Kinetic model for reactivity in quaternary water-in-oil microemulsions

Article abstract of DOI:10.1002/chem.200600091

A study was carried out on the nitrosation of piperazine (PIP) and N-methylbenzylamine (MeBzAm) by N-methyl-N-nitroso-p-toluenesulfonamide (MNTS) in quaternary microemulsions of tetradecyltrimethylammonium bromide (TTABr)/isooctane/alcohol/water, varying the nature and the concentration of the following alcohols: 1-pentanol, 1-hexanol, 1-heptanol, 1-octanol and 1-decanol keeping the [1-alcohol]/[TTABr] = 4 relationship constant. In addition a study was carried out on the influence of the alcohol concentration, working with molar relationships [1-hexanol]/[TTABr] = 3, 4 and 5. On the basis of the molar volumes of the alcohol and surfactant and the concentration of alcohol at the interface it was possible to calculate the change in its volume with as varying compositions of the microemulsion. In order to interpret the experimental results a kinetic model was devised which takes into account the distribution of the reactants between the different pseudophases and the change in the volume of the interface. The rate constants at the interface of the microemulsion are lower than in pure water and are independent of the nature of the alcohol used as a cosurfactant and the molar relationship [alcohol]/[TTABr]. This independence indicates that the main role of the cosurfactant is to increase the volume of the interface with the consequent dilution of the reactants.

Full text of DOI:10.1002/chem.200600091

DOI: 10.1002/chem.200600091
Kinetic Model for Reactivity in Quaternary Water-in-Oil Microemulsions
[
a]
Luis García-Río* and Pablo Hervella
Abstract: A study was carried out on
the nitrosation of piperazine (PIP) and
N-methylbenzylamine (MeBzAm) by
N-methyl-N-nitroso-p-toluenesulfona-
mide (MNTS) in quaternary micro-
lationships [1-hexanol]/
A
H
R
U
[TTABr]=3, 4
the reactants between the different
pseudophases and the change in the
volume of the interface. The rate con-
stants at the interface of the microe-
mulsion are lower than in pure water
and are independent of the nature of
the alcohol used as a cosurfactant and
and 5. On the basis of the molar vol-
umes of the alcohol and surfactant and
the concentration of alcohol at the in-
terface it was possible to calculate the
change in its volume with as varying
A
C
H
T
R
E
U
N
G
emulsions of tetradecyltrimethylammo-
nium bromide (TTABr)/isooctane/alco-
hol/water, varying the nature and the
concentration of the following alco-
compositions of the microemulsion. In
A
H
R
U
G
order to interpret the experimental re-
sults a kinetic model was devised which
takes into account the distribution of
the molar relationship [alcohol]/[T-
A
C
H
T
R
E
U
N
G
TABr]. This independence indicates
that the main role of the cosurfactant
is to increase the volume of the inter-
face with the consequent dilution of
the reactants.
hols: 1-pentanol, 1-hexanol, 1-hep
-octanol and 1-decanol keeping the
1-alcohol]/[TTABr] = 4 relationship
A
H
R
U
G
A
H
R
U
1
[
ACHTREUNG
Keywords: cosurfactant · kinetics ·
microemulsions · reaction mecha-
nism · reactivity
constant. In addition a study was car-
ried out on the influence of the alcohol
concentration, working with molar re-
Introduction
face, the addition of other surface-active substances is often
required.
Microemulsions are thermodynamically stable systems that
are composed of a surfactant, an organic solvent, and a
small amount of water. These systems appear as homogene-
ous, transparent solutions that can solvate a wide range of
hydrophilic and hydrophobic compounds. The formation of
microemulsions requires the presence of water, which is
solubilized in a polar core, forming the so-called water pool.
These structures are then spherical droplets of water dis-
persed in oil. Their spontaneous curvature arises from the
energetically favorable packing configuration of the surfac-
tant molecules at the water/oil interface and depends basi-
cally on the molecular geometry of the surfactant molecules.
One of the surfactants often used to form microemulsions is
sodium bis(2-ethylhexyl)sulfosuccinate (AOT). However, to
attain the appropriate packing of amphiphiles at the inter-
Although quaternary microemulsions have not been in-
vestigated to the same degree as AOT microemulsions, re-
searchers have probed a range of properties in these sys-
tems. For example, the size and shape characteristics of ce-
tyltrimethylammonium bromide microemulsions formed in
alkanes with alcohol cosurfactants have been determined
through light scattering, pulse field gradient NMR, near-IR
spectroscopy, and conductivity.
[1]
These cosurfactants (usually medium-chain linear alco-
hols) partition themselves among the oil, water and the in-
terface domains. Therefore, without a quantitative descrip-
tion of the dependence of the partition equilibrium on the
system composition, a full understanding of quaternary mi-
croemulsions cannot be attained. For the estimation of the
distribution of the cosurfactants between the interfacial and
[2]
the bulk oil phase, small-angle neutron scattering, conduc-
[3]
[2b,4]
tivity, interfacial tension
and small-angle X-ray scatter-
[2b]
ing and dynamic light scattering methods have been used.
The alcohols added have marked effects on the microemul-
sion properties such as the critical micelle concentration of
three-component microemulsions, the degree of counterion
binding of microemulsions with composed ionic surfactants,
the aggregation number, and the capacity of microemulsions
[
a] Prof. L. García-Río, P. Hervella
Departamento de Química Física
Facultad de Química
Universidad de Santiago, 15782 Santiago (Spain)
Fax : (+34)981-595-012
E-mail: qflgr3cn@usc.es
8284
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8284 – 8295

FULL PAPER
[
5]
to dissolve oil. Alcohol partition or binding constants are
estimated by a variety of methods. Much of this work has
been carried out in aqueous three-component microemul-
sions, some in four-component water-in-oil microemul-
made up of isooctane/tetradecyltrimetylammonium bromide
(TTABr)/1-alcohol/water. To this end we have chosen a re-
action which is particularly well known to us, namely the ni-
troso group transfer from N-methyl-N-nitroso-p-toluenesul-
[6]
[7]
[11]
sions. Romsted and Yao have devised a new method
based on the product yields from dediazoniation of 4-hexa-
decyl-2,6-dimethylbenzenediazonium tetrafluoroborate to
estimate the changes in the composition (concentrations of
water and alcohol) at the interface of microemulsions con-
sisting of cetyltrimethylammonium bromide, water, hexade-
cane and an alcohol (1-butanol or 1-hexanol), on varying
the composition of the microemulsion. Their results have
enabled them to calculate the distribution constants of the
fonamide (MNTS) to secondary amines
(Scheme 1). As
secondary amines we have used piperazine, and, to a lesser
extent, N-methylbenzylamine. The choice of these amines is
based on their different solubilization properties on AOT/
[13]
isooctane/water microemulsions. Hence the comparison of
the results obtained for quaternary TTABr microemulsions
with those previously obtained in ternary AOT microemul-
sions will enable us to obtain more information about the
influence of the cosurfactant on the reactivity of these
media. The alcohols used have varied from 1-pentanol to 1-
1
-butanol and 1-hexanol between the interface and the con-
tinuous medium of the microemulsion.
decanol keeping the [1-alcohol]/
constant. In the case of the 1-hexanol we have also worked
with the molar relationships [1-alcohol]/
5.
A
H
R
U
G
[8]
Lang et al. have studied the effect of alcohol chain
length on surfactant aggregation number, radius of the drop-
let water core, intensity of attractive interdroplet interac-
tions, onset of percolation of electrical conductivity and the
rate constant for the exchange of material between droplets
of quaternary microemulsions made up of water/chloroben-
zene/cationic surfactants/1-alcohol. They found that the ag-
gregation number, the radius of the droplet water core and
the intensity of attractive interactions decrease when the al-
cohol chain length increases. The value of the rate constant
for exchange of material between droplets at the percolation
threshold is independent of the chain length of the alcohol
or surfactant. The percolation threshold increases with the
alcohol chain length.
Results
The experiments were carried out under the same condi-
tions as those described in previous articles.
equations obtained for the reactions in microemulsions were
in all cases similar to those in water, with first-order terms
in MNTS and total amine concentration (figures not
shown). The experimental results obtained from studying
the influence of the variation of the microemulsion composi-
tion on kobs will depend heavily on the way in which the
amine is distributed through the pseudophases of the micro-
emulsion, that is, on the hydrophobicity of the amine.
The significant capacity of solvation, together with the
specific properties of trapped water in the microemulsions
has given rise to a huge amount of kinetic studies on these
types of reaction media. The greater simplicity of the AOT
microemulsions implied that the majority of kinetic studies
have been carried out on AOT/alkane/water microemul-
Nitrosation of piperazine: A series of experiments were per-
formed at 25.08C and fixed amine concentration and varia-
ble TTABr concentration (typically between 0.1 and 0.8m)
with the water/TTABr mol ratio, W, fixed at values ranging
from W=10 to 45. Two experiments were carried out. In the
first experiment the molar relationship remained constant
[9]
sions. However, the number of kinetic studies carried out
[10]
on microemulsions with four components is very small. In
fact, we are not aware of any kinetic study to date which
has systematically researched the influence of the chain
length of the cosurfactant or its concentration on the reac-
tivity in microemulsions.
In this article we will describe a systematic study of the in-
fluence of a variety of alcohols on a well-characterized reac-
tion in the presence of cationic microemulsions. The fact
that reaction kinetics are influenced by the nature of the
medium suggests that such studies may throw light on the
changes in the microemulsion properties caused by the addi-
tion of alcohols. In particular we have studied the influence
of the cosurfactant on the reactivity in microemulsions
[alcohol]/[TTABr]=4 regardless of the microemulsion com-
A
H
E
N
position. These experiments were carried out using microe-
mulsions in which the length of the chain of the alcohol
varied between 1-pentanol and 1-decanol. In the second ex-
periments the influence of the molar relation [alcohol]/
TABr] on kobs was studied using molar relationships [1-hexa-
nol]/[TTABr]=3, 4 and 5.
AHCTREUNG
Figure 1 shows an example of observed experimental be-
havior. In all microemulsions it has been observed that the
pseudo-first-order rate constant, kobs, for the nitrosation of
Scheme 1.
Chem. Eur. J. 2006, 12, 8284 – 8295
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8285

