Ia

CD

h

•

•

•

•

•

a

ARRAA

KDDEOCD

h0

Colloids and Surfaces A: Physicochem. Eng. Aspects 460 (2014) 176–183

Contents lists available at ScienceDirect

Colloids and Surfaces A: Physicochemical andEngineering Aspects

j ourna l h om epa ge: www.elsev ier .com/ locate /co lsur fa

nterfacial activity of amino acid-based glycerol ether surfactantsnd their performance in stabilizing O/W cosmetic emulsions

.D. Ampatzidis, E.-M.A. Varka, T.D. Karapantsios ∗

epartment of Chemistry, Aristotle University of Thessaloniki, University Box 166, 541 24 Thessaloniki, Greece

i g h l i g h t s

Emulsions containing PhGE12 showbetter stability.Increase of surfactant’s concentrationlead to increased stability.PhGE12’s emulsions have smallerdroplets.Interfacial tension has approached itsequilibrium value after 1800 s.Interfacial rheology seems not toaffect emulsion’s stability.

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 10 December 2013eceived in revised form 11 February 2014ccepted 11 February 2014vailable online 22 February 2014

eywords:ynamic interfacial tensionilatational viscoelasticitycofriendly surfactants/W emulsionsreaming Indexroplet size distribution

a b s t r a c t

Biocompatible and biodegradable molecules of Phenyl- and Tyrosine-glycerol ether surfactants are goodcandidates for eco-friendly cosmetic emulsions. Both molecules have a hydrophobic alkyl chain with 12carbon atoms, but different amino acids in the hydrophilic head: phenylalanine (PhGE12) and tyrosine(TyrGE12). This study reports on the activity of liquid/liquid interfaces with the glycerol ether surfac-tants being dissolved in the organic phase (olive oil) and the aqueous phase containing the non-ionicpoly(oxyethelene) (20) sorbitan monooleate (trade name Tween 80). Measurements refer to a) dynamicsurface tension by drop profile tensiometry and drop volume tensiometry and b) interfacial dilatationalviscoelasticity by oscillating drop profile analysis. It is shown that PhGE12 attains much lower interfacialtension values and exhibits faster interfacial adsorption than TyrGE12 whereas interfacial dilatationalviscoelasticity values are comparable between the two surfactants. Oil-in-water emulsions are preparedand their stability is assessed by monitoring the time evolution of creaming index. To interpret observa-tions, measurements of the emulsions’ initial droplet size distributions and apparent bulk viscosity areemployed. It is found that at similar surfactant concentrations, PhGE12 emulsions exhibit better stabilitythan TyrGE12 emulsions. This might be attributed to the lower interfacial tension in PhGE12 emulsions

that yield droplet size distributions richer in intermediate (∼50 �m) and small size (∼0.5 �m) dropletsand to their higher apparent viscosities. On the other hand, increasing the concentration of either surfac-tant leads to a substantial increase of emulsions stability which is in line with the lower interfacial tensionand higher apparent viscosity measurements but does not agree with the insensitive versus surfactantconcentration droplet size distributions and interfacial dilatational viscoelasticity.

© 2014 Elsevier B.V. All rights reserved.

∗ Corresponding author. Tel.: +0030 2310 99 7772; fax: +0030 2310 99 7772.E-mail address: karapant@chem.auth.gr (T.D. Karapantsios).

ttp://dx.doi.org/10.1016/j.colsurfa.2014.02.033927-7757/© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Emulsions are omnipresent. They dominate our daily liferanging from cosmetic products, foods, cleaning and pharmaceu-tical products to paint and oil industries [1]. Knowledge of the

A: Phy

poc

top[filhotctr

tspetsdatamm

sdpofiNopmcv

s[wfvahede

ismsteatccit

C.D. Ampatzidis et al. / Colloids and Surfaces

hysicochemical properties of such formulations is essential toptimize manufacturing conditions, provide cosmetic elegance andontrol the delivery of cosmetic agent to the skin.

Emulsions are heterogeneous mixtures consisting of at leastwo immiscible liquids [2]. Emulsions are made stable by the usef emulsifiers, i.e., surface active agents (surfactants), amphiphilicolymers or proteins, which sufficiently reduce interfacial tension3]. Such molecules stabilize the oil droplets at the time of manu-acture by the formation of an interfacial film, and also control thenterfacial rheological properties of the formulation between wideimits. Therefore, many researchers in the physical-chemistry fieldave tried to assess the efficiency of emulsification and the stabilityf emulsions based on measurements of static/dynamic interfacialension and of interfacial viscoelasticity [4–9]. However, in mostases systematic measurements over a broad range of interface life-imes (using independent techniques) is missing and so results arearely conclusive.

