Pickering emulsions stabilized by stacked catanionic micro-crystals controlled by charge regulation [original]

Pickering emulsions stabilized by stacked catanionic micro-crystals controlled

by charge regulation†

Natascha Schelero,

Antonio Stocco,

Helmuth M€

ohwald

and Thomas Zemb

Received 18th April 2011, Accepted 8th August 2011

DOI: 10.1039/c1sm05689a

In this paper the mechanism behind the stabilization of Pickering emulsions by stacked catanionic

micro-crystals is described. A temperature-quench of mixtures of oppositely charged surfactants

(catanionics) and tetradecane from above the chain melting temperature to room temperature produces

stable oil-in-water (o/w) Pickering emulsions in the absence of Ostwald ripening. The oil droplets are

decorated by stacks of crystalline discs. The stacking of these discs is controlled by charge regulation as

derived from conductivity, scattering and zeta potential measurements. Catanionic nanodiscs are ideal

solid particles to stabilize Pickering emulsions since they present no density difference and a structural

surface charge which is controlled by the molar ratio between anionic and cationic components. The

contact angle of catanionic nanodiscs at a water/oil interface is also controlled by the non-

stoichiometry of the components. The resulting energy of adhesion and the repulsion between droplets

is much larger than kT. As a consequence of these unique properties of nanodiscs, this type of

emulsions presents an extremely high resistance towards coalescence and creaming, even in the presence

of salt.

1. Introduction

Emulsions stabilized by particles are called solid-stabilized

emulsions or Pickering-emulsions, because in 1907, Pickering

generalized the initial observations by Perrin and Ramsden that

fine solid particles wetted by water rather than oil act as emul-

sifiers for o/w emulsions by residing at the interface.

Since then,

properties of micro and nano particles which are ideally suited

for stabilizing Pickering emulsions have been identified. The

adsorption of particles at the oil/water interface strongly depends

on the particle size and shape, wettability, and interparticle

interaction.

2,3

Furthermore, such particles must have (1) negli-

gible solubility in water and in oil, even with a high specific

surface, so that a transition from particle to molecules via

leaching or chemical dissolution does not occur, (2) a density at

best in between water and oil to avoid density gradients, and (3)

known contact angles with water and oil close to 90in order to

induce surface tension reduction between the two phases of the

emulsion.

Interestingly, the known properties of catanionic crystals

produced in the salt-free system made from weak carboxylic

acids and alkyltrimethylammonium hydroxides, uniquely fulfil

these four requirements:

1. the density gradient is negligible, since their components

have densities between water and many oils;

2. the osmotic repulsion between discs is known;

3. the local crystalline in plane order is known;

4. the ternary phase prism important for the determination of

the temperature range for preparing and storing the emulsions is

known.

6,7

For these reasons, catanionics seem to be a powerful emulsi-

fier. In general, the term catanionics describes on the one hand,

systems where cationic and anionic surfactants are mixed in non-

equimolar ratios with their counterions present.

In these

mixtures, the catanionic surfactant coexists as an individual

chemical species with an equimolar amount of salt and excess of

one of the ionic components. On the other hand, catanionics are

formed from pairing two oppositely charged surfactants with

removal of the inorganic counterions by ion-exchange, precipi-

tation or extraction at equimolar quantities.

Even though it was

empirically known since the early seventies that mixtures of

oppositely charged surfactants and lipid formulations are

extremely efficient emulsifiers, emulsions stabilized by cata-

nionics have been rarely mentioned in the literature.

10–16

This is

even more astonishing since it is proposed that catanionics might

combine the advantages of particle stabilization of emulsions and

the amphiphilicity of molecular surfactants to afford better

emulsion stabilization. This unique property is else only known

for the so-called ‘‘Janus’’ particles.

Previously, the authors

ascertained the proposed extraordinary high lifetime of Pickering

Stranski-Laboratorium, Department of Chemistry, TU Berlin, Strasse des

17. Juni 124, 10623 Berlin, Germany

Max-Planck-Institut f€

ur Kolloid- und Grenzfl€

achenforschung, Am

M€

uhlenberg 1, 14424 Potsdam, Germany

Institut de Chimie S

eparative de Marcoule UMR 5257, CEA/CNRS/

UM2/ENSCM, Marcoule, France

† Electronic supplementary information (ESI) available. See DOI:

10.1039/c1sm05689a

10694 | Soft Matter, 2011, 7, 10694–10700 This journal is ªThe Royal Society of Chemistry 2011

Dynamic Article LinksC

Soft Matter

Cite this: Soft Matter, 2011, 7, 10694

www.rsc.org/softmatter PAPER

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online

/ Journal Homepage

/ Table of Contents for this issue

emulsions stabilized by catanionic crystals due to particle-by-

particle counting via light scattering.

