Micellization.pdf

M

Micellar Emulsion

▶Microemulsions

Micellar Systems

Tharwat Tadros

Wokingham, Berkshire, UK

Synonyms

Microemulsions

Definition

Micellar systems or microemulsions are a special

class of “dispersions” (transparent or translucent)

which are better described as “swollen micelles.”

A convenient way to describe microemulsions is

to compare them with micelles. The latter which

are thermodynamically stable may consist of

spherical units with a radius that is usually less

than 5 nm. Two types of micelles may be consid-

ered: normal micelles with the hydrocarbon tails

forming the core and the polar head groups in

contact with the aqueous medium and reverse

micelles (formed in nonpolar media) with

a water core containing the polar head groups

T. Tadros (ed.), Encyclopedia of Colloid and Interface ScienDOI 10.1007/978-3-642-20665-8, # Springer-Verlag Berlin

and the hydrocarbon tails now in contact with

the oil. The normal micelles can solubilize oil in

the hydrocarbon core forming oil-in-water (O/W)

microemulsions, whereas the reverse micelles

can solubilize water forming a water-in-oil

(W/O) microemulsion. A rough guide to the

dimensions of micelles, micellar solutions, and

macroemulsions is as follows: micelles, R<5 nm

(they scatter little light and are transparent);

macroemulsions, R >50 nm (opaque and milky);

and micellar solutions or microemulsions,

5–50 nm (transparent, 5–10 nm, translucent,

10–50 nm). A thermodynamic definition of

microemulsions can be obtained from a consid-

eration of the energy and entropy terms for for-

mation of microemulsions. The increase in

surface area when forming a microemulsion is

DA, and the surface energy increase is equal to

DA g12. The increase in entropy is T DSconf (sincea large number of droplets can arrange them-

selves in several ways compared with one oil

drop, which has much lower entropy). According

to the second law of thermodynamics, the

free energy of formation of microemulsions

DGm is given by the following expression:

DGm ¼ DA g12 �TD Sconf . With microemulsions

DA g12 � T DSconf (this is due to the ultralow

interfacial tension accompanied with

microemulsion formation) and DGm �0. The

system is produced spontaneously, and it is ther-

modynamically stable. Thus, microemulsions

are better described as “swollen micelles.”

ce,Heidelberg 2013

M 686 Micelle Formation

The best definition of microemulsions is as fol-

lows: “System of Water + Oil + Amphiphile

that is a single Optically Isotropic and Thermo-

dynamically Stable Liquid Solution.” Amphi-

philes refer to any molecule that consists

of hydrophobic and hydrophilic portions, e.g.,

surfactants, alcohols. The driving force for

microemulsion formation is the low interfacial

energy which is overcompensated by the entropy

of dispersion term. The low (ultralow) interfa-

cial tension is produced in most cases by com-

bination of two molecules, referred to as the

surfactant and cosurfactant (e.g., medium-chain

alcohol).

Cross-References

▶ Interfacial Tension

▶Microemulsions

▶ Surfactants

Micelle Formation

▶Micellization

Micellization

Tharwat Tadros

Wokingham, Berkshire, UK

Synonyms

Micelle formation

Definition

Micellization is a process of aggregation of sur-

factant ions or molecules above a critical con-

centration (that is referred to as the critical

micelle concentration, c.m.c.) to form self-

assembly structures referred to as micelles.

The process of micellization is evident from

consideration of the solution properties of sur-

factants which show abrupt changes at a critical

concentration that is consistent with the fact that

above this concentration (the c.m.c.), surface

active ions or molecules in solution associate to

form larger units that are referred to as self-

assembled structures or micelles. The latter can

be spherical, rod-like, or lamellar structures. The

c.m.c. depends on the surfactant structure and the

medium in which they are present. For

a surfactant series with the same polar head

group (ionic or nonionic), the c.m.c. decreases

with increase of the alkyl chain length of the

surfactant molecule. For a given alkyl chain,

the c.m.c increases with increase of the size

of the polar head group (e.g., number of ethylene

oxide units). For ionic surfactants, the c.m.c.

increases slowly with increase of temperature,

and it decreases with addition of electrolyte. The

driving force of the process of micellization is the

entropy increase on association. Two main

sources of entropy increase can be described.

The first is due to the so-called hydrophobic

effect. The surfactant monomers contain “struc-

tured” water around their hydrocarbon chain. On

transfer of such monomers to a micelle, these

water molecules are released and they have

a higher entropy. The second source of entropy

increase on micellization may arise from

the increase in flexibility of the hydrocarbon

chains on their transfer from an aqueous to

a hydrocarbon medium. The orientations and

bendings of an organic chain are likely to be

more restricted in an aqueous phase compared to

an organic phase. It should bementioned that with

ionic and zwitterionic surfactants, an additional

entropy contribution, associated with the ionic

head groups, must be considered. Upon partial

neutralization of the ionic charge by the counter-

ions when aggregation occurs, water molecules

are released. This will be associated with an

entropy increase which should be added to the

entropy increase resulting from the hydrophobic

effect mentioned above. However, the relative

contribution of the two effects is difficult to assess

in a quantitative manner.

Microcapsules and Controlled Release 687 M

Cross-References

▶Micelle Formation

▶ Self-assembly Structures

▶ Surfactants

Microcapsule

Tharwat Tadros

Wokingham, Berkshire, UK

Synonyms

Microcapsules and controlled release

M

Definition

Microcapsules with particles in the size range

1–100 mm consist of a distinct capsule wall

(mostly a polymer) surrounding a biologically

or chemically active core. They are applied

for the controlled release of the active ingredient

(a.i.), its protection against the environment

in which they are dispersed or eliminating the

material encapsulated from interacting with

other substances in the formulation. Microencap-

sulation is mainly carried out by interfacial

condensation, in situ polymerization, and coacer-

vation. Interfacial condensation is perhaps the

most widely used method for encapsulation in

industry. The a.i. which may be oil soluble, oil

dispersible, or oil itself is first emulsified in water

using a convenient surfactant or polymer.

A hydrophobic monomer A is placed in the oil

phase (oil droplets of the emulsion), and

a hydrophilic monomer B is placed in the aque-

ous phase. The two monomers interact at the

interface between the oil and the aqueous phase

forming a capsule wall around the oil droplet.

The role of surfactant in this process is

crucial since an oil-water emulsifier (with high

hydrophilic-lipophilic balance, HLB) is required.

Alternatively, a polymeric surfactant such as

partially hydrolyzed polyvinyl acetate (referred

to as polyvinyl alcohol, PVA) or an poly(ethylene

oxide-propylene oxide-ethylene oxide) (PEO-

PPO-PEO, Pluronic) block copolymer can be

used. The emulsifier controls the droplet size

distribution and hence the size of capsules

formed. On the other hand, if the material to be

encapsulated is water soluble, a water-in-oil

(W/O) emulsion is prepared using a surfactant

with low HLB number or an A-B-A block copol-

ymer of polyhydroxystearic acid-polyethylene

oxide-polyhydroxystearic acid (PHS-PEO-

PHS). In this case, the hydrophilic monomer is

dissolved in the aqueous internal phase droplets.

In interfacial polymerization, the monomers

A and B are polyfunctional monomers capable

of causing polycondensation or polyaddition

reaction at the interface. Examples of oil soluble

monomers are polybasic acid chloride, bis-

haloformate, and polyisocyanates, whereas

water soluble monomers can be polyamine or

polyols. Thus, a capsule wall of polyamide, poly-

urethane, or polyurea may be formed. Some

trifunctional monomers are present to allow

cross linking reactions. If water is the second

reactant with polyisocyanates in the organic

phase, polyurea walls are formed. The latter

modification has been termed in situ interfacial

polymerization. One of the most useful microen-

capsulation processes involves reactions that

produce formation of urea-formaldehyde (UF)

resins.

Cross-References

▶Agrochemical Formulations

▶Emulsions

▶ Polymeric Surfactant

▶ Surfactants

Microcapsules and ControlledRelease

▶Microcapsule

M 688 Microemulsions

Microemulsions

Julian Eastoe1, Marios Hopkins Hatzopoulos1

and Rico Tabor2

1University of Bristol, Bristol, UK2School of Chemistry, Monash University,

Clayton, Australia

Synonyms

Micellar emulsion; Nanoemulsion; Swollen

micelle

Keywords

Enhanced oil recovery; Light scattering;

Microemulsions; Nanoparticles; Percolation;

Phase behavior; Self-diffusion NMR; Small-angle

neutron scattering; Soil remediation; Stability

Definition

Microemulsions are dispersions of two or more

immiscible or partially miscible fluids stabilized

by added surfactants. The dispersed domains are

generally in the nanometer size range. Visually,

they are transparent or translucent, and their

appearance does not change with time.

Microemulsions are considered to be thermody-

namically stable.

Overview

Microemulsions may be considered as a subset of

emulsions, exhibiting certain unique properties.

Added surfactants perform a key role to reduce

interfacial tension sufficiently to promote ther-

modynamic stability of microemulsions. At the

same time, microemulsion droplets are short

lived; they are amenable to investigation by an

array of techniques, such as electrical conductiv-

ity, self-diffusion NMR, light scattering, and

small-angle neutron scattering. Microemulsions

may be employed in processes of nanoparticle

synthesis, soil remediation, and enhanced oil

recovery. Microemulsion science is a steadily

developing and highly promising field with

many recent technological advancements.

Theory

Introduction

Microemulsions comprise two or more immisci-

ble or partially miscible fluids, stabilized by

added surfactants. The surfactants must promote

sufficiently low interfacial tensions to permit the

formation of nanometer scale domains. This

characteristic nanometer length scale explains

the transparency (translucency), which is

a distinguishing feature of microemulsions. The

visual appearance of microemulsions does not

change with time. Microemulsions are distin-

guished from emulsions, which are generally

opaque due to of light scattering from the dis-

persed domains as a result of their size and refrac-

tive index difference from the continuous fluid.

The visual appearance of emulsions changes with

time as the dispersed and continuous fluids

resolve. Therefore, it can be reasonably assumed

that microemulsions are thermodynamically sta-

ble systems, whereas emulsions are thermody-

namically unstable.

The visual appearance of microemulsions,

transparency (or translucency) and stability over

time, suggests that microemulsions could be con-

sidered as single-phase (monophasic) systems.

This immediately provokes consideration of

what is meant by “single phase.”

A phase is a region of a material that has the

same composition throughout, persisting up to

a boundary with another phase. A phase may

contain one or more components but is uniform

in composition. In the case of a pure fluid

(or a mixture), of a single phase, the time average

molecular distribution is constant. The Gibbs

phase rule (Eq. 1) indicates the number of phases

in equilibrium, based on the number of compo-

nents and on the degrees of freedom. The degrees

of freedom are intensive properties of the

system which are independent from each other.

Microemulsions 689 M

Intensive properties are also independent of the

size of the system. For this purpose, chemical

compositions are defined in terms of mole frac-

tions which are clearly intensive properties.

M

F ¼ C� Pþ 2 (1)

Where F is the degrees of freedom, C the

number of components, and P is the number of

phases (Rock 1969).

It is commonly stated that microemulsions

form single phases; however, that does not hold

at all length scales, since microemulsions consist

of nanometer-sized dispersed domains in a bulk

fluid. Indeed, based on visual observations over

macroscopic length scales, microemulsions

are apparently monophasic. However, when

interrogated at microscopic length scales (i.e.,

1–100 nm), because of the presence of identifi-

able dispersed nanometer-sized domains, it

becomes clear that microemulsions are not

monophasic but are in fact biphasic. In

microemulsions, domains (or droplets) of one

phase are dispersed in another phase, separated

by amonolayer of surfactant stabilizer molecules.

Microemulsions were first identified in the

early 1940s by Schulman et al. (Hoar and

Schulman 1943). Initially, they were referred to

as hydrophilic “oleomicelles” or “oleophillic

hydromicelles,” “swollen micelles,” and “micel-

lar emulsions,” though these terms have largely

been discontinued nowadays. The term

“microemulsion” was coined in the late 1950s,

but until the mid-1970s, they were viewed as

something of a scientific curiosity with little

research being conducted on them. Research

interest picked up during the “oil crisis” in the

early 1970s because microemulsions can be used

in tertiary oil recovery (i.e., the partial removal of

the residual oil remaining in the well rock) but

faded again as the oil crisis receded and tertiary

oil recovery became commercially unrealistic

due to its high cost.

The unique properties of microemulsions

make them important systems for drug delivery

(Wei et al. 2005), synthesis of polymers and

high-value nanoparticles, in catalysis, enhanced

oil recovery, liquid–liquid extractions, and

nanoparticle recovery (Abecassis et al. 2009;

Hollamby et al. 2010; Faizan et al. 2011), as

well as applications in pharmaceuticals (Lv

et al. 2006), detergency (Kling 1949), and lubri-

cation (Hone et al. 2000). Therefore, the field

remains sufficiently important to continue to

attract considerable research interest. For that

reason, a great deal of progress has been made

in the last 20 years in understanding

microemulsion properties. In particular, interfa-

cial film stability and microemulsion structures

can now be characterized in detail owing to the

development of new and powerful techniques

such as small-angle neutron scattering (SANS,

as described below). The following sections

deal with the fundamental microemulsion prop-

erties of formation and stability, surfactant films,

classification, and phase behavior.

To understand which kind of systems may

(and may not) be considered as microemulsions,

it is useful to compare them with normal

emulsions.

Emulsions and Microemulsions

Emulsions are systems of with at least one fluid

dispersed in another fluid: emulsions may be

called macro-, mini-, nano-, and microemulsions.

The fluids must be immiscible or partially

miscible. Most emulsions can be categorized as

oil-in-water (o/w) or water-in-oil (w/o), although

other more complex kinds are known. In this

field, the characteristic size of the internal

domain constitutes another important way to

distinguish between categories: microemulsions

have domains between 1 and 100 nm;

miniemulsions, between 0.1 and 1 mm; and for

macroemulsions, the internal domains (droplets)

are typically >1 mm.

Hence, the term microemulsion can be mis-

leading, as the droplets tend to be submicron

sized being the 1–100-nm-size range. As

a result,microemulsions are transparent (or trans-

lucent) which, confusingly, has also led to an

increased use of the term “nanoemulsions.”

Other names have been used for microemulsions

over the years, including “swollen micelles” and

“micellar emulsions,” though these terms have

largely been discontinued nowadays.

M 690 Microemulsions

The principal difference between macro-

emulsions/miniemulsions and microemulsions

is explained in terms of stability: all other

emulsions are kinetically stable but thermody-

namically unstable, whereas microemulsions are

thermodynamically stable but kinetically unsta-

ble. This is evident when it is considered how the

different systems are prepared. Emulsions

(macro-, mini-, nano-) are prepared by three

main routes:

(a) Batch comminution, which requires energy

being supplied to the two bulk fluids until

one becomes dispersed within the other.

(b) Continuous comminution by which one fluid

is forced through a capillary (or frits) emerg-

ing as droplets in the second fluid.

(c) Nucleation and growth is a method in which

two fluids are completely miscible under cer-

tain conditions. By change of one of the

intensive properties (such as temperature) in

a controlled manner, the fluids become

immiscible, and in the presence of surface-

active stabilizers, a dispersion will form

(Vincent et al. 1998).

Emulsions are kinetically stable which means

the impact of droplet-droplet collisions is mini-

mized. However, given time, the drops will even-

tually coalesce, the larger drops growing at the

expense of the smaller ones and the two fluids

finally separate into two distinct phases. The rate

of this phase separation process (emulsion reso-

lution) depends on numerous physicochemical

factors, including the degree of miscibility (or

immiscibility) and on the viscosity of the disper-

sion. The process of emulsion resolution nor-

mally takes longer if the initial droplets are

smaller. To decrease the time of phase separation

and hence maintain the dispersion for longer

periods of time, stabilizers are employed. Stabi-

lizers act by adsorbing on the interface between

the two fluids. Appropriate stabilizers can be

surfactants, polymers, and colloidal- and

nanoparticles, which serve to enhance electro-

static interactions (surfactants, polymers, parti-

cles) or steric interactions (polymers, nonionic

surfactants). The result is to promote repulsive

and hinder attractive interactions between the

dispersed fluid drops, hence retarding droplet

coalescence so that the drops remain dispersed

for longer. Macroemulsions and miniemulsions

may form in the absence of stabilizers; however,

for microemulsions, the presence of surfactants is

essential. The reason for this becomes apparent

on consideration of the thermodynamics of for-

mation (see section “Formation”). Under appro-

priate conditions and in the presence of added

stabilizers, microemulsions form “spontane-

ously,” whereas emulsions do not.

Microemulsions comprise of three (or more)

components: two immiscible or partially miscible

fluids and surfactant or polymer stabilizers. The

origin of the thermodynamic stability is the sig-

nificant reduction in fluid–fluid interfacial ten-

sion go/w brought about by adsorption of highly

efficient surfactants (amphiphiles). As a result of

promoting ultralow interfacial tensions (< 0.10

mN m�1), larger internal surface areas are gener-

ated (surface area/volume � 105 m�1). It is often

found that standard surfactants (e.g., sodium

dodecylsulfate) are not capable of reducing go/wto the required levels, and an additional

cosurfactant or cosolvent is needed, for example,

medium-chain-length aliphatic alcohols such as

pentanol and hexanol. Studies of microemulsions

indicate that chemical structure(s) of the stabi-

lizer is all important, being one major factor

determining stability and formation. The other

factors are the chemical nature/composition of

the two fluid components and thermodynamic

variables temperature and pressure (Nave

et al. 2000a).

Thermodynamics

Formation

As stated above, the surfactant(s) lowers the

interfacial tension go/w between the two immisci-

ble fluids sufficiently such that nanometer-sized

droplets are formed. The droplet formation can be

understood in terms of the free energy changes on

forming a dispersion DGform. Given that the dis-

persed phase forms small droplets, the configura-

tional entropy change will be given by (Overbeek

1978):

DSconf ¼�nkB lnfþ 1�ff

� �ln 1�fð Þ

� �(2)

Microemulsions 691 M

M

where n and f are the number of droplets and the

volume fraction of the dispersed fluid, respec-

tively. The free energy of formation (microemul-

sification) is then the sum of this entropic term

and the energy required to create new interfacial

area A, given in terms of the change in interfacial

area per droplet, DA (which would be equal to

n4pr2 for n spherical droplets of radius r) and theinterfacial tension go/w

DGform ¼ DAgo=w � TDSconf (3)

For microemulsion formation to be spontane-

ous,DGform< 0, and hence DAgo/w> TDSconf. Ondispersion, the droplet number increases and so

DSconf is positive. Typically, the increase in inter-facial area is of the order 104–105, and the natural

fluid–fluid interfacial tension of surfactant

free systems, go/w, is of the order 50 mN m�1

(for the common example of water–oil). This

suggests DAgo/w is somewhere in the region of

1,000 kBT, hence, there is a need for the surfac-

tant to lower the interfacial tension significantly,

to the region of 0.01 mN m�1. Some surfactants

(double chain ionics (Kunieda and Shinoda 1980;

Chen et al. 1984) and some nonionics (Kahlweit

et al. 1990)) can produce extremely low interfa-

cial tensions – in the region of 10�2 to 10�4 mN

m�1 – but in most cases, such low values cannot

be achieved by a single surfactant. An effective

way to further decrease go/w, so as to promote

formation of a microemulsion, is to include a

second surface-active species (either a surfactant

or medium-chain alcohol), that is, a cosurfactant.