At the other end of the line, one can find a vast number ofechnology oriented publications where emulsification and emul-ion stability are assessed by macroscopic quantities such as theropagation of the phase separation front, droplet size distribution,mulsion apparent viscosity etc. These quantities are often enougho appraise the production of emulsions and the subsequent phaseeparation due to the gradual flocculation and aggregation of theispersed phase droplets leading ultimately to a cracked or sep-rated emulsion. However, they cannot explain the mechanismshat dictate the specific properties and shelf-life of emulsions. It ispparent that a combination of physicochemical and macroscopiceasurements is advantageous to shed light on the underlyingechanisms.It has long been recognized that mixed emulsifying agents (co-

urfactants) often yield more stable emulsions than single oneso, and, in most cases the high stability is ascribed to the com-lex formation at the interface, which results in an interfacial filmf great strength which prevents coalescence of droplets. Usually,lm strength is quantified by interfacial elasticity measurements.evertheless, when emulsifying agents are present in both phasesf the emulsion then the situation is more complicated and theossibility cannot be excluded that there is interaction betweenolecules across the interfaces affecting also the bulk rheologi-

al properties of the system e.g. increase the emulsion apparentiscosity.

Two amino-acid glycerol ether surfactants have been synthe-ised and chemically characterised in detail in previous work10,11]. These are Phenylalanine (PhGE12) and Tyrosine (TyrGE12)ith 12 carbon atoms in their hydrophobic chain. Air/water inter-

acial activity (dynamic surface tension and interfacial dilatationaliscoelasticity) of both surfactants has been dealt with in two recentrticles [12,13]. These molecules dissolve readily in water only atigh pH value and this limits their use only to specific applications.g., mercerization of fabrics. On the other hand, both moleculesissolve easily in organic solvents revealing strong potential formulsions applications.

This work first examines the interfacial activity at the oil/waternterface of Phenylalanine (PhGE12) and Tyrosine (TyrGE12) dis-olved in olive oil in the presence of poly(oxyethelene) (20) sorbitanonooleate (Tween 80) in the aqueous phase. This includes mea-

urements of dynamic interfacial tension by two independentechniques (drop profile tensiometry and drop volume tensiom-try), which in combination cover a wide range of interface age,nd of interfacial dilatational viscoelasticity. Then, emulsions par-icularly rich in oil content—such as those in skin and health

are applications—are produced in a broad range of surfactantsoncentrations and their stability versus time is registered by mon-toring the displacement of the creaming front (creaming index) inime. Measurements of initial droplet size distributions and initial

sicochem. Eng. Aspects 460 (2014) 176–183 177

emulsion apparent viscosity have been combined with measure-ments of oil/water interfacial properties to interpret observations.An effort is made to provide insight on what might explain thebehavior of such oil rich surfactants.

2. Materials and methods

2.1. Preparation of solutions for interfacial measurements

Organic solutions of amino-acid glycerol ether surfac-tants (PhGE12 and TyrGE12) are prepared at concentrations of1 × 10−6–5 × 10−4 M (mol/L of oil phase). Due to an extra hydroxylanion (phenolic group instead of a simple aromatic ring), TyrGE12is more hydrophilic than PhGE12. Olive oil (Altis, Elanthy S.A., 2012)is the organic solvent. Olive oil is a high valued natural producttypically used in foods and cosmetics. Dilution of surfactants inoil is carried out at low heating and mild stirring. Olive oil ispurchased from the market and used as received, viz. purified fromthe company that bottled it.

Poly(oxyethelene) (20) sorbitan monooleate (Tween 80) (syn-thesis grade, Merck) is used at a constant concentration of 40% w/vin the aqueous phase for all measurements (C = 3.05 × 10−1 mol/Lof aqueous phase). Its trade name, Tween 80, is used henceforth forconvenience. It derives from polyoxylated sorbitol and oleic acid. Itis a nonionic surfactant with HLB 15, which means that it is solublein water and not in oil. Dilution of Tween 80 in Millipore water isaccomplished at mild heating and stirring. The pH value of aqueousphase is 7.3 (at 24.2 ◦C).

2.2. Interfacial (liquid/liquid) measurements

2.2.1. Dynamic interfacial measurementsMeasurements are conducted by two independent techniques

spanning different ranges of interfacial lifetimes. This is important,first, to show that measurements are complementary and so arenot technique dependent and, second, to extend information overa broad range interfacial lifetimes.

