemulsion range. It is generally agreed that
below this value there is no bulk but only bound water.7,24 Above
w ) 24, the corresponding AOT-isooctane-water mixtures
were not homogeneous but turbid, due to the reduction of the
L2 area in the phase diagram25 of this system by addition of
0.1 M sodium bromide to the water component.
-
Br2 + Br- a Br3
(1)
dure, high bromination rate constants can be measured using
-
very small bromine concentrations since ꢀBr at 290 nm is ca.
3
104, whereas that of molecular bromine (λmax ) 415 nm) is 102
times smaller. It was confirmed that, when bromination was
not very fast (entry 1 of Table 1), the rate constants obtained
by monitoring bromine consumption at 290 and 415 nm were
identical within experimental error. Moreover, perchloric acid
(10-4 M) was added to the water content of the microemulsion
in order to minimize the formation of the nonbrominating
hypobromic acid21 via the fast equilibrium 2.
Discussion
Br2 + H2O a BrOH + H+ + Br-
(2)
1-Octene bromination does not occur in the isooctane
microphase since the reaction is first-order in bromine, as
observed in any protic solvent.15 In contrast, second-order in
bromine kinetics was observed in nonprotic solvents such as
halogenated hydrocarbons.26 This is currently attributed to
differences in the reaction mechanisms. In protic solvents, the
formation of the bromonium-bromide ion-pair is solvent-
The second-order rate constants were obtained from the
pseudo-first-order rate constants measured at several alkene
concentrations, with a reproducibility better than (2%.
typical second-order plot is shown in Figure 1. In all the
microemulsions, these plots were always rigorously linear
A

12640 J. Phys. Chem., Vol. 100, No. 30, 1996
Blagoeva et al.
SCHEME 2
results not only from differences in the hydration of the reaction
environment but also from different brominating agents. In
CTAB micelles, the brominating agent is not molecular bromine
but the poorly electrophilic tribromide ion, due to strong
interactions33 between bromide ions and ammonium head groups
at the interface which shift eq 1. In AOT microemulsions, there
is evidence that molecular bromine is the main reagent. In
agreement with the negligible interactions33 between bromide
and a sulfate head group, a constant for equilibrium 1 of 0.2-
0.4 in the range of investigated w values, as compared to 16 in
pure water, has been measured spectroscopically.34 It is,
therefore, reasonable to conclude that the reactivity data of Table
1 correspond mainly to the reaction of molecular bromine.
The Pseudophase Model for Reaction Kinetics in Micro-
emulsions. The experimental rate coefficients shown in Table
1 are not usual rate constants related to a given elementary
reaction and are, therefore, meaningless as regards the reactivity-
determining factors. In agreement with the pseudophase model
widely applied to reactions in direct micelles,11 local reagent
concentrations17 and local rate constants are the only parameters
relevant to the understanding of reactivity. In microemulsions,
a kinetic model implying competition between reactions in three
phases, oil, water, and interface, can be constructed10,12 with
the same assumptions as those used for reactions in micellar
media. In particular, it is assumed that the reaction time scale
is large with respect to that of the dynamics of the medium,16,17,19
so that three different phases can be considered. This hypothesis
is reasonable for bromination since the absence of any induction
period in the first-order kinetic signals clearly excludes any
influence of percolation between the aqueous microdroplets
initially containing one of the two reagents. Therefore, Scheme
2, where the subscripts 1, 2, o, i, and w refer to alkene, bomine,
oil, interface, and water, respectively, can be assumed. Ki and
Kii, with i ) 1 or 2, are the transfer coefficients of the reagents
from isooctane to interface and from water to interface,
respectively, expressed in terms of molar ratios in each phase
(eqs 3-6) and ki, the local rate constants in the o, i, and w
phases.