This can be understood in terms of the Gibbs

equation extended to multicomponent systems

(Hunter 1994). It relates the interfacial tension

to the surfactant film composition and the chem-

ical potential, m, of each component in the sys-

tem, that is,

dgow¼ �

Xi

Gidmið Þ � �Xi

GiRTdlnCið Þ (4)

whereCi is the molar concentration of component

i in the mixture and Gi the surface excess

(mol m�2). Assuming that surfactants and cosur-

factants, with concentration Cs and Cco,

respectively, are the only adsorbed components

(i.e., Gwater ¼ Goil ¼ 0), Eq. 4 becomes:

dgo=w ¼ GsRTd lnCs � GcoRTd lnCco (5)

Integration of Eq. 5 gives:

go=w ¼ goo=w

�Z Cs

0

GsRTd lnCs�Z Cco

0

GcoRTd lnCco

(6)

Gs and Gco are the surface excesses of surfac-

tant and cosurfactants, respectively. Equation 6

shows that go/w is decreased by both the surfactantand cosurfactant so their effects are additive.

Stabilization

An important quantity defining the composition

of a microemulsion is the w value, which in the

case of a w/o system is defined in terms of the

molar concentrations of water and surfactant, and

is effectively the stabilization efficiency:

w ¼ H2O½ �AOT½ � (7)

For o/w systems, the water concentration in

Eq. 7 would be replaced by the molar concentra-

tion of the dispersed oily component.

Based on simple spherical droplet geometry,

an approximate droplet (domain) size can then be

calculated fromw, as pointed out by Hilfiker et al.

(1987):

r ¼ 3wvwao

þ lc (8)

where r is the radius of the microemulsion drop-

let, vw is the volume of a dispersed phase mole-

cule (� 30 A3 in the case of water in w/o systems,

but depending on molecular structure for O/W

counterparts), and lc and ao are the length of the

fully extended surfactant hydrophobic chain and

the area per head group of the surfactant, respec-

tively. The range of w values accessible is depen-

dent on the surfactant and solvent combination;

100(a) (a)(b) (b)

(c) (c)

50R

00 15 30 45 20

0

50

DC8

C7D C6 C8C7 C6

100

a b

T/°C40 60

Microemulsions,Fig. 1 Phase behavior of

AOT-stabilized w/o

microemulsions. A:

(a) n-octane, (b) isooctane,(c) oct-1-ene; B: cyclic

alkanes and decalin,

(D) at fixed

[AOT] ¼ 0.10 mol dm�3

(Reprinted from Fletcher

et al. 1987 copyright RSC)

M 692 Microemulsions

some systemsmay only support very loww values

(below 10), but for others, values of 70–80 are not

uncommon. One added advantage in introducing

w is that it can simplify the presentation of the

phase behavior of microemulsion systems, by

reducing the number of degrees of freedom, F,

in the Gibbs phase rule (Eq. 1). For instance,

a ternary compositional plot (water-oil-surfac-

tant) can be reduced to a pseudobinary plot by

fixing the surfactant concentration and varying

the amount of water – i.e., a convenient and

useful plot of w versus temperature T could then

be presented. This is shown in Fig. 1 for the case

of w/o microemulsions in some common organic

solvents, stabilized by Aerosol-OT (AOT,

sodium bis(2-ethylhexyl)sulfosuccinate).

These phase diagrams show a typical funnel

shape, which is characteristic of AOT

microemulsions. The lower phase boundary is

termed the water solubilization boundary, where

the macroscopically single phase system will

undergo separation to form a so-called Winsor II

system – that is, separation into a water-in-oil

microemulsion phase in equilibrium with an

excess water part. At the upper temperature

phase boundary, the transition is into a Winsor I

system which is an oil-in-water droplet phase in

equilibrium with excess oil. The two alternative

possible microemulsion phases are a Winsor III

system, with a (bicontinuous) microemulsion

phase in equilibrium with both excess water and

excess oil, and a Winsor IV system, which is

a macroscopically single microemulsion.

Kinetic Aspects

This structural and compositional complexity of

microemulsions means that there are potentially

numerous relaxation processes. Despite this,

microemulsion kinetics have been relatively

well researched, and there have been several

important reviews of the area (Lopez-Quintela

et al. 2004a; Moulik and Paul 1998). The

dynamic processes which can occur can be sum-

marized as: (1) motion of interfacial molecules,

(2) exchange of droplet contents, and (3) coales-

cence of droplets. Motion of interfacial mole-

cules refers to the mobility of surfactant

molecules adsorbed at the interface. Ahlnas

determined that this occurs on the picosecond

timescale (Ahlnas et al. 1983). Exchange pro-

cesses can be of surfactant, cosurfactant, or dis-

persed solvent. For dispersed water droplets in

w/o systems, the exchange rate of immobilized or

“bound” water (in a layer associated with the

polar surfactant head groups) and free water was

determined to be on the millisecond timescale

using NMR (Hansen 1960).

Exchange rates of cosurfactant molecules

between interface and bulk solution seem to

vary with molecular size (mobility) and environ-

ment, but a typical rate of around 108 s�1

was found by Lang et al. for butan-1-ol in

Microemulsions 693 M

M

water/toluene/SDS microemulsions (Lang et al.

1975). Exchange of surfactant molecules may be

on a similar timescale but seems to be greatly

affected by many other parameters (Tondre

2005). It also seems that the exchange times

found in different studies are dependent on the

method of measurement, so the picture is not yet

complete. Exchange of material in droplets seems

to occur via a process of coalescence and then

rapid fission, with data supporting this theory

coming from Abe and coworkers (1987). They

found when mixing two three-component

microemulsions with differently sized droplets

a unimodal distribution was spontaneously

reached without stirring, demonstrating that

there must be an exchange process occurring

more rapidly than Ostwald ripening, which is

most plausibly interpreted as fusion–fission. The

seminal paper in this field came from Fletcher,

Howe, and Robinson, who emphasized that

droplet exchange (via coalescence/fission) is

rapid and continuous (Fletcher et al. 1987).

They were even able to determine the activation

barrier to water exchange for AOT-stabilized

water-in-oil microemulsions (of the order of

70–100 kJ mol�1), with an associated

rate constant for water exchange of

106–108 mol�1 dm3 s�1.

Therefore, for (typical) millimolar concentra-

tions of droplets, the internal contents of the

microemulsion domains exchange, typically on

the millisecond timescale (Fletcher et al. 1987,

1990). If collisions are sufficiently violent, then

the surfactant film may rupture thereby facilitat-

ing droplet exchange, which is why the droplets

are kinetically unstable. Hence, there is an asso-

ciated “activation energy” for droplet fusion. The

mechanism of droplet coalescence has been

reported for AOT w/o microemulsions (Fletcher

et al. 1987); the droplet exchange process was

characterized by a second-order rate constant kex,

which appears to be activation controlled (Milner

and Safran 1987a) (hence the activation energy,

Ea, barrier to fusion) and not purely diffusion

controlled. Other studies (Bancroft 1913a)

have shown that the dynamic aspects of

microemulsions are affected by the flexibility

of the interfacial film, that is, film rigidity

(see section “Film Bending Rigidity”), through

a significant contribution to the energy barrier.

Under the same experimental conditions, differ-

ent microemulsion systems can have different kexvalues (Fletcher et al. 1987): for AOT w/o system

at room temperature, kex is in the range

106–109 dm3 mol�1 s�1, and for nonionics CiEj,

108–109 dm3 mol�1 s�1 (Fletcher et al. 1987,

1990; Bancroft 1913a).

Predicting Microemulsion Type

A further microemulsion categorization (other

than w/o and o/w) is the aforementioned system

proposed by Winsor (1948) who introduced four

types of phase equilibria where two (or more)

phase portions are present in equilibrium:

• Type I: The system consists of a small portion

of organic component (oil) and surfactant

coexisting with the larger volume of

a microemulsion dominated by the polar fluid

(water): for example, a oil-in-water (o/w)

microemulsion in equilibrium with excess dis-

persed oil (Winsor I).

• Type II: The system is in effect the reverse of

type I, for example, with a w/o microemulsion

coexisting in equilibrium with excess water

(Winsor II).

• Type III: For common water + oil

microemulsions, this consists of a three

regions, with the top region the oil phase and

the bottom phase being the polar fluid phase.

The middle region is a microemulsion stabi-

lized by the majority of surfactant present

(Winsor III or middle-phase microemulsion).

• Type IV: It is a homogenous system, compris-

ing only one macroscopic phase (biphasic

on nanometer-length scales, see above)

(Winsor IV).

Phase transitions from Winsor I to Winsor II

via Winsor III “phases” can brought about by

increasing either electrolyte concentration (in

the case of ionic surfactants) or temperature (for

nonionic amphiphiles). To account for the type of

emulsion formed by a particular surfactant, Ban-

croft (1913a) and Clowes (1916) considered the

adsorbed film in emulsion systems to have an

inner and an outer interfacial tension acting inde-

pendently (Adamson 1960). The interface would

M 694 Microemulsions

curve in such a way as for the inner surface to

always have the higher tension. Bancroft’s rule

states the type of emulsion formed is dictated by

the preferred solubility of surfactant (i.e., the

tendency of a given surfactant to partition more

or less strongly into one or other of the fluids). If

the surfactant is preferentially water soluble, then

an oil-in-water (micro)emulsion will form, and

conversely if the surfactant is preferentially oil

soluble, then a water-in-oil (micro)emulsion is

expected. Other approaches supporting the Win-

sor and Bancroft classifications are discussed

below.

The R Ratio

Winsor (1948) proposed the R ratio as a means to

determine the type of (micro)emulsion formed.

The R ratio (Eq. 12) considers the strength of

interaction of the amphiphile with the two fluids

hence the dictating type of emulsion. The total

interaction of surfactant with the two fluids (Axy)

can be expressed as:

Axy ¼ ALxy þ AHxy (9)

where ALxy and AHxy are the interactions of the

nonpolar parts of the two molecules (dispersion

forces) and the polar interactions (such as hydro-

gen bonding and Coulombic interactions),

respectively. In the case of an oil–water system,

the interactions are:

Aso ¼ ALso þ AHco (10)

ASW ¼ ALsw þ AHsw (11)

where AHco and ALsw are small and often ignored.

Aso is the cohesive energy responsible for the

solvation of the surfactant in oil. ASW is that

responsible for solvation in water. Other cohesive

interactions are the following:

Aww, the interaction of water molecules between

themselves

Aoo, the interaction of oil molecules between

themselves

ALL, the interactions of the hydrophobic parts of

the surfactant molecules

AHH, the interaction of the hydrophilic parts of

the surfactant molecules

The above parameters oppose the miscibility

of the surfactant in oil and water. For the (micro)

emulsion to prove stable, that is, to prevent phase

separation from occurring, the difference

between Aso and ASW should not be large. In

other words, the surfactant should have sufficient

miscibility in both fluids. Winsor expressed this

balance as:

R ¼ Aso

ASW(12)

An extension of the theory (Bourrel and

Schechter 1988) accounts for all the cohesive

energies with:

R ¼ Aso � Aoo � ALLð ÞASW � Aww � Awwð Þ (13)

Based on the R ratio, the type of emulsion

formed can be predicted. When R > 1, the inter-

face increases its interaction with the oil phase at

the expense of water interaction and hence a w/o

emulsion forms. When R < 1, o/w emulsions

form, and when R ¼ 1, a type III or IV system

may form.

Hydrophilic–Lipophilic Balance (HLB)

Another concept relating molecular structure to

interfacial packing and film curvature is hydro-

philic–lipophilic balance (HLB). The HLB scale

supports Bancroft’s rule (Bancroft 1913b;

McBain 1950) which states that the type of emul-

sion formed is dependent on the preferential

partitioning of the emulsifier in either one of the

fluids. If the emulsifier is soluble in water more so

than in oil, then o/w emulsions form and vice

versa. The HLB is generally expressed as an

empirical equation based on the relative propor-

tions of hydrophobic and hydrophilic groups

within the molecule. As such, HLB is an attempt

at quantifying hydrophilicity or lipophilicity

(hydrophobicity) of any given amphiphile. In

the early days, the majority of work on the HLB

was carried out on nonionic surfactants which

Microemulsions 695 M

were of primary industrial importance. Griffin

(1949) characterized a number of surfactants

and derived an empirical equation for nonionic

alkyl polyglycol ethers (CiEj) based on the sur-

factant chemical composition (Griffin 1954):

M

HLB ¼ Ej þ OHwt%

5(14)

where Ejwt% and OHwt% are the weight percent

of ethylene oxide and hydroxide groups,

respectively.

For bicontinuous structures, that is, zero cur-

vature, it was shown thatHLB� 10 (Israelachvili

1994). Then, w/o (micro)emulsions form when

HLB < 10, and o/w (micro)emulsions when

HLB > 10.

Davies (1957) attempted to predict the type of

emulsion formed when water and oil were agi-

tated in the presence of an emulsifier. Davies’s

approach accounted for both ionic and nonionic

surfactants. The experiments conducted of rela-

tive coalescence rates for a particular oil and

surfactant, so that the rates of coalescence of an

oil drop and that of a water drop to the interface

were measured. From these rates, hydrophilic and

hydrophobic numbers were derived which were

proportional to the number of hydrophobic and

hydrophilic units of the surfactant. Introducing

these to Eq. 15, the HLB of the surfactant can

be calculated with good agreement in most cases

to those derived by Griffin.

HLB ¼ 7þ SnH � SnL (15)

This approach is popular in industry to guide

the preparation of (micro)emulsions and for the

choice of surfactants for a particular (micro)

emulsion system. Equation 15 can be rewritten

as:

HLB ¼ 7

þ S Hydrophilic group numbersð Þ� n group number per� CH2�ð Þ (16)

where n is the number of hydrophobic carbons in

the alkyl tail of the single chain surfactant, as

methyl and methylene groups are assigned the

same group number.

Lin (Lin et al. 1973) tried to extend Davies’s

work by relating HLB to the cmc of the amphi-

philes. It was noted by Klevens et al. (Klevens

1953) that

log cmcð Þ ¼ a� bn (17)

Lin et al. (1973) combined Eqs. 16 and 17

leading to

log cmcð Þ ¼ a0 þ b0 HLBð Þ (18)

where a and b are empirical constants for

a homologous series and from which a0 and b0

can be derived. Given a value for cmc, the HLB of

any member of a surfactant homologous series

can be calculated. Shinoda (Shinoda and Friberg

1986) however issued warning at this correlation

of the cmc to the HLB, as the latter can stay

constant with size of hydrophile and lipophile

while the cmc will increase geometrically. How-

ever, Lin only examined ionic surfactants and

their aqueous cmcs with the derived HLB num-

bers were in very reasonable agreement with

those predicted by previous methods.

Phase Inversion Temperature (PIT) or HLB

Temperature

HLB has been a widely used tool in the prepara-

tion of emulsions. However, it primarily deals

with the solubility of the surfactant in a general

water and oil emulsion. The HLB method, as

discussed in the previous section, does not

account for the behavior of the surfactant in dif-

ferent oils, polar solvents, temperatures, and

additives in the polar and nonpolar fluids.

Shinoda (Shinoda and Saito 1969) used

a characteristic property of the emulsions as an

indicator of the HLB of the surfactant. Nonionic

surfactants form microemulsions and emulsions

that are highly sensitive to temperature. At

a particular, system-specific point, known as the

phase inversion temperature (PIT) or HLB tem-

perature, the surfactant film curvature (explained

below) changes from positive (around oil) to

M 696 Microemulsions

negative (around water), and phase inversion

takes place from o/w to w/o. The origin of this

structural inversion is thought to be a result of

changes in the hydration of the hydrophilic sur-

factant moieties, which generally weakens with

increasing temperature. The reduced hydration

then effectively decreases the HLB value, as

pointed out by Shinoda et al. (Shinoda and Saito

1969):

• If T < PIT, an oil-in-water microemulsion

forms (Winsor I)

• I T> PIT, a water-in-oil microemulsion forms

(Winsor II)

• At T ¼ PIT, a middle-phase microemulsion

exists (Winsor III)

Packing Parameter and Microemulsion Structures

Recall that the chemical nature of the surfactant is

one of the important factors determining the type

of (micro)emulsion formed. Israelachvili et al.

(1976) studied changes in film curvature that

followed certain geometrical parameters of

the surfactant molecules. When considering sur-

factant in aqueous systems, the interactions

at play are the hydrophobic effect of the surfac-

tant tails promoting molecular aggregation and

the repulsive electrostatic and steric effects of

the head groups. As such, the hydrophobic

effect works toward decreasing the contact

between the polar environment, while the elec-

trostatic and steric forces work toward increas-

ing contact with the polar fluid. The most

favored geometry is formed when these interac-

tions are balanced.

To introduce the concept of “packing param-

eter,” consider first of all an aqueous micelle,

comprising Nagg surfactant molecules. Since the

structure is built up of many individual mole-

cules, it can be appreciated that the most favored

aggregation geometry will be governed by

a balance of intermolecular interactions, being

dominated by an optimal head group area ao(representing repulsive electrostatic and/or steric

interactions) and the hydrocarbon tail volume v

(being the limiting volume of more weakly

interacting and incompressible hydrocarbon

chains). The tails are also assumed to be at

a maximum extension lc, which is slightly less

than fully extended equivalent hydrocarbon chain

length lmax. Therefore,

lc � lmax (19)

According to Tanford (1978), the volume of

liquid hydrocarbon chains are essentially the

same as those for the same fragment in micellar

aggregates, as determined by X-ray scattering.

Assuming additivity and exclusion of water

from the hydrophobic core, the volume is propor-

tional to the number of surfactants through Nagg,

and the volumes of the hydrocarbon tails further.

Hence,

v nm3� � ¼ 0:0274þ 0:0269nc (20)

and

lmaxðnmÞ ¼ 0:154þ 0:126nc (21)

with nc being the number of carbon atoms in the

hydrocarbon chain and v and lmax having units of

nm3and nm, respectively. With these variables

considered, then the packing parameter can be

defined:

P ¼ v

aolc(22)

Geometrical constraints dictate:

P � 13spherical micelles

13< P � 1

2cylindrical or rod-like micelles

12< P � 1 vesicles and flexible bilayers

P ¼ 1 planar extended bilayers

P > 1 reverse micelles

The conditions that the aggregate must satisfy

irrespective of its shape are the following:

(a) No point in the aggregate can be further than

lc from the water interface.

(b) The total hydrocarbon volume or the aggre-

gate and the total interface area must follow

the relationship Vv ¼ A

ao¼ Nagg, where Nagg is

the aggregation number.