2.2.1.1. Drop volume tensiometry. Dynamic interfacial tensionmeasurements at short interface lifetimes (less than 3 s) are per-formed with a drop volume tensiometer (DVT; TVT 2, LAUDA). DVTprovides estimation of the interfacial tension of a freshly formedinterface via an aqueous drop created into the oil phase. The instru-ment is calibrated against Millipore water and clearness of glasssyringe and stainless steel capillary is checked with measurementwith Millipore water. All measurements are carried out with freshlymade solutions. The sample cell is filled with the organic solution ofamino acid glycerol ether surfactant. A graded glass syringe (2.5 ml)with a stainless steel capillary, which has a diameter of 1.385 mm, isfilled with Tween 80 aqueous solution. Then the syringe is adjustedto the test cell of the tensiometer. The temperature of both test celland syringe is kept constant at 25 ± 0.1 ◦C. Initial drop time is 0.4 sand the subsequent drops time is 0.42 s. The experimental scenarioconsists of 3 drops per cycle for 10 consecutive cycles.

2.2.1.2. Drop profile analysis. Dynamic interfacial tension measure-ments at long adsorption times, (several seconds to hours), areemployed via drop profile tensiometer (DPT; PAT-1S SINTERFACE).The instrument is placed on an anti-vibration table, (HalcyonicsMicro 40/60/80). The tensiometer allows long-time experimentskeeping either the volume or the surface area of a drop constant.Calibration took place against Millipore water. Due to the long times

involved, surface lifetime is explicitly measured. The temperatureof the solutions is kept constant at T = 25 ± 0.1 ◦C. In this study apendant drop (V = 4 mm3) of Tween 80 aqueous solution is used.The drop is created by a computer driven dosing system at the tip

1 A: Physicochem. Eng. Aspects 460 (2014) 176–183

oce

2

omAmttfiatsfdt03

2

pewtvtttsitoc

tai(tsasa4t

stpdsmTc

2

i

78 C.D. Ampatzidis et al. / Colloids and Surfaces

f a stainless steel capillary (d = 2 mm), immersed in a sealed glassuvette, which contains the organic solution of amino acid glycerolther surfactant.

.3 Interfacial dilatational viscoelasticity

Complex interfacial dilatational modulus is measured by anscillating drop profile analysis (ODPA) technique using a dedicatedode of the drop profile tensiometer (ODPA; PAT-1S Sinterface).nalysis of data is carried out on the basis of a simple rheologicalodel using just one relaxation frequency. This approach assumes

hat the relaxation frequency and the intrinsic elastic modulus ofhe adsorption layers are independent from the applied oscillationrequency. The apparatus and operational procedures are describedn detail elsewhere [14]. As with DPT, a pendant drop of Tween 80queous solution is created at the tip of the capillary immersed inhe organic solution of amino acid glycerol ether surfactant. Theoftware allows controlling the surface volume of the drop as aunction of time and thus imposing harmonic oscillations of therop volume. The volume oscillation amplitude is 5% of the ini-ial drop volume (V = 4 mm3). The oscillation frequencies are 0.001,.005, 0.01, 0.02, 0.05, 0.1 and 0.2 Hz. Measurements are taken after000 s of adsorption to allow quasi-equilibrium conditions.

.4 Emulsification process

Oil in water emulsions are prepared using olive oil as the organichase (45% w/w of the emulsion) in which the amino-acid glyc-rol ether surfactants are dissolved. Tween 80 solution in Milliporeater is used as the aqueous phase (55% w/w of the emulsion). Ini-

ially, proper amount of an amino acid glycerol ether surfactant (atarious concentrations, mol/L (M) of the oil phase) is dissolved tohe oil and Tween 80 (always 16% w/v of emulsion) is dissolved tohe aqueous phase in different beakers. The above proportion ofhe two phases is realistic for cosmetic applications e.g. skin emul-ions [15–18]. On the other hand, the concentration of Tween 80n the aqueous phase is selected lower than usual in such applica-ions in order to yield emulsions that destabilize in a reasonablebservation period (less than two days) and so allow easy and fastomparisons between the two surfactants.

Beakers are heated at 70 ◦C in a water bath, until both solu-ions become clear. Emulsification is implemented by mixing thequeous phase (35 ml) with the oil phase (25 ml) using a Rushtonmpeller (d = 4.6 cm), rotating at the central axis of a glass vesseldint= 9.5 cm). The impeller is placed 1.0 cm above the bottom ofhe vessel. Four buffles are placed at 90o intervals around the ves-el, preventing vortexing and air suction. Emulsification starts bydding slowly the hot aqueous phase into the hot oil. Agitationpeed is initially set at 150 rpm. When water addition is completed,gitation speed is increased to 350 rpm for 1.5 min and then to500 rpm for 30 min, time sufficient for the emulsion to cool downo ambient temperature(25 ± 1 ◦C).