The aim of this paper is to

gain a general understanding of the mechanism governing the

stability of emulsions stabilized by catanionic crystals.

Furthermore, the resilience of such emulsions against external

influences like salt is tested.

2. Experimental methods

2.1 Materials

We chose tetradecane as oil because its low solubility renders

negligible the ripening mechanism by molecular exchange.

Tetradecane (Fluka, p.a > 99%) was used as received. Water was

passed through a reverse osmosis unit and a Milli-Q reagent

water system. A 10 wt% solution of hexadecyl-

trimethylammonium hydroxide (CTAOH) in water (Fluka) was

freeze-dried, recovered in a minimum amount of ethanol and

recrystallized twice from anhydrous ether in order to remove the

impurities due to self-decomposition of CTAOH, mainly trime-

thylamine and hexadecane resulting from Hoffmann elimination.

Myristic acid (Fluka, p.a. >99%) was recrystallized twice from

hot acetonitrile.

2.2 Sample preparation

2.2.1 Catanionic solutions. The catanionic solutions were

prepared following the standard procedure as reported in several

publications.

4,6,7,19–22

In general, catanionic systems are specified

by three characteristic quantities: (1) the total amount of dry

surfactant cin wt%, (2) the molar ratio of anionic surfactant r,

and (3) the fraction of counterions exchanged from original OH



to Cl



x. In order to adjust the desired amount of ions – in the

presented cases x¼5.3%, CTAOH and CTACl were mixed as

powders in the required ratio, solved in small amounts of water

and then lyophilized. To prepare the catanionic solutions, the

desired amount of CTAOH/CTACl is weighed, then myristic

acid is added to receive the desired r-value and finally the missing

amount of water is filled up in order to gain the wanted

concentration cof surfactant. After that the samples have to be

heated above 50 C to dissolve myristic acid. Short heating to

65 C allows homogenisation of the sample and rearrangement

of the surfactant chains.

2.2.2 Emulsions. Depending on the oil/water ratio the needed

volume was put in a tube. The samples containing oil, water and

catanionics were mixed for 2 min with a vortex-apparatus and

then emulsified for 7 min by using an Ultra-Turrax T8 (Ika) while

heating the sample in a water bath. The used dispersing machine

works with a rotor–stator configuration, with a 5 mm head

operating at a shearing speed of 49 700 s

1

. During dispersion the

samples were heated above the chain melting temperature since

the used catanionic formulations are in the crystallized state (L

)

at room temperature. As already known from earlier work, the

chain melting temperature depends on the molar ratio of the

anionic surfactant.

Thus, the samples with r< 0.5 were heated

up to 35–40 C and the samples with r$0.5 above 65 C. After

dispersion the samples were cooled to room temperature and

stored at a constant temperature of 20 C. To perform

measurements the emulsions had to be diluted with water by

different factors depending on the method used. Observation of

various oil droplets showed that the characteristics of such

droplets has never changed by diluting with water by different

dilution factors.

2.3 Methods

2.3.1 Confocal laser scanning microscopy (CLSM). The

emulsion droplets were visualized by an inverted confocal

microscope IX 61 from Olympus with a FV 1000 confocal head

equipped with a 100oil immersion objective. In the present

experiments we employed a combination of fluorescence and

confocal microscopy. Therefore a fluorescent marker was added

to all examined samples. The fluorescence of Oregon Green was

excited by a 488 nm laser line. To reduce the high optical density

for any confocal microscopic observation, the emulsions were

diluted either 1 : 100 or 1 : 1000 by water. Five drops of Oregon

green solution per 1 ml diluted emulsion were added. The

concentration of the dye depends on the dilution factor of the

emulsion. One drop of the prepared dilution containing the dye

was put on a glass slide and covered with a second glass slide. The

glass slides were stuck together to avoid too fast evaporation of

the small sample under the microscope.

2.3.2 Electrophoretic mobility. The electrophoretic mobility

was achieved by using a Zetasizer Nano Z (Malvern Instru-

ments). For measuring, the emulsions were diluted by water by

a factor 1 : 1000 and measured in a Folded Capillary cell DTS

1060—a maintenance-free capillary cell. For comparison, several

experiments were repeated with a Zetasizer 3000 HS (Malvern

Instruments). The Smoluchowski formula

UE¼z330

h(1)

where his the the viscosity of the solvent, 3

is the dielectric

permittivity in vacuum, and 3is the relative dielectric permittivity

of the solvent, was used to relate the electrophoretic mobility,

, with the droplet’s z-potential.