L. García-Río and P. Hervella
PIP by MNTS, decreases as the TTABr concentration in-
creases and as the water content of the microemulsion in-
[13]
creases
(increase in W) (see Figure 1). The decrease in
kobs as W increases is not very marked, and decreases almost
threefold as W increases from 10 to 40. If we compare the
influence of [TTABr] on kobs in experiments carried out with
W=10 and with different alcohols no clear tendency is ob-
servable (not shown) although there seems to exist a slight
tendency for kobs to decrease as the length of the hydrocar-
bon chain of the alcohol increases.
Figure 2. Influence of TTABr concentration upon kobs for the nitrosation
of piperazine by MNTS at W=10 in TTABr/1-hexanol/isooctane/water
ꢀ
2
microemulsions. [piperazine]tot =5.45”10 m; (*) [ROH]tot/[TTABr]=5;
*) [ROH]tot/[TTABr]=4; (&) [ROH]tot/[TTABr]=3. Lines are drawn
(
for clarity. T=25.08C.
havior in the nitrosation of PIP and MeBzAm, respectively,
must be due to their different hydrophobicity.
Figure 1. Influence of the TTABr concentration upon kobs for the nitrosa-
tion of piperazine by MNTS at constant W in TTABr/1-hexanol/isooc-
tane/water microemulsions. [ROH]tot/[TTABr]=5. [piperazine]tot =5.45”
ꢀ
2
1
0
m; (*) W=10; (*) W=15; (&) W=20; (
&
) W=40. Lines are drawn
for clarity. T=25.08C.
Figure 2 shows the variation of kobs with TTABr concen-
tration for experiments carried out at W=10 with different
Figure 3. Influence of TTABr concentration upon kobs for the nitrosation
of MeBzAm by MNTS at constant W in TTABr/1-pentanol/isooctane/
molar relationships [1-hexanol]/[TTABr]. It can be seen that
A
H
R
U
G
the pseudo first-order rate constant, kobs, increases as the
concentration of alcohol present in the reaction medium de-
creases. This variation of kobs with the alcohol content in the
medium can be interpreted as being due to the dilution of
the reactants. As more alcohol is added to the microemul-
sion the quantity of alcohol which will be incorporated into
the interface increases and, hence, the volume of the inter-
face increases. This increase in volume causes the MNTS
and amine to be diluted and therefore kobs decreases. More-
over, incorporation of small molecules into the interfacial
water microemulsions. [ROH]tot/[TTABr]=4. [N-methylbenzylamine]tot
.10m; (*) W=10; (*) W=15; (&) W=20; (
) W=30 and (~) W=35.
Lines are drawn for clarity. T=25.08C.
=
0
&
Alcohol distribution in the microemulsion: As mentioned
above, it is important to be aware of the distribution of the
alcohol between the different pseudophases of the microe-
mulsion. Given that the alcohol acts as a solubilized sub-
stance, we can transform the four-component microemulsion
into a microemulsion consisting of three pseudocomponents
(Figure 4).
As we can see from Figure 2 we neglect the amount of
solubilized alcohol in the aqueous phase. This simplification
seems correct given the low solubility of the alcohols used
[14]
region of microemulsions can increase the rigidity of the
interface layer affecting the reaction rates.
Nitrosation of N-methylbenzylamine: The study of the influ-
ence of the microemulsion composition on kobs in the nitro-
sation of MeBzAm by MNTS has been more limited.
Figure 3 shows an example of the experimental behavior ob-
served when studying the influence of TTABr concentration
on kobs in series of experiments in which W remained con-
stant. The observed behavior is the opposite of that shown
in the nitrosation of PIP by MNTS. Hence, kobs for the nitro-
sation of MeBzAm by MNTS does indeed increase along
with the surfactant concentration and the water content of
the microemulsion. The difference between the observed be-
Figure 4.
8286
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8284 – 8295

Water-in-Oil Microemulsions
FULL PAPER
ROH
in water and the small fraction in volume of the aqueous
phase. The distribution constants of the alcohol between the
pseudophases of the microemulsion can be defined as:
Table 1. Values of Koi for alcohol distribution between the continuous
medium and the interface of TTABr-based microemulsions.
ROH
Koi
ROH
Koi
ROH
[
a]
[b]
[b]
[b]
[b]
[b]
[b]
1
1
1
-butanol
-pentanol
-hexanol
43
34
35
44
48
37
36
31
32
28
26
23
[
a]
½
ROHŠ ½H OŠ
½ROHŠ
½ROHŠ
[a]
[a]
ROH
i
2
i
Kwi
¼
¼
¼
W
ð1Þ
ð2Þ
½
ROHŠ ½TTABrŠ
w
w
1-heptanol
1
1
[
[
a]
a]
-octanol
-decanol
½
ROHŠ ½isooctaneŠ
½ROHŠ
½ROHŠ
ROH
i
i
Koi
¼
Z
½
ROHŠ ½TTABrŠ
[a] Obtained by the Schulmanꢁs titration method. [b] Extrapolated from
aqueous micelles.
o
o
where W and Z are the parameters of the microemulsion
composition, W=[H O]/[TTABr] and Z=[isooctane]/[T-
A
H
R
U
G
ACHTREUNG
2
TABr].
Schulmanꢀs titration method: For the existence of a stable
w/o microemulsion, a critical amount of the cosurfactant has
to be present in the oil and at the interface. Other condi-
tions remain the same, the ratios between the number of
o
a
moles of alcohol in oil (n ) and the number of moles of oil
(
n ), and between the mole fraction of alcohol at the inter-
o
i
o
o
face (X ) and that in the oil (X ) are fixed: K = n /n and
K = X /X , where K and K are the appropriate constants,
a
a
a
o
i
o
a
d
a
d
the latter being the distribution constant. Addition of oil in
the system would change K and K by abstracting the alco-
d
hol and making the system unstable. Since the stability of
the microdispersion of water is affected by a threshold pop-
ulation of alcohol at the interface and consequently in the
bulk oil, the addition of alcohol can reestablish the thresh-
old stable condition of the system by restoring the required
magnitudes of both K and Kd.
t
a
Figure 5. Plot of n /n versus n
0
/n
s
according to Equation (3) for w/o
s
TTABr/isooctane/water/alcohol microemulsion systems at 25.08C with
different TTABr concentration: (*) 0.0019m; (*) 0.0030m; (~) 0.0043m;
( ) 0.0060m and (&) 0.0078m.
&
Representing the number of moles of alcohol in water
Extrapolation from aqueous micelles: The evaluation of the
distribution constants is possible on the basis of the values
of the association constants of the alcohols to direct micelles
w
i
and at the interface by n and n , respectively, by using K=
a
a
0
a
n /n , the total number of moles of alcohol in the microe-
0
t
a
mulsion system, n per mole of added surfactant can be rep-
resented as:
(water/tensioactive systems) and the distribution constants
[15]
w
o
of the alcohols between water and isooctane K .
The values of the distribution constants of the alcohols to
direct TTABr micelles are expressed as a molar relationship
between the alcohol associated with the micelle and the al-
t
w
i
a
na
na þ n
no
¼
þ K ns
ð3Þ
ns
n
s
[16]
m
w
cohol present in an aqueous medium, KX = X
/X
.
ROH
ROH
These distribution constants of the alcohol can be trans-
formed into the association constants of the alcohol to the
where n is the number of moles of surfactant present in the
s
t
system. In the dilution experiment at a constant n , n and n
s
a
o
t
micelle, K , which is more usual in the kinetic studies, as in
S
are varied so as to get a series of n /n and n /n values
a
s
o
s
KS = K v¯ H O, where v¯ H O is the molar volume of water.
X
2
2
whose graphical plotting according to Equation (3) can yield
i
a
o
a
w
a
The transformation of KS, KS
=
[ROH] /[ROH] -
m w
A
C
H
T
R
E
U
N
G
the values of n and n from the intercept and slope if n is
ROH
[surfactant], to the corresponding distribution constant of
negligible. The alcohol distribution constant, Koi , between
the continuous medium and the interface of the microemul-
sions can be obtained as the ratio between the intercept/
slope of plots such as shown in Figure 4:
ROH
the alcohol in the microemulsion, Kwi , is immediate,
½
ROHŠi ½H OŠ
ROH
2
Kwi
¼
¼ K ½H OŠ
ð5Þ
S
2
½
ROHŠw ½TTABrŠ
i
½
ROHŠi
n n
intercept
slope
ROH
a
o
where we have considered [H O] = 55.5m for direct micel-
Koi
¼
Z ¼
¼
ð4Þ
2
o
½
ROHŠo
n n
a
s
lar systems, since in such systems the volume occupied by
the surfactant may be negligible. Table 2 shows the calculat-
ROH
From plots analogous to those of Figure 5 we can get the
distribution constants for the different alcohols between the
isooctane and the interface of TTABr microemulsions.
ed values for Kwi for the different alcohols used in this
ROH
study. The values of Kwi for the 1-octanol and 1-decanol
were obtained by extrapolation on the basis of a linear rep-
resentation of logKwi against the number of carbon atoms
ROH
ROH
Values of Koi are shown in Table 1.
Chem. Eur. J. 2006, 12, 8284 – 8295
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8287