Figure 2. Dependence of the experimental rate constants on water
concentration: (+) w ) 13.3; (O) w ) 15.9; (×) w ) 19.9; (4) w )
24.
assisted, whereas in nonprotic solvents, a second bromine
molecule assists bromide ion departure by forming a tribromide
ion, markedly more stable than bromide in these media. The
absence of reaction in the isooctane phase was expected a priori
since the bromination rate constant in apolar hydrocarbons can
be estimated to be as small as 10-2 M-2 s-2, from a previously
published Kirkwood-Westheimer relationship for solvent ef-
fects on bromination rates.26 In fact, in hydrocarbons homolytic
bromination by bromine radicals22 is expected to be faster than
polar reactions. Nevertheless, the monoexponential kinetic
signals observed in AOT microemulsions do not correspond to
a radical reaction. The bromination kinetics are, therefore, in
agreement with a reaction occurring in an aqueous environment,
although the reagent concentrations in the water phase are very
small. The coefficients for 1-octene27 and bromine transfers28
from isooctane to water are 1.5 × 10-5 and 8.8 × 10-3
,
respectively.
The Role of the Water Content. 1-Octene bromination is
strongly inhibited in AOT microemulsions as compared with
the same reaction in pure water, where the rate constant29 is
2.3 × 107 M-1 s-1 (or 1.1 × 109 s-1 in the same units as those
in microemulsions), or in water containing 0.1 M sodium
bromide, k ) 1.1 × 107 M-1 s-1 (or 6 × 108 s-1). As shown
in Figure 2, there is a sharp increase in the apparent rate constant
as the water content increases. Despite the approximate linearity
of this plot, since extrapolation of the linear regression between
k2,app and water concentration would lead to a kw value in pure
water ([H2O] ) 55 M) of 2 × 103 s-1, i.e. much smaller than
that experimentally observed, the relationship cannot be linear.
[Al]i[IO]
Analogous inhibition has already been found in micellar
solutions. In micelles of SDS,30 a surfactant with the same polar
head group as AOT, the rate constant for 1-octene bromination
is 2 × 105 M-1 s-1. The inhibition (km/kw ) 10-2) in this
medium, where [H2O] ) 55 M, is much smaller than in
microemulsions where at the highest water content, [H2O] )
12.5 M, the inhibition is as large as 10-5. This comparison
exhibits clearly the key role of water. A different situation was
found in the bromination of unsaturated fatty esters,31 for which
K1 )
(3)
(4)
(5)
(6)
[AOT][Al]o
[Br2]i[IO]
K2 )
[Br2]o[AOT]
[Al]i[H2O]
K11
)
[AOT][Al]w
the inhibition is extremely large, 10-8
. In agreement with
McClelland’s conclusion, this enormous deceleration suggests
strongly that bromination of these substrates occurs in a very
poorly hydrated region of the micelle. In CTAB solutions32 (k
) 6.9 M-1 s-1) the inhibition is also very large. However, it
[Br2]i[H2O]
K22
)
[AOT][Br2]w
From Scheme 2, the experimental rate, Vapp, is the sum of

Pseudophase Model for Reactivity in Reverse Micelles
J. Phys. Chem., Vol. 100, No. 30, 1996 12641
the local rates in each phase. Since for bromination, there is
no reaction in isooctane, Scheme 2 is expressed by eq 7, where
the subscript T refers to the total reagent concentration, as shown
in eqs 8 and 9.
K1K2
z2 + (K2 + K1)z
k
2,app[AOT] ) ki
(15)
on the water concentration or on w, since w is related to it.
This dependence can be expressed by eq 15 if ki is w-dependent.
Secondly, the k2,app dependence on z is expressed by the ratio
in the right-hand-side term of eq 15 and does not depend on w.
Therefore, linear relationships should be observed between k2,app
[AOT] values at various w values, the slope of which would be
the ratios of ki values at these w.
ki
Vapp ) k2,app[Br2]T[Al]T )
[Br2]i[Al]i +
[AOT]
kw
-
[Br2]w[Al]w (7)
[H2O]
[Br2]T ) [Br2]o + [Br2]i + [Br2]w
[Al]T ) [Al]o + [Al]i + [Al]w
(8)
(9)
(ki)w
(k2,app[AOT])w)a
)
a(k2,app[AOT])w)b
(16)
(ki)wb
If the microemulsion composition is defined by w ) [H2O]/
[AOT] and z ) [IO]/[AOT], the local reagent concentrations
expressed in terms of total concentrations are given by eqs 10
and 11.