2.3.3 Conductivity measurements. The conductivity of the

emulsions and the catanionic solutions was in all cases deter-

mined using a CDM 210 digital conductivity meter with Pt/Pt

electrodes for micro samples. The emulsions were kept in tubes

for at least 20 days which were turned upside down and the

different phases were extracted carefully, filled in Eppendorf cups

and measured with this electrode for microsamples.

2.3.4 Interfacial tension and contact angle experiments.

Interfacial tension was measured by the capillary pressure tech-

nique (DPA, Sinterface, Germany). In this method, the interfa-

cial tension is calculated from the pressure measured by

a pressure sensor. Additionally, the shape and the volume of the

droplet formed at the tip of a capillary can be monitored in time

by means of a camera.

First, we measured the interfacial

tension of the bare water–tetradecane interface (g

¼52.9 mN

1

) to check the purity of the liquids used. Then, we measured

the interfacial tensions of concentrated aqueous catanionic

solutions against tetradecane.

This journal is ªThe Royal Society of Chemistry 2011 Soft Matter, 2011, 7, 10694–10700 | 10695

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online

The contact angles of tetradecane droplets on catanionic

substrates were measured by a commercial apparatus (G10,

Kr€

uss, Germany).

3. Results

As mentioned above, the obtained emulsions remain re-dispersi-

ble irrespective of the ratio of the catanionic solution and the

formulation. The reason for this particular behavior is that the

anionic and cationic surfactants used as catanionics interact with

typically 10 kT per amphiphilic pair, resulting in a monomer

concentration range from 1 to 10 mM. Due to these ultra-low

‘‘cmc’’ (critical micelle concentration) or ‘‘cac’’ (critical aggregate

concentration) values for salt-free catanionics,

the investigated

emulsions can be diluted since only very little amounts of the

micro crystals present are solubilizied.

Fig. 1 depicts the observation of diluted samples by CLSM

microscopy which is straightforward in case of droplets in the

range of 1 to 5 mm.

The catanionic mixture used contains a tiny amount of

a fluorescent lipid-based dye (Oregon green). The mono-

dispersity of the size can be estimated from the low magnification

picture, while at higher magnification, it appears that each

fluorescent patch has typically five neighbors, meaning that

twelve patches typically cover an oil droplet. This can be either in

oder to minimize the bending energy or because of a topological

reason. On a molecular scale, a six fold symmetry in the surface

plane must be broken at least twelve times to form a closed

sphere. The typical solid angle covered by each apparent particle

from the center is therefore around 4p/12.

From the phase diagram and WAXS/WANS (wide angle

X-ray/neutron) experiments it is known that these aggregates are

crystalline. All emulsions are prepared at temperatures above the

melting temperature as shown in the complete phase prism.

Moreover, semi-flexible discs are the ideal shape for stabilizing

Pickering emulsions since they can accommodate any contact

angle due to the high curvature of their hydrophilic edges.

According to the literature, the thickness of one catanionic

bilayer is 4 nm while SANS experiments using contrast variation

have shown that less than ten discs are stacked in one ‘‘particle’’

stabilizing the emulsions.

In order to relate both measurements

more quantitative information about the amount of material

present at the liquid/liquid interface is needed.

Direct application of the Gibbs equation is not possible due to

the low solubility. However, surface tension measurements at the

air/water interface showed a strong increase in surface pressure

P¼g

g¼45–46 mN m

1

¼72 mN m

1

Here, the

water/oil interfacial tension as a function of time is deduced from

the capillary pressure of a water droplet in oil for two different

compositions (Fig. 2). The capillary pressures which have been

established slowly, are reduced by a factor of ten in the presence

of catanionic micro crystals. As can be seen in Fig. 4, the zeta-

potential of catanionic nanodiscs decreases from around +47 mV

for r¼0.4 with increasing molar ratio of anionic surfactant, r,to

25 mV for r¼0.7. Thus, catanionic nanodiscs containing an

excess of the cationic component should be more attracted to the

tetradecane/water interface since the water/oil interface is nega-

tively charged.

26,27

However, the results of the interfacial tension

measurements (Fig. 2) indicate the opposite. It is assumed that

a part of the myristic acid which is not needed after the rear-

rangement of the nanodiscs at the water/oil interface, dissolves in

tetradecane since myristic acid easily dissolves in the oil phase

but hardly in the water phase. This can lower the interfacial

tension beside the adsorption of the catanionic nano discs.