When discussing microemulsions, the above

conditions cease to hold. For instance, when oil is

introduced to a surfactant–water system where

the aggregates are cylindrical (and as a result

Microemulsions 697 M

M

stressed), the aggregate will become spherical,

thus favoring an o/w microemulsion

(Israelachvili et al. 1976).

• If ao > v/lc, oil-in-water microemulsion.

• If ao < v/lc, water-in-oil microemulsion.

• If ao � v/lc, a middle-phase microemulsion is

the preferred system.

Becher (1984) argued that the packing param-

eter, HLB, and R ratio approaches to understating

microemulsions were all consistent. The packing

parameter could be considered as an inverse HLB

number expressed in terms of volume fractions of

the hydrophobic parts.

Surfactant Film Properties

A further approach to understanding the forma-

tion of a particular type of (micro)emulsion is to

consider the mechanical properties of a surfactant

film at an oil–water interface. The three phenom-

enological constants governing the film behavior

are the tension, bending rigidity, and spontaneous

curvature. The extent of influence of each of these

parameters depends on the constraints felt by the

film. These parameters are of importance since

the film behavior governs the microemulsion

behavior and response in terms of phase behavior

and stability, structure, and solubilization

capacity.

Ultralow Interfacial Tension In creating

a microemulsion, the interfacial area rises by 4–5

orders ofmagnitude, and (as stated above) to permit

this, the interfacial tension needs to be considerably

lowered. Microemulsion formation requires

ultralow interfacial oil–water tensions, go/w, usuallyof the order of 10�2 to 10�4 mN m�1. Typical

surfactants are incapable of reaching such low

interfacial tension values, and for this reason,

addition of cosurfactant as well as electrolyte

and variation of temperature, pressure, and oil

chain length are required, often in combination.

Several studies have been reported on the effects

of composition variables on go/w. In particular,

Aveyard and coworkers performed systematic

interfacial tension studies with both ionic

(Aveyard et al. 1986a, b) and nonionic surfactants

(Aveyard et al. 1989), varying oil chain length,

temperature, and electrolyte (NaCl) content. For

example, in the systemwater–AOT–n-heptane, at

constant surfactant concentration (above its cmc),

a plot of go/w as a function of electrolyte concen-

tration shows a deep minimum that corresponds

to the Winsor phase inversion, that is, upon addi-

tion of NaCl, go/w decreases to a minimum critical

value (Winsor III structure) and then increases to

a limiting value close to 0.2–0.3 mN m�1 (Win-

sor II region). At constant electrolyte concentra-

tion, varying temperature (Aveyard et al. 1986a),

oil chain length, and cosurfactant content

(Aveyard et al. 1986b) have a similar effect.

With nonionic surfactants, a similar tension

curve and phase inversion were observed, but

instead with increasing temperature rather than

[NaCl] (Aveyard et al. 1989). In addition, when

increasing surfactant chain length, the interfacial

tension curves shift to higher temperatures and

the minimum in go/w decreases (Sottman and

Strey 1996). Ultralow interfacial tensions cannot

be measured with standard techniques such as Du

Nouy ring, Wilhelmy plate, or drop volume

(DVT), and spinning drop tensiometry (SDT) or

surface light scattering (Langevin 1992) must be

used.

Spontaneous Curvature Stability and struc-

tural transformations in microemulsions can

also be understood by considering the energy

associated with forming and bending of stabiliz-

ing interfacial films. In fact, this surface energy

contribution may be nonnegligible, because

interfacial areas in microemulsions are very

large (S/V � 105 m�1).

Spontaneous (natural or preferred) curvature

Co is defined as the curvature adopted a surfactant

film in the presence of equal amounts of water

and oil. This equilibrium condition imposes no

constraints on the film, which is then free to attain

the lowest free energy state. Relaxing the condi-

tion that the oil/water ratio is unity means

a deviation from Co. Every point on a surface

possesses two principal radii of curvature, R1

and R2, so that the associated principal curvatures

are C1 ¼ 1/R1 and C2 ¼ 1/R2 (see Fig. 2). The

separate mean and Gaussian curvatures are used

to define the bending of surfaces, defined as fol-

lows (Hyde et al. 1997):

R2R1

a

b c d

n

P

R1 < 0R2 > 0

R1 > 0R2 > 0 R2 > 0

R1 = ∝

Microemulsions,Fig. 2 Principal

curvatures of different

surfaces. (a) Intersection of

a surfactant film surface

with planes containing the

normal vector (n) to the

surface at the point p.(b) convex curvature,

(c) cylindrical curvature,

(d) saddle-shaped

curvature (After Hyde et al.

1997)

M 698 Microemulsions

Mean curvature : C ¼ 1=2 1=R1 þ 1=R2ð Þ (23)

Gaussian curvature : k ¼ 1=R1 � 1=R2 (24)

If a circle is placed tangentially to a point p on

the surface and if the circle radius is chosen so

that its second derivative at the contact point

equals that of the surface in the direction of the

tangent (of normal vector, n), then the radius of

the circle is a radius of curvature of the surface.

The curvature of a surface is described by two

such circles chosen in orthogonal (principal)

directions as shown in Fig. 2a. For a sphere, R1

andR2 are equal and positive (Fig. 2b): for a cylin-

drical surface, R2 is indefinite (Fig. 2c); whereas,

for a plane, both R1 and R2 are indefinite. One

special case is a saddle structure, R1 ¼ �R2, that

is, at every point the surface is both concave and

convex (Fig. 2d), so that both a plane and saddle

have the property of zero mean curvature.

In this model, the curvature of the surface Co

depends both on the composition of the phases it

separates and on surfactant type. Considering the

apolar side of the interface, it is possible for oil

molecules to penetrate to some extent between

the surfactant hydrocarbon tails. The more exten-

sive the penetration, the more curvature is

imposed toward the polar side. This would result

in a decrease of Co since, by convention, positive

curvature is toward oil (negative toward water).

The longer the oil chains, the less they penetrate

the surfactant film and the smaller the effect on

Co. Eastoe et al. have studied the extent of solvent

penetration in microemulsions stabilized by

dichained surfactants, using SANS with selective

deuteration. Results suggested that oil penetra-

tion is only a subtle effect, depending on the

chemical structures and architecture of both

surfactant and oil. In particular, unequal surfac-

tant chain lengths for d-chain compounds,

Microemulsions 699 M

M

(Eastoe et al. 1996a, b, 1997a, b) or the presence

of C ¼ C bonds (Bumajdad et al. 1998) result in

a more disordered surfactant/oil interface,

thereby providing a region of enhanced oil

mixing. For symmetric dichained surfactants

(e.g., didodecyldimethylammonium bromide

DDAB and the anionic AOT), however, no evi-

dence for oil mixing into the interfacial region

was found (Eastoe et al. 1997a). The effect of

alkane structure and molecular volume on the oil

penetration was also investigated with n-heptane

and cyclohexane. The results indicate that hep-

tane is essentially absent from the layers, but the

more compact cyclohexane has a slightly greater

penetrating effect (Eastoe et al. 1997b).

Surfactant type, and nature of the polar head

group, also influences Co through different inter-

actions with the polar (aqueous) phase:

• For ionic surfactants, electrolyte content and

temperature affect Co in opposite ways.

Increasing salt concentration screens electro-

static head group repulsions – i.e., decreases

head group area – so the film curves more

easily toward water, leading to a decrease in

Co. Increasing temperature has two effects:

(1) an increase in electrostatic repulsions

between head groups due to higher counterion

dissociation, so Co tends to increase, and

(2) more gauche conformations are induced

in the surfactant chains, which become more

coiled, decreasing Co. Therefore, the com-

bined effects of temperature on the apolar

chains and on electrostatic interactions are

competitive. Although the electrostatic term

is believed to be slightly dominant, Co also

increases weakly with increasing temperature.

For nonionic surfactants, unsurprisingly, elec-

trolytes have very little effect on Co, whereas

temperature is a critical parameter due to the

strong dependence on ethyleneoxide head group

hydration. For nonionic CiEj alkyl polyethy-

leneoxide surfactants, as temperature increases,

water becomes a poorer solvent for the hydro-

philic units and penetrates less into the surfactant

layer. In addition, on the other side of the film, oil

can penetrate further into the hydrocarbon chains,

so that increasing temperature for this type of

surfactant causes a strong decrease in Co.

Thus, by changing parameters such as temper-

ature and pressure, the nature of the oil, or elec-

trolyte concentration, the spontaneous curvature

can be tuned, so driving transitions betweenWin-

sor systems.

Film Bending Rigidity The film bending

energy concept was first introduced by Helfrich

(1973) and is now considered as an important

model for understanding microemulsions. The

interfacial film is described by two elastic moduli

(Kellay et al. 1994), which are measures of the

energy required to deform the interfacial film

from Co:

• The mean bending elasticity (or rigidity), K,associated with the mean curvature, which

represents the energy required to bend unit

area of surface by unit amount. K is positive,

that is, spontaneous curvature is favored.

• On the other hand, the factor �K is associated

with Gaussian curvature, and hence account-

ing for film topology. �K is negative for spher-

ical structures or positive for bicontinuous

cubic phases.

Theoretically, it is expected that bending mod-

uli should depend on surfactant chain length

(Safran and Tlusty 1996), area per surfactant

molecule in the film (Szleifer et al. 1990), and

electrostatic head group interactions

(Winterhalter and Helfrich 1992).

The film rigidity theory is based on the inter-

facial free energy associated with film curvature.

The total free energy, F, of a surfactant layer at

a liquid interface can be considered a composite

of an interfacial energy term, Fi, a bending

energy term, Fb, and an entropic term, Fent. For

a droplet type structure, the total can be written as

(Gradzielski et al. 1996):

F¼ FiþFbþFent ¼ gA

þZ

K

2C1þC2�2C0ð Þ2þ �kC1C2

� �dA

þ nkBTf ðfÞ

(25)

where g is the interfacial tension, A is the total

surface area of the film, K is the mean elastic

bending modulus, �K is the Gaussian bending

M 700 Microemulsions

modulus, C1 and C2 are the two principal curva-

tures, Co the spontaneous curvature, n is the num-

ber of droplets, kB is the Boltzmann constant, and

f(f) is a function accounting for the entropy of

mixing of microemulsion droplets, where ’ is the

droplet core volume fraction. For dilute systems

where f < 0.1, it was shown that f(f)¼(ln(f)�1)

(Gradzielski et al. 1996). As noted above,

microemulsion formation is associated with

ultralow interfacial tension, go/w so the go/w A

term is small compared to Fb and Fent and can be

ignored as an approximation.

The curvatures C1, C2, and Co can be

expressed in terms of radii as 1/R1, 1/R2, and

1/Ro, respectively. For spherical droplets,

R1 ¼ R2 ¼ R, and the interfacial area is

A ¼ n4pR2. Note that R and R0 are core radii

rather than droplet radii (Gradzielski et al.

1996). Solving Eq. 25 and dividing by area A,

the total free energy, F, for spherical droplets

(of radius R) is expressed as:

F

A¼ 2K

1

R� 1

R0

� �2

þ�K

R2þ kBT

4pR2f fð Þ

� �(26)

For systems at a solubilization boundary (WI

or WII region), the droplets have achieved

their maximum size Ravmax. Under this condition,

minimization of the total free energy leads

to a relation between the spontaneous radius,

R0, and the elastic constants K and �K

(Safran 1992):

Ravmax

R0

¼ 2K þ K

2K� kB T

8pKf fð Þ (27)

A number of techniques have been used to

measure K and �K separately, in particular,

ellipsometry, X-ray reflectivity, and small-angle

X-ray scattering (SAXS) (Meunier and Lee 1991;

Kegel et al. 1995; Sicoli et al. 1993). De Gennes

and Taupin (1982) have developed a model

for bicontinuous microemulsions. For Co ¼ 0,

the layer is (Tondre and Derouiche 1991) sup-

posed to be flat in the absence of thermal fluctu-

ations. They introduced the term xK, the

persistence length of the surfactant layer that

relates to K via:

xK ¼ aexp 2pK kBT=ð Þ (28)

where a is a molecular length and xK is the cor-

relation length for the normals to the layer, that is,

the distance over which this layer remains flat in

the presence of thermal fluctuations. xK is

extremely sensitive to the magnitude ofK. Exper-imental data show that K is typically between 100

kBT for condensed insoluble monolayers

(Daillant et al. 1989) and about 10 kBT for lipid

bilayers (Schneider et al. 1984; Engelhardt et al.

1985; Bivas et al. 1987) but can decrease below

kBT in microemulsion systems (Di Meglio et al.

1985; Farago et al. 1990).

An alternative, more accessible, method to

quantify film rigidities is to calculate the com-

posite parameter (2K þ �K), which may be done

via tensiometry and SANS measurements. Two

expressions can be derived from Eqs. 26 and 27.

1. From interfacial tension go/w and the maxi-

mum mean core radius Ravmax (measured by

SANS):

go=w ¼ 2K þ �K

Ravmax

� �2 þ kBT

4p Ravmax

� �2 f fð Þ (29)

which gives for the bending moduli:

2K þ �K ¼ go w= Ravmax

� �2 � kBT

4pf fð Þ (30)

2. Using the Schultz polydispersity width

r ¼ s Ravmax

�obtained fromSANSdata analysis:

Thermal fluctuations of the microemulsion

droplets drive droplet polydispersity, which

relates to the bending moduli. Safran (1983)

and Milner (Milner and Safran 1987b)

described the thermal fluctuations by an

expansion of the droplet deformation in

terms of spherical harmonics. In the case of

the two-phase equilibria at maximum solubi-

lization (WI or WII), this polydispersity, p,

may be expressed as a function of K and �K:

p2 ¼ u2o4p

¼ kBT

8p 2K þ �Kð Þ þ 2kBTf fð Þ (31)

where uo is the fluctuation amplitude for the

I¼ 0 mode. This polydispersity is given by the

Microemulsions 701 M

SANS Schultz polydispersity parameter

s Ravmax

�(Eastoe et al. 1997c), and Eq. 31 can

be written:

Microemulsions, Fig. 3 The phase prism describing the

phase space for a system at constant pressure

M

2K þ �K ¼ kBT

8p s Ravmax

�� �2 � kBT

4pf fð Þ (32)

Now, Eqs. 30 and 32 give two experimentally

accessible expressions for the sum (2K þ �K)

using data from SANS and tensiometry. The

composite (2K þ �K) has been determined for

nonionic films in WI systems (Gradzielski et al.

1996; Gradzielski and Langevin 1996) and also

cationic (Eastoe et al. 1997c) and zwitterionic

(Eastoe and Sharpe 1997) layers in WII

microemulsions, as a function of surfactant

alkyl carbon number n-C. These trends are in

line with statistical mechanical theories

(Szleifer et al. 1990), indicating K should vary

as n-C2.5 to n-C3, whereas there is only a small

effect on �K.

Phase Behavior

Microemulsion phase stability depends on the

nature and concentration of the components and

also on the thermodynamic variables pressure

and temperature. Phase stability diagrams (or

phase maps), and location of the different struc-

tures formed within these mixed systems in terms

of variables of interest are, therefore, very impor-

tant. Several types of phase diagram can be gen-

erated depending on the number of variables

under consideration. In using an appropriate

graphical representation, it is possible to describe

the limits of existence of the single and

multiphase regions and to characterize equilibria

between phases (tie-lines, tie-triangles, critical

points, etc.).

Microemulsion Equilibria and the Phase Rule

The phase rule (Eq. 1) enables the identification

of the number of variables (or degrees of free-

dom) depending on the system composition and

conditions. In the following, the phase rule is

applied strictly to the macroscopic appearance

of microemulsions (rather than the micro-/

nanoscopic structure).

A system is called invariant, monovariant,

bivariant, and so on, according to whether the

number of degrees of freedom F is zero, 1, 2,

and so on. For example, in the simplest case of

a system composed of three components and two

phases, F is unity at a fixed temperature and

pressure. This means that the mole or weight

fraction of one component in one of the phases

can be specified to define the equilibrium, but all

other compositions in both phases are fixed. In

general, microemulsions contain at least three

components: oil (O), water (W), and amphiphile

(S), and as mentioned previously a cosurfactant

(alcohol) and/or an electrolyte are usually added

to tune the system stability. These can be consid-

ered as simple O–W–S systems: if a cosurfactant

is used, and the ratio oil/alcohol is kept constant,

and assuming that alcohol does not interact with

any other component, then the mixture can be

treated (to a first approximation) as a three-

component system. Hence, at constant pressure,

the composition–temperature phase behavior can

be presented in terms of a phase prism, as illus-

trated in Fig. 3.

From an experimental point of view, the con-

struction of 3-D phase maps is rather complex

and time consuming, so it is convenient to sim-

plify the system by selecting only specific phase

cuts. Then, the number of variables F can be

reduced, either by keeping one term constant

and/or by combining two or more variables.

Any system with a composition within the

two-phase region (e.g., point o in Fig. 4a, c) will

exist as two phases with compositions linked at

water

water

oil

oil

cA

2φ

2φm

2φ

1φ

1φ

o p

n

surfactant

surfactanta

b

c

water

Salinity

Winsor type I III II

oil

2φm

1φ

op

n

surfactantB

3φ q

Microemulsions,Fig. 4 Ternary diagram

representations of two- and

three-phase regions formed

by simple water–oil–

surfactant systems at

constant temperature and

pressure. (a) Winsor I type,

(b) Winsor II type,

(c) Winsor III type systems

M 702 Microemulsions

the ends of the “tie-line,” that is, a segment

formed by phases m and n. Therefore, every

point on a particular tie-line has identical

coexisting phases (m and n) but of different rela-

tive volumes. When the two conjugate phases

have the same composition (m ¼ n), this corre-sponds to a critical (or plait) point, p.

If three phases coexist (Fig. 4b) corresponding

to aWIII system, at constant T and P, there are noindependent degrees of freedom. This region of

three-phase invariant compositions is triangular

in form and called “tie-triangle” (Bourrel and

Schechter 1988). Any overall composition, such

as point q (Fig. 4b), lying within the tie-triangle

will divide into three phases having compositions

corresponding to the vertices A, B, and C of the

triangle.

Binary Phase Diagrams

As mentioned previously, ternary diagrams can

be further simplified by fixing some parameters

and/or combining two (or more) variables

together (examples are: defining water and elec-

trolyte “brine” at a fixed concentration or com-

bining water and oil into a fixed “water-to-oil

ratio”). The effect is to conveniently reduce the

degrees of freedom F, permitting determination

of a simplified phase diagram which reduces to

a study of a planar section through the phase

prism. Examples of typical pseudobinary dia-

grams are given in Figs. 5–7 for nonionic and

anionic surfactants.

The third example (Fig. 7) concerns the

anionic surfactant Aerosol-OT, which can be

conveniently studies in water-in-oil system by

maintaining a fixed surfactant concentration, giv-

ing F ¼ 2, for a single-phase region at constant

pressure. Then, the two experimental variables

are temperature and w, the water-to-surfactant

molar ratio defined as w ¼ [water]/[surfactant].