Figure 3 shows clearly that relationships 16 are not linear and,
therefore, that eq 15 does not apply. In conclusion, bromination
must also occur in the water phase.
Partition Coefficients and Local Rate Constants. Since
bromination does not occur exclusively at the interface, the
kinetics data of Table 1 were analyzed in terms of eq 14, which
assumes competition between reactions at the interface and in
water. A nonlinear least squares regression procedure gives the
results shown in Table 2. The calculated rate constants agreed
well with the experimental values of k2,app (Table 1 and Figure
4), and the assumption that 1-octene is brominated in two
microemulsion phases appears to be justified. In particular, it
is noticeable that the two terms of eq 14 (kiK1K2 and
kwKt1rKt2r) are of similar magnitude. The first striking result of
the calculation is that the coefficient of bromine partition, K2,
between isooctane and the interface is large and very close to
6.8 since K1 for 1-octene is necessarily much smaller than K2
(vide supra). The bromine concentration at the interface is,
therefore, quite significant. This result is not unexpected.
Bromine is a highly polarizable species that would be stabilized
in the polar interfacial microenvironment better than in the
nonpolar isooctane phase. Accordingly, complexation between
bromine and AOT, analogous to that observed for iodine,35 is
suggested by preliminary spectroscopic studies.34 Similar
complexation between AOT and 1-octene is less probable. Even
if the 1-octene preference for the interface were as large as that
of bromine, the K1 value for the alkene could not be greater
than 1 × 10-2, i.e. negligible compared to K2 for bromine. (This
estimation assumes proportionality between the transfer coef-
ficients of the two reagents and the coefficients of their partition
between isooctane and the interface.) From these data, it is
possible to calculate the bromine partition coefficients (K2 and
K22 of Scheme 1) and to estimate those of 1-octene. These
results are shown in Table 3.
w
z
[Al]T ) [Al]i 1 +
+
(10)
(11)
(
)
K11 K1
w
z
[Br2]T ) [Br2]i 1 +
+
(
)
K22 K2
With these definitions, the k2,app dependence on microemul-
sion composition is given by eq 12, obtained by combining eqs
7-11.
ki + (1/K11K22)kww
k2,app
)
(12)
w
z
w
z
[AOT] 1 +
+
1 +
+
(
)(
)
K11 K1
K22 K2
In eq 12, the w/Kii terms are negligible with respect to z/Ki
since the Kii constants are necessarily very large because of the
small reagent concentrations in the water phase. Therefore, eq
12 simplifies into eq 13, where K1/K11 and K2/K22 are the
experimentally obtained transfer coefficients27,28 of alkene and
bromine, i.e., K1tr and Kt2r, from isooctane to water.
K1K2ki + Kt1rK2trkww
k2,app
)
(13)
[AOT](K1 + z)(K2 + z)
Another simplification that makes the statistical data analysis
easier and that is justified by the following discussion can be
made assuming that in the denominator of eq 13 the product
K1K2 is negligible as compared to the terms in z2 and z. Finally,
eq 14 will be used to analyze the bromination rates.
The lower limit for the interfacial rate constant, ki, can be
estimated from the calculated kiK1K2 term of eq 14. Term ki
is, at least, in the range of 6 × 104 s-1 (or 1 × 103 M-1 s-1).
This is a small value compared to that of bromination of the
same alkene in bulk water29 (k ) 6 × 108 s-1) but much larger
than that found for diphenylmethyl chloride solvolysis13 in the
same microemulsions, which leveled off at 9 × 10-3 s-1 for
high water contents (w g 30). Nevertheless, this last result is
not surprising since it is well-known that alkene bromination is
much faster than the solvolytic processes. The interfacial
bromination rate constant in the 103 M-1 s-1 range corresponds36
to that in poorly hydrated bulk methanol (MeOH/H2O, 95/5,
v/v). This is in agreement with an interface polarity significantly
smaller than that of bulk water37 but large enough to promote
the fast ionization of the bromine-ofefin charge transfer
complex.