Therefore, the interfacial tension of r¼0.6 appears lower than

that of r¼0.4. In the present case, the surface pressure P¼g

g¼49–53 mN m

1

(see gin Fig. 2) is very high and approaches

the value of the bare interfacial tension (g

¼52.9 mN m

1

Fig. 1 Tetradecane emulsion droplets stabilized by stacks of nanodiscs as seen by confocal microscopy with three different magnifications. Discs are

labeled with Oregon green, a lipid based fluorescent dye which mixes with the catanionic bilayer. Emulsions decorated with catanionic crystals with r¼

0.4 (a) and r¼0.6 (b).

10696 | Soft Matter, 2011, 7, 10694–10700 This journal is ªThe Royal Society of Chemistry 2011

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online

Whatever the molecular origin is, extreme reduction of water/

tetradecane interfacial tension leads to long term emulsions

stability, since the driving force towards coarsening is reduced.

Another way to obtain an average thickness of the discs is to

follow the conductivity of the emulsions. In salt-free catanionics,

conductivity is governed by ‘‘free’’ counter-ions since the

surfactants in their monomer form as well, as the micro crystals

as such do not contribute appreciably to the conductivity. As is

usual with surfactant aggregates, ions partition between a

physisorbed layer and ‘‘free’’ counter-ions in the double layer. In

the absence of ‘‘Hofmeister’’ effect increasing the amount of

bound cosmotropic ions, the free ions around micelles are typi-

cally 2/3 of the total ions present.

The ratio of expected conductivity from a known amount of

free counter-ions of the pure catanionic solutions versus

measured conductivity of the emulsion indicates the amount of

counter-ions adsorbed between stacked discs due to electro-

neutrality, and hence the average number of discs. The raw data

of the conductivity of emulsions and the corresponding cata-

nionic solution without tetradecane droplets are given as ESI.† It

is crucial to note that when two nanodiscs stack in a three-

dimensional crystal, all counter-ions of the two stacked discs

must be adsorbed. Thus stacking occurs best with the two

compositions where local order is close to the one of anti-ferro-

magnetic materials with hexagonal symmetry (i.e. 1/3 and 2/3).

As can be seen in Fig. 3, the relative conductivity significantly

increases between r¼0.4 and r¼0.6. The observed increase in

relative conductivity corresponds to 60% (r¼0.4) and 90% (r¼

0.6) of trapped ions in the catanionic aggregates around the oil

droplet. Consequently, for r¼0.6 typically ten discs are stacked

which then form the nanoparticle firmly adsorbed at the surface

of the oil droplets.

The corresponding thickness of the particle stabilizing the

Pickering droplets is 40 nm, while the lateral expansion is of the

order of 1 mm. The ratio between the three axis of the solid

particles is therefore 1-25-25. Any contact angle between 45 and

135in the three tension line can be accommodated without

attraction between particles or increase in water/oil contact area.

Moreover, the stacking of discs does not change the zeta

potential. As shown in Fig. 4, the zeta potentials measured as

a function of molar ratio rdemonstrates that the zeta potential is

controlled by the composition of individual catanionic crystals.

With increasing molar ratio of anionic component r, the zeta-

potential decreases from +47 mV to 25 mV. In the present case,

the anionic component was myrsitic acid, a weak acid. Myristic

acid has a pK

5 and tends to form dimers. Therefore, one can

assume that in the catanionic samples considered, the myristic

acid was not fully dissociated. For this reason, the amount of

negative charges caused by the addition of anionic component

can by no means be proportional to the amount of myristic acid

added and charge reversal of the zeta potential occurs at r[0.5.

After an exhaustive characterization of the stabilizing cata-

nionic particles and long term stability experiments,

we can

now turn to elucidating the mechanism of governing the stability

of emulsions stabilized by such catanionic crystals.

As a first step, the energy of adsorption, in kT per particle, is

derived from interfacial tension data and contact angle

measurements.

Fig. 2 Interfacial tensions as a function of drop aging for r¼0.4 and 0.6.

Central insets show an aqueous droplet in tetradecane during the capil-

lary pressure experiment. Right insets show the contact angles of tetra-

decane (T) droplets on catanionic substrates. Top insets refer to r¼0.4

and the lower to r¼0.6.

Fig. 4 Zeta- potential zas a function of the ratio of anionic surfactant r

of diluted catanionic solutions (filled circles) and the corresponding

emulsions (open squares). The catanionic solutions and emulsion were

diluted by a factor 1 : 1000.

Fig. 3 Relative conductivities of emulsions as a function of the ratio of

anionic surfactant in the catanionic solutions (c¼1 wt%) for three

different values of oil/catanionic solution ratio. The relative conductivi-

ties were calculated by the conductivity of the emulsion over the

conductivity of the corresponding catanionic solution.