Conductivity measurements confirm that in the produced emul-ions the aqueous phase is the continuous phase and olive oil ishe dispersed phase. At the end of the emulsification period, theroduced emulsion is transferred into two glass beakers (50 mmiameter, 70 mm height) in order to perform volumetric mea-urements. Additionally, a small quantity of emulsion is taken foreasurements of apparent viscosity and droplet size distribution.

he pH value is approximately 7 (23.6 ◦C ± 0.1) for all the measuredoncentrations of surfactants.

.5 Creaming index from volumetric measurements

Direct visual observations are employed to determine thenstantaneous heights of the emulsion and of the aqueous phase

Fig. 1. Schematic representation of creaming index estimation.

inside the beakers during phase separation and from these valuesestimate the creaming index according to formula (1) (Fig. 1)

CI = Haq

Htot× 100 (1)

Creaming index represents the global (i.e. over the whole liquidvolume) separated water fraction. It is the most customary measureof emulsion destabilization reflecting phase separation. Neverthe-less, there are two disadvantages associated with the creamingindex: (i) the interface separating the cream phase (droplet-rich)from the serum phase (droplet-depleted) is never sharp except inthe occasion—unlike for real systems of monodispersed dropletsize distributions and (ii) it is only possible to obtain informationabout the location of the interface between the cream phase andserum phase and not about the full vertical concentration profile ofdroplets.

2.6 Droplet size distribution (DSD)

DSD is obtained with Mastersizer Hydro 2000 MU system(Malvern Instruments Ltd) based on laser diffraction. The detec-tion range of the instrument varies from 0.020 to 2000 �m. Theapparatus is appropriated to perform measurements at low con-centrations of particles/droplets in a transparent environment butalso in slightly dark media. Due to the opaqueness of the emulsionsamples a diluent is employed. The main purpose of dilution is toavoid multiple scattering by getting droplets far away from oneanother. A relative refractive index of 1.4582 (oleic acid), a particleabsorbance of 0.1 and a continuous phase refractive index of 1.33(distilled water) are the parameters chosen in order to calculate thesize distribution of droplets [19].

2.7 Emulsion apparent viscosity

Emulsions’ apparent (bulk) viscosity is measured by a rheome-ter (Physica MCR301, Anton Paar) fitted with a cone-plate system(rcone = 2.5 cm, �cone = 1o). A small amount of emulsion is put onthe plate and then the cone comes down to a specified distance(d = 0.049 mm). Measurements are performed in a shear rate rangeof 1–10000 s−1 for three concentrations of both amino acid glycerolether surfactants, namely 5 × 10−6, 5 × 10−5 and 5 × 10−4 M.

3. Results and discussion

It is useful to report first data regarding the equilibrium (static)interfacial tension of the system obtained by the Wilhelmy plate

◦

technique at 25 C [20]. The air/liquid surface tension of the aque-ous phase (Millipore water with Tween 80) is 35 ± 1 mN/m. Thecritical micelle concentrations (CMC) for PhGE12 and TyrGE12 at theoil/water system are 2.66 × 10−5 M and 3.05 × 10−5 M, respectively.

C.D. Ampatzidis et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 460 (2014) 176–183 179

Fws

3

3

tftmlitttmb

3

oiMsetactspohcttb

Fig. 3. Dynamic interfacial tension of organic solutions of (a) PhGE and (b) TyrGE ,

ig. 2. Dynamic interfacial tension of organic solutions of (a) PhGE12 and (b) TyrGE12,ith Tween 80 in the aqueous phase as measured by Drop Volume Tensiometry for

hort interfacial lifetimes at 25 ◦C.

.1. Dynamic interfacial tension

.1.1. Drop Volume TensiometryFigure 2 displays the variation of dynamic interfacial tension of

he two amino acid glycerol ether surfactants with respect to inter-ace lifetime as measured by Drop Volume Tensiometry (DVT). Withhe exception of the two higher concentrations, where measure-

ents are comparable between the two ethers, PhGE12 presentsower interfacial tension values than TyrGE12. In addition, dropnterfacial lifetime (time up to drop detachment) is smaller for solu-ions of PhGE12, which means that this molecule adsorbs faster athe tip of the capillary. Both observations indicate that PhGE12 solu-ions present a higher surface activity than solutions of TyrGE12 for

edium and low concentrations but for C ≥ 2 × 10−4 M the activityecomes comparable between the two molecules.