K1K2ki + Kt1rK2trkww
z2 + (K2 + K1)z
k2,app[AOT] )
(14)
Interfacial Bromination Only? Equation 14 involves two
terms related to the reaction at the interface and in the water
phase, respectively, reagent partition between isooctane and the
interface being taken into account in the denominator. It can
be assumed, firstly, that the reaction takes place essentially at
the interface. A priori, this assumption seems reasonable since
the reagent concentrations in water are almost negligible because
of the very small values of their transfer coefficients. A similar
hypothesis has recently been postulated for the hydrolysis of
diphenylmethyl chloride in the same microemulsions.13 How-
ever in the case of bromination, this assumption does not hold.
If the reaction in the water phase is negligible, eq 14 reduces
to eq 15. Firstly, it is obvious from Figure 2 that k2,app depends
Finally, kw, the rate constant in the aqueous core of the
microdroplets, is obtained from the coefficient of the w term in

12642 J. Phys. Chem., Vol. 100, No. 30, 1996
Blagoeva et al.
on the experimental rate constant in bulk water and the
assumptions of the pseudophase model. Nevertheless, the
differences in the chemical properties of bulk and microemul-
sion-encased water, which can be viewed as a reagent in
bromination, are frequently mentioned although not yet well-
known.4,7,24
Competition between Bromination at the Interface and
in the Water Phase. With the rate constants and partition
coefficients obtained by application of the pseudophase model,
it is now possible to estimate the importance of the competition
between the reactions in the two microphases from the relative
values of the two terms of eq 14 at various w values. At the
highest investigated w (w ) 24), the two terms of eq 14 are in
a ratio of 4.5; that is, more than 80% of the reaction occurs in
the water phase. This may seem surprising since the reagent
concentrations in this phase are very small. However, it is still
more surprising that at the smallest w (w ) 13), the limit
between reverse micelles and microemulsions,7 about 70% of
the reaction still takes place in a water phase. This means that,
although the reagent concentration is significant at the interface
as shown by the non-negligible K1 and K2 values, the rate
constant ki is not large. In fact, the competition is controlled
not only the relative concentrations but also by the relative rate
constants. In bromination, the rate constant ratio, kw/ki, is large,
in the range of 105. Consequently, even though the concentra-
tions in the water phase are exceedingly small, the reaction still
prefers this phase. Nevertheless, in Figure 4, some scatter for
a given w is observed. This probably comes from the neglect
of the w/Kii terms in eq 12, which are related to the concentra-
tions in the water phase. This small scatter would, therefore,
indicate that these concentrations are not as negligible as
deduced from the pseudophase model. In contrast, in the
diphenylmethyl chloride solvolysis,13 kw/ki is much smaller,
about 102, so that the competition is controlled to a greater extent
by the localization of the reagents within the several mi-
crophases.
Figure 3. Comparison of rate data at various constant w values, eq
16. The curvature in the plots shows that bromination does not occur
exclusively at the interface; see text.
TABLE 2: Parameters of the Nonlinear Least Squares
Fitting of Eq 14 to the Experimental Data (ø2 ) 0.002)
kiK1K2
4 200 ( 950
800 ( 70
kwKt1rKt2r
K1 + K2
6.8 ( 0.6
The differences between the two reactions, bromination and
solvolysis, which are both water-promoted and exhibit similar
solvent effects in homogeneous media,36 point out the need of
a better understanding of water behavior in reverse micelles and
w/o microemulsions.