This journal is ªThe Royal Society of Chemistry 2011 Soft Matter, 2011, 7, 10694–10700 | 10697

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online

Equilibrium interfacial tensions g

0.4, TW

¼01mNm

1

and

0.6, TW

¼41mNm

1

were measured for r¼0.4 and r¼0.6 by

a capillary pressure technique at a relatively high catanionic

concentration c¼1 wt% (aqueous droplets in tetradecane are

shown in Fig. 2). Moreover, the macroscopic contact angles of

tetradecane drops on r¼0.4 and r¼0.6 substrates are Q

0.4, AT

38.0.3 and Q

0.6, AT

¼32.00.1, respectively (see insets in

Fig.2). Hence, the relations between surface energies for cata-

nionic layers, air (A) and tetradecane (T), can be written as:

0.4, A

¼s

0.4, T

cos(Q

0.4, AT

) (2)

0.6, A

¼s

0.6, T

cos(Q

0.6, AT

) (3)

At the air/water interface, the surface tensions of catanionic

mixtures show plateau values in a relatively large range of

concentrations pointing to a compact catanionic layer at the

surface.

Thus, we estimated the surface energies s

for cata-

nionics from the equilibrium surface tensions g

by applying

Connor’s equation:

30–32

si¼gi

1þbð1FÞ

1að1FÞ(4)

where Fis the molar fraction and aand bare two empirical

constants close to 1.

0.4,A

¼g

(F/1) ¼25 1mNm

1

(5)

0.6,A

¼g

(F/1) ¼26 1mNm

1

(6)

From eqn (2)–(6), one obtains s

0.4, T

¼41mNm

1

and

0.6, T

¼3.5 1mNm

1

. In order to estimate the energy of

adsorption one just needs to calculate the surface tension

between catanioinics and water. Following the procedure

described by Binks and Clint,

one obtains s

0.4, W

¼26.8 1

mN m

1

and s

0.6, W

¼27.9 1mNm

1

(see ESI).†

The energy of adsorption per particle as a function of the disc-

particle location (z-direction) is plotted in Fig. 5. Wetting ener-

gies are calculated following Pieranski’s model and by supposing

that the particle has an oblate shape of 1 mm in diameter and

a thickness of 40 nm so that the oblate exposes at maximum its

surface to the interface.

Hence, a particle composed by a stack

of nanodiscs adsorbs on an oil/water interface with an energy in

the order of 10

kT/particle.

As a second step, the contact energy barrier can be estimated

from the DLVO theory. The result is much larger than kT. Using

the Schmoluchovki theory, the lifetime versus coalescence or

creaming would be infinite. This means that in catanionic

stabilized emulsions, lifetime is limited by Ostwald ripening.

Finally, the resistance of the considered emulsion against

external stimuli, in this case against salt, is tested. Therefore, the

stability of two types of emulsions, namely, r¼0.4 and r¼0.6

with 10

2

M NaCl, has been followed for more than nine months.

The practical stability of emulsions has been determined by

monitoring droplet size distribution by SPLS

and concentra-

tion during storage as described by Weiss.

Fig. 6 shows that in case of positively charged stacks of

nanodiscs, the stability against coalescence and creaming after

one month in the salt-free catanionic system is limited to certain

oil/catanionic solution ratios. Namely, emulsions with a fraction

of 25 to 55% of catanionic solution remain completely stable.

Indeed, those emulsions have been stable for more than 90

days.

Adding 10

2

M of NaCl induces a widening of the

stability range. In this case, emulsions containing a fraction of

23% to almost 100% of catanionic solution are remarkably

stable. The authors are not aware of any other example of such

extremely stable oil-rich emulsions in the absence of strongly

adsorbed ionomers where the stability is enhanced by salt. The

reason for the observed increase in stability due to salt addition is

that stacking of nanodiscs is easier with salt. Moreover, the

average thickness of stacks increases when salt is added. The

energy barrier towards coalescence is still high, while the energy

of adsorption is still [kT. The observed salt-resilience is

a peculiarity of an emulsion stabilized by catanionic crystals,

since all other known cases of oil-in-water emulsions present

a stability either insensitive or decreasing when salt is added. In

contrast to this, catanionic based Pickering emulsions stabilized

by charged nanodiscs are much more stable in physiological

conditions than in pure water.

By putting all presented results together, a clear picture of the

stabilization mechanism associated with the structure of the

catanionic crystals emerges (see Fig. 7). The shape ratio 1-25-25

of the solid catanionic crystals stabilizing the presented Pickering

emulsions allows contact angle conditions which can be satisfied

at a negligible cost of free energy.

4. Discussion

The mechanism of stacking between nanodics is easily found in

a displacement of half a lattice vector in local hexagonal

arrangements.