L. García-Río and P. Hervella
Table 2. Equilibrium constants for the distribution of alcohols between
the different pseudophases of the microemulsion.
several concentrations of 1-pentanol, 1-hexanol, 1-heptanol
and 1-octanol. Variation of the alcohol chain length had
little apparent effect on microemulsion structure where the
volume of the aqueous pseudophase can be taken as a con-
ꢀ1
ROH
Kwi
Kow
ꢀ1
ROH
K
S
/m
v¯ ROH/m
ꢀ
2
1-butanol
1-pentanol
1-hexanol
1-heptanol
1-octanol
1-decanol
0.90
3.24
9.9
38.6
127.8
1523
50
180
550
2142
7096
84528
0.72
0.173
5.9”10
1.3”10
3.7”10
2.7”10
9.15”10
0.109
0.126
0.141
0.157
stant.
A tendency of Koi
ꢀ
ꢀ
ꢀ
ꢀ
2
2
3
4
ROH
to decrease when the number of
carbon atoms in the alcohol increases has also been found in
[7]
0.191
experiments by Romsted and Yao when studying product
yields from dediazoniation of 4-hexadecyl-2,6-dimethylben-
zenediazonium tetrafluoroborate in microemulsions of cetyl-
trimethylammonium bromide (CTABr)/water/hexadecane/
alcohol (1-butanol or 1-hexanol). Their results indicate that
of the alcohol. This representation was constructed with the
ROH
values of Kwi calculated for 1-butanol, 1-pentanol, 1-hexa-
CTABr mixes ideally with both alcohols. The values of
ROH
nol and 1-heptanol (not shown, R=0.9994).
Koi
for 1-butanol and 1-hexanol are similar, but that for
On the basis of the values of the distribution constant of
butanol is slightly larger, probably because butanol is less
hydrophobic than hexanol and it associates more strongly
with the microemulsion interface. This trend has also been
w
the alcohols between water and isooctane, K , the value of
o
ROH
w
Koi was calculated. The values of Ko expressed as a molar
relationship between water and isooctane, K = X
w
o
w
ROH
[6]
ROH
/
observed by others. Constant values of Koi suggest that
ROH distributions between the interfacial and oil regions of
cationic w/o microemulsions depend primarily on the hydro-
phobic effect and that both butanol and hexanol mix ideally
with CTABr in the microemulsion aggregates. Thus, ROH
binding in these microemulsions does not depend upon spe-
cific interactions, for example changes in hydration or
charge–charge interactions, between ROH and CTABr head
groups that would depend upon ROH/CTABr head group
ratios in the interfacial region. However, as we mentioned
previously results obtained by Moulik and co-workers for al-
cohol incorporation to the interface of hexadecyl pyridinium
chloride/hexane/water/alcohol microemulsions show the op-
o
X
, can be expressed as:
ROH
ROH
oi
ROH
½
ROHŠ
Z
K
K
w
w
Ko
¼
¼
ð6Þ
½
ROHŠ W
o
wi
w
Ko values between isooctane and water were estimated con-
sidering that they must be in the vicinity of the published
values for alcohols distribution between octane/water,
decane/water and dodecane/water for 1-butanol, 1-pentanol,
[17]
w
1
-hexanol and 1-heptanol. Ko values for 1-octanol and 1-
decanol were obtained by extrapolation of a representation
of DG of transfer of the alcohol between water and isooc-
tane against the number of carbon atoms of the alcohol (not
[15f]
posite behavior.
w
The incorporation of alcohols into AOT/heptane/water
microemulsions shows a similar behavior to the nature of al-
shown, R=0.998). Table 2 shows the values of Ko and
ROH
Table 1 those of Koi for the different alcohols used in this
[6b,18]
cohol.
The incorporation of the alcohol increases when
study.
the length of the alkyl chain decreases. However, this de-
pendence is small, the partition constants changing by less
than a factor of 6 between ethanol and decanol. These re-
sults are interpreted in terms of an interfacial localization of
the polar head of the alcohols in the interface, with the alkyl
Discussion
ROH
Koi
Alcohol distribution constants: It is important to
chains extended toward the bulk solvent.
ROH
oi
ROH
point out two conclusions drawn from the K
data in
On the basis of Koi
values (Table 1) it is possible to
ROH
Table 1. Firstly, the Koi values obtained by the Schulmanꢁs
titration method and those extrapolated from aqueous mi-
celles agree quite well. Discrepancies are less than 10% for
butanol, pentanol and hexanol and higher for heptanol, oc-
tanol and decanol. The differences may be attributed to un-
transfer the four-component microemulsion into one of
three pseudo components. In the following discussion we
ROH
will use the Koi
values obtained by the Schulmanꢁs titra-
tion method in order to avoid any assumption derived from
ROH
the Koi
values extrapolated from aqueous micelles. This
certainties in alcohol binding constants to aqueous micelles
transformation means that we must redefine the composi-
tion parameters of the system. Given that the alcohol can be
considered as a solubilized substance at the interface, we
will apply Stilbsꢁ approximation considering that the sur-
factant, TTABr, is the only component which determines
the properties of the interface. Hence, the composition pa-
ROH
for very hydrophobic ones. Importantly the Koi
obtained
values are close to published values for the incorporation of
[19]
butanol, pentanol and hexanol from hexane to hexadecyl
ROH
oi
pyridinium chloride w/o microemulsions: K
and 48, respectively.
= 35, 34
[15f]
ROH
Secondly, the Koi values obtained
from the Schulmanꢁs titration method show no dependence
of Koi on the length of the hydrocarbon chain of the alco-
rameter, W, W = [H O]/[TTABr], is not modified. Howev-
er, the composition of the continuous medium is altered by
the presence of the alcohol and therefore we will redefine
A
H
R
U
G
2
ROH
hol. This result is consistent with those obtained by small
[6c]
angle neutron scattering measurements with microemul-
sions comprising hexadecane/potassium oleate/water, plus
the parameter Z, Z = [isooctane]/[TTABr], as
A
H
R
U
G
8288
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8284 – 8295