Concluding Remarks
Insofar as reverse micelles and microemulsions are of interest5
as microreactors for water-promoted reactions, it is essential to
understand the chemical properties of the encased water. In
this respect, the numerous spectroscopic investigations4,6,7,24 on
the nature of bound water, i.e. on the interactions of water
molecules and surfactant head groups or their counterions, are
of great significance. Nevertheless, the relationship between
these interactions and the availability of water as an electrophilic
or nucleophilic reagent for performing chemical reactions38 is
not clear. The only reactions which have presently been
investigated with this aim are acid-base-catalyzed reactions4,39
involving protonated water, H3O+, or the hydroxyl anion, OH-.
However, the question of the proton existence in reverse micelles
is still open,40 because of the difficulties (pH meaning, probe
localization, etc.) of the approaches to acid-base equilibria7,39
in these media. In this regard, the present results on bromination
and those recently published on hydrolytic processes13 can open
a new way to tackle these problems.
Figure 4. Agreement between experimental and calculated kinetic data
for 1-octene bromination. The straight line is calculated from the
pseudophase model, eq 14.
TABLE 3: Partition Coefficients and Local Bromination
Rate Constants of Scheme 2, Calculated from the
Pseudophase Model
Kt1ra
Kt2rb
K1
K2
K11
K22
c
d
e
d,e
1.5 × 10-5 8.8 × 10-3 e1 × 10-2 6.8 g7 × 102 7.7 × 102
d
f
kic
(kw)µE
(kw)H O
2
g6 × 104 s-1
6 × 109 s-1
6 × 108 s-1
a Reference 27. b Reference 28. c Estimated, see text. d Standard error
tr
e
less than 10%. Kii ) Ki/Ki . f Reference 29.
eq 14. The value found from the two-phase calculations, kw )
6 × 109 s-1 (or 1 × 108 M-1 s-1 in concentration units), is
quite consistent with that measured in 0.1 M NaBr water, k )
1.1 × 107 M-1 s-1. The discrepancy between the two values
is probably not significant, in view of the large error ((20%)
However, measurement of rate data meaningful in terms of
microscopic or local reactivity is an obligatory preliminary. Our
present results show that the extension of the pseudophase model
to w/o microemulsions can be useful to obtain them. More work

Pseudophase Model for Reactivity in Reverse Micelles
J. Phys. Chem., Vol. 100, No. 30, 1996 12643
(4) (a) Zhu, X. X.; Bardez, E.; Dallery, L.; Larrey, B.; Valeur, B. New
J. Chem. 1992, 16, 973. (b) Aliotta, F.; Migliardo, P.; Donato, D. I.; Turco-
Liveri, V.; Bardez, E.; Larrey, B. Progr. Colloid Polym. Sci. 1992, 89, 258.
(5) Luisi, P. L.; Giomini, M.; Pileni, M. P.; Robinson, B. H. Biochim.
Biophys. Acta 1988, 947, 209.
is in progress to extend this model to the understanding of the
reactivity of bound water molecules in these microorganized
media.
(6) Llor, A.; Rigny, P. J. Am. Chem. Soc. 1986, 108, 7533.
(7) El Seoud, O. A.; El Seoud, M. I.; Mickiewicz, J. A. J. Colloid
Interface Sci. 1994, 163, 87.
(8) Breslow, R. Acc. Chem. Res. 1991, 24, 159. Li, C. J. Chem. ReV.
1993, 93, 2023.
(9) Blokzijl, W.; Engberts, J. B. F. N. Angew. Chem., Int. Ed. Engl.
1993, 32, 1545.
(10) Valiente, M.; Rodenas, E. J. Phys. Chem. 1991, 95, 3368. Moya,
M. L.; Izquierdo, C.; Casado, J. Ibid. 1991, 95, 6001. Khmelnitsky, Y. L.;
Noverova, I. N.; Polyakov, V. I.; Grinberg, V. Y.; Levashov, A. V.;
Martinek, K. Eur. J. Biochem. 1990, 190, 155.
(11) Bunton, C. A.; Nome, F.; Quina, F. H.; Romsted, L. S. Acc. Chem.
Res. 1991, 24, 357. Bunton, C. A.; Savelli, G. AdV. Phys. Org. Chem.
1986, 22, 213 and references therein.