It is more surprising that stacks of discs do not

facet appreciably the tetradecane droplets, although bending

moduli are of the order of 1 GPa.

In principle, the stress g/r

induced by droplets of r¼1mm and surface pressure P¼4mN

1

is 1000 Pa. It is likely that defects in the crystal match the

surface curvature radius during growth.

With regard to the very low water/oil surface tension, it is

important to note that only the presence of patches of crystalline

catanionic nanodiscs at the oil water interface should not

Fig. 5 Energy of adsorption measured for an oblate particle (1 mm

diameter and 40 nm thick) at the water/tetradecane interface. E(z)¼s

(z)+s

(z)g

(z), where r¼0.4 or 0.6; A

(z), A

(z) and

(z) are the areas of the oblate in water, tetradecane and the area

removed to the bare interface, respectively. A contact angle close to 90

(z¼0) and high adsorption energy E[kT are found for r¼0.4 and

r¼0.6.

10698 | Soft Matter, 2011, 7, 10694–10700 This journal is ªThe Royal Society of Chemistry 2011

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online

produce such low surface tensions at equilibrium, i.e. after a few

minutes as reported in Fig. 2. The origin of the observed trend of

water–oil surface tension could be a consequence of the following

points:

1. As shown in Fig. 7, the water/oil interface is covered by

a monolayer between the stacks of nanodiscs. According to the

segregation model explaining the shapes of such crystals,

this

monolayer should be enriched by the excess surfactant (nega-

tively charged carboxylic acid on one side (r> 0.5), and cationic

alkyl trimethylammonium surfactant on the other side (r< 0.5)).

The surface pressure is higher on the positively charged side due

to a higher surface charge.

2. Moreover, even in the case of faces of discs near equi-

molarity, the surface pressure is high and the surface tension is

low. One surface phase and two surface phase situations can be

distinguished using fluorescence micrographs since they corre-

spond to plateau leveling off in surface pressure isotherms.

37,38

Due to the fact that the linear solvent tetradecane matches the

chain length of the last monolayer, the oil can ‘‘penetrate’’ into

the monolayer. This situation has been encountered and

thoroughly described for lipids at the oil/water interface by

ellipsometry and reflectivity in case of lipids where a bulky

phosphtidylcholine head-group is involved.

According to

Thoma and coworkers, even in the liquid condensed phase, linear

matched oil penetration significantly increases the surface pres-

sure. These mechanisms could probably be also the origin of the

low surface tension observed for emulsions leading to their long

term stability. Moreover, these mechanisms of stabilization are

not sensitive to the presence of salt, since molecules are closely

packed and the Debye screening length remains larger than the

distance between adjacent amphiphiles in the plane of discs, even

in the presence of 10

2

M of salt.

The zeta-potentials measured indicate that particles have

a monolayer on the oil side. This assumption is strengthened by

the value frequently measured as triple layers as seen by SANS

for the stack.

From this one can conclude that the stack of

nanodiscs must be covered by a monolayer on the oil side of the

stacks, as shown in Fig. 7. If we consider the stack of nanodiscs

covering tetradecane droplets as well-defined colloids, those

catanionic crystals might be designated as self-organized Janus-

type particles. The win in surface energy is superior to the

entropy of mixing as soon as typical dimensions of charged

bilayers are larger than 100 nm.

To our knowledge, this is the

first time that such a kind of ‘‘Janus’’-particles produced by

segregation of oppositely charged surfactants is used for creating

ultra-stable emulsions.

Another interesting feature of the catanionic disc used as

emulsifier is that the resulting emulsions are insensitive towards

the impact of salt. Resilience against the presence of salt is

a major problem in emulsion stability. For example, the impact

of salt on the stability of emulsion stabilized by globular proteins

has been extensively studied because of the large number of

important applications.

According to the authors, two regimes

appear in this type of micro-emulsions: On the one hand,

a surfactant-rich regime is observed where only a part of the

surfactant is required for covering all droplets and micelles

remain in solution. On the other hand, one can also determine

a surfactant-poor regime, where the equilibium size is not

dependent on the initial dispersion energy. All emulsions

described in this paper are in what one could call the interme-

diary regime: the vast majority of surfactant is essentially located

in the flat particles, but the equilibrium size depends more on

initial break-up during preparation in emulsifying mill than on

the initial concentration.

For micron-sized droplets with a surface potential in the order

of 20 mV as found in the present case, the electrostaic interaction

is screened by typically 10 to 100 mM of salt. Above an ion

Fig. 6 Stability against creaming (left hand ordinate) and coalescence (right hand ordinate) for emulsions stabilized by positively charged catanionics

prepared without (a) and with 10

2

M NaCl (b) as a function of the oil/catanionic solution ratio. The marked domains correspond to the oil content

admissible in formulation for stable emulsions showing neither creaming on the top nor rejection of water at the bottom after more than 90 days of shelf

storage.