.1.2 Drop profile analysisFigure 3 presents the variation of dynamic interfacial tension

f the two amino acid glycerol ether surfactants with respect tonterface lifetime as measured by Drop Profile Tensiometry (DPT).

easurements indicate that interfacial tension has dropped con-iderably after 1800 s (30 min), time used for the preparation ofmulsions. When increasing surfactant concentration, interfacialension drops more quickly to very low values. Nevertheless, event very long interface lifetime (3000 s) the slopes of the curves indi-ate that �d is not equal to equilibrium values. This is probably dueo steric repulsion between adsorbed surfactant molecules from theame side or both sides of interface. Tween 80 has three hydrophilicolyoxyethylene groups with hydroxyl moieties and a hydrophobicne (oleic acid). Therefore, it can be settled at the interface with theydrophobic polyethylene group oriented toward the oil phase or

an be folded over the interface. Surfactant molecule TyrGE12, withhe phenolic ring on the amino acid’s side chain, takes a position athe interface with its hydrophilic head towards the aqueous phase,ecause the hydroxyl group of the phenolic ring prefers the contact

12 12

with Tween 80 in the aqueous phase, with Drop Profile Tensiometry technique forlong interfacial lifetimes at 25 ◦C.

with water [21]. Since the hydrophilic head of TyrGE12 is considered“bulky” due to the phenolic ring, steric interaction between Tween80 and TyrGE12 may be strong. On the contrary, surfactant moleculePhGE12, which has a simple aromatic ring, tends to be folded downtowards the oil phase [21]. Although the folded molecule occupiesappreciable area inside the oil phase, it is likely that it does notinterfere sterically with the hydrophobic chain of Tween 80.

Olive oil’s main component is oleic acid, which has a carboxylmoiety on its head group. Although it has a very low solubility inwater (6 × 10−9 mol/cm3) [22], molecules at the interface are ori-ented with their carboxyl groups towards the aqueous phase. Sincethe oleic acid has a carbon chain with 18 carbon atoms [23,24],molecules are placed close to each other, occupying a fairly largearea of the interface [24].

In the presence of micelles (C > CMC), fast micelle dissociationand/or disintegration is an additional driving force which acceler-ates adsorption [25,26].

Once more, with the exception of the two higher concentrations,where measurements are comparable between the two ethers,PhGE12 presents lower interfacial tension values than TyrGE12,therefore, expressing a higher surface activity.

Measurements from DVT and DPT although refer to differentinterface lifetimes both support that phenylalanine is more sur-face active than tyrosine, for concentrations smaller than the CMCwhereas for C ≥ 2 × 10−4 M the two molecules have comparableperformance.

3.2 Interfacial dilatational viscoelasticity

Figures 4 and 5 illustrate the dependence of the storagemodulus (Er real part, dilatational elasticity) and loss modulus(Ei imaginary part, dilatational viscosity), respectively, of thetwo surfactants’ adsorbed layers on surfactant concentration andoscillation frequency. Interfacial dilatational viscoelasticity values

for the two surfactants do not show a big difference to each otherand so one would not expect differences in emulsion’s stabilitybetween the two surfactants due to dilatational viscoelasticity.Both the storage modulus and loss modulus show an increasing

180 C.D. Ampatzidis et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 460 (2014) 176–183

a)

0

10

20

30

40

50

1.E-06 1.E-05 1.E-04 1.E-0 3

C, M

r, m

N/m

0.001 Hz0.005 Hz0.010 Hz0.020 Hz0.050 Hz0.100 Hz0.200 Hz

b)

0

10

20

30

40

50

1.E-06 1.E-0 5 1.E -04 1.E -03

C, M

r, m

N/m

0.001 Hz0.005 Hz0.010 Hz0.020 Hz0.050 Hz0.100 Hz0.200 Hz

FTw

tt

a

FTw

a)

0

10

20

30

40

50

60

0 500 1000 1500 2000 2500t, min

%C

.I

1E-6 M5E-6 M1E-5 M5E-5 M1E-4 M2E-4 M5E-4 M

b)

05

1015202530354045505560

0 500 1000 1500 2000 2500t, min

%C

.I

1E-6 M5E-6 M1E-5 M5E-5 M1E-4 M2E-4 M5E-4 M

ig. 4. Storage modulus (Er real part, dilatational elasticity) of (a) PhGE12 and (b)yrGE12 adsorbed layers versus surfactant concentration and oscillation frequencyith Tween 80 in the aqueous phase.

rend with frequency indicating that the relaxation frequency ofhe system is above the tested oscillation frequencies [27].