Characterization

Methods and Characterization

Electrical Conductivity

Percolation is a phenomenon where conductivity

(s) of a reversed (w/o) microemulsion increases

rapidly by orders of magnitude when a specific

threshold has been crossed. As such, percolation

can be induced by varying water content, ionic

strength of the dispersed water and temperature.

There are two proposed explanations of the phe-

nomenon, static percolation and dynamic perco-

lation. The static percolation model attributes the

increase in conductivity to the formation of sys-

tem-spanning water channels above the percola-

tion threshold. In dynamic percolation, ions cross

Microemulsions, Fig. 5 shows the schematic phase dia-

gram for a nonionic surfactant–water–oil ternary system

Microemulsions 703 M

M

the surfactant layer when the attractive

interdroplet interactions, which arise, become

enhanced. However, the exact mechanism is

still under debate. As such, the study of percola-

tion can give insight into interfacial surfactant

film properties:

(a) Temperature has been found to induce perco-

lation for systems which would otherwise not

exhibit significant conductivity (Borkovec

et al. 1988). For ionic surfactants, tempera-

ture increases the level of interdroplet inter-

actions, and hence there is an increase in

conductivity. This however does not hold

for nonionic surfactants where the reverse is

observed: conductivity decreases with

increasing temperature (Eicke and Meier

1996). This behavior was also observed for

certain microemulsions with ionic surfactants

(John and Rakshit 1994).

(b) Salinity has been shown to decrease conduc-

tivity in microemulsions stabilized by ionic

surfactants. Tondre et al. (Tondre and

Derouiche 1991) examined water/AOT/dec-

ane microemulsions with increasing electro-

lyte concentration from 0 to 0.3 M. They

found that the maximum water content wmax

sharply increased and then dropped abruptly

leading to a Winsor II system, at which point

the conductivity dropped significantly. In

examining the effect of cation size, Garcia-

Rio et al. (1994) found that the temperature of

percolation increased with increasing

hydrated cation radius.

(c) Organic salt additives are of particular inter-

est since microemulsions may be employed

in drug delivery and extraction processes.

Furthermore, these actives often show

a degree of surface activity, and thus might

coadsorb into the surfactant film, thereby

affecting the properties. Hydrotropes are

such a case, and many drugs can be broadly

classed as “hydrotropic.” Moulik et al. (Ray

et al. 1993) studied the effect of a variety of

hydrotropes and of the bile salt sodium cho-

late on percolation dynamics and clustering

energetics in water/AOT/heptane (Ray et al.

1993), water/AOT/isooctane, and water/

AOT/decane w/o microemulsions (Hait

et al. 2001). The maximum gradient of the

sigmoidal s�f and s� y plots is consideredthe transition point of the percolation process.

The points at which these occur are known as

the threshold volume fraction ft and temper-

ature yt. Composition, pressure, and additives

can have an effect on the both ft and yt. Thethreshold temperature yt was shown to be

water-content dependent. The yt decreasedwith increasing w in both heptane and decane

microemulsions in the absence of additives

Aromatic solutes such as sodium salicylate

were shown to increase both yt and yt, henceaffecting the onset of percolation. On the

other hand, bile salts decrease ft and yt, byassisting in channel formation. Hydroqui-

none (Hq), pyrogallol (Pg), resorcinol (Rc),

and catechol (Cc) reduced the yt, in the orderHq < Pg < Rc < Cc, by assisting the

droplet fusion. Sodium salicylate, a-napthol,and b-napthol blocked droplet fusion,

thereby increasing yt. Hait et al. (2002) inves-tigated percolation in water/AOT/decane

microemulsions containing hydroxyl and

methoxy analogues of phenol. The monohy-

droxy compounds increased yt, whereas

Microemulsions,Fig. 6 Binary phase

behavior in ternary

microemulsion systems

formed with nonionic

surfactants. (a) Illustration

of a section at constant

surfactant concentration

through the phase prism.

(b) Schematic phase

diagram, plotted as

temperature versus volume

fraction of oil, jo, at

constant surfactant

concentration. Also shown

are various microstructures

found in different

microemulsion regions,

M. At higher temperatures,

a liquid phase is in

equilibrium with excess

water (M + W), at lower

temperatures with excess

oil (M + O). At

intermediate temperatures,

lamellar phases may be

stable at higher water

contents and higher oil

contents, respectively

(After Olsson and

Wennerstrom 1994)

M 704 Microemulsions

dihydroxy compounds decreased the

value, with the relative efficiency at inducing

percolation following the order ortho < meta

< para. The threshold temperature was

shown to be hydrotrope concentration

dependent.

Spin-Echo Diffusion NMR

The study of self-diffusion provides information

on the connectivity of a system and of the size of

dispersed domains. Diffusion depends on factors

such as the size and shape of the diffusing enti-

ties, the viscosity of the medium and temperature.

In a dilute system of dispersed spherical

droplets of effective radius R, the limiting infinite

dilution diffusion coefficient Do is given by the

Stokes–Einstein equation (Eq. 33):

Do ¼ kT

6p�R(33)

where k is the Boltzmann constant, T the

temperature, and � the viscosity of the medium.

Because of the size dependence of Do, diffusion

coefficient measurements can give an impression

of the underlying structure and domain size in

a system:

(a) In a w/o microemulsion, the diffusion coeffi-

cient of the confined water molecules will be

much lower than that of the external oil

medium.

(b) In o/w microemulsions, the reverse of (a)

holds.

(c) In a bicontinuous system, both oil and water

are expected to relatively have high diffusion

4515 30T/°C

00

50 11

10

9

9

88

77 66

55

12

w

100

Microemulsions, Fig. 7 Pseudobinary phase diagram in

ternary microemulsion systems formed with the anionic

surfactant Aerosol-OT (AOT) in various straight-chain

alkane solvents. The water-to-surfactant molar ratio, w,is plotted versus temperature at constant surfactant con-

centration and pressure. Alkane carbon numbers are indi-

cated; ringed numbers correspond to the lower

temperature (solubilization) boundary, TL, and unringed

numbers to the upper temperature (haze) boundary, TU.The single-phase microemulsion region is located

between TL and TU. Below TL, the system consists of

a microemulsion phase in equilibrium with excess water

(WII type), and above TU, the single microemulsion phase

separates into a surfactant-rich phase and an oil phase

(After Fletcher et al. 1987)

Microemulsions 705 M

M

coefficients, owing to the interpenetrating

and system-spanning nature of the underlying

structure.

(d) In a chaotic molecular solution, in the

absence of any colloidal-scale self-assembly

structures, the diffusion coefficients of the

components will be greater than any of

the previous three cases. Obviously in this

case, D should tend to that for the neat

solvent.

The collective diffusion domains will be

affected by collisions by other components. To

a first approximation, for spherical domains,

obstruction can be accounted for by Eq. 34:

D

Do1� 2f (34)

With D being the diffusion coefficient at

a given volume fraction f.Self-diffusion can be studied using spin-echo

NMR experiments. The technique relies on two

radio frequency pulses at phase angles of 90 and180 with respect to the main magnetic field Bo

which are separated by a delay time t. At 2t, thesystem generates an echo. Two gradient field

pulses of duration d and intensity g are added to

the spin-echo sequence. The first one labels the

spin by introducing a phase shift in the magneti-

zation. The second one reverses the phase shift.

The separation between gradient pulses is known

as the diffusion time. If no diffusion occurs, the

amplitude of the echo will not change with

respect to the signal after the first 90o pulse. If

diffusion does occur, a reduction in the echo

amplitude will be detected as a function of gradi-

ent strength. The amplitude of the echo is related

to the self-diffusion coefficient by Eq. 35:

A 2tð Þ ¼ Að0Þ exp �2tT2i

� �

� exp �g2G2d2 D� d3

� �Di

� �(35)

where A(2t) and A(0) are the echo amplitudes in

the presence and absence of the field gradient. T2iis the transverse relaxation time for the nuclei i, gis the gyromagnetic ratio, G is the gradient

strength, d is the length of the field gradient

pulse, D is the distance between gradient pulses,

and Di is the self-diffusion coefficient.

Lindman et al. (1981) examined an array of

microemulsions including those stabilized by

SDS, sodium octanoate, sodium octylbenzene

sulfonate, AOT, potassium oleate, and the non-

ionic C10E4 in the presence of a number of cosur-

factants and in different oil media. In these

systems, the self-diffusion of added short chain

alcohols, butanol and pentanol, and in certain

cases the water and oil, approached values of

the neat components. This was attributed to the

alcohol bearing similar solubility in both oil and

water and absence of extended well-defined

M 706 Microemulsions

domains. The nonionic surfactant systems

exhibited similar behavior to the ionic surfactant

systems in the presence of short chain alcohols.

The remainder of the systems, those with long-

chain cosurfactants and those not requiring the

presence of alcohol for microemulsion forma-

tion, showed distinct domains in the media,

judged by a distinct difference in the magnitudes

of the determined diffusion coefficients. Geuring

and Lindman (1985) continued probing such sys-

tems by looking into water/SDS/Butanol/Tolu-

ene microemulsions by scanning salt

concentration and replacing butanol for pentanol.

Low salinity levels showed that the self-diffusion

coefficients of water and toluene (Dw and Dtoluene

respectively) differed, with Dw > > Dtoluene,

while at higher salinity levels, the reverse

occurred, Dw < < Dtoluene. At intermediate salin-

ity, Dwater � Dtoluene indicative of a bicontinuous

phase. The behavior of the self-diffusion coeffi-

cients of each of the components of the

microemulsions with increasing salinity is sum-

marized in Fig. 8.

A further study (Lindman et al. 1983) focused

primarily on a quaternary system of water/

sodium octylbenzene sulfonate/alcohol/toluene,

wherein the effect of alcohol chain length was

investigated further. Nonanol and decanol sys-

tems showed behavior attributed to clear segre-

gated domains. Once again, butanol and pentanol

exhibited what was referred to as a “structure-

breaking effect” which paralleled the behavior of

the composition extent of the isotropic phase

region observed in the phase behavior. The

expected picture of distinct domains was in con-

tradiction with the component self-diffusion

observations. It was argued that this indicated

the presence of disrupted structures. In a note

supporting these two studies, Stilbs and Rapacki

(1983) examined the effect of alcohol chain

length from methanol through to octanol on

different combinations of surfactant and oil.

The weight ratios of the components were

kept the same for all systems produced qualita-

tively the same picture as keeping the molar

ratios constant. It was noted that the self-

diffusion of hydrocarbons remained relatively

fast irrespective of the alcohol chain length.

Surfactant diffusion increased with increasing

alcohol chain length but was always slower in

comparison to other components. The self-

diffusion of the cosurfactants appeared to

decrease with increasing tail length. The self-

diffusion of water showed a decreasing trend

with increasing alcohol length. The coefficient

decreased by a factor of 2–3 with every additional

methylene group in the alcohol. Overall, the

report was in agreement with the previous exper-

iments which showed that lower chain alcohols

give rise to less structured solutions which they

described as in effect bicontinuous.

To examine the effect of molecular identity of

cosurfactant alcohol, Stilbs et al. (1984) exam-

ined medium- and long-chain alcohols as well as

branched and aromatic analogues. The water dif-

fusion coefficient Dw differed appreciably with

different cosurfactant. In agreement with previ-

ous reports, Dw was smallest for the decanol

system. For phenol, butanol, and ethylene glycol

monobutyl ether, Dw was highest and

approaching that of neat water. The pentanol

system showed slightly decreased water mobility

compared to butanol, indicating a more struc-

tured system than was obtained when using the

smaller chain alcohols. Interestingly, branched

isomers of pentanol exhibited behavior similar

to the smaller alcohols. It was noted that the

alcohol mobility increased with dilution for the

aromatics, which was attributed to the lowering

of viscosity (Sjoblom and Henriksson 1983).

Water mobility on the other hand decreased, and

it was argued that this was due to alcohol being

redistributed resulting in the domain interfaces

becoming depleted of cosurfactant, which appar-

ently hinders water diffusion.

Dynamic Light Scattering

Dynamic light scattering (DLS, also known as

quasielastic light scattering, QELS, and photon

correlation spectroscopy, PCS), measures tempo-

ral fluctuations in the intensity of laser light

which is scattered by a sample at a fixed angle;

this information can then be analyzed to provide

data on the diffusion of colloidal particles, such

as found in droplet microemulsions. Hence, DLS

is an important characterization method for

Butanol

Butanol

Toluene

Water

D/m2s–1

10–10

10–9

10–11

3

SALINITY ( g / 100 cm3)

SDS

4 5 6 7 8 9 10

Microemulsions,Fig. 8 Self-diffusion

coefficients of components

of brine/SDS/butanol/

toluene microemulsion

(Guering and Lindman

1985)

Microemulsions, Fig. 9 Schematic exponential decay

in correlation function for large (slow-diffusing) and

small (fast-diffusing) microemulsion droplets

Microemulsions 707 M

M

microemulsions, providing initial (lab-based)

evidence for the presence of nanometer-sized

dispersed domains or droplets. The basic princi-

ple of DLS relies on static Rayleigh scattering,

but now, the intensity is studied as a function time

(more strictly a “delay time”). For an ensemble of

particles, the intensity measured at a given angle

will be a product of the constructive and destruc-

tive interference from light scattered by all of the

particles. As the particles experience Brownian

motion, the measured intensity fluctuates over

time: these fluctuations contain information

about the rate at which particles are diffusing in

the system. The fluctuating intensity data is han-

dled by a computerized correlator, which com-

pares I(t) (the intensity signal at time t) with that

a small time t + t later (where t is usually around5� 10�7s). The rate at which the signal intensity

changes is proportional to the rate of diffusion of

the microemulsion droplets. The correlation

function, g(t), describes this rate of change

(Wu 2008):

g tð Þ ¼ I tð Þ I tþ tð Þh i (36)

For an ensemble of microemulsion droplets

undergoing Brownian motion, this function

takes the form of an exponential decay, with

faster diffusing (smaller) droplets showing

a more rapid decay (Fig. 9):

g tð Þ ¼ A 1þ B g1 tð Þ½ �2

(37)

In Eq. 37, A is the baseline of the correlation

function as t ! 1, B is a parameter which

depends on the detection geometry in the system

(determined by calibration with a known

M 708 Microemulsions

standard), and the function g1(t) contains contri-butions from all exponential decays contained in

the correlation function. Hence, the decay rate is

proportional to the effective droplet diffusion

coefficient. Information on size data is then

obtained by fitting a model function to the exper-

imental correlation function data (Koppel 1972;

Provencher 1982; Roig and Alessandrini 2006).

These fitting methods are incorporated as algo-

rithms in the control software for modern DLS

instruments, and hence, analysis is highly

automated.

Using the Stokes–Einstein equation (Eq. 33),

assuming monodispersity and the absence of

interdroplet interactions, this diffusion coeffi-

cient allows calculation of the effective hydrody-

namic radius of the droplets:

In the case of concentrated or interacting sys-

tems, modifications to this approach must be

made (Cazabat et al. 1980). A limitation of DLS

(as presented) is that it does not explicitly sepa-

rate the effect of interparticle interactions on dif-

fusion. However, as samples becomemore dilute,

interactions should decrease, and hence, a series

of measurements at increasing dilution is often

performed. By extrapolating the obtained diffu-

sion coefficient to zero particle concentration,

a value for the infinitely dilute (noninteracting)

case, and hence a better estimate of the particle

size, can be obtained (Ricka et al. 1991).

An early example of light scattering experi-

ments on microemulsions was by Schulman et al.

(Schulman and Friend 1949), at a stage when

uncertainty reigned over the properties of trans-

parent and ternary water–amphiphile–oil mix-

tures. These experiments were carried out to

support findings from X-ray diffraction experi-

ments contained in an earlier publication

(Schulman and Riley 1948): both concluded that

dispersed droplets surrounded by a monolayer of

amphiphiles were present in both o/w and w/o

apparently monophasic, transparent systems.

The difference between micelles and

microemulsions were long contested (points

discussed in section “Thermodynamics”) with

conventional techniques unable to provide

a clear picture. Zulauf et al. (Zulauf and Elcke

1979) systematically investigated the effect of

surfactant and water concentration and tempera-

ture on the geometry of reverse water/AOT/iso-

octane microemulsions. AOT micelles in

isooctane showed a size invariance (15.0 A

radius) within the concentration range of

8–200 mM and a temperature range

20.0–95.0 C. It was noted that the length of

AOT molecules is in fact 12.0 A which lead to

the conclusion that, despite purification, water

molecules were still present and associated with

the surfactant head group. Karl Fischer titrations

indicated that there were sufficient amounts of

water to speculate that they might act as linking

agents between surfactant head groups. This, they

proposed, was the reason for the lack of a cloud

point even in the very low temperatures.

When water was added atw< 10, the apparent

radius determined by DLS increased, continuing

with increasing [H2O]. It is interesting to note

that when the samples were examined in open

cuvettes at atmospheric pressure, the dispersed

drops decreased in size down to the original

micelle size if given sufficient time. The rate of

this process increased with increased T.

In microemulsions where w > 10, an increase

in radii was observed in the range of

20.0–50.0 C, above which the solutions became

turbid. These observations were fully reversible.

Dilution with isooctane produced a pronounced

decrease in radius. Again when the solutions

were examined in open conditions, a rapid

decrease in sizes were observed down to the

values measured for the w < 10 regime. When

the temperature was lowered to 18 C and below,

the drops increased in size and eventually coa-

lesced to produce a macroscopically “biphasic”

system.

Clarke and Nicholson (1984) examined the

behavior of water stabilized by AOT in a range

of alkanes and concluded that the radii of struc-

tures present increased linearly with increasing

w, a behavior almost unique to AOT.

Cazabat et al. (1980) studied amatrix of water/

SDS/pentanol/cyclohexane microemulsions at

SDS/water ratios of 0.69 (A), 0.345 (B), and

0.23 (C) by varying the volume fraction f.They noted that the volumes of alcohol to

cyclohexane followed a linear relationship

3

4

0

1

2

.1 .2 .3 .4

A

Microemulsion

BCC′

.5

Φ

D×107 cm2/sMicroemulsions,Fig. 10 Diffusion

coefficients D against

volume fraction f for A

Microemulsions 709 M

M

which was obeyed in the smaller f regions but

not for higher, which was attributed to

a discontinuity in the phase quaternary diagram.

Figure 10 shows that diffusion D initially

decreased with increasing f as would be

expected from Eq. 33. The particles increase in

size and hence R and to � increase, thereby

decreasing the value of D. At larger f (�0.5),

D showed a maximum which was attributed to

a transition between inverse to regular

microemulsions.

Light scattering has been proven to be

a particularly useful technique and a first refer-

ence in attaining the size of dispersed systems;

however, it has a finite resolution. For example,

Tavacoli et al. (2008) investigated calcium car-

bonate nanoparticles (CaCO3) dispersed in

dodecane and cyclohexane stabilized by indus-

trial surfactants. The purpose of the experiments

was to determine the locus of trace water in

engine oils, which commonly contain dispersed

nanoparticles. One fraction of dispersed CaCO3

was placed in a desiccator under vacuum in the

presence of silica gel, so that any water present

was removed, and a second fraction was placed in

a desiccator under vacuum in the presence of an

open H2O container. Both fractions were kept

under these conditions for 65 days and removed

only to carry out the measurements. DLS showed

that the particle radii in both fractions remained

the same. Fourier transform infrared spectros-

copy (FTIR) showed an increasing water peak

with passing time. For the purpose of contrast

for SANS measurements, aliquots of D2O were

introduced to dry dispersed CaCO3 nanoparticles.