Fig. 7 Mechanism of stabilization by a stack of catanionic nanodics.

The contact angle of a micro sized oil droplet with the water–oil triple line

can be accommodated without deformation. Electric stabilization occurs

via the outer bilayer, while the effective bending radius of the droplet

interface is a combination of nanodiscs rigidity and surface to volume

ratio.

This journal is ªThe Royal Society of Chemistry 2011 Soft Matter, 2011, 7, 10694–10700 | 10699

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online

concentration of 100 mM, ion specific effects come into play

leading to an enhanced adsorption of chaotropic ions.

Furthermore, ions which adsorb are hydrated and secondary

hydration forces come into play at this high salt concentration.

The presence of such secondary hydration forces indeed stabi-

lizes emulsion again.

In the case described here, the resilience of

the Pickering emulsion against screening of electrostatic forces is

unique to our knowledge, since the angle of contact as well as the

anisometry of the stacks are controlled by non-stoichiometry of

the components.

5. Conclusion and outlook

Within this work, a simple and robust method describing the

preparation of Pickering emulsions is shown and the corre-

sponding mechanism governing the stability of such emulsions

has been elucidated in detail. Although the method of stabilizing

emulsions by segregation of oppositely charged surfactants has

been known for more than 20 years in the field of pharmacy, few

papers concerning this topic have been published. This is

surprising since the technology was—better said is—still freely

accessible, despite of existing patents in the field of emulsions

stabilized by oppositely charged surfactants. Furthermore, the

resulting catanionic emulsions are extremely difficult to break

and their properties can be directly linked to spontaneous

‘‘unbreakable’’ emulsions obtained by diluting a reverse micelle

with water. The extensive investigation of the stabilizing mech-

anism showed that the used nanodiscs have the unique property

to break in presence of a hydrophobic interface so that they

adsorb at the offered interface via a monolayer. Due to this

property one can consider such catanionic nanodiscs as self-

induced ‘‘Janus’’ particles. In addition, the investigated cata-

nionic emulsions are extremely stable, since the contact angle at

the oil–water surface can be adjusted by a suitable choice of the

composition of the catanioncs (r-ratio). Due to the hydrophilic

edges of such nanodiscs, it is possible to get contact angles close

to 90which would be the ideal case. For these reasons, cata-

nionic nanodiscs (micro crystals) are a versatile tool to prepare

ultra-stable emulsions. This might be interesting for biotechno-

logical applications such as peptides or steroids adsorbed on

discs and hydrophobic anti-inflammatory drugs with slow

release, as well as for industrial applications as the emulsification

of hydrophobic perfumes without alcohol.

Acknowledgements

A.S. thanks Dr Aliyar Javadi for interfacial tension experiments.

N.S. thanks the European Master ‘‘Complex Condensed Mate-

rials and Soft Matter’’ (EMASCO-COSOM) and the DFG

(French-German Network ‘‘Thin films between two and three

dimensions’’).

References

1 S. U. Pickering, J. Chem. Soc., 1907, 91, 2001.

2 S. Melle, M. Lask and G. Fuller, Langmuir, 2005, 21, 2158.

3 S. Abend, N. Bonneke, U. Gutschner and G. Lagaly, Colloid Polym.

Sci., 1998, 276, 730.

4 A. Meister, M. Dubois, L. Belloni and T. Zemb, Langmuir, 2003, 19,

7259.

5 D. Carri

ere, L. Belloni, B. Dem

e, M. Dubois, C. Vautrin, A. Meister

and T. Zemb, Soft Matter, 2009, 5, 4983–4990.

6 T. Zemb, M. Dubois, B. Dem

e and T. Gulik-Krzywicki, Science,

1999, 283, 816–819.

7 T. Zemb and M. Dubois, Aust. J. Chem., 2003, 56, 971–979.

8 E. Maurer, L. Belloni, T. Zemb and D. Carriere, Langmuir, 2007, 23,

6554–6560.

9 A. Khan; E. Marques. Specialists Surfactant; Blackie Academic and

Professional, an imprint of Chapman & Hall, London, 1997.

10 M. Schwuger, Kolloid. Z. Z. Polym., 1971, 243, 129–135.

11 G. May-Alert. Ph.D. thesis, Fachbereich Pharmazie an der

Universit€

at Marburg, Germany, 1985.

12 G. May-Alert and P. List, Pharm. Unserer Zeit, 1986, 15, 1–7.

13 B. Joensson, P. Jokela, A. Khan, B. Lindman and A. Sadaghiani,

Langmuir, 1991, 7, 889–895.