For all examined concentrations of PhGE12 and for frequenciesbove 0.05 Hz the storage modulus is higher than the loss modulus.

a)

0

5

10

15

20

25

1.E-0 6 1.E-0 5 1.E-04 1.E-03

C, M

i, m

N/m

0.001 Hz0.005 Hz0.010 Hz0.020 Hz0.050 Hz0.100 Hz0.200 Hz

b)

0

5

10

15

20

25

1.E-06 1.E-0 5 1.E-04 1. E-03

C, M

i, m

N/m

0.001 Hz0.005 Hz0.010 Hz0.020 Hz0.050 Hz0.100 Hz0.200 Hz

ig. 5. Loss modulus (Ei imaginary part, dilatational viscosity) of (a) PhGE12 and (b)yrGE12 adsorbed layers versus surfactant concentration and oscillation frequencyith Tween 80 in the aqueous phase.

Fig. 6. Creaming Index (%C.I) versus time for emulsions of (a) PhGE12 and (b) TyrGE12

with Tween 80 in the aqueous phase.

At frequencies below 0.05 Hz, the storage modulus is again higherthan the loss modulus but chiefly for the higher concentrations. Onthe contrary, for all the examined concentrations and frequencies ofTyrGE12—apart from scattered exceptions—the storage modulus ishigher than the loss modulus. The tendency at high frequencies thestorage modulus to be higher than the loss modulus manifest thedomination of the elastic character of the adsorption layer whichis a good sign for emulsification process. On the contrary, at lowfrequencies there is tendency the loss modulus to be comparableto the storage modulus and, furthermore, both moduli to attain lowvalues. These findings imply unfavorable conditions for emulsionstability (phase separation occurs at low shear rates).

Both surfactants exhibit a general decreasing trend in both thestorage modulus and loss modulus with increasing concentration.The trend diminishes as frequency decreases where both moduliappear more and more insensitive to concentration variations.Small peaks at specific concentrations must not be overvalued asthey may result from experimental peculiarities and, in addition, donot really affect the general trend. The above trend is opposite toobservations (see below) where emulsions get considerably morestable as surfactant concentration increases. Therefore, dilatationalviscoelasticity may not be the governing factor of emulsion stabil-ity.

3.3 Emulsion stability assessed by the creaming index

The interface separating the cream from the serum phase is notsharp. The cream phase presents a gradual denser opacity from theseparating interface to the top whereas the serum phase is every-where turbid, although less and less with time. Even so, it is stillpossible to identify the location of the boundary with the highestcontrast between the two phases and this is considered as the sepa-

rating interface to estimate the creaming index (Section 2.6) shownin Fig. 6.

All examined emulsions gradually destabilize and this isreflected in an increasing creaming index. Separation of phases

A: Physicochem. Eng. Aspects 460 (2014) 176–183 181

iccegtttcppvttcs(et

sPPcvs

ptwmbbbsch

3

aaadac

si8aaAvm

cscsmotd

a)

0

2

4

6

8

10

0.1 1 10 100 1000Droplet size ( m)

Volu

me

(%)

5E-6 M

5E-4 M

0

0.2

0.4

0.6

0.8

0.1 1 10

b)

0

2

4

6

8

10

0.1 1 10 100 1000Droplet size ( m)

Volu

me

(%)

5E-6 M5E-4 M

0

0.2

0.4

0.6

0.8

0.1 1 10

bility when increasing surfactant concentration since the dropletsize distributions vary only slightly with surfactant concentra-tion.

b)

05

1015202530354045

Num

ber (

%)

5E-6 M

5E-4 M

00.020.040.060.080.1

10 10 0

a)

05

1015202530354045

0.1 1 10 100Droplet size ( m)

Num

ber (

%)

5E-6 M5E-4 M

00.020.040.060.080.1

10 100

C.D. Ampatzidis et al. / Colloids and Surfaces

s fast for emulsions with low surfactant concentration, but moreoncentrated emulsions take longer time to destabilize. As the con-entration of PhGE12 and TyrGE12 in the organic phase increases,mulsion stability increases; yet, not in a linear fashion. Threeroups of emulsions are roughly identified depending on surfac-ant concentration: those with concentrations up to 1 × 10−5 M,hose with concentrations from 5 × 10−5 M to 2 × 10−4 M andhose with concentration 5 × 10−4 M. In the group of lower con-entrations, which correspond to the most unstable emulsions,hase separation proceeds rapidly. For tyrosine, the separationrocess begins right after decanting the emulsion into the obser-ation glass beakers and is almost complete (∼53% v/v is theotal proportion of the aqueous phase) in about six hours. Onhe contrary, in the group of the highest concentration, whichorresponds to the most stable emulsions, separation is verylow and is not complete at the end of the observation period2500 min). Within each group, there are no clear differences inmulsion stability between the different surfactant concentra-ions.