Water-in-Oil Microemulsions
FULL PAPER
½
isooctaneŠ þ ½ROHŠ
Z ½ROHŠ
ROH
o
tot
Z * ¼
¼ Z þ
½
TTABrŠ
ðKoi þ ZÞ ½TTABrŠ
ð7Þ
where [ROH]tot refers to the total alcohol concentration
added to the microemulsion. In experiments carried out in
this study we showed that [ROH]tot/
A
H
R
U
G
system TTABr/1-hexanol/isooctane/water was also studied,
varying the molar relationship [ROH]tot/
and 5.
A
H
R
U
G
Variation of the volume of the interface: In simple three
component microemulsions, that is, AOT/isooctane/water,
the molar volume of the interface can be considered to
remain constant. However, in the presence of cosurfactants
its incorporation into the interface causes the molar volume
of the interface to be equal to the molar volume of the sur-
factant plus the molar volume of the bonded alcohol. The
application of the formalism developed in the previous sec-
tion implies that the number of moles of alcohol incorporat-
ed into the interface increases along with the concentration
Figure 6. Influence of TTABr concentration upon Vtot/V
TTABr/1-hexanol/isooctane/water microemulsions. [ROH]tot/[TTABr]=
i
at constant W in
5. (*) W=10; (*) W=15; (&) W=20; (
W=40. Lines are drawn for clarity.
&
) W=30; (~) W=35 and (~)
[20]
of the surfactant. The alcohol incorporation into the inter-
face means that its volume varies along with the composi-
tion of the microemulsion. The volume of the interface can
be calculated considering it to be equal to the sum of the
volumes of the surfactant and the alcohol incorporated into
the interface. This approximation is similar to that used in
[21]
simple micellar systems (water/surfactant).
Figure 7. Influence of TTABr concentration upon Vtot/V
TTABr/1-hexanol/isooctane/water microemulsions.
(*)
TTABr]=3; (*) [ROH]tot/[TTABr]=4; (&) [ROH]tot/[TTABr]=5. Lines
are drawn for clarity.
i
at W=10 in
Vi ¼ Vꢀ TTABr ½TTABrŠ þ V
ꢀ
ð8Þ
ð9Þ
i
ROH
½ROHŠi
[ROH]tot/
[
ROH
Vtot
Vi
1
Koi þ Z
¼ Vꢀ
þ Vꢀ
ROH
½TTABrŠ
ROHKoi ½ROHŠ
TTABr
tot
As for the molar volume of the surfactant, TTABr, the
the volume of the interface with the surfactant concentra-
tion and with the molar relationship [ROH]tot [TTABr].
ꢀ
1
¯
value of V
=0.336m was used, which corresponds with
/
ACHTREUNG
TTABr
[22]
the complete micellar volume of the TTABr micelles. The
molar volume of the alcohol, v¯ ROH, was calculated on the
basis of its molecular density and weight (values in Table 2).
Figure 6 shows how the relationship V /V [values calcu-
Nitrosation of piperazine: The significant effect of the alco-
hols on the formation of w/o microemulsions is well docu-
mented. For example, the addition of alcohol molecules of
appropriate chain length to a ternary system may considera-
bly enlarge the microemulsion range. Alcohols should be
considered as cosolvents that are distributed between the
pseudophases of the microemulsion, thereby decreasing the
tot
i
lated according to the Eq. (9)] decreases as the surfactant
concentration increases. The decrease in V /V with the
tot
i
TTABr concentration is a consequence of the alcohol incor-
poration into the interface and its consequent increase in
volume. Hence, the calculated values for V /V are inde-
[23,24]
effective hydrophilicity of the amphiphile
and increasing
tot
i
pendent of the W parameter. This behavior is due to the
quantity of solubilized alcohol in the aqueous pseudophase
of the microemulsion being negligible. Figure 7 shows how
V /V varies with the surfactant concentration for different
the hydrophilicity of the oil pseudophase. The increase in
the hydrophilic character of the continuous medium can
alter the distribution equilibrium of the MNTS and amine
through the different pseudophases of the microemulsion.
Indeed, we have confirmed that piperazine is distributed be-
tween the three pseudophases of the microemulsion when
tot
i
molar relationships [ROH]tot/
A
H
R
U
G
molar relationship [ROH]tot/[TTABr] decreases, so does the
AHCTREUNG
[12]
alcohol concentration incorporated into the interface and
an alcohol is used as a cosurfactant. This behavior is dif-
ferent from that observed in AOT/isooctane/water microe-
hence its volume decreases, increasing the V /V relation-
tot
i
[13]
ship.
mulsions and is consistent with the increase in hydrophi-
licity of the continuous medium. Hence, we can propose a
kinetic model (Figure 8) to explain the experimental behav-
ior observed in the nitrosation of piperazine by MNTS.
The kinetic results obtained should be explained by
taking into account the alcohol incorporation into the inter-
face of the microemulsion and consequently the variation of
Chem. Eur. J. 2006, 12, 8284 – 8295
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8289

L. García-Río and P. Hervella
Vtot ½PIPŠtot
Table 4. Slopes and intercepts at the origin obtained upon fitting Vi
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-hexanol/isooctane microemul-
sions at 258C.
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
1
1
2
2
0
5
0
5
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
3
3
3
3
3
3
3
391ꢁ7
37ꢁ1
106
8.2
6.8
5.7
5.5
5.0
4.4
434ꢁ45
597ꢁ59
577ꢁ47
647ꢁ32
701ꢁ41
906ꢁ57
53ꢁ5
88ꢁ5
101ꢁ10
118ꢁ9
139ꢁ7
204ꢁ9
Figure 8.
30
4
5
0
0
According to the kinetic model put forward the pipera-
zine will be distributed between the three pseudophases of
the microemulsion. Its concentration in each pseudophase
can be calculated on the basis of the corresponding distribu-
tion constants.
Vtot ½PIPŠtot
Table 5. Slopes and intercepts at the origin obtained upon fitting Vi
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-hexanol/isooctane microemul-
sions at 258C.
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
½
PIPŠi
½PIPŠ
PIP
wi
PIP
i
K
¼
W
Koi
¼
Z *
ð10Þ
10
15
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
4
4
4
4
4
4
4
291ꢁ3.3
26.1ꢁ0.3
42.5ꢁ0.8
60.9ꢁ1.4
86.6ꢁ1.5
101ꢁ2.4
109ꢁ3.4
125ꢁ5.6
11.17
9.25
7.15
6.63
6.12
5.84
4.93
½
PIPŠw
½PIPŠ
w
394ꢁ15.2
435ꢁ25.3
575ꢁ28.4
620ꢁ43.2
637ꢁ59.4
614ꢁ94.8
2
0
where [PIP] , [PIP] and [PIP] are the amine concentrations
30
w
i
o
35
40
45
in the aqueous pseudophase, interface and continuous
medium respectively, with reference to the total volume of
the microemulsion. The composition parameter of the mi-
croemulsion Z*, Z* = ([isooctane]+[ROH] )/[TTABr],
A
H
R
U
G
o
has been introduced to reduce the four component microe-
mulsion to one of three pseudo components. The minimal
solubility of MNTS in an aqueous medium suggests that
MNTS is only found distributed between the interface and
Vtot ½PIPŠtot
Table 6. Slopes and intercepts at the origin obtained upon fitting Vi
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-hexanol/isooctane microemul-
sions at 258C.
MNTS
the continuous medium. Its distribution constant, Koi , is
MNTS
oi
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
defined in the same way as that of the amines, K
=
i
10
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
1-hexanol
5
5
5
5
5
5
420ꢁ7
34.6ꢁ0.7
55ꢁ2
12.1
8.3
7.8
5.5
6.5
5.4
(
[MNTS] Z*)/[MNTS] . The rate constant k corresponds to
A
C
H
T
R
E
U
N
G
i
o
2
1
2
3
5
0
0
456ꢁ26
524ꢁ8
the bimolecular rate constant for the nitrosation of pipera-
zine by MNTS at the interface of the microemulsion. Given
67ꢁ1
593ꢁ29
753ꢁ69
775ꢁ89
108ꢁ2
115ꢁ4
144ꢁ7
that the kinetic model, which will be set out below, includes
35
40
i
the volume of the interface, the rate constant k will be a bi-
2
ꢀ
1
ꢀ1
molecular rate constant with units m s .
From the proposed kinetic model it can be observed that
the reactants will be in contact at the interface and in the
continuous medium, in such a way that two possible reactive
zones will exist in the microemulsion. However, the rate of
the nitroso group transfer reactions from alkyl nitrites or N-
methyl-N-nitrosobenzenesulfonamides to amines considera-
Vtot ½PIPŠ
tot
Table 7. Slopes and intercepts at the origin obtained upon fitting Vi
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-heptanol/isooctane microemul-
sions at 258C.
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
[25]
bly decreases along with the polarity of the medium, al-
lowing the mechanism of the reaction to be modified. This
behavior means that we can disregard the possibility that
1
0
1-heptanol
1-heptanol
1-heptanol
1-heptanol
1-heptanol
1-heptanol
4
4
4
4
4
4
303ꢁ8
27.4ꢁ0.8
39ꢁ2
11.1
10.2
7.67
6.56
6.21
5.21
15
20
30
400ꢁ26
453ꢁ21
553ꢁ7
59ꢁ2
[26]
84ꢁ1
the reaction may take place in the continuous phase.
3
4
5
0
615ꢁ65
594ꢁ14
99ꢁ7
114ꢁ4
Vtot ½PIPŠtot
Table 3. Slopes and intercepts at the origin obtained upon fitting Vi
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-pentanol/isooctane microemul-
sions at 258C.
Bearing this in mind, and on the basis of the kinetic model
of Figure 9, we can obtain the following expression for kobs
.
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
10
15
20
30
1-pentanol
1-pentanol
1-pentanol
1-pentanol
4
4
4
4
275ꢁ12
325ꢁ43
398ꢁ44
499ꢁ38
24ꢁ1
38ꢁ4
51ꢁ4
88ꢁ2
11.1
8.5
7.8
MNTS
oi
PIP
PIP
Vtot
K
Koi Kwi ½PIPŠ
tot
i
2
kobs ¼ k
MNTS
PIP
oi
PIP
PIP
PIP
Vi ðKoi
þ Z Þ ðK Kwi þ Koi W þ Kwi Z Þ
*
*
5.6
ð11Þ
8290
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8284 – 8295