The oil phase was then removed. The SANS pro-

files were fitted with a Schultz polydisperse hard-

sphere model. Analysis showed that introducing

water to the system actually increases the particle

size. The core of the particles was 2.79 nm with

the fitting providing a shell (D2O) thickness of

0.54 nm. Contrast matching to include the surfac-

tant layer in the particle shell thickness showed

that the shell was larger than the surfactant

length, reiterating that a D2O layer must be pre-

sent on the particle surface. This increase of shell

thickness corresponded well with an increase of

the radii ascertained from the effective volume

fractions as determined from viscosity measure-

ments. The slight increase in viscosity

corresponded to an increased volume fraction.

The disparity between DLS and SANS was

attributed to the errors in viscosity as determined

by DLS due to the presence of some base oil in

the overbased nanoparticles. Such issues do

not arise with SANS, as it is a firsthand report

Microemulsions, Table 1 Scattering lengths of selected

nuclei and scattering length densities of some common

compounds

Nucleus b 10�12cm� ��

Compound r 1010cm�2� ��

1H �0.374 H2O �0.5602H(D) 0.667 2H2O(D2O) 6.35612C 0.665 Toluene 0.94116O 0.580 D-toluene 5.66214N 0.936 TX-100 0.51932S 0.285 AOT 0.542

M 710 Microemulsions

on the geometry of the scatterer. However, DLS

is still an important technique in studies of

microemulsions.

Small-Angle Neutron Scattering

Over recent decades, small-angle neutron scatter-

ing (SANS) has become one of the most valuable

techniques for probing structure and interactions

in colloidal dispersions. Like any scattering

experiment, it relies on the interaction of a beam

of radiation with the sample of interest; neutrons

are scattered by interaction with the atomic nuclei

in the sample. By measuring the intensity of

scattered radiation as a function of scattering

angle, information can be extracted on the size

and shape of structures present in the sample

(provided they are on an appropriate length

scale for the wavelength of radiation used)

and the interactions between them. Colloidal

structures such as small nanoparticles and

microemulsion droplets fall within a size range

that is ideal for study by neutrons (wavelength,

l ¼ 0.1–30 A). Therefore, SANS is a very

important method for characterization of

microemulsions.

Basic Principles The magnitude of the interac-

tion between the incoming neutron beam and the

nucleus depends on a property of the nucleus

known as scattering length, b. For bulk materials,

it is more convenient to use a summation of the

scattering lengths for all nuclei per unit volume.

This property is known as scattering length den-

sity, r, and is calculated as:

r ¼ DNA

MR

Xibi ¼

Pi bi

Vm(38)

In Eq. 38, D is the bulk density, MR is the

molar mass, NA is Avogadro’s number, and Vm

is the molecular volume. The scattering lengths

and scattering length densities of some com-

monly encountered nuclei and compounds are

shown in Table 1.

The substantial difference in scattering length

between hydrogen (1H) and deuterium (2H, D), is

a useful and much exploited property. By selec-

tively deuterating parts of a microemulsion,

contrast can be generated in order to show scat-

tering from certain structures (e.g., the core of

a microemulsion droplet) by enhancing the dif-

ference in scattering length densities between

components. This is shown schematically in

Fig. 11.

The downside to this powerful tool is that

deuteration can be extremely expensive, it may

not be feasible to deuterate certain compounds,

and deuteration sometimes produces slight

changes in the phase behavior of a system.

An important experimental quantity is related

to both the scattering angle, y, and the incident

neutron wavelength, l:

Q ¼ 4plsin

y2

(39)

The scattered intensity, Is measured at the

detector, is simply the squared modulus of the

amplitude:

Is Q

¼ As Q ��� ���2

� (40)

For a system of n particles, the scattering aver-

aged over all orientations, o, and shapes, s, this

becomes:

Is Q

¼ As Q ��� ���2

� o

� s

(41)

Data Analysis Many systems studied by SANS

(such as micelles and microemulsion droplets)

core contrast: ρ1 ≠ ρ2 = ρ3

1 2 3 1 2 3 1 2 3

shell contrast: ρ1 = ρ3 ≠ ρ2 drop contrast: ρ1 = ρ2 ≠ ρ3

Microemulsions,Fig. 11 Schematic of the

different possible neutron

scattering contrasts for

a microemulsion droplet

Microemulsions 711 M

comprise spherical particles dispersed in the sol-

vent medium. A physical description of the pat-

tern of scattered radiation from these structures is

therefore useful. Scattering from monodisperse

spheres can be described by:

M

IðQÞ ¼ fp rp � rs� �2 Vp P Q;Rð Þ

SðQÞ þ Binc

(42)

In Eq. 42, fp is the sphere volume fraction

(calculated as the number density of particles, Np

multiplied by the volume of one particle Vp), rpand rs are the scattering length densities of the

particle and solvent, P(Q,R) and S(Q) are the form

factor and structure factor, and Binc is the incoher-

ent background. The proportionality factor which

relates the intensity to the form and structure

factors is known as the scale factor, SF:

SF ¼ fp rp � rs� �2 Vp (43)

This parameter is useful when modeling

SANS data, as it can separately calculated and

fitted, giving an estimate of the fit accuracy.

The form factor, P(Q), describes scattering

which arises from the nuclei within structures,

and hence, the interference patterns give infor-

mation on the size and shape of structures. The

form factor for a homogeneous sphere is

P Q;Rð Þ ¼ 3 sin QRð Þ � QR cos QRð Þð ÞQR3

� �2

(44)

Systems such as microemulsion droplets sta-

bilized by a surfactant layer may be seen as

a central core (the dispersed solvent component

of a microemulsion) surrounded by a shell of

surfactant (Fig. 11). In this case, a more complex

form factor is needed to describe the scattering

which arises from such structures (Markovic and

Ottewill 1986).

In real systems such as micelles and

microemulsions, there is a distribution of sizes

around an average value. This polydispersity has

a significant effect on the scattering and needs to

be accounted for in modeling. The Schultz distri-

bution function is a commonly used polydisper-

sity factor, described by an average particle

radius and root mean square deviation

(Kotlarchyk et al. 1984):

s ¼ Rav

Z þ 1ð Þ1=2(45)

where Rav is the average particle radius and Z is

a width parameter.

The structure factor, S(Q), describes

interaggregate/particle scattering and hence

gives information on the interactions between

structures. It is dependent on the type of interac-

tions which the structures experience (e.g.,

excluded volume, attractive, repulsive). For

water-in-oil (w/o) microemulsion droplets far

from any phase boundary, the hard-sphere inter-

action potential is often appropriate, as attractive

interactions are minimal (Ashcroft and Lekner

1966; de Kruif et al. 1988):

SðQÞ ¼ 1

1� Np

� � f rd;fp

� � (46)

When a phase boundary is approached in

microemulsions (e.g., when droplets are close to

coalescence), attractive forces start to become

30

20

I(Q

) (r

elat

ive

units

)

10

00 0.04

Q/Å−1

0.02 0.06 0.08

120 bar

92 bar

500 bar

200 bar

Microemulsions,Fig. 12 SANS profiles at

increasing pressure at

constant temperature of

AOT w/o microemulsions

at w ¼ 30 (Ornstein and

Zernike 1914)

M 712 Microemulsions

important, and the hard-sphere potential is no

longer obeyed. In this case, the Ornstein–Zernike

potential is commonly used (Ornstein and

Zernike 1914; Zemb 1991):

SðQÞ ¼ 1þ NpkBTw

1þ Q2x2

� �(47)

In Equation, kB is the Boltzmann constant, T is

the temperature, w is the isothermal compressibil-

ity and x is a correlation length.

Eastoe et al. (1990) studied AOT-stabilized

w/o microemulsions in alkanes, from propane to

decane. Microemulsions in propane to pentane

are considered to be “low-density systems,”

while microemulsions in hexane to decane are

referred to as “high-density systems.” Pressure–

temperature phase diagrams were determined for

a range of w values. Viscosity and density of the

alkanes is a function of temperature and pressure.

Higher pressure at constant temperature causes

an increase in viscosity and density.

An increase of temperature at constant pres-

sure results in a decrease of viscosity and density.

SANS revealed that the effect of pressure on the

microemulsion radius could be considered negli-

gible. This appeared to be the case for all alkanes

and w values. It would have been expected that

a change in pressure might affect the equilibrium

between surfactant adsorbed at the water/oil

interface. However, if the concentration of

adsorbed surfactant was high and that of the free

surfactant was low, then changes in density and

consequently a shift of the equilibrium toward the

adsorbed surfactant would be small, as indeed

shown in the decreased scattering intensity with

increasing pressure in Fig. 12. Changing the oil

alkane length again was thought to have an influ-

ence on the droplet radius, as smaller alkanes

would be better at penetrating the adsorbed sur-

factant layer than the higher-chain-length

alkanes. However, once more there appeared to

be no change in the radii of the corresponding

w values across the alkane series. When the

microemulsions were forced close to the phase

boundary, by pressure variation, an effect on the

scattering profiles was observed. An increase in I

(Q) was observed with decreasing pressure. At

w ¼ 30 at 500 bar analysis showed a P(Q) con-sistent to a non interacting spherical model.

When pressure decreased, analysis showed

a progressively stronger S(Q), which indicates

that the droplets are experiencing long-range

attractive interactions. It was concluded that

near the phase boundary, large transient clusters

of droplets are formed.

It is readily appreciated that AOT-stabilized

microemulsions are suitable model systems for

understanding microemulsions in general and

their properties. Since microemulsion properties

10

1φ

2φL2-H2O

L2-microemulsion

20

20

30

di-C6SS

ooo

oo

o

o

40

T / °C

T / °C

40

w

w

50 60

60

80

20

40

60

80

70

10 20 30 40 50 60 70

SO3− Na+

SO3− Na+

SO3− Na+

SO3− Na+

SO3− Na+

SO3− Na+

SO3− Na+

5

66

4

2

5

3

11 2 43

o

oo

oo

oo

oo

oo

oo

oo

o

o

oo

oo

Microemulsions, Fig. 13 Phase behavior and structural analogues of AOT (Zemb 1991)

Microemulsions 713 M

M

are governed by the stabilizing film at the o/w

interface, investigating the effect of molecular

architecture variance proves fruitful. Nave et al.

(2000a) investigated eleven AOT-related com-

pounds. This library of surfactants consisted of

linear dichain (di-C6SS) and branched dichain

analogues (AOTs) (Fig. 13).The linear counterparts required the presence

of hexanol for the formation of microemulsions,

while the branched surfactants did not require

a cosurfactant. The phase behavior of the

branched analogue microemulsions differed on

in the locus of the phase boundary along the

temperature scale. SANS was performed on the

branched analogues, as comparison to the linear

analogues would not yield a direct comparison

due to the presence of cosurfactant. SANS on

water/AOTs/heptane reverse microemulsion sys-

tems were carried out at temperatures that were of

equivalent distance to the phase boundaries.

Analysis of the topologies of the microemulsion

was attained by carrying out core-shell-drop

contrast series experiments (CSD, Fig. 14).

The experiments revealed that the two most

highly branched analogues had smaller solubili-

zation capacities with a decrease in droplet radii

of 3–6 A relative to the rest. It was also noted that

the two branched analogues with the longest tails

also showed a thicker surfactant layer which is

testament toward the high-resolution capabilities

of the technique. The obtained radii scaled line-

arly with w value, a behavior shown to occur for

regular AOT microemulsions in previous publi-

cations (Kotlarchyk et al. 1982).

Following up on this investigation, Nave et al.

(2000b) examined AOT analogues with linear

and branched phenyl-tipped tails and compared

these to AOT itself (Fig. 15). Water in oil

microemulsion phase stability diagrams were

carried out in n-heptane and toluene.

While AOT formed stable single-phase

microemulsions in both media, the phenyl-tipped

10

50

100

I(Q) / cm–1

Q / Å–1

5

0.5

0.1

0.05

0.004 0.008 0.02 0.04 0.07

AOT 4

0.3

1

drop

core

shell

Microemulsions,Fig. 14 Contrast-matched

scattering data and fitting

lines for AOT analogue

(Zemb 1991)

Aerosol-OT

O

O

O

O

di-PhC4SS

O

O

O

O SO3− Na+

SO3− Na+

SO3− Na+

SO3− Na+SO3

− Na+

di-PhC5SS

O

O

O

O

br-di-PhC5SS

O

O

O

O

br-di-PhC3SS

O

O

O

OMicroemulsions,Fig. 15 Regular and

phenyl-tipped AOT

analogues (Kotlarchyk

et al. 1982)

M 714 Microemulsions

analogues only formed Winsor II systems in hep-

tane. Single-phase systems were attained in tolu-

ene though at significantly lower w values than

typical AOT microemulsions in aliphatic media.

Interestingly, the phenyl-tipped analogues only

showed Winsor II system formation at higher

w values. This boundary increased slightly with

temperature. This boundary was found at higher

w values for the shorter of the two linear phenyl-

tipped analogues, while the reverse was observed

0.01

0.05

0.1

0.5

1

5

10

50

0.02 0.03 0.05

Q / Å−1

I(Q

) / c

m−1

0.08 0.28

Microemulsions,Fig. 16 Scattering profiles

of di-PhC4SS (circles),di-PhC5SS (crosses), andbr-PhC5SS (triangles) atw ¼ 16 (intensities have

been multiplied by 3.5 for

di-PhC5SS and 20 for

br-PhC5SS for clarity)

Microemulsions 715 M

M

for the branched ones. The boundary for AOT

itself lay between the two aforementioned clas-

ses. SANS measurements were carried out and

analyzed in the same manner as the previous

AOT analogue study. As was the case with the

previous set of experiments, at constant w values,

the droplet radius was shown to decrease slightly

with increasing alkyl tail chain length, while the

polydispersity was more or less constant

(Fig. 16).

This similarity in polydispersities follows the

argument of the film bending rigidity model, in

which the surfactant layer thickness scales

inversely with polydispersity. As the chain

lengths are not different in a pronounced way,

with chain length differences of 2–3 carbon

atoms in the alkyl tail, the approximately constant

polydispersities observed can be seen to be in line

with this argument. Reiterating the sensitivity of

this technique, surfactant layer thicknesses did

scale according to the increase in chain length.

Once more, the core radii were also shown to

scale linearly with increasing w values. Although

the w values were small, head group areas were

found to correlate well with the extent of

branching in the molecules. The two branched

molecules (AOT and the longest chain phenyl-

tipped AOT analogue) were found to pack less

efficiently than the linear examples, producing

higher head group areas. This reduced packing

efficiency can account for the lower water

solubilization in toluene. The head group areas

of all compounds in microemulsions appeared

similar to those estimated by surface tension

measurements at the air–water interface,

suggesting little difference between the behavior

of the surfactants at the two interfaces.

Another example of the sensitivity and hence

suitability of SANS as a technique toward the

study of such systems is highlighted in the work

of Tabor et al. (2010). Here when water was

added to AOT-stabilized silica nanoparticles dis-

persed in toluene, the formation of reverse

microemulsion droplets occurred. An AOT-

stabilized silica sol was able to accommodate up

to 2.5 wt% water, remaining as an isotropic and

clear, homogeneous phase. Any further addition

resulted in a cloudy appearance that eventually

resolved into a sol solution coexisting with an

excess water phase. Tavacoli et al. (2008) previ-

ously showed that water uptake was indeed pos-

sible however at much lower amounts than that

observed in this study. DLS measurements

showed that addition of water yielded an increase

in the apparent effective diameter. While water

can be expected to form a film around the

nanoparticles as was the case for the study

(Tavacoli et al. 2008), this is not enough to pro-

duce the physical increase seen on this occasion.

Carrying out careful contrast matching variations

the scattering profiles exhibited two form factors

(P(Q)), one from the surfactant around the

M 716 Microemulsions

particles at low Q and another at high Q owing to

the presence of micelles/microemulsions. The

silica sol particles did not appear to increase in

size with addition of water; however, the back-

ground microemulsions observed were of similar

size to microemulsions containing the same

amount of water, in the absence of the dispersed

nanoparticles. In other words, the microemulsion

droplets in the wet organosol are the same as in

a background silica-free microemulsion.

Hence, SANS indeed has proven itself over

the years as an essential and highly versatile

technique in the study of micellar and

microemulsion systems and of soft matter as

a whole. Hydrogen–deuterium contrast variation

represents the main advantage of SANS, permit-

ting a unique insight into internal structure of the

dispersed domains in microemulsions.

Applications

Microemulsions as Media for Synthesis and

Processing of Nanoparticles

Synthesis

The topic of nanoparticle synthesis has seen

a rapid rise in publications since 1980. Research

into inorganic nanoparticles has soared owing to

the special properties of nanomaterials compared

to bulk metals, especially with respect to their

photochemical and semiconductor properties.

Noble metal nanoparticles have attracted signifi-

cant interest due to their application in electronics

(McConnell et al. 2000), catalysis (Xia et al.

2003), and in potential medical applications

(El-Sayed et al. 2005), as well as for their bacte-

ricidal properties. The properties that are so desir-

able in metal nanoparticles arise from their

particle size and size distribution, since nanoscale

sizes show remarkable optical properties as well

as lower melting points, making the engineering

of devices more feasible. Silver nanoparticles are

especially interesting due to both their optical

properties, their use in the fabrication of fine-

line electronic circuits.

Reverse microemulsions lend themselves as

suitable “nanoreactors” for the synthesis of

nanoparticles, as the water pool is of the same

dimensions as the desired nanoparticle size.

Inorganic salts, the starting materials for

nanoparticles, can be dispersed in the

microemulsion medium by dissolution inside

water pools. Microemulsions are dynamic sys-

tems, and as described in the section discussing

percolation, the droplets collide as a result of

Brownian motion, hence facilitating exchange

of materials allowing the reactions to take place.

This mechanism is of fundamental importance in

the synthesis of nanoparticles, and there are two

ways in which it can be employed. In the first, two

microemulsions are prepared, one containing an

inorganic salt and another containing a reducing

agent, which are then mixed (Fig. 17). The reac-

tion has been shown to start at the interface and

proceed toward the center of the droplet (Li et al.

2005). The rate limiting step appears to be the

droplet diffusion. Control of the exchange can be

achieved by tuning the surfactant film rigidity.

The second method, and perhaps the most com-

mon, involved the preparation of one

microemulsion containing the inorganic salt.

The reducing agent is then introduced directly

into the microemulsion medium.

A commonly encountered issue is that of par-

ticle size and shape control. Through the exten-

sive research carried out in the field, parameters

that influence the size of the nanoparticles have

been identified as (a) the type of solvent, (b) the

surfactant and/or cosurfactants, (c) added electro-

lyte, (d) water content, and (e) concentration of

reagents. All of these parameters have an effect

on the film rigidity of the microemulsion. This is

commonly referred to as “interfacial fluidity” and

is a key concept in the exchange of materials

between microemulsions.