(12) Garcia-Rio, L.; Leis, J. R.; Pena, M. E.; Iglesias, E. J. Phys. Chem.
1993, 97, 3437.
Experimental Section
Materials. AOT (Sigma) was of the highest purity available
and used without further purification after drying in a vacuum
desiccator over P2O5 for two days. 2,2,4-Trimethylpentane
(isooctane, Aldrich) was of spectrophotometric grade. 1-Octene
was used as supplied. Bromine and sodium bromide were
Suprapur Merck. A stock solution of bromine (1 mL) in 25
mL of CCl4 was prepared shortly before use under protection
from external light.
Typical Preparation of Microemulsion. Microemulsions
with a constant z value and variable w values were prepared by
addition of the calculated amounts of 0.1 M NaBr solution in
distilled water to an isooctane-AOT solution of known mo-
larity. The microemulsion composition was checked by weigh-
ing. Usually, 50 mL of AOT solution in isooctane was weighed
in a closed vessel and weighed again after addition of the water
solution. The microemulsion obtained was pretreated by
addition of small amounts of bromine solution in CCl4 in order
to eliminate traces of bromine-consuming impurities. The
“presaturation” with bromine was controlled by UV at 270 nm.
This “mother” microemulsion was further used for the prepara-
tion of the series of three different concentrations of alkene used
for the determination of k2app. The density was determined by
weighing 10 mL of the microemulsion in volumetric flasks, and
the concentration of the alkene by weighing a syringe containing
40-120 µL of 1-octene. The final concentrations of 1-octene
were in the 10-2 M range.
Typical Kinetic Runs. All kinetic runs were performed on
a Hi-tech stopped-flow spectrophotometer thermostated at 25.0
( 0.1 °C and equipped with a data acquisition system. One
syringe of the apparatus contained microemulsion with bromine
and the other microemulsion with 1-octene. The reaction was
followed by the decrease of absorbance due to Br3- at 290 nm
instead at the maximum of absorbance at 270 nm as better
conditions for the work of the photomultiplier could be obtained.
Pseudo-first-order rate constants kexp were calculated from the
monoexponential kinetic signals by the Marquardt procedure41
for least squares nonlinear curve fitting.
(13) Garcia-Rio, L.; Leis, J. R.; Iglesias, E. J. Phys. Chem. 1995, 99,
12318.
(14) Lopez, P.; Rodriguez, A.; Munoz, E.; Gomez-Herrera, C.; Sanchez,
F.; Moya, M. L. J. Colloid Interface Sci. 1994, 166, 503.
(15) Ruasse, M. F. AdV. Phys. Org. Chem. 1993, 28, 207.
(16) Chevalier, Y.; Zemb, T. Rep. Prog. Phys. 1990, 53, 279.
(17) Pernot, P.; Pansu, R.; Levy, B.; Faure, J. J. Chem. Phys. 1993,
177, 793.
(18) Ruasse, M. F.; Motallebi, S.; Galland, B. J. Am. Chem. Soc. 1991,
113, 3440.
(19) Jada, A.; Lang, J.; Zana, R. J. Phys. Chem. 1989, 93, 10. Jada,
A.; Lang, J.; Zana, R.; Makloufi, R.; Hirsch, E.; Candau, S. Ibid. 1990, 94,
387. Lang, J.; Mascolo, G.; Zana, R.; Luisi, P. L. Ibid. 1990, 94, 3069.
(20) Ruasse, M. F.; Aubard, J.; Galland, B.; Adenier, A. J. Phys. Chem.
1986, 90, 4382.
(21) Eigen, M.; Kustin, K. J. Am. Chem. Soc. 1962, 84, 1355.
(22) Byrnell, C. J. A.; Coombes, R. G.; Hart, L. S.; Whiting, M. C. J.
Chem. Soc., Perkin Trans 2 1983, 1079.
(23) Warner, P.; La Rose, R.; Schleis, T. Tetrahedron Lett. 1976, 4443.