Water-in-Oil Microemulsions
FULL PAPER
Vtot ½PIPŠtot
Table 8. Slopes and intercepts at the origin obtained upon fitting Vi
Tables 3–9 show the values of the intercepts and slopes
obtained by applying Equation (12) in the nitrosation of PIP
by MNTS in TTABr microemulsions stabilized by different
alcohols. Figure 10 shows how well Equation (13) is fulfil-
led for the nitrosation of PIP by MNTS in TTABr/1-hexa-
nol/isooctane/water microemulsions where the relationship
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-octanol/isooctane microemul-
sions at 258C.
[12]
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
10
15
20
25
30
35
40
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
4
4
4
4
4
4
4
313ꢁ9
24.6ꢁ0.9
37ꢁ3
12.7
10.6
8.6
8.2
6.5
388ꢁ10
435ꢁ64
561ꢁ84
528ꢁ83
570ꢁ59
643ꢁ72
[
1-hexanol]/[TTABr]=5 has remained constant.
A
H
R
U
G
50ꢁ3
68ꢁ2
81ꢁ2
94ꢁ4
6.0
5.5
117ꢁ5
Vtot ½PIPŠtot
Table 9. Slopes and intercepts at the origin obtained upon fitting Vi
kobs
versus Z* [Eq. (12)] to the experimental kinetic data for the reactions of
piperazine with MNTS in water/TTABr/1-decanol/isooctane microemul-
sions at 258C.
W
ROH
[ROH]tot/[TTABr]
Intercept
Slope
Int/Slp
10
15
20
25
1-decanol
1-decanol
1-decanol
1-decanol
4
4
4
4
310ꢁ67
371ꢁ71
464ꢁ71
533ꢁ37
27ꢁ2
49ꢁ3
63ꢁ2
9ꢁ3
11.5
7.5
7.4
5.9
Figure 10. Variation of the relationship intercept/slope of the data of
Figure 7, Table 6, in accordance with Equation (13) (see text).
[27]
Equation (11) can be rewritten as:
The kinetic model represented in Figure 8 [Eq. (12)] pre-
dicts the existence of a linear dependency between the inter-
cepts of Figure 9 (data in Table 6) and W according to Equa-
tion (14):
ꢀ
ꢁ
Vtot ½PIPŠ
1
W
tot
¼
þ
þ
i
i
2
PIP
wi
Vi
k
obs
k
k K
2
ꢀ
ꢁ
ð12Þ
1
1
W
þ
þ
Z *
i
PIP
oi
i
MNTS
oi
i
2
MNTS
oi
PIP
Kwi
k K
k K
k K
2
2
1
W
intercept ¼ ki
þ
ð14Þ
This Equation predicts the existence of a linear dependency
i
2
PIP
k K
wi
Vtot ½PIPŠtot
2
between
versus Z*. The results shown as an exam-
Vi
kobs
ple in Figure 9 for nitrosation of piperazine by MNTS in mi-
croemulsions of TTABr/1-hexanol/isooctane/water, main-
Figure 11 shows by way of example that the values of the in-
tercepts of Table 6 (Figure 9) comply with the behavior pat-
tern predicted by the kinetic model of Figure 8, Equa-
tion (14), for the nitrosation of piperazine by MNTS in
TTABr/1-hexanol/isooctane/water microemulsions, in which
taining a relationship of [1-hexanol]/[TTABr] = 5, entirely
A
H
R
U
G
fulfill the behavior pattern predicted by Equation (12).
This Equation predicts the existence of a linear dependen-
cy between the relationship intercept/slope with 1/W accord-
ing to Equation (13):
the relationship [1-hexanol]/[TTABr]=5 remains constant.
A
H
R
U
G
Likewise, on the basis of the kinetic model, Equation (12),
the existence of a linear dependence can be observed be-
tween the intercepts of the representations of Figure 9 (data
intercept
slope
1
W
MNTS
MNTS
PIP
¼
Koi
þ Koi
Kwi
ð13Þ
Figure 9. Linearization of the data of Figure 1 in accordance with Equa-
tion (12) (see text) for nitrosation of piperazine by MNTS in TTABr/1-
hexanol/isooctane/water microemulsions. [1-hexanol]/[TTABr]=5. * W=
Figure 11. Variation of the intercepts and slopes of Figure 7 in accordance
1
0; (*) W=15; (&) W=20; (
&
) W=30; (~) W=35 and (~) W=40.
with Equations (14) and (15), Table 6, (see text).
Chem. Eur. J. 2006, 12, 8284 – 8295
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8291

L. García-Río and P. Hervella
in Table 6) and W, in accordance with the following Equa-
tion.
ꢀ
ꢁ
1
1
slope ¼
þ
i
2
PIP
oi
i
2
MNTS
oi
k K
k K
ð15Þ
W
MNTS
oi
1
W
MNTS
oi
þ
ꢂ
þ
k K
i
2
PIP
Koi
i
2
MNTS
i
2
PIP
Kwi
k K
k K
oi
Figure 12.
Figure 11 also shows that Equation (15) is fulfilled for the
nitrosation of PIP by MNTS in TTABr/1-hexanol/isooctane/
water microemulsions confirming, therefore, the validity of
the kinetic model developed in Figure 8.
constants observed in the microemulsions, showing that, as
in the case of piperazine, in w/o microemulsions the reaction
effectively takes place only at the interface, so that the reac-
tion rate in the continuous medium is considered negligible
To obtain the values of the distribution constants of the
PIP
MNTS
[26]
amine, Kwi , and of the MNTS, Koi , as well as the rate
compared with the reaction at the interface.
i
constant for the reaction at the interface, k , we have fol-
From Figure 12 and applying a similar methodology to
that developed for the nitrosation of piperazine we can
obtain rate Equation (16):
2
lowed the following scheme. The relationship between the
MNTS
slopes of Equations (14) and (15) gives Koi . On the basis
MNTS
of the slope of Equation (13) and the value of Koi
we can
MNTS
oi
MeBzAm
oi
obtain the distribution constant of the amine between the
Vtot
Vi
K
K
½MeBzAmŠ
ðKoi þZ *Þ
i
2
tot
PIP
kobs ¼ k
ð16Þ
MNTS
MeBzAm
aqueous pseudophase and the interface, Kwi . From the
ðKoi þZ *Þ
MNTS
oi
slope of Equation (15) and with the known values of K
PIP
and Kwi we can obtain the bimolecular rate constant for
the reaction in the interface, k . Table 10 shows the values
This Equation can be transformed into:
i
2
MNTS
MeBzAm
oi
MNTS MeBzAm
K
oi oi
Vtot ½MeBzAmŠ
ðKoi þZ *ÞðK
þZ *Þ
tot
¼
ð17Þ
i
Vi
k
k K
obs
2
Table 10. Kinetic parameters and partition coefficients estimated by fit-
ting Equations (12)–(15) to the experimental kinetic data for the reaction
of PIP with MNTS in water/TTABr/alcohol/isooctane and in AOT/isooc-
tane/water microemulsions at 258C.
The application of Equation (17) predicts the existence of a
Vtot ½MeBzAmŠtot
linear and quadratic dependence of
versus Z*
Vi
kobs
MNTS
Koi
PIP
Kwi
i
2
ꢀ1 ꢀ1
ROH
[ROH]tot/[TTABr]
k /m
s
independently of the values of W. As way of example
Figure 13 shows the fulfillment of the Equation (17) for the
nitrosation of MeBzAm in TTABr/1-pentanol/isooctane/
water microemulsions for values of W between 10 and 45
and values of [TTABr] between 0.1 and 0.7, confirming the
validity of the proposed kinetic model.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
3
3
3
3
3
3
3
3
1
1
1
1
1
1
1
-pentanol
-hexanol
-hexanol
-hexanol
-heptanol
-octanol
-decanol
4
3
4
5
4
4
4
3.5
3.0
3.6
3.5
3.5
3.4
3.8
11
22.4
25.0
21.3
23.9
22.6
28.0
23.4
9.5
3.93”10
3.39”10
4.79”10
3.40”10
4.42”10
3.45”10
2.81”10
5.33”10
To fit Equation (17) to the experimental results we took
AOT/isooctane/water microemulsion
MNTS
as a constant the value of Koi
previously obtained in the
MNTS
PIP
i
2
of Koi , Kwi and k obtained for the different systems
studied as the length of the hydrocarbon chain of the alco-
hol varies and as the molar relationship [1-hexa-
nol] /[TTABr] varies. In all cases the results obtained are
A
C
H
T
R
E
U
N
G
tot
consistent with the kinetic model proposed in Figure 8 and
with the transformation of a quaternary microemulsion into
one of three pseudo components where a variation in the
volume is produced in the interface as the composition of
the microemulsion varies.
Nitrosation of N-methylbenzylamine: The minimal solubility
of N-methylbenzylamine in water means that it is found in
the continuous medium and the interface of the microemul-
sion, as shown in Figure 12.
Vtot ½PIPŠtot
Figure 13. Plot of
versus Z* for nitrosation of N-methylbenzyla-
Vi
kobs
mine by MNTS in TTABr/1-pentanol/isooctane/water microemulsions.
1-pentanol]/[TTABr]=4. The solid line is the theoretical line predicted
by Equation (17) (parameters in Table 11). [N-methylbenzylamine]=
.10m. * W=10; (*) W=15; (&) W=20; ( ) W=30; (~) W=35; (~
W=40 and (!) W=45.
[
The observed rate constants for reaction of MNTS with
N-methylbenzylamine (and other amines) in pure isooc-
0
&
)
[25]
tane are several orders of magnitude slower than the rate
8292
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8284 – 8295