(a) The solvent effect is thought to play a lesser

role when compared to the other parameters

(Eastoe and Sharpe 1997), with noticeable

effects only for large changes in oil chain

lengths, with large chain lengths producing

more rigid films. Furthermore, solvents that

interact more with the surfactant tails are

better able to stabilize larger particles

(Lopez-Quintela et al. 2004b).

(b) The effects of surfactant and cosurfactant

type are still under considerable debate.

Microemulsion I

Aqueous PhaseMetal Salt

(FeCl3, FeCl2,CuCl2, etc.

Aqueous PhaseReducing Agent(NH4OH, N2H4,

NaBH4, etc.)

Oil PhaseMix Microemulsions I and IIOil Phase

Percolation

Precipitate

(Metal or Metal Oxide)Chemical Reaction Occurs

Collision andCoalescence of

Droplets

Microemulsion IIMicroemulsions,Fig. 17 Proposed

mechanism for

nanoparticle preparation in

microemulsions

(Capek 2004)

Microemulsions 717 M

M

guished. Longer surfactant hydrophobic

chains make for more rigid films than shorter

chains (Szleifer et al. 1990; Gradzielski et al.

1996; Eastoe et al. 1997c; Eastoe and Sharpe

1997). Smaller surfactant head group molecu-

lar areas generally give rise to smaller particle

sizes. However, the opposite effect was noted

by Lee et al. (2005) for certain nonionic sur-

factant-stabilized microemulsions. The use of

cosurfactants appears to yield smaller particles

due to increased film fluidity (Uskokovic and

Drofenik 2005). The size of the nanoparticles

also decreases with increasing cosurfactant

chain length (Charinpanitkul et al. 2005).

(c) Added electrolyte does not seem to affect

the size of the final particles; however,

initial growth rates appeared to be greatly

enhanced. Both observations were attributed

to the destabilizing effect of the electrolytes

on the microemulsions (Uskokovic and

Drofenik 2005).

(d) Studies on the effect of water content

have failed to provide a generalized rule.

In many cases, the final particle size showed

a dependence on the initial w0 (Pileni 1997;

Lisiecki and Pileni 1995; Pileni et al.

1992); however, in many other cases, such

control was not observed. Irrespective of the

w0, when allowed to go to completion, the

reactions produced similarly sized particles.

Instead, it was argued that the rate of reaction

was affected by w0. The nanoparticle

growth rate was lower at lower w0 as

a result of high film rigidity with most of

the water molecules being bound to the sur-

factant head group. Increasing w0 allows

for greater exchange rate between

microemulsion droplets. However, after

a certain w0, the rate decreases or reaches

a plateau, attributed to a concentration effect.

At this point, the reagents are too dilute for

optimum reaction rates.

(e) Reagent concentration has been found to

have a significant role in particle size with

size increasing with increasing reagent

concentration (Lisiecki and Pileni 2003;

Eastoe et al. 1996c; Maillard et al. 2003).

0.110−3

10−2

10−1

100

1

t = 0.136 s

t = 15.476 s

q (nm−1)

Inte

nsity

(m

m−1

)

10

Microemulsions, Fig. 18 SAXS profile of

a microemulsion during nucleation and growth of gold

nanoparticles at t ¼ 0.136 s (red) and t ¼ 15.476 s (blue)

M 718 Microemulsions

Further still, polydispersity appeared reduced

when the concentration of one of the reagents

was increased (Eastoe and Cox 1995).

Despite extensive research into the effect of

these parameters, these observations serve more

as possible guidelines than governing laws

toward attaining specific particle geometries.

Another technique in producing nanoparticles

is the use of radiation and, in particular, light.

Luisa Marin et al. (2008) employed UV to

photoinduce free radicals from photosensitive

organic molecules (ketones) for the preparation

of Au nanoparticles in regular micelles. There

are now other examples (Oliveira et al. 2011),

where the surfactant also acts as a reducing

agent, being itself photoreactive. The generation

of nanoparticles in microemulsions using such

photoreactive surfactants holds promise for

future developments (Oliveira et al. 2011).

There are only a few publications employing

light to produce nanoparticles in microemulsions.

An early example was by Kurihara et al. (1983)

who used pulse radiolysis to generate Au

nanoparticles in water/pentaethylene glycol

dodecyl ether (PEGED)/n-hexane reverse

microemulsions. It was concluded that particle

growth was primarily governed by exchange of

the irradiated content between microemulsion

droplets.

Recovery

In the previous section, it was demonstrated that

microemulsions can serve as nanoreactors for

the production of nanoparticles. In making

nanoparticles, a crucial step is the nanoparticle

recovery and separation from by-products. Typi-

cal techniques are ultracentrifugation (Germain

et al. 2005), solvent evaporation (Steingerwald

et al. 1988), and the addition of appropriate

antisolvent to induce precipitation or “phase”

separation (Chen and Wu 2000). These are

high-energy, costly, and environmentally

unfriendly processes.

In recent times, physicochemical aspects of

microemulsions have been employed for the

recovery of nanoparticles. Abecassis et al.

(2009) used reverse microemulsions based on

the catanionic surfactant octylammonium

octanoate in octane. Gold nanoparticles were

generated in this system by reducing gold hydro-

chloride with sodium borohydride. The reaction

and recovery were followed by small-angle X-ray

scattering (SAXS) using a stopped-flow device.

The SAXS profiles (Fig. 18) showed that the

microemulsion only changed slightly upon gold

nanoparticle formation, as evidenced by the

slight increase in intensity in the medium

Q-range. The nanoparticles generated were

spherical and the reaction did not appear to affect

the microemulsion geometry at any stage. Upon

completion of the reaction, the systemwas cooled

to 18 C from room temperature, causing a Win-

sor II phase separation (Fig. 19).

This gave preferential partitioning of the

nanoparticles into the top oil “phase.” Hollamby

et al. (2010) prepared Au nanoparticles bymixing

two microemulsion solutions consisting of water/

CTAB/butanol/octane, one of which contained

KAuCl4 and the other sodium borohydride,

respectively. After the formation of the Au

nanoparticles, further water was added forcing

a Winsor II transition. Employing NMR, compo-

sitions were determined to within a 1 wt% error.

The upper oil phase consisted of 96 % octane and

4% butanol. Surfactant and water quantities were

too low to be detected. The lower phase had

a composition of 61 % water, 20 % octane, 9 %

butanol, and 10 % CTAB. The Au nanoparticles

again showed a strong preferential partitioning

T = 25°C T = 15°C

Microemulsions, Fig. 19 Appearance of a gold nano-

particle containing microemulsion Winsor IV system

(25 C) and a temperature induced a Winsor II system

(15 C)

Microemulsions 719 M

M

toward the upper oil phase (82 % of the original

system). The difference to the work of Abecassis

et al. is in the method of inducing phase separa-

tion. Rather than energy-intensive cooling of the

system, environmentally benign water was intro-

duced. This approach was extended, successfully

exploring the production/recovery and reuse of

nanoparticles by microemulsion physicochemi-

cal properties by adding nonadsorbing polymer

(Faizan et al. 2011) and fine tuning the quality of

the solvent with respect to microemulsion stabil-

ity (Myakonkaya et al. 2011).

Soil Remediation

The approach of using microemulsions in soil

remediation (i.e., the removal of environmental

pollutants from topsoil) bears similarity with

approaches intended for oil recovery, which are

discussed below. This is partly due to the behav-

ior of soil contaminants, which in most cases tend

to be aromatic and chlorinated organic mole-

cules. Issues that arise in the use of

microemulsions in soil remediation are that

unlike oil wells, contaminant deposits have

much lower temperatures, obviously affecting

microemulsion phase behavior but also bulk

properties such as viscosity. Surfactants and

microemulsions will also adsorb onto soil parti-

cles, and hence their substituent parts need to be

biodegradable so as to avoid exchanging one

contaminant for another. Microemulsions can be

used in mobilizing the contaminants as well as

solubilizing them. Which one of the two pro-

cesses dominates is system and contaminant

dependent. For example, adsorbed solid or vis-

cous matter cannot be mobilized and solubiliza-

tion is required, while low-density contaminants

can be mobilized and solubilized. In both cases,

ultralow interfacial tensions are required such as

those exhibited in Winsor III systems, as contam-

inants tend to be strongly adsorbed in porous

matter. Mobilization can be problematic as it

can lead to the contaminants being displaced to

greater a depth, which is speculated to have

occurred in at least one large scale field test

(Fountain et al. 1991; Oostrom et al. 1999). The

solubilization by microemulsions is seen as sys-

tem dependent, with contaminants being solubi-

lized in some microemulsions systems but not

others, and salinity scans and use of cosurfactants

are usually required, especially for polar contam-

inants such as chloroform, 1,2-dichlorobenzene

and trichloroethylene (Baran et al. 1994; Shiau

et al. 1996).

Soil remediation processes often involve abra-

sion of soil particles. However, washing is not

effective when the contaminants are organic mol-

ecules with low water solubility and hence strong

adsorption onto inorganic matter. Solvent extrac-

tions, though effective, normally are time consum-

ing and require large solvent/soil ratios

(Khodadoust et al. 2000). Supercritical fluids

have also been employed, such as supercritical

carbon dioxide containing 10 % methanol, which

extracted up to 90 % polycyclic aromatic hydro-

carbons (PAH) from soil at 50 C (Librando et al.

2004). However, it is readily understandable that

this is an energy-intensive process. Soil washing

with surfactant solutions showed that polycyclic

aromatic hydrocarbons can be removed up to

approximately 60 % by employment of commer-

cially common nonionic surfactants (Ahn et al.

2008), though studies have shown that pollutant

swollen micelles can also adsorb into soil (Zhou

and Zhu 2007). An early report by Clemens et al.

(1994) demonstrated that surfactants can influence

the adsorption of pyrene (a model PAH).

M 720 Microemulsions

The nonionic surfactant C12E4 below its cmc wasfound to have no effect on the solubility of pyrene

in water, and only a minor effect on its adsorption

on layer silicates. Sodium dodecylsulfate (SDS)

and dodecyltrimethylammonium bromide

(DTAB) however enhanced adsorption leading to

accumulation at the solid/liquid interface. For the

nonionic surfactant above the cmc and in the pres-

ence of isooctane, solubilization of pyrene was

much higher than plain micellar solubilization,

leading to the conclusion that microemulsions

should be investigated for soil remediation appli-

cations. Suitable microemulsions should be com-

posed of biodegradable surfactants so that

reintroduction of the soil avoids any toxicity

risks. Clemens et al. (1998) investigated three

types of microemulsions formed by nonionic

surfactants, which would take up a model contam-

inant and release it through the formation of a

Winsor I system as a function of temperature.

The microemulsion exhibited slight differences in

phase behavior, which played a detrimental role

in the selection of microemulsion for this process.

Calcium bentonite and a real contaminated soil

sample were used to test the efficiency of the

microemulsions in removing pyrene. Three

microemulsions were prepared: the first two types

contained two nonionic surfactants, and it was

found that the phase behavior of these systems

played a vital role in determining the optimal

extraction. In particular, avoidance of the forma-

tion of lamellar phases was crucial. All three

microemulsions showed remarkable extraction

efficiencies, surpassing Soxhlet extractions. Typi-

cally, the recovery of pyrene, in both model

and real samples, with microemulsions reached

approximately 100 % on the first step of extraction

with temperature increase offering little assistance

to the process.

A maximum contaminant solubilization was

observed in Winsor III systems. As discussed in

section “Predicting Microemulsion Type”, the

middle phase of the Winsor III system comprises

equal volumes of oil and water which coexists

with excess upper oil and excess lower water

phases. In the middle phase, interfacial tension

is at a minimum, resulting in optimal solubiliza-

tion of contaminants therein. Winsor III systems

can be produced by both ionic and nonionic

surfactants, by tuning water salinity and temper-

ature respectively. This enhanced solubilization

by microemulsions is often referred to as

“supersolubilization,” and it is seen to occur in

Winsor III systems and near the transition from

Winsor I to Winsor III.

Graciaa et al. (1993a) investigated the effect

of the often necessary cosurfactant on the solubi-

lizing capability of microemulsions. Ethoxylated

alkylphenols were employed in the formation of

brine/hexadecane and brine/ethyl oleate systems.

Oil and brine were present in equal amounts

(water–oil ratio, WOR ¼ 1). A scan of the alkyl

tail lengths for the hexadecane system showed

that in order to maintain maximum solubilization

(i.e., Winsor III system), the optimum HLB and

the optimum solubilization parameter, SP*

(Vsurfacant/Vwater), had increasing trends with

increasing tail length. In the ethyl oleate system,

although the optimum HLB followed the same

trend as for the hexadecane system, the SP* pro-

duced a minimum. The experiments were

repeated by using mixed surfactant systems

aiming at keeping the ethylene oxide number

(EON) constant at optimum (EON*) using surfac-

tants of different ethylene oxide lengths. The

minimum in SP* was again observed, though

this time was shifted slightly higher than the

single surfactant systems. To further investigate

the causality of the effect, a model system of

brine/octylphenolethoxylates/ethyl oleate was

chosen. They found that although the apparent

EON was kept constant for all surfactant mix-

tures, the SP* varied considerably with the differ-

ence in EON for each of the two surfactant mixes.

In other words, solubilization increased with

increasing bimodality of the surfactants. As the

apparent EON and tail length were kept constant

at a constant brine/ethyl oleate ratio of unity, the

solubilization trend was thought to be due to

mixing effects. Also noted was that solubilization

was higher when stronger fractionation was tak-

ing place. The smaller EON surfactants become

strongly solubilized in the oil phase. These

smaller EON surfactants were given the name

lipophilic linkers. EO scans were carried out for

surfactants in the presence of a hydrophobic

Microemulsions 721 M

M

linker to ascertain the behavior of the EON*. The

results showed that the nonethoxylated

octylphenol linker and those bearing 0.5 and 1

EO units forced the EON* higher, which indi-

cated that the oil medium has increased in polar-

ity. The more lipophilic the linker was, the

stronger the effect observed. The presence of

1 % octylphenol in the oil doubled the amount

of oil solubilized in the microemulsion, which in

turn had to be balanced by water–surfactant inter-

actions, leading to the higher apparent EON

required for the optimum phase to exist. It was

concluded that the linkers provided ordering and

enhanced oil–oil and oil–surfactant interactions.

Though the interactions are extended, the order-

ing is not strong enough to allow for liquid crystal

formation. Graciaa et al. (1993b) went on to

examine the effect of lipophilic linkers in the

form of alcohols. For a WOR of unity, EON

scans were carried out in the presence of alcohol.

At the optimum formulation, the EON* should

shift to higher values to compensate for the

increased oil interactions; in other words more

hydrophilic surfactants are required for the for-

mation of theWinsor II system. Bymeasuring the

amount of alcohol in the upper oil and lower

water phase, the amount in the middle phase can

be determined. Only small amounts of alcohols

were found in the middle phase, partitioning

favorably and increasingly toward the oil phase

with increasing alcohol tail length. This also

suggested that the higher alcohols actually act as

cosolvents, a notion introduced in the first study.

The SP* variedmost with the small alcohols (buta-

nol to hexanol) which makes them also the most

surface active. Large increases in solubilization

were observed for medium length and long-chain

alcohols (hexanol–hexadecanol) which is counter

intuitive as the HLB* was kept constant. The find-

ings were interpreted as follows:

(a) C2–C6 alcohols are more surface active and

adsorb at the water/oil interface at the

expense of surfactant. As a consequence, sol-

ubilization is low. This also explains the ini-

tial minima observed (also seen in the first

study (Graciaa et al. 1993a)).

(b) C6–C10 alcohols are less surface active and

do not adsorb to the same extent as in case (a)

but provide additional surfactant–oil interac-

tions leading to higher required EON* values.

(c) C10–C16 alcohols have negligible interfacial

presence. The EON* values showed invari-

ance with increasing alcohol tail length. The

increase in solubilization is then attributed to

the lipophilic effect described in the previous

study. Furthermore, solubilization appeared

to be proportional to the alcohol length.

As mentioned previously, the nature of the

contaminant has a strong effect on its solubiliza-

tion by microemulsions, with chlorinated con-

taminants posing the biggest problem. Acosta

et al. (2002) demonstrated that use of equimolar

amounts of hydrophilic and lipophilic linkers in

the form of sodium dimethylnapthalenesulfonate

and dodecanol in sodium dihexylsulfosuccinate

microemulsions produced the best results for

polar contaminants. This reinforces the concept

that strongly binary systems exhibit higher solu-

bilization capabilities and demonstrating that the

linkers need not belong to the same family of

amphiphiles.

Oil Recovery

Oil fields contain, among many other substances,

a diverse mixture of hydrocarbons. These natural

resources have formed over many years from the

decomposition of biological organic matter

trapped inside the earth, under high temperatures

and pressures. Petroleum deposits (oil fields) can

span large areas and are usually within porous

rock enclaves which are surrounded by imperme-

able rock, thus confining oil within that area.

Initially, in certain deposits, oil may leak out

from the rock under its own pressure. In early

reports on oil, local populations especially

around the Caspian Sea would describe sponta-

neously occurring oil fountains which would

ignite, lasting even months (LeVine 2007). This

stage of oil recovery can release up to 20 % of the

reserve. When oil flow ceases, the next stage

involves well flooding with surfactant–polymer

solutions, injected under high pressure, mobiliz-

ing any easily accessible and displaceable oil in

the pores. This procedure can yield a further 20%

of the total oil available in the well. The role of

the polymer is to increase the viscosity of the

M 722 Microemulsions

medium, and importantly the polymer displaces

the surfactant from the bulk to the interface

and reduces its cmc. Prior to injecting the

surfactant–polymer solution, however, another

solution precedes it with the aim of removing

salts that would affect the surfactant. Such vol-

umes of injected fluids irrespective of their nature

are referred to as “slugs.”

Thus the majority of the oil remains under-

ground, adsorbed in the porous rock and too vis-

cous to be mobilized by water. The oil exists held

in the pores in the form of ganglia trapped by

capillary action. The ganglia, which can be

thought of as large oil drops may not be mobi-

lized due to bypassing of the water through least

resistance capillary paths or by an action known

as “snap-off,” where the ganglia become discon-

nected from the pore walls and unable to pass

through the capillary themselves, allow the less

viscous water to pass around it. This occurs when

the pore in which the oil is trapped is larger than

the pore throat. Hence, it can be seen that there

are two opposing forces, those of viscous forces

of the injected water and the capillary forces

trapping the ganglia. The dimensionless capillary

number (Nc) is used to describe the physical

properties of the particular site:

Nc ¼ mv go w=

.(48)

where m is the viscosity and n the velocity of the

injected fluid and go/w is the oil–water interfacial

tension. A high Nc is necessary to obtain a higher

extraction yield. Hence, research has focused on

increasing the viscosity of the fluid injected and

decreasing the interfacial tension. At this point,

efforts toward recovering the remaining oil are

referred to as enhanced oil recovery (EOR).