14 E. Melzer Ph.D. thesis, Universit€

at Braunschweig, 2000.

15 A. Lind, J. Andersson, S. Karlsson, P. Agren, P. Bussian,

H. Amenitsch and M. Linden, Langmuir, 2002, 18, 1380–1385.

16 N. Schelero, H. Lichtenfeld, H. Zastrow, H. M€

ohwald, M. Dubois

and T. Zemb, Colloids Surf., A, 2009, 337, 146–153.

17 J.-W. Kim, D. Lee, H. C. Shum and D. A. Weitz, Adv. Mater., 2008,

20, 3239–3249.

18 L. Taisne, P. Walstra and B. Cabane, J. Colloid Interface Sci., 1996,

184, 378–390.

19 M. Dubois and T. Zemb, Curr. Opin. Colloid Interface Sci., 2000, 5,

27–37.

20 M. Dubois, B. Dem

e, T. Gulik-Krzywicki and T. Zemb, Nature, 2001,

411, 672–675.

21 C. Vautrin, T. Zemb, M. Schneider and M. Tanaka, J. Phys. Chem. B,

2004, 108, 7986–7991.

22 M. Dubois, B. Dem

e, T. Gulik-Krzywicki and T. Zemb, C.R. Acad.

Sci. Paris, 1998, S

erie IK, 567–575.

23 A. Javadi, J. Kraegel, P. Pandolfini, G. Loglio, V. Kovalchuk,

E. Aksenenko, F. Ravera, L. Liggieri and R. Miller, Colloids Surf.,

A, 2010, 365, 62.

24 Y. Michina, D. Carriere, C. Mariet, M. Moskura, P. Berthault,

L. Belloni and T. Zemb, Langmuir, 2009, 25, 698–706.

25 A. Stocco, D. Carriere, M. Cottat and D. Langevin, Langmuir, 2010,

26, 10663–10669.

26 K. Ciunel, M. Armelin, G. H. Findenegg and R. von Klitzing,

Langmuir, 2005, 21, 4790–4793.

27 J. K. Beattie, A. M. Djerdjev and G. G. Warr, Faraday Discuss., 2009,

141, 31–39.

28 B. Binks and J. H. Clint, Langmuir, 2002, 18, 1270–1273.

29 F. Talens-Alesson, J. Phys. Chem. B, 2009, 113, 9779–9785.

30 K. A. Connors and K. A. Wright, Anal. Chem., 1989, 61, 194.

31 P. Lavi and J. Marmur, J. Colloid Interface Sci., 2000, 230, 107.

32 A. Zdziennicka and B. Janczuk, J. Colloid Interface Sci., 2010, 349,

374.

33 P. Pieranski, Phys. Rev. Lett., 1980, 45, 569–572.

34 J. Weiss, Curr. Prot. Food Analytical Chem., 2002, D3.4.1–D3.4.17.

35 N. Delorme, M. Dubois, S. Garnier, A. Laschewsky, R. Weinkamer,

T. Zemb and A. Fery, J. Phys. Chem. B, 2006, 110, 1752–1758.

36 M. Dubois, V. Lizunov, A. Meister, T. Gulik-Krzywicki,

J.-M. Verbavatz, E. Perez, J. Zimmerberg and T. Zemb, Proc. Natl.

Acad. Sci. U. S. A., 2004, 101, 15082–15087.

37 M. Thoma and H. M€

ohwald, J. Colloid Interface Sci., 1994, 92, 340–

349.

38 M. Thoma, T. Pfohl and H. M€

ohwald, Langmuir, 1995, 11, 2881–

2888.

39 M. Thoma, M. Schwendler, H. Baltes, C. A. Helm, T. Pfohl and

H. R. M€

ohwald, Langmuir, 1996, 12, 1722–1728.

40 M. Dubois, L. Belloni, T. Zemb, B. Dem

e and T. Gulik-Krzywicky,

Prog. Colloid Polym. Sci., 2000, 115, 238–242.

41 S. Tcholakova, N. Denkov, I. Ivanov and B. Campbell, Adv. Colloid

Interface Sci., 2006, 123–126, 259–293.

42 Specific Ion Effects, ed. W. Kunz, World Scientific Publishing, 2010.

43 V. Parsegian and T. Zemb, Curr. Op. Colloids Interfaces, 2011, DOI:

10.1016/j.cocis.2011.06.010.

10700 | Soft Matter, 2011, 7, 10694–10700 This journal is ªThe Royal Society of Chemistry 2011

Published on 04 October 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 12:48:28.

View Article Online