Water-in-Oil Microemulsions
FULL PAPER
MNTS
[20]
nitrosation of piperazine (values of Koi
in Table 10).
mulsions have shown that the polarity of the interface is
always substantially less than that of pure water. This reduc-
tion of the polarity of the interface with regard to that of
MeBzAM
Values of the distribution constants of the amine, Koi
and those of the bimolecular rate constants at the interface
of the microemulsions, k , (values in Table 11) can be ob-
,
i
2
i
pure water is the factor responsible for the decrease in k2
tained for TTABr-based microemulsions stabilized by 1-pen-
tanol, 1-hexanol and 1-heptanol.
with regard to the bimolecular rate constant in water, k .
2
MNTS
oi
PIP
Table 10 shows the values for the parameters K
and k obtained when studying this reaction in AOT/isooc-
, Kwi
i
2
[13]
tane/water microemulsions. The values of the distribution
constants of the amine and the MNTS are not comparable
between the AOT and TTABr microemulsions since surfac-
tants of a different nature are being dealt with. However, it
should be possible to establish a comparison between the
Table 11. Kinetic parameters and partition coefficients estimated by fit-
ting Equation (15) to the experimental kinetic data for the reaction of
MeBzAm with MNTS in water/TTABr/alcohol/isooctane and in AOT/
isooctane/water microemulsions at 258C.
MNTS
MeBzAM
i
2
ꢀ1 ꢀ1
ROH
[ROH]/[TTABr]
Koi
Koi
k /m
s
values of the rate constant for the reaction at the interface.
i
2
ꢀ
ꢀ
ꢀ
ꢀ
4
4
4
3
From k values obtained in the quaternary microemulsions
1
1
1
-pentanol
-hexanol
-heptanol
4
4
4
3.7
3.5
3.9
11
62.4
103
100
25.6
8.14”10
6.99”10
8.50”10
3.10”10
with the different alcohols and different molar relationships
i
2
[
3
alcohol]/
A
T
E
N
[TTABr] allows us to obtain a mean value of k =
AOT/isooctane/water microemulsion
ꢀ3 ꢀ1 ꢀ1
.74”10 m s . This value is very close to that obtained in
i
ꢀ3 ꢀ1 ꢀ1
the AOT microemulsions, k = 5.33”10
m s , although
2
slightly lower. This slight reduction could be due to the dis-
tribution of the alcohol between the pseudophases making
the interface of the microemulsion more hydrophobic.
Comparison of results: Tables 10 and 11 show the values of
MNTS
PIP
MeBzAM
i
MNTS
MeBzAM
i
Koi , Kwi , Koi
and k obtained for the nitrosation of
Table 11 shows the values of Koi , Koi
and k ob-
2
2
piperazine and N-methylbenzylamine in the different micro-
tained for the nitrosation of N-methylbenzylamine by
emulsions used. The values of Table 10 show that the distri-
MNTS in quaternary microemulsions of TTABr and tertiary
PIP
MNTS
bution constants of the amine, K , and of the nitrosating
ones of AOT/isooctane/water. Koi
values remained con-
wi
MNTS
agent, Koi , obtained in the nitrosation of PIP are inde-
stant and equal to those obtained in the nitrosation of PIP
MeBzAM
i
pendent of the nature of the alcohol used as a cosurfactant
by MNTS. Neither the values of Koi
nor those of k2
and of the molar relationship [1-hexanol]/
values of Kwi are in the vicinity of 23 in all cases. If we ana-
A
H
R
U
G
show any dependence on the nature of the alcohol used as a
PIP
i
cosurfactant. The values of k are approximately 50 times
2
lyze the variation of the distribution constant of the nitrosat-
lower than the value of the bimolecular rate constant ob-
MNTS
ꢀ2 ꢀ1 ꢀ1
ing agent, Koi , a similar behavior pattern is observed. A
tained in pure water, k = 4.10”10 m s . As in the nitro-
2
MNTS
mean value of Koi
can be obtained, of around 3.5 for all
sation of piperazine, this reduction is considered to be due
to the reduction of polarity in the interface. It is important
the microemulsions. It is worth noting that none of the ki-
netic parameters depend to any significant extent on the
presence of alcohol, which must therefore affect the experi-
mental rate constant solely by increasing the reaction
volume, that is, by diluting the reactants at the interface of
the microemulsion.
to point out that in the nitrosation of the piperazine an
i
2
eight-fold reduction is observable in k with regard to the
value of k obtained in pure water, while in the nitrosation
2
of N-methylbenzylamine this reduction is approximately
fifty-fold. This difference could be a consequence of the fact
that the piperazine has two positions in which nitrosation
can take place, which constitutes a statistical advantage for
the nitrosating process. This advantage must be more impor-
tant at the interface of the microemulsion than in pure
From the point of view of reactivity it is more interesting
to study the variation of the rate constant of nitrosation of
piperazine by MNTS at the interface of the microemulsions,
i
2
k . This reaction has been thoroughly researched in water in
our laboratory and we have obtained a value of the bimolec-
water due to the high microviscosity of the interface of the
ꢀ
2
ꢀ1 ꢀ1
[32]
ular rate constant in pure water of k2 = 2.98”10
m
s .
microemulsions.
This high microviscosity gives rise to a
The values obtained at the interface of the microemulsion,
lower degree of mobility of the reactants and hence the
steric impediments or the statistical advantages of the reac-
tion will be more significant than in pure water.
i
k , are always lower than the value obtained in pure water,
2
and an eight-fold decrease can be observed in the rate con-
stant. The state of water within tertiary, AOT, microemul-
sions and the influence of W on the micellar properties have
It is also important to point out the difference observed in
k when we compare microemulsions of three components
i
2
[28]
been investigated using many techniques.
CTABr has
(AOT/isooctane/water) and those of four components
(TTABr/alcohol/isooctane/water). The incorporation of the
alcohol at the interface of the latter displaces water mole-
cules and thus increases its hydrophobicity. This increase
could be responsible for the reaction rate being less at the
interface of TTABr microemulsions than in AOT microe-
been known for several years to be capable of forming mi-
croemulsions in various solvents such as n-hexanol, chloro-
[
29]
[30]
form and dichloromethane.
H NMR spectral
There is infrared
and
1
[29e,31]
evidence for a change of the proper-
ties of water in these cationic reverse micelles with an in-
crease in W. Likewise recent studies carried out in our labo-
ratory on solvolytic processes in TTABr and SDS microe-
i
2
AOT
i TTABr
2
mulsions, (k ) /(k )
ꢂ3.9 and 1.42 for nitrosation of
MeBzAm and PIP respectively.
Chem. Eur. J. 2006, 12, 8284 – 8295
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8293