As mentioned in the previous section, ultralow

interfacial tensions are exhibited in Winsor III

systems which are obtained by increasing the

surfactant hydrophobicity, by raising the temper-

ature for nonionic surfactants, or increasing salin-

ity for ionic surfactants. Also at this point, equal

volumes of water and oil are solubilized in the

middle region microemulsion. In oil extraction,

the components (surfactants and brine) are usu-

ally injected into the well in which they can

form microemulsions. Ideally, the components

should form Winsor III systems once in the

well. However, there are issues with the use

of microemulsions since in many cases a

cosurfactant is necessary which may separate

the solution as the mixture proceeds down

through the porous rock. As a result, research

efforts have been made toward attaining the

desirable behavior by use of a single surfactant.

Graciaa et al. showed that introducing a polar part

in the long alkyl tails of a sulfonate surfactant

allowed it to form Winsor III systems in both

polar and nonpolar oils, also exhibiting remark-

able solubilization capability (Minana-Perez

et al. 1995).

Furthermore, the surfactants need to be solu-

ble in the presence of brine and other salt ions and

to not adsorb strongly to the mineral surfaces.

Hence, sulfonate surfactants are often employed

for this purpose.

Doe et al. (1977a) varied the chain length, and

isomers were varied for alkylbenzenesulfonates

(ABS) to examine their effect on the ability of

these compounds to effect ultralow interfacial

tensions. The ABS studied are depicted below

(Fig. 20) and can be thought of as a compound

containing m + n carbon atoms with the benzene

ring attached to (m+ 1)th carbon withm< n. The

compound can be visualized as a normal Cn ABS

with a Cm branch on the first carbon of the Cntail. Their results could be helpful in designing

surfactants depending on the desired application.

Measuring the interfacial tension of standard

surfactant solutions against a series of alkanes

provides important information on the minimum

interfacial tension (gmin) and the alkane carbon

number for minimum tension (nmin). Knowledge

of these variables can be of importance, espe-

cially in well flooding for the extraction of

petroleum.

The same authors investigated the linear

alkylbenzenesulfonates with additional alkyl

substituents for producing low interfacial tension

(Doe et al. 1978). The surfactants employed were

linear ABS having 8 to 16 carbon atom linear

alkyl chains. These molecules had one or two

alkyl substituents that could have methyl, ethyl,

propyl, i-propyl, butyl, or t-butyl groups on the

SO3Na

H2m+1Cm Cn−1H2n−1

H

C

SO3Na

H15C7

H

C C8H17

Microemulsions, Fig. 20 Alkylbenzene sulfonates

(Doe et al. 1978)

Microemulsions 723 M

M

benzene ring. The dialkylbenzenesulfonates were

grouped into two categories: those with a C12 or

C16 main chain length and those having C8 and

C10. For the first category, the general trend

followed was that the nmin increased with increas-

ing surfactant molecular weight. For the second

category, the shift in nmin for an n-alkyl group on

the benzene ring shifted almost linearly with the

number of carbon atoms it contained, though an

added methyl group had less of an effect than

previously reported. This they attributed to the

first carbon atom effect being masked by the

benzene ring. The branched alkyl group had

a lesser effect than the linear groups. The authors

noticed that the shift in nmin for a linear butyl

group was larger than increasing the main alkyl

tail by four carbon atoms. Trialkylbenzene-

sulfonates were divided in the same way as the

dialkylbenzenesulfonates. For the high molecular

weight group the general trend again indicated

a nmin increase with increasing surfactant molec-

ular weight. In the lower molecular weight group,

they observed that the introduction of two ethyl

groups had less of an effect than a single butyl

group (as in the case of dialkylbenzene-

sulfonates). Thus, they concluded that

dialkylbenzenesulfonates had the better molecu-

lar weight efficiency.

Doe et al. (1977b) revisited the study of

di- and trialkylbenzenesulfonates with an empha-

sis on the group in the ortho position on

the benzene ring relative to the sulfonate

group and molecular weight efficiency. The mol-

ecules investigated were p-di-n-alkylbenzene-sulfonates, o-di-n-alkylbenzenesulfonates, and

tri-n-alkylbenzenesulfonates. The p-di-n-alkylbenzenesulfonates had a butyl or higher

n-alkyl group in the ortho position with

respect to the sulfonate group; thus, the main

tail was in the para position. Similarly, o-di-n-

alkylbenzenesulfonates had the main tail in

the ortho position and the minor in the para

position. Of these three compounds, the p-di-n-

alkylbenzenesulfonates with the additional

n-alkyl chain proved to be the most molecular

weight efficient. Furthermore, the molecules

with relatively large ortho substituent gave their

lowest interfacial tensions at the lowest alkane

lengths a trend that held true even for the low

molecular weight structures.

In formulating water-based flooding fluids, the

parameters taken into account primarily revolve

around the surfactants employed. High tempera-

tures pose a danger of surfactant decomposition

as the fluid may have a well residency time of up

to a year in some cases. It affects solubility and

phase behavior as well as electrolyte dissolution,

viscosity, density, and the equivalent alkane car-

bon number (EACN) of the oil in the well. Hence,the oil tested under laboratory conditions may

have different EACN than that inside the well.

Other factors include salinity which has a primary

role when ionic surfactants are used as it

increases surfactant hydrophobicity and allows

the formation of Winsor III systems. However,

the presence of salts in the well may have adverse

effects when dissolved in the flooding fluid. Diva-

lent ions may cause the precipitation of part of the

surfactant, rendering them ineffective. To avoid

such an occurrence, ion scavengers have been

employed, as well as sacrificial surfactant flushes

and salinity gradients in the injected fluid.

In summary, the overall process of EOR by

microemulsions would involve three steps:

preflushing, which conditions the well for the sec-

ond flooding with the surfactants. The third step

requires a viscous polymer solution which will act

in almost a piston-like fashion to push the surfac-

tant fluid through the pores. When this occurs, the

surfactant front forms Winsor III systems at the

frontier of the slug. This allows desorption of

the ganglia which coalesce to form an oil bank

that can then be pushed through the pore throats.

M 724 Microemulsions

Increasing the rate of delivery and viscosity of

the slugs is very costly, and significant forces are

required to mobilize the ganglia in poorly acces-

sible pores. Injecting viscous fluids at high rates

can also cause the cracking of the rock and hence

offer preferred paths of high porosity, hence low-

ering the efficiency of any additional input fluids.

As an alternative to oil/water-based systems,

supercritical carbon dioxide (scCO2) has been

evaluated as a displacing fluid, as it can reach

otherwise inaccessible pores. scCO2 is also mis-

cible with an array of lower hydrocarbons, which

circumvents the issues found in water-based

flooding. As such, scCO2 has very little solubility

in water, and there is practically no loss of CO2

into brine. The disadvantage with the use of

scCO2 is that it is less viscous than the oil for

displacement, so that leakage and channeling

around the viscous oil deposits may occur with

a resulting loss of efficacy. Since the 1980s, aca-

demic attention has been directed toward increas-

ing the viscosity of scCO2 (Cummings et al.

2011). A further issue is that the vast majority

of conventional surfactants do not dissolve in

scCO2 due to its low polarity and poor solvent

quality. Fluorinated surfactants have shown high

solubility in scCO2, but they are expensive and

environmentally persistent and damaging. This

can be avoided or partially avoided by lowering

the fluorine content in the molecules. Introducing

fluorinated groups into the surfactants is also

known to drastically decrease the cmc in water.

To examine the potential of self-assembly to pro-

vide enhanced viscosities, Eastoe et al. (Trickett

et al. 2009) employed a partially fluorinated AOT

analogue and examined the effect of three differ-

ent surfactant counterions (Na+, Ni2+, Co2+).

Phase behavior significantly changed when the

divalent ions were used. The pressure instability

point (above which the system is stable, relating

to scCO2 bulk density) increased by 150 bar at

25 C. Introducing water to the micelles, thereby

increasing w, required an increase in temperature

to yield a stable dispersion. High-pressure SANS

measurements showed that the sodium analogue

formed near spherical microemulsions while the

divalent ions forced the formation of rigid, thin,

long, rod-shaped aggregates. High-pressure

viscometry revealed an increase of up to 40 %

in scCO2 viscosity with the Ni2+ counterion sur-

factant. However, further research is required to

eliminate fluorine from such compounds and to

further increase their thickening capabilities. The

challenge has been to produce a fluorine-free

surfactant that is also soluble in scCO2 and

which also shows a workable solubility in

water. To address this problem, Mohamed et al.

(2010) designed a CO2-compatible surfactant,

TC14, being a fully hydrogenated tert-butyl-

tipped trichain Aerosol-OT analogue. This

TC14 not only showed considerable solubility

in water, organic, and scCO2 media but also

formed micelles and microemulsions (Fig. 21

below).

In summary, the use of microemulsions in

EOR is still an emerging field with a steadily

increasing technology and promising results so

far that have found applications in detergency

and remediation processes.

Assessment

New and Future Developments

Microemulsions with Green and Novel Solvents

There is a need to reduce volatile organic solvents

and compounds (VOCs) employed in industrial

processes, since they pose environmental

threats. This has led to a drive to find suitable,

green, non-VOC solvents. Since they can be con-

sidered “universal solvents,” microemulsions

offer attractive prospects in this area. Two main

approaches have been explored: (a) supercritical

carbon dioxide (scCO2) (Eastoe et al. 2006)

and (b) room temperature ionic liquids (RTILs)

(Eastoe et al. 2005) as such replacements.

As such, scCO2 represents an excellent green

solvent due to the ease of solvent removal (by

reducing pressure to effect evaporation),

tuneability and recyclability of solvent quality

by temperature and pressure, the easily accessible

critical point (Pc ¼ 72.8 bar, Tc ¼ 31.1 C), andthe fact that it is nonflammable, nontoxic, envi-

ronmentally benign, biocompatible, cheap, and

finally abundant. On the other hand, RTILs

are salts made of sterically mismatched ions.

0.10 0.15 0.20

Q /Å−10.05

AOT

TC4

TC14CO2 w = 0

d-heptane w = 5

AOT4

0.00

0.02

0.04

0.06

0.10 0.15 0.20Q /Å−1

0.05

0

2

4

6

8

10

12

D2O

0.10 0.15

0.05

AOT4

AOT

TC14

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.10 0.15 0.20

0.20Q /Å−1

0.05

0.0

0.2

0.4

0.6

0.8

1.0 AOTa

b

c

TC14

AOT4

AOT

AOT4

TC4

TC14

I(Q

) / c

m−1

I(Q

) / c

m−1

I(Q

) / c

m−1

Q /Å−1

I(Q

) / c

m−1

I/CT

Microemulsions,Fig. 21 SANS profiles for

TC14, TC4, AOT4, and

AOT showing changes in

aggregate structure in three

different solvents:

(a) normal micelles in

water at 0.10 M surfactant.

The inset shows I(Q) scaledby a reduced concentration

Cr ¼ (C – cmc)/cmc toaccount for differing levels

of micellized surfactants

owing to the widely

different cmcs and for

presentation purposes.

(b) Reversed micelles in

heptane for w ¼ [water]/

[surf] ¼ 5 and a surfactant

concentration of 0.10 M.

(c) Dry reverse micelles in

CO2 for w ¼ 0 and

[surf] ¼ 0.04 M obtained at

380 bar and 25 C(Mohamed et al. 2010)

Microemulsions 725 M

M

This structure hinders crystallization, and the

materials remain “trapped” in the liquid state.

RTILs are also green solvent candidates due to

tuneability, polar solvation properties, and zero

volatility. By forming microemulsions with these

unusual liquids, their properties can be enhanced:

0.10 0.15 0.20 0.25

Q /Å−1

0.05

2

4

6

8TC14 w5

TC14 w0

AOT4 w5

AOT w0

2RmicI(

Q)

× 10

2 / c

m−1

Microemulsions,Fig. 22 SANS profiles of

TC14-stabilized dry

(w ¼ 0) and hydrated

(w ¼ 5) micelles obtained

in liquid CO2 at 360 bar,

25 C. The scatteringobtained from formulated

AOT4 w5 and dry AOT

w ¼ 0 systems is also

shown for comparison.

Smooth lines represent

model fits to a spherical

form factor scattering

model consistent with 11 A

radii (�10 %) for both

TC14-stabilized micelles

(Hollamby et al. 2009)

M 726 Microemulsions

water-in-CO2 systems (Eastoe et al. 2006) serve

to extend the capability with polar solutes, and

oil-in-RTIL microemulsions provide compatibil-

ity of the highly polar RTIL with organic and

hydrophobic components. The development of

custom-made CO2-philic and RTIL-philic surfac-

tants (which are not necessarily also very hydro-

philic or hydrophobic) has greatly stimulated

research in these fields. As an example, SANS

data proving micelles of a custom-made CO2-

philic surfactant (TC14) are formed in dense

CO2, shown in Fig. 22. On the other hand, the

normal AOT-like compounds do not show any

evidence for aggregation in CO2.

Hydrofluorocarbon (HFC) solvents are recog-

nized as attractive alternatives to fully fluorinated

solvents, owing to both low toxicity and flamma-

bility. A significant development is replacement

of CFC’s by HFC’s as refrigerants, as propellants

in metered-dose inhalers and drug delivery

devices for respiratory tract infections. However,

these HFC carrier solvents are hydrophobic

and, being partially fluorinated, are also

oleophobic, creating solubility problems for the

pharmaceutically active components. Using suit-

able HFC-compatible surfactants could allow for

the dispersion of aqueous drug solutions as

microemulsions. Recently, progress has been

made (Patel et al. 2003) toward designing surfac-

tants HFC microemulsions for potential drug

delivery applications.

With regard to drug delivery and encapsula-

tion, the first commercialized oil-in-water

microemulsion formation is for the immunosup-

pressant drug cyclosporine, which is marketed as

Neoral (UK Patent No 2 222 770). The success of

Neoral points to a future role for microemulsions

in the pharma and allied medical sciences, an area

popularly termed “nanomedicine.”

To wrap up, microemulsion science is highly

developed: the initial intense phase of thermody-

namic and phase behavior studies have paved the

way for detailed structural investigations by light,

X-ray, or neutron scattering, as well as NMR,

among other techniques. The underlying self-

assembly structures are well understood and

documented, and the relationships between

chemical structure of the components (surfac-

tants, oils) and phase stability, as well as colloi-

dal-scale structure, are now well advanced.

The future research directions in this field will

be toward practical applications of

Microemulsions 727 M

microemulsions in diverse fields such as

nanomaterials, drug delivery, energy efficiency,

and environmental cleanup.

Cross-References

▶Emulsions

▶Micellar Systems

▶ Phase Behavior of Surfactants

▶ Surfactant Adsorption

▶ Surfactants

M

References

Abe M, Nakamae M, Ogino K, Wade WH (1987) Chem

Lett 16(8):1613–1616

Abecassis B, Testard F, Zemb T (2009) Soft Matter

5:974–978

Acosta E, Trans S, Uchiyama H, Sabatini DA, Harwell JF

(2002) Environ Sci Technol 36:4618–4624

Adamson AW (1960) Physical chemistry of surfaces.

Interscience, New York, p 393

Ahlnas T, Soderman O, Hjelm C, Lindman B (1983)

J Phys Chem 87:822–828

Ahn CK, Kim YM, Woo SH, Park JM (2008) J Hazard

Mater 154:153–160

Ashcroft NW, Lekner J (1966) Phys Rev 145:83–89

Aveyard R, Binks BP, Clarke S, Mead J (1986a) J Chem

Soc Faraday Trans I 82:125–142

Aveyard R, Binks BP, Mead J (1986b) J Chem Soc Fara-

day Trans I 82:1755–1770

Aveyard R, Binks BP, Fletcher PDI (1989) Langmuir

5:1210–1217

Bancroft WD (1913a) J Phys Chem 17:501–519

Bancroft J (1913b) Phys Chem 17:514–518

Baran JR, Popr GA, Wade WH, Weerasooriya V, Yapa

A (1994) Environ Sci Technol 28(7):1361–1366

Becher P (1984) J Disp Sci Technol 5:81–96

Bivas I, Hanusse P, Botherel P, Lalanne J, Aguerre-

Chariol OJ (1987) Physique 48:855–867

Borkovec M, Eicke H-F, Hammerich H, Gupta BD

(1988) J Phys Chem 92:206–211

Bourrel M, Schechter RS (1988) Microemulsions and

related systems. Marcel Dekker, New York

Bumajdad A, Eastoe J, Heenan RK, Lu JR, Steytler DC,

Egelhaaf S (1998) J Chem Soc Faraday Trans

94:2143–2150

Capek I (2004) Adv Colloid Interface Sci 110:49–74

Cazabat AM, Langevin D, Pouchelon A (1980) J Colloid

Interface Sci 73:1–12

Charinpanitkul T, Chanagul A, Dutta J, Rungsardthong U,

Tanthapanichakoon W (2005) Sci Technol Adv Mater

6:266–271

Chen DH, Wu SH (2000) Chem Mater 12:1354–1360

Chen SJ, Evans FD, Ninham BW (1984) J Phys Chem

88:1631–1634

Clarke JHR, Nicholson JP (1984) In: Mittal K,

Lindman B (eds) Surfactants in solution III. Plenum,

New York

Clemens WD, Haegel F-H, Schwuger MJ (1994) Lang-

muir 10:1366–1369

ClemensWD, Haegel F-H, SchwugerMJ (1998)Micelles,

microemulsions and monolayers: science and technol-

ogy. Marcel Dekker, New York, pp 215–231

Clowes GHA (1916) J Phys Chem 20:407–451

Cummings S, Trickett K, Enick R, Eastoe J (2011) Phys

Chem Chem Phys 13(4):1276–1289

Daillant J, Bosio L, Benattar JJ, Meunier J (1989)

Europhys Lett 8:453–458

Davies JT (1957) Proceeding of the 2nd congress surface

activity. Butterworths, London

De Gennes PG, Taupin C (1982) J Phys Chem

86:2294–2304

de Kruif CG, Briels WJ, May RP, Vrij A (1988) Langmuir

4:668–676

Di Meglio JM, Dvolaitzky M, Taupin C (1985) J Phys

Chem 89:871–874

Doe PH, Wade WH, Schechter RS (1977a) J Colloid

Interface Sci 59(3):525–531

Doe PH, El-Emary M, Wade WH, Schechter RS (1977b)