Emulsions containing PhGE12 are more stable than emul-ions containing TyrGE12, at equal concentrations. In other words,hGE12 is a more effective emulsion stabilizer than TyrGE12. InhGE12 emulsions, even for the three lowest concentrations thereaming index does not exceed ∼45% even at the end of the obser-ation period which indicates that a portion of the two phases istill stably dispersed.

In emulsions with so high volumetric proportion of the dis-ersed phase like the present ones (47% v/v) larger droplets ofhe dispersed phase are in contact to each other forming a net-ork inside the continuous phase. This prevents oil droplets fromoving freely one past another and so delays phase separation by

uoyancy (the role of emulsions’ apparent viscosity is examinedelow). Unless some local rearrangement occurs which allows arief movement of smaller droplets the main mechanisms for phaseeparation in the examined time frame are flocculation and coales-ence of adjacent droplets, which create larger entities/droplets ofigher buoyant force.

.4 Droplet size distribution (DSD)

Figures 7 and 8 show the droplet size distribution in % volumend % number, respectively, of emulsions containing the two aminocid glycerol ether surfactants. Measurements are performed rightfter the end of emulsification so they refer to the initial droplet sizeistribution of the emulsions. The values shown in Figs. 7 and 8re averages computed from four repetitions. The mean varianceoefficient (stdev/average) among repetitions is 0.1.

As can be seen, all emulsions prepared with the employed emul-ification process have comparable droplet size distributions. Thiss not surprising for systems rich in emulsifier (in our case Tween0) to adequately stabilize the droplets [28–30]. Generally, therere two major categories of droplets: оne with predominant sizeround 100 �m and another with predominant size around 1 �m.pparently, droplets with size around 100 �m occupy the largerolume of the droplets, but droplets with size around 1 �m areore numerous.Comparing the two surfactants together appears that emulsions

ontaining phenylalanine have slightly smaller droplet sizes in themall size category (∼0.5 �m). In addition, the PhGE12emulsionsontain more droplets of intermediate size (∼50 �m). The aboveupport the notion that emulsions with phenylalanine might be

ore stable than emulsions with tyrosine. However, it remains

pen if the above differences in droplet size distribution betweenhe two surfactants are alone adequate to explain the observedifferences in emulsions stability. Furthermore, droplet size

Fig. 7. Droplet size distribution by volume% for emulsions of (a) PhGE12 and (b)TyrGE12 in the presence of Tween80 in the aqueous phase.

distributions cannot really explain the increase in emulsions sta-

0.1 1 10 10 0Drop let size ( m)

Fig. 8. Droplet size distribution by number% for emulsions of (a) PhGE12 and (b)TyrGE12 in the presence of Tween80 in the aqueous phase.

182 C.D. Ampatzidis et al. / Colloids and Surfaces A: Phy

a)

0

100

200

300

400

500

600

1 10 10 0 100 0 1000 0

s, s-1

, mPa

*s

5E-6 M5E-5 M5E-4 M

b)

0

100

200

300

400

500

600

1 10 10 0 100 0 1000 0

s, s-1

, mPa

*s

5E-6 M5E-5 M5E-4 M

Fa

3

ew1doo

ioeSdo

ivpsts

pTv2c

tfvw

projects: FASES (Fundamental and Applied Studies of Emulsion Sta-

ig. 9. Apparent bulk viscosity (�) versus shear rate (�) for emulsions of (a) PhGE12

nd (b) TyrGE12 in the presence of Tween80 in the aqueous phase at T = 25 ◦C.

.5 Emulsions apparent viscosity

Figure 9 displays the variation of apparent (bulk) viscosity ofmulsions containing the amino acid glycerol ether surfactantsith respect to shear rate. Viscosity measurements last about

5 min, so no appreciable change in the droplet size distributionuring measurement is expected. Higher viscosity is an indicationf more stable emulsions, since viscosity acts to retard the motionf droplets.