L. García-Río and P. Hervella
Conclusions
shortly afterwards. N-methyl-N-nitroso-p-toluenesulfonamide (MNTS)
was supplied by Merck.
Microemulsions of the desired compositions were prepared from stock
water/surfactant/isooctane/alcohol microemulsions by addition of appro-
priate amounts of isooctane and/or water. During the experiments the
molar relationship [alcohol]/[TTABr] remained constant and equal to 4
and the nature of the alcohol varied. Likewise the influence of the
changes in the molar relationship was studied [1-hexanol]/[TTABr]
making it equal to 3, 4 and 5. Densities were measured with a pycnome-
ter.
The presence of the cosurfactant not only modifies the prop-
erties of the interface but also increases the hydrophilic
character of the continuous medium. This change in the
properties can cause a distribution of the reactants between
the pseudophases of the microemulsion, which is different
from that observed in microemulsions of three components.
There follows a summary of the main conclusions of this
study.
We quantified the distribution of the cosurfactant through
the different pseudophases of the microemulsion on the
basis of the Schulmanꢁs titration method and by using the
association constants of the alcohols to simple micellar sys-
tems (water/surfactant) and the distribution constants of the
alcohols between water and isooctane. The calculated results
are compatible with those experimentally obtained for the
composition of the interface of the microemulsion.
The transnitrosation reactions were carried out using a Varian Cary 500
Scan UV/Vis-NIR spectrophotometer fitted with thermostated cell hold-
ers (all experiments were carried out at 25.08C). Kinetic measurements
were carried out following the disappearance of the absorbance at
2
60 nm due to MNTS consumption for nitrosation of PIP and the initial
ꢀ
4
MNTS concentration was 1.91”10 m. In the case of N-methylbenzyla-
mine, its high molar absorptivity precluded us from studying the reaction
at 260 nm; it was therefore necessary to follow the reaction at 392 nm
using an initial MNTS concentration of 2.0”10 m. In all cases MNTS
was in deficit against amine concentration. The kinetic data always fitted
the first order integrated rate Equation satisfactorily (R>0.999); in what
follows, kobs denotes the pseudo first order rate constant.
ꢀ
3
The incorporation of the alcohol into the interface of the
microemulsion causes an increase in the volume of the inter-
face with the consequent dilution of the MNTS and amine.
The volume of the interface was calculated on the basis of
the molar volumes of the surfactant and the alcohol incor-
porated into the interface considering that an ideal mixture
is formed where the volumes are additives. The variation in
the volume of the interface was introduced into the kinetic
model proposed to explain the experimental behavior.
The application of a kinetic model developed on the basis
of the formalism of the micellar pseudophase enabled us to
obtain the values of the distribution constants of the reac-
tants between the different pseudophases of the microemul-
sion and the values of the rate constants of the reaction at
Acknowledgements
This work was supported by Ministerio de Ciencia y Tecnología (Project
CTQ2005-04779) and Xunta de Galicia (PGIDT03-PXIC20905PN and
PGIDIT04TMT209003PR).
[
1] a) P. K. Das, A. Chaudhuri, Langmuir 1999, 15, 8771–8775; b) M.
Giustini, G. Palazzo, G. Colafemmina, M. Della Monica, M. Giomi-
ni, A. Ceglie, J. Phys. Chem. 1996, 100, 3190–3198; c) B. Sengupta,
J. Guharay, P. K. Sengupta, Spectrochim. Acta Part A 2000, 56,
1
433–1441; d) A. M. Vinogradov, A. S. Tatikolov, S. M. B. Costa,
Phys. Chem. Chem. Phys. 2001, 3, 4325–4332; e) G. Gonzalez-Gaita-
no, M. Valiente, G. Tardajos, G. Montalvo, E. Rodenas, J. Colloid
Interface Sci. 1999, 211, 104–109; f) O. A. El Seoud, J. Mol. Liq.
1997, 72, 85–103; g) I. M. Cuccovia, L. G. Dias, F. A. Maximiano, H.
Chaimovich, Langmuir 2001, 17, 1060–1068; h) G. Palazzo, F.
Lopez, M. Giustini, G. Colafemmina, A. Ceglie, J. Phys. Chem. B
i
i
the interface, k . The values of k are in all cases lower than
2
2
those obtained in pure water, which has been interpreted as
a consequence of the reduction in polarity of the interface
in comparison with that of pure water. The values of the dis-
tribution constants of the reactants and the rate constants,
2
003, 107, 1924–1931; i) F. Lopez, G. Cinelli, L. Ambrosone, G. Co-
i
2
k , are independent of the length of the hydrocarbon chain
lafemmina, A. Ceglie, G. Palazzo, Coll. Surf. A Physicochem. Eng.
Aspects 2004, 237, 49–59; j) G. Palazzo, L. Carbone, G. Colafemmi-
na, R. Angelico, A. Ceglie, M. Giustini, Phys. Chem. Chem. Phys.
of the alcohol used as a cosurfactant and the molar relation-
ship [alcohol]/[TTABr], indicating that the main effect of
the cosurfactant is the increase in volume of the interface
2
004, 6, 1423–1429.
[
2] a) E. Caponetti, A. Lizzio, R. Triolo, W. L. Griffin, J. S. Johnson,
Langmuir 1990, 6, 1628–1634; b) E. Caponetti, A. Lizzio, R. Triolo,
W. L. Griffin, J. S. Johnson, Langmuir 1992, 8, 1554–1562.
with the consequent dilution of the reactants. The values of
i
2
k
obtained in TTABr/alcohol/isooctane/water microemul-
sions are always lower than those obtained in AOT/isooc-
tane/water microemulsions. This result is considered to be
due to the incorporation of the alcohol into the interface of
the microemulsion displacing the water molecules and thus
increasing its hydrophobic character.
[3] a) C. Petit, A. S. Bommarius, M. P. Pileni, T. A. Hatton, J. Phys.
Chem. 1992, 96, 4653–4658; b) M. Lagues, C. Sauterey, J. Phys.
Chem. 1980, 84, 3503–3508; c) S. R. Bisal, P. K. Bhattacharya, S. P.
Moulik, J. Phys. Chem. 1990, 94, 350–355.
[
4] W. K. Kegel, G. A. van Aken, M. N. Bouts, H. N. W. Lekkerkerker,
J. Th. G. Overbeek, P. L. de Bruyn, Langmuir 1993, 9, 252–256.
5] a) R. Zana, S. Yiv, C. Strazielle, P. Lianos, J. Colloid Interface Sci.
[
1
981, 80, 208–223; b) K. Shinoda, B. Lindman, Langmuir 1987, 3,
1
35–149.
[
6] a) L. Damaszewski, R. A. Mackay, J. Colloid Interface Sci. 1984, 97,
66–175; b) E. A. Lissi, D. Engel, Langmuir 1992, 8, 452–455; c) E.
Experimental Section
1
Caponetti, A. Lizzio, R. Triolo, W. L. Griffith, J. S. Johnson, Lang-
muir 1992, 8, 1554–1562.
[7] J. Yao, L. S. Romsted, J. Am. Chem. Soc. 1994, 116, 11779–11786.
[8] J. Lang, N. Lalem, R. Zana, J. Phys. Chem. 1991, 95, 9533–9541.
[9] See for example: a) M. Valiente, E. Rodenas, J. Phys. Chem. 1991,
95, 3368–3370; b) M. L. Moyµ, C. Izquierdo, J. Casado, J. Phys.
Tetradecyltrimethylammonium bromide (TTABr) was supplied by Sigma
and used without further purification. Alcohols and isooctane were Al-
drich products of the highest degree of purity available commercially. Pi-
perazine (PIP) (Merck) was used without further purification. N-Methyl-
benzylamine (MeBzAm) (Aldrich) was distilled under argon and used
8294
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8284 – 8295