(24) Kawai, T.; Hamada, K.; Shindo, N.; Kon-No, K. Bull. Chem. Soc.
Jpn. 1992, 65, 2515.
(25) Zulauf, M.; Eicke, H. F. J. Phys. Chem. 1979, 83, 480.
(26) Bellucci, G.; Bianchini, R.; Ambrosetti, R.; Ingrosso, G. J. Org.
Chem. 1985, 50, 3313.
(27) Schantz, M. N.; Martire, D. E. J. Chromatogr. 1987, 391, 35.
(28) Tohyama, I.; Otoza¨ı, K. Frezenius Anal. Chem. 1978, 293, 282.
(29) Measured by couloamperometry at (20% (Dubois, J. E.; Ruasse,
M. F.; Poupard, D. J. Electroanal. Chem. 1983, 152, 85). This value agrees
with that previously measured for 1-pentene bromination in the same solvent
(Garnier, F.; Dubois, J. E. Bull. Soc. Chim. Fr. 1968, 3797).
(30) Ruasse, M. F.; Blagoeva, I. B.; Krys, S.; Sebastian-Gambaro, M.
A. J. Chem. Soc., Perkin Trans. 2 1993, 1283.
(31) Lennox, R. B.; Mc Clelland, R. A. J. Am. Chem. Soc. 1986, 108,
3771.
(32) Cerichelli, G.; Grande, C.; Luchetti, L.; Mancini, G. J. Org. Chem.
1991, 56, 3025.
(33) Cerichelli, G.; Grande, C.; Luchetti, L.; Mancini, G.; Bunton, C.
A. J. Org. Chem. 1987, 52, 5167.
(34) Garcia-Rio, L.; Mejuto, J.; Ruasse, M. F. Unpublished results. A
strong charge transfer band, λmax ) 290 nm, appears in the bromine spectrum
when AOT is added to bromine solutions in isooctane.
(35) Biasutti, M. A.; Sereno, L.; Silber, J. J. J. Colloid Interface Sci.
1994, 164, 410.
Acknowledgment. We gratefully acknowledge Professor J.
R. Leis and his group for helpful suggestions and stimulating
discussion. We are indebted to Professor A. Lattes and Dr. I.
Rico-Lattes, whose pioneer work inspired this research. I.B.B.
would like to thank CNRS and the University of Paris 7 for the
award of a Research Associateship.
(36) Ruasse, M. F.; Motallebi, S. J. Phys. Org. Chem. 1991, 4, 527.
(37) Wong, M.; Gra¨tzel, M.; Thomas, J. K. J. Am. Chem. Soc. 1976,
98, 2391. Keh, E.; Valeur, B. J. Colloid Interface Sci. 1981, 79, 465. Zinsli,
P. E. J. Phys. Chem. 1983, 83, 3223.
(38) Rico, I.; Halvorsen, K.; Dubrule, C.; Lattes, A. J. Org. Chem. 1994,
59, 415. Gautier, M.; Rico, I.; Lattes, A. Ibid. 1990, 55, 1500.
(39) Chinelatto, A. M.; Fonseca, M. T. M.; Kiyan, N. Z.; El Seoud, O.
A. Ber. Bunsen-Ges. Phys. Chem. 1990, 94, 882.
References and Notes
(1) The term “reverse micelles” is used here for aggregates in which
the water is interfacial only, i.e. in strong interaction with surfactant head
groups and counterions. “Microemulsions”, sometimes called swollen
micelles, refers to aggregates containing a core water phase more or less
analogous to bulk water.
(2) Schwuger, M. J.; Stickdorn, K.; Schoma¨cker, R. Chem. ReV. 1995,
95, 849.
(40) Bardez, E.; Monnier, E.; Valeur, B. J. Phys. Chem. 1985, 89, 5031.
(41) Marquardt, D. W. J. Soc. Ind. Appl. Math. 1963, 11, 431.
(3) Structure and Reactivity in Reverse Micelles; Pileni, M. P., Ed.;
Elsevier: Amsterdam, 1989.
JP960772A