J Am Oil Chem Soc 55(5):513–520

Doe PH, El-Emary M, Wade WH (1978) J Am Oil Chem

Soc 55:505–512

Eastoe J, CoxAR (1995) Colloid Surf A Physicochem Eng

Asp 101:63–76

Eastoe J, Sharpe D (1997) Langmuir 13:3289–3294

Eastoe J, Young WK, Robinson BH (1990) J Chem Soc

Faraday Trans 86(16):2883–2889

Eastoe J, Dong J, Hetherington KJ, Steytler DC, Heenen

RK (1996a) J Chem Soc Faraday Trans 92:65–72

Eastoe J, Hetherington KJ, Sharpe D, Dong J, Heenan RK,

Steytler DC (1996b) Langmuir 12:3876–3880

Eastoe J, Stebbing S, Dalton J, Heenan RK (1996c) Col-

loid Surf A Physicochem Eng Asp 119:123–131

Eastoe J, Hetherington KJ, Sharpe D, Dong J, Heenan RK,

Steytler DC (1997a) Colloid Surf A 128:209–215

Eastoe J, Hetherington KJ, Sharpe D, Steytler DC,

Egelhaaf S, Heenan RK (1997b) Langmuir

128:2490–2493

Eastoe J, Sharpe D, Heenan RK, Egelhaaf S (1997c)

J Phys Chem B 101:944–948

Eastoe J, Gold S, Rogers SE, Paul A, Welton T, Heenan

RK, Grillo I (2005) J Am Chem Soc 217:7302

Eastoe J, Gold S, Steytler DC (2006) Langmuir 22:9832

Eicke H-F, Meier W (1996) Biophys Chem 58:29–37

El-Sayed IH, Huang X, El-Sayed MA (2005) Nano Lett

5:829–834

M 728 Microemulsions

Engelhardt H, Duwe HP, Sackmann E (1985) J Phys Lett

46:395–400

Faizan M, Myakonkaya O, Shah SS, Eastoe J (2011)

J Colloid Interface Sci 354:624–629

Farago B, Huang JS, Richter D, Safran SA, Milner ST

(1990) Prog Colloid Polym Sci 81:60–63

Fletcher PDI, Howe AM, Robinson BH (1987) J Chem

Soc Faraday Trans I 83:985–1006

Fletcher PDI, Clarke S, Ye X (1990) Langmuir

6:1301–1309

Fountain JC, Klimek A, Beikirch MG, Middleton TM

(1991) J Hazard Mater 28(3):295–311

Garcia-Rio L, Leis JR, Mejuto JC, Pena ME (1994) Lang-

muir 10:1676–1683

Germain V, Brioude A, Ingert D, Pileni M-P

(2005) J Chem Phys 122(124707):1–8

Graciaa A, Lachaise J, Cucuphat C, Bourrel M, Salager JL

(1993a) Langmuir 9:669–672

Graciaa A, Lachaise J, Cucuphat C, Bourrel M, Salager JL

(1993b) Langmuir 9:3371–3374

Gradzielski M, Langevin D (1996) J Mol Struct

383:145–156

Gradzielski M, Langevin D, Farago B (1996) Phys Rev

E 53:3900–3919

Griffin WC (1949) J Cosmet Chem 1:311–326

Griffin WC (1954) J Cosmet Chem 5:249–256

Guering P, Lindman B (1985) Langmuir 1:464–468

Hait SK, Moulik SP, Rodgers MP, Burke SE, Palepu

R (2001) J Phys Chem B 105:7145–7154

Hait SK, Sanyal A, Moulik SP (2002) J Phys Chem

B 106:12642–12650

Hansen RS (1960) J Phys Chem 64:637–641

Helfrich WZ (1973) Naturforsch 28c:693–703

Hilfiker R, Eicke H-F, Hammerich H (1987) Helv Chim

Acta 70:1531–1536

Hoar TP, Schulman JH (1943) Nature (Lond)

152:102–103

Hollamby MJ, Tricket K, Mohamed A, Cummings S,

Tabor S, Myakonkya O, Gold S, Rogers S,

Heenan RK, Eastoe J (2009) Angewandte Chem

48:4993–4995

Hollamby MJ, Eastoe J, Chemelli A, Glatter O,

Rogers SE, Heenan RK, Grillo I (2010) Langmuir

10:6989–6994

Hone DC, Robinson BH, Steytler DC (2000) Langmuir

16:340–346

Hunter RJ (1994) Introduction to modern colloid science.

Oxford University Press, Oxford

Hyde S, Andersson K, Larsson K, Blum Z, Landh S,

Ninham BW (1997) The language of shape. Elsevier,

Amsterdam

Israelachvili JN (1994) Colloid Surf A 91:1–8

Israelachvili JN, Mitchell DJ, Ninham BW (1976) J Chem

Soc Faraday Trans 2(72):1525–1568

John AC, Rakshit AK (1994) Langmuir 10:2084–2087

Kahlweit M, Strey R, Busse G (1990) J Phys Chem

94:3881–3894

Kegel WK, Bodnar I, Lekkerkerker HNW (1995) J Phys

Chem 99:3272–3281

Kellay H, Binks BP, Hendrikx Y, Lee LT, Meunier

J (1994) Adv Colloid Interface Sci 9:85–112

Khodadoust AP, Bagchi R, Suidan MT, Brenner RC,

Sellers NG (2000) J Hazard Mater B80:159–174

Klevens H (1953) J Am Oil Chem Soc 30:74–80

Kling C (1949) Kolloid-Z 115:37–44

Koppel DE (1972) J Chem Phys 57:4814–4820

Kotlarchyk M, Chen S-H, Huang JS (1982) J Phys Chem

86:3273–3276

Kotlarchyk M, Chen S-H, Huang JS, Kim MW

(1984) Phys Rev A 29:2054–2069

Kunieda H, Shinoda K (1980) J Colloid Interface Sci

75(2):601–606

Kurihara K, Kizling J, Stenius P, Fendler J (1983) J Am

Chem Soc 105:2574–2579

Lang J, Zana R, Bauer R, Hoffman H, Ulbricht W (1975)

J Phys Chem 79:276–283

Langevin D (ed) (1992) Light scattering by liquid surfaces

and complementary techniques. Marcel Dekker,

New York

Lee MS, Park SS, Lee G-D, Ju CS, Hong S-S (2005) Catal

Today 101:283–290

LeVine S (2007) The oil and the glory. Random House,

New York

Li L, Quing-Sheng W, Ya-Ping D, Pei-Ming W (2005)

Mater Lett 59:1623–1626

Librando V, Hutzinger O, Tringali G, Aresta M (2004)

Chemosphere 54:1189–1197

Lin IJ, Friend JP, Zimmels Y (1973) J Colloid Interface

Sci 45(2):378–385

Lindman B, Stilbs P, Moseley ME (1981) J Colloid Inter-

face Sci 83(2):569–582

Lindman B, Ahlnas T, Soderman O, Walderhaug H,

Rapacki K, Stilbs P (1983) Faraday Discuss Chem

Soc 76:317–329

Lisiecki I, Pileni M-P (1995) J Phys Chem 99:5077–5082

Lisiecki I, Pileni M-P (2003) Langmuir 19:9486–9489

Lopez-Quintela MA, Tojo C, Blanco MC, Rio LG,

Leis JR (2004a) Curr Opin Colloid Interface Sci

9:264–278

Lopez-Quintela MA, Tojo C, Blanco MC, Garcia Rio L,

Leis JR (2004b) Curr Opin Colloid Interface Sci

9:264–278

Luisa Marin ML, McGilvray KL, Scaiano JC (2008) J Am

Chem Soc 130(49):16572–16584

Lv FF, Li N, Zheng LQ, Tung CH (2006) Eur J Pharm Sci

62:288–294

Maillard M, Giorgio S, Pileni M-P (2003) J Phys Chem

B 107:2466–2470

Markovic I, Ottewill RH (1986) Colloid Polym Sci

264:454–462

McBain JW (1950) Colloid Science. D.C. Health, Boston

McConnell WP, Nowak JP, Brousseau LC,

Tenent RC, Feldheim DLJ (2000) Phys Chem

B 104:8925–8930

Meunier J, Lee LT (1991) Langmuir 46:1855–1860

Milner ST, Safran SA (1987a) Phys Rev A 36:

4371–4379

Milner ST, Safran SA (1987b) Phys Rev A 36:4371–4379

Miniemulsions 729 M

M

Minana-Perez M, Graciaa A, Lachaise J, Salager JL

(1995) Colloid Surf A 100:217–224

Mohamed A, Trickett K, Chin SY, Cummings S,

Sagisaka M, Hudson L, Nave S, Dyer R, Rogers SE,

Heenan RK, Eastoe J (2010) Langmuir 26(17):

13861–13866

Moulik SP, Paul BK (1998) Adv Colloid Interface Sci

78:99–195

Myakonkaya O, Eastoe J, Mutch KJ, Grillo I (2011) Phys

Chem Chem Phys 13:3059–3063

Nave S, Eastoe J, Heenan RK, Steytler D, Grillo I (2000a)

Langmuir 16:8741–8748

Nave S, Paul A, Eastoe J, Pitt AR, Heenan RK (2000b)

Langmuir 21:10021–10027

Oliveira RJ, Brown P, Correia G, Rogers SE, Heenan RK,

Grillo I, Galembeck A, Eastoe J (2011) Langmuir

27:9277–9284

Olsson U, Wennerstrom H (1994) Adv Colloid Interface

Sci 49:113–146

Oostrom M, Hofstee C, Walker RC, Dane JH

(1999) J Contam Hydrol 37(1–2):179–197

Ornstein LS, Zernike F (1914) Proceeding of the

Koninklijke Akademia van Weten-schappen te

Amsterdam, vol 17, pp 793–806

Overbeek JTG (1978) Faraday Discuss Chem Soc

65:7–19

Patel N, Marlow M, Lawrence MJ (2003) J Colloid Inter-

face Sci 258:345

Pileni M-P (1997) Langmuir 13:3266–3276

Pileni M-P, Motte L, Petit C (1992) Chem Mater

4:338–345

Provencher SW (1982) Comp Phys Commun 27:229–242

Ray S, Bisal SR, Moulik SP (1993) J Chem Soc Faraday

Trans 89(17):3277–3282

Ricka J, Borkovec M, Hofmeier U (1991) J Chem Phys

94(12):8503–8509

Rock PA (1969) Chemical thermodynamics. Macmillan,

London

Roig AR, Alessandrini JL (2006) Particle and particle

characterization 23:431–437

Safran SA (1983) J Chem Phys 78:2073–2076

Safran SA (1992) In: Chen SH, Huang JS, Tartaglia P

(eds) Structure and dynamics of strongly interacting

colloids and supramolecular aggregates in solution,

vol 369, NATO Advanced Study Institute, Series C:

Mathematical and physical sciences. Kluwer, Dortrecht

Safran SA, Tlusty T (1996) Ber Bunsenges Phys Chem

100:252–263

Schneider MB, Jenkins JT, Webb WW (1984) Biophys

J 45:891–899

Schulman JH, Friend JA (1949) J Colloid Sci

4(5):497–509

Schulman JH, Riley DP (1948) J Colloid Sci 3:383–405

Shiau B-J, Sabatini DA, Harwell JH, Vu DQ (1996) Envi-

ron Sci Technol 30(1):97–103

Shinoda K, Friberg S (1986) Emulsions and solubilization.

Wiley-Interscience, New York

Shinoda K, Saito H (1969) J Colloid Interface Sci

34:258–263

Sicoli F, Langevin D, Lee LT (1993) J Chem Phys

99:4759–4765

Sjoblom E, Henriksson U (1983) In: Mittal KL,

Lindman B (eds) Surfactants in solution. Plenum

Press, New York, p 1867

Sottman T, Strey R (1996) Ber Bunsenges Phys Chem

100:237–241

Steingerwald ML, Alivisatos AP, Gibson JM, Harris R,

Kortan R,Muller AJ, Thayer AM, Duncan DC, Douglas

DC, Brus LE (1988) J Am Chem Soc 110:3046–3050

Stilbs P, Rapacki K (1983) J Colloid Interface Sci

95(2):583–585

Stilbs P, Warnheim T, Sjoblom E, Henriksson U (1984)

J Phys Chem 88:5420–5425

Szleifer I, Kramer D, Ben-Shaul A, Gelbart WM, Safran

S (1990) J Chem Phys 92:6800–6817

Tabor RF, Eastoe J, Dowding PJ, Grillo I, Rogers SE

(2010) J Colloid Interface Sci 344:447–450

Tanford C (1978) The hydrophobic effect. Wiley,

New York

Tavacoli JW, Dowding PJ, Steytler DC, Barnes DJ, Routh

AF (2008) Langmuir 24:3807–3813

Tondre C (2005) Dynamics in microemulsion systems. In:

Zana R (ed) Dynamics of surfactant assemblies,

vol 125. Francis and Taylor, Boca Raton

Tondre CJ, Derouiche A (1991) J Disp Sci Technol

12:517–530

Trickett K, Xing D, Enick R, Eastoe J, Hollamby MJ,

Mutch KJ, Rogers SE, Heenan RK, Steytler DC

(2009) Langmuir 26(1):83–88

UK Patent No. 2 222 770 “Pharmaceutical compositions

containing cyclosporine”

Uskokovic V, Drofenik M (2005) Surf Rev Lett

12:239–277

Vincent B, Kiraly Z, Obey TM (1998) In: Binks BP (ed)

Modern aspects of emulsion science. Royal Society of

Chemistry, London, p 100

Wei L, Sun P, Nie S, Pan W (2005) Drug Dev Ind Pharm

31:785–794

Winsor PA (1948) Trans Faraday Soc 44:376–398

Winterhalter M, Helfrich W (1992) J Phys Chem

96:327–330

Wu C (2008) In: Seidel A (ed) Characterisation and

analysis of polymers. Wiley, New York

Xia Y, Yang P, Sun Y, Wu Y, Mayers B, Gates B, Yin Y,

Kim F, Yan H (2003) Adv Mater 15:353–389

Zemb T (1991) In: Lidner P, Zemb T (eds) Neutron, X-ray

and light scattering: introduction to an investigativetool for colloidal and polymeric scientists. North

Holland, Amsterdam, pp 177–198

Zhou W, Zhu L (2007) Environ Pollut 147:66–73

Zulauf M, Elcke H-F (1979) J Phys Chem 83(4):480–486

Miniemulsions

▶Nanoemulsions

M 730 Mixed Film Theory

Mixed Film Theory

Tharwat Tadros

Wokingham, Berkshire, UK

Synonyms

Mixed film theory for microemulsion formation

Definition

The film (which may consist of surfactant and

cosurfactant molecules) is considered as a liquid

“two dimensional” third phase in equilibrium

with both oil and water. Such a monolayer

could be a duplex film, i.e., giving different prop-

erties on the water side and oil side. The initial

“flat” duplex film has different tensions at the oil

and water sides. This is due to the different pack-

ing of the hydrophobic and hydrophilic groups

(these groups have different sizes and cross-

sectional areas). It is convenient to define a two-

dimensional surface pressure p: p ¼ go �g; go isthe interfacial tension of the clean interface,

whereas g is the interfacial tension with adsorbedsurfactant. One can define two values for p at the

oil and water phases, po and pw, which for a flat

film are not equal, i.e., p0o¼ p0w. As a result of thedifference in tensions, the film will bend until

po ¼ pw. If p0o > p0w, the area at the oil side hasto expand (resulting in reduction of p0o) until

po ¼ pw. In this case, a W/O microemulsion is

produced. If p0w > p0o, the area at the water sideexpands until pw ¼ po. In this case, an O/W

microemulsion is produced. According to the

duplex film theory, the interfacial tension gT is

given by the following expression:

gT ¼ gðO=WÞ �p, where (go/w)a is the interfacial

tension that is reduced by the presence of the

alcohol. The value of (go/w)a is significantly

lower than go/w in the absence of the alcohol

(e.g., for hydrocarbon/water go/w is reduced

from 50 to 15–20 mNm�1 on the additional of

a significant amount of a medium-chain alcohol

like pentanol or hexanol). Contributions to p are

considered to be due to crowding of the surfactant

and cosurfactant molecules and penetration of the

oil phase into the hydrocarbon chains of the inter-

face. If p > (go/w)a, gT becomes negative and this

leads to expansion of the interface until gTreaches a small positive value. Since (go/w)a is

of the order of 15–20 mNm�1, surface pressures

of this order are required for gT to approach

a value of zero. This above duplex film theory

can explain the nature of the microemulsion: The

surface pressures at the oil and water sides of the

interface depend on the interactions of the hydro-

phobic and hydrophilic portions of the surfactant

molecule at both sides, respectively. If the hydro-

phobic groups are bulky in nature relative to the

hydrophilic groups, then for a flat film, such

hydrophobic groups tend to crowd forming

a higher surface pressure at the oil side of the

interface; this results in bending and expansion at

the oil side forming a W/O microemulsion. An

example for a surfactant with bulky hydrophobic

groups is Aerosol OT (dioctyl sulfosuccinate). If

the hydrophilic groups are bulky such as is the

case with ethoxylated surfactants containing

more than five ethylene oxide units, crowding

occurs at the water side of the interface. This

produces an O/W microemulsion.

Cross-References

▶ Interfacial Tension

▶Microemulsions

▶ Surfactants

Mixed Film Theory forMicroemulsion Formation

▶Mixed Film Theory

Mixing Interaction Free Energy

▶Osmotic Repulsion

Mouthfeel and Food Texture 731 M

Mixtures of Suspensions andEmulsions

▶ Suspoemulsions

Molecular Mixing

▶Diffusion of Particles

Monolayer Phases

▶ Surfactant Monolayers

Mouthfeel

Tharwat Tadros

Wokingham, Berkshire, UK

Synonyms

Mouthfeel and food texture

Definition

Mouthfeel is a sensory perception of food texture.

Food products are generally designed with an

optimum “consistency” for application in cutting,

slicing, spreading, or mixing. During eating and

mastication, the food loses its initial “consis-

tency,” at least partially. The mouthfeel of food

products may be related to the loss of this initial

“consistency.” During the first stage of this mas-

tication process, the food is comminuted by the

action of the teeth into particles (few millimeters

in size). At this stage, the food is close to its initial

“consistency.” Thus, in the first stages of

mastication, the mouthfeel may be related to its

rheological characteristics. It is, therefore, possi-

ble to relate the mouthfeel during the first stages

of mastication to the rheological parameters such

as yield value, creep compliance, and storage

modulus. After the initial stages of comminution,

the food particles “soften” as a result of temper-

ature rise and moisture uptake in the oral cavity.

This results in significant reduction in “consis-

tency” which may reach values of stresses com-

parable to the level encountered by the saliva

flow in the oral cavity. When these stresses are

reached, the food particles will be broken down to

much smaller size that is determined by the

hydrodynamics of the “flowing” saliva. The

flow in the saliva is rather complex, and calcula-

tion of shear stresses is not straightforward.When

the above stage is reached, the food product will

form a “homogeneous” mix with the saliva, and

the mouthfeel will appear smooth. It is clear that

if the “consistency” of the product does not

decrease to a sufficient degree (such that the

stresses are comparable to those encountered by

the saliva flow), the masticated food will remain

“thicker” and the mouthfeel becomes unaccept-

able to the consumer (feel of “graininess,” “stick-

iness,” or “waxiness”). Control of the

“consistency” (rheological characteristics) of

food products is essential for consumer accept-

ability, and this may require sophisticated mea-

surements and interpretation of the results

obtained.

Cross-References

▶ Food Colloids

▶Rheology

Mouthfeel and Food Texture

▶Mouthfeel