Water-in-Oil Microemulsions
FULL PAPER
Chem. 1991, 95, 6001–6004; c) L. García-Rio, J. R. Leis, J. C.
Mejuto, J. Phys. Chem. 1996, 100, 10981–10988; d) F. P. Cavasino,
C. Sbriziolo, M. L. Turco Liveri, J. Phys. Chem. B 1998, 102, 3143–
[18] S. Pꢂrez-Casas, R. Castillo, M. Costas, J. Phys. Chem. B 1997, 101,
7043–7054.
[19] P. Stilbs, J. Colloid Interface Sci. 1982, 87, 385–394.
[20] a) L. García-Rio, J. R. Leis, C. Reigosa, J. Phys. Chem. B 1997, 101,
5514–5520; b) L. García-Río, J. R. Leis, Chem. Commun. 2000,
455–456; c) L. García-Río, J. R. Leis, J. A. Moreira, J. Am. Chem.
Soc. 2000, 122, 10325–10334; d) L. García-Río, J. R. Leis, J. C.
Mejuto, Langmuir 2003, 19, 3190–3197.
3
146; e) F. P. Cavasino, C. Sbriziolo, M. L. Turco Liveri, J. Phys.
Chem. B 1998, 102, 5050–5054, and references therein; f) E. N. Du-
rantini, C. D. Borsarelly, J. Chem. Soc. Perkin Trans. 2 1996, 719–
7
23.
[
10] a) M. C. R. Franssen, J. G. J. Weijnen, J. P. Vincken, C. Laane, H. C.
Van der Plas, Biocatalysis 1988, 1, 205–216; b) R. Hilhorst, R.
Spruijt, C. Laane, C. Veeger, Eur. J. Biochem. 1984, 144, 459–466;
c) B. Hoffmann, H. Jackle, P. L. Luisi, Biopolymers 1986, 25, 1133–
[21] C. Bravo, J. R. Leis, M. E. PeÇa, J. Phys. Chem. 1992, 96, 1957–
1
961.
22] A. K. Yatsimirski, K. Martinek, I. V. Berezin, Tetrahedron 1971, 27,
855–2868.
23] M. H. G. M. Penders, R. Strey, J. Phys. Chem. 1995, 99, 10313–
0318.
24] M. Kahlweit, R. Strey, G. Busse, J. Phys. Chem. 1991, 95, 5344–
352.
25] a) L. García-Río, J. R. Leis, E. Iglesias, J. Org. Chem. 1997, 62,
[
[
[
[
2
1
1
156; d) J. P. Samama, K. M. Lee, J. F. Biellmann, Eur. J. Biochem.
987, 163, 609–617; e) C. A. Martin, P. M. McCrann, M. D. Ward,
1
G. H. Angelos, D. A. Jaeger, J. Org. Chem. 1984, 49, 4392–4396;
f) C. Minero, E. Pramauro, E. Pelizzetti, Langmuir 1988, 4, 101–
5
1
9
05; g) R. D. R. Pereira, D. Zanette, F. Nome, J. Phys. Chem. 1990,
4, 356–361; h) S. Adhikari, R. Joshi, C. Gopinathan, Int. J. Chem.
4
1
701–4711; b) L. García-Río, J. R. Leis, E. Iglesias, J. Org. Chem.
997, 62, 4712–4720; c) L. García-Río, J. R. Leis, J. A. Moreira, D.
Kinet. 1998, 30, 699–705.
[
11] a) A. Castro, J. R. Leis, M. E. PeÇa, J. Chem. Soc. Perkin Trans. 2
Serantes, Eur. J. Org. Chem. 2004, 614–622.
1
989, 1861–1866; b) L. García-Río, E. Iglesias, J. R. Leis, M. E.
[
26] We have experimentally observed that the rate constant for nitroso
group transfer from MNTS to secondary amines in isooctane is sev-
eral orders of magnitude smaller than that observed in the TTABr/
alcohol-based microemulsions. Moreover we have tested that addi-
tion of moderate quantities of alcohol to isooctane have a small
effect on the rate constant. These results allow us to neglect the pos-
sibility of nitroso group transfer taking place in the continuous
PeÇa, A. M. Rios, J. Chem. Soc. Perkin Trans. 2 1993, 29–37.
12] L. García-Río, J. R. Leis, J. Phys. Chem. B 2000, 104, 6618–6625.
13] L. García-Río, J. R. Leis, M. E. PeÇa, E. Iglesias, J. Phys. Chem.
[
[
1
993, 97, 3437–3442.
[
14] The mechanism by which cosurfactants stabilize micelles is still not
very clear, although much research has been carried out. Han and
co-workers (D. Shen, R. Zhang, B. Han, Y. Dong, W. Wu, J. Zhang,
J. Li, T. Jiang, Z. Liu, Chem. Eur. J. 2004, 10, 5123–5128 and D.
Shen, B. Han, Y. Dong, W. Wu, J. Chen, J. Zhang, Chem. Eur. J.
medium of the microemulsion.
27] The inequality is due mainly to the high value of Koi .
PIP
[
[
28] a) H. F. Eicke, Top. Curr. Chem. 1980, 87, 85–145; b) P. L. Luisi,
L. J. Magid, CRC Crit. Rev. Biochem. 1986, 20, 409–474; c) Y.
Chevalier, T. Zemb, Rep. Prog. Phys. 1990, 53, 279–371; d) A. Goto,
H. Yoshioka, M. Manabe, R. Goto, Langmuir 1995, 11, 4873–4875;
e) T. K. De, A. Maitra, Adv. Colloid Interface Sci. 1995, 59, 95–193.
29] a) P. Ekwall, L. Mandell, K. Fontel, J. Colloid Interface Sci. 1969, 29,
2
005, 11, 1228–1234) have demonstrated that compressed CO
the function of a cosurfactant that stabilizes Triton X-100 or AOT-
based microemulsions. It has been proposed that CO , a small and
2
has
2
linear molecule, can penetrate into the surfactant tail region and
thus push the surfactant head groups together. Suitable penetration
of compressed CO may increase the rigidity of the surfactant film,
2
which decreases the attractive interaction between droplets effec-
tively, and lead to a larger critical droplet radius. However, as is
pointed by the authors, extrapolation of these results to convention-
al cosurfactants is risky because alcohols and other cosurfactants
contains both a polar group and a hydrocarbon chain, and it is very
[
6
39–646; b) M. Seno, K. Sawada, K. Araki, K. Iwamoto, H. Kise, J.
Colloid Interface Sci. 1980, 78, 57–64; c) P. D. I. Fletcher, M. F.
Galal, B. H. Robinson, J. Chem. Soc. Faraday Trans. 1 1985, 2053–
2
1
065; d) J. Lang, G. Mascolo, R. Zana, P. L. Luisi, J. Phys. Chem.
990, 94, 3069–3074; e) R. Germani, G. Savelli, G. Cerichelli, G.
Mancini, L. Luchetti, P. P. Ponti, N. Spreti, C. A. Bunton, J. Colloid
Interface Sci. 1991, 147, 152–162.
2
difficult to clarify their functions, while compressed CO is a nonpo-
lar molecule that solely has the function of the hydrocarbon chain in
a conventional cosurfactant.
[
[
30] a) P. D. Profio, R. Germani, G. Onori, A. Santucci, G. Savelli, C. A.
Bunton, Langmuir 1998, 14, 768–772; b) J. Sunamoto, K. Iwamoto,
S. Nagamatsu, H. Kondo, Bull. Chem. Soc. Jpn. 1983, 56, 2469–
[
15] a) J. E. Bowcott, J. H. Z. Schulman, Z. Elektrochem. 1955, 59, 283–
2
2
88; b) W. Gerbacia, H. L. Rosano, J. Colloid Interface Sci. 1973, 44,
42–248; c) V. K. Bansal, D. O. Shah, J. P. OꢁConel, J. Colloid Inter-
2
472; c) H. Kondo, I. Miwa, J. Sunamoto, J. Phys. Chem. 1982, 86,
4
826–4831.
face Sci. 1980, 75, 462–475; d) K. S. Birdi, Colloid Polym. Sci. 1982,
6, 628–631; e) H. N. Singh, S. Swarup, R. P. Singh, S. M. Saleem,
31] a) R. Germani, P. P. Ponti, N. Spreti, G. Savelli, A. Cipiciani, G. Cer-
2
ichelli, C. A. Bunton, V. Si, J. Colloid Interface Sci. 1990, 138, 443–
Ber. Bunsen Ges. 1983, 87, 1115–1120; f) S. P. Moulik, L. G. Digout,
W. M. Aylward, R. Palepu, Langmuir 2000, 16, 3101–3106; g) S. K.
Hait, S. P. Moulik, Langmuir 2002, 18, 6736–6744; h) M. Giustini, S.
Murgia, G. Palazzo, Langmuir 2004, 20, 7381–7384.
4
50; b) R. Germani, P. P. Ponti, T. Romeo, G. Savelli, N. Spreti, G.
Cerichelli, L. Luchetti, G. Mancini, C. A. Bunton, J. Phys. Org.
Chem. 1989, 2, 553–558.
[32] a) M. Hasegawa, T. Sugimura, Y. Suzaki, Y. Shindo, A. Kitahara, J.
Phys. Chem. 1994, 98, 2120–2124; b) E. Keh, B. Valeur, J. Colloid
Interface Sci. 1981, 79, 465–478; c) P. E. Zinsli, J. Phys. Chem. 1979,
83, 3223–3231.
[
16] a) R. Zana, S. Yiv, C. Strazielle, P. Lianos, J. Colloid Interface Sci.
1
981, 80, 208–223; b) H. Hoiland, A. M. Blokhus, O. J. Kvammen, S.
Backlund, J. Colloid Interface Sci. 1985, 107, 576–578; c) I. V. Rao,
E. Ruckenstein, J. Colloid Interface Sci. 1986, 113, 375–387.
Received: January 20, 2006
[
17] R. Aveyard, R. W. Mitchell, J. Chem. Soc. Faraday Trans. 1 1969, 65,
Revised: March 22, 2006
2
645–2653.
Published online: August 9, 2006
Chem. Eur. J. 2006, 12, 8284 – 8295
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8295