The present emulsions show a shear-thinning behavior which isntense at low shear rates but fades out at high shear rates. More-ver, an increase on surfactant concentration yields an increase onmulsions’ viscosity. This becomes more so as shear rate decreases.ince emulsion’s destabilization takes place at low shear rates, theifferent apparent viscosity appears a possible explanation of thebserved differences in emulsions stability.

As pointed out by a number of authors [31–35], emulsion viscos-ty is a very complex function of the viscosity of the external phase,olume fraction of the dispersed phase, viscosity of the internalhase, droplet size distribution, nature of the interfacial film, emul-ifier concentration, the extent of flocculation, etc. From the above,hat droplet size dictates largely the rheology (bulk viscosity andtorage moduli) of emulsions [36].

The higher apparent viscosity of PhGE12 emulsions may beartly explained by the somewhat smaller droplets compared toyrGE12 emulsions. Indeed, there is evidence that the apparentiscosity of concentrated emulsions (dispersed phase larger than0% v/v) containing small droplets is higher than that of emulsionsontaining large droplets [37].

In recent publications we examined the interfacial activity ofhese same two glycerol-ether surfactants at the air/water inter-

ace. It was shown that TyrGE12 attains much lower surface tensionalues and exhibits faster interfacial adsorption than PhGE12hereas it takes higher viscoelasticity values for concentrations

sicochem. Eng. Aspects 460 (2014) 176–183

below the CMC [12,13,38]. This is completely different from thepresent findings regarding the water/oil interface. This is not sur-prising since it is known that not only the chemical structure of thesurfactant but also the nature of the interface dictate the type andconformation of the assembled structures. On this account, the OHgroup in TyrGE12 prefers contact with water whereas, in PhGE12,the phenyl ring, lacking the hydrophilic hydroxyl group, is foldeddown into the oil. It has been shown that this behavior, confers ioniccharacter to PhGE12 in contrast to the nonionic character of TyrGE12[39]. It is therefore possible that the ionic character of PhGE12 inthe organic phase combined with the non-ionic Tween 80 in theaqueous phase yield more stable liquid/liquid interfaces (i.e. emul-sions) than the combination of the non-ionic molecules TyrGE12and Tween 80. This is in line with the argument that at high waterfraction emulsions extensive hydrogen bonds can form leading todecrease in the molecular distances of the emulsion system as wellas to increase of resistance to flow [40].

On the other hand, since the droplet size distributions are com-parable among surfactant concentrations (Figs. 7 and 8), it is notobvious why emulsions’ viscosity changes so much. For such a largevolume content of the oil phase (∼47% v/v), large droplets are incontact with each other (packed) inside the aqueous phase. Frictionbetween droplets, and therefore emulsion viscosity, might increaseas surfactant concentrations increases depending on the mech-anism of droplet flocculation and colloidal interactions betweendroplets [25]. This might be reflected in an increasing interfacialshear viscoelasticity of the adsorbed layer at the liquid-liquid inter-face but such measurements are beyond the capacity of this work.

4 Conclusions

Drop volume tensiometry and drop profile analysis are used tomeasure dynamic interfacial tension for PhGE12 and TyrGE12 solu-tions in olive oil in the concentration range 1 × 10−6–5 × 10−4 M.Oscillating drop profile analysis is used to study the interfa-cial dilatational rheology of solutions in the frequency range0.001–0.2 Hz. Independent measurement of three experimentalparameters (creaming index values, emulsion apparent viscosityand droplet size distribution) are employed for the assessmentof the stability of oil-in-water emulsions produced with differ-ent surfactant compositions and in the presence of Tween 80 inthe aqueous phase. PhGE12 attains lower interfacial tension val-ues and exhibits faster adsorption kinetics than TyrGE12. Emulsionscontaining PhGE12 show better stability compared to those withTyrGE12, as seen from creaming index measurements. Droplet sizedistributions reveal that in PhGE12 containing emulsions there area bit more small (∼0.5 �m) and intermediate droplets (∼50 �m),which might explain their higher apparent viscosity and their bet-ter stability. Increasing surfactant’s concentration for both glycerolether molecules lead to a substantial increase of stability of theproduced emulsions. This is in accordance to trends versus surfac-tant concentration of interfacial tension and apparent viscosity butopposes trends of interfacial dilatational viscoelasticity and dropsize distributions. Overall, these observations make these surfac-tants, especially the PhGE12 molecule, good candidates for use incosmetic emulsions e.g., cleaning fluids, personal hygiene creamsetc.

Acknowledgement

This work was supported by ESA (European Space Agency)

bility) and PASTA (Particle Stabilized Emulsions and Foams). Thework was performed under the umbrella of COST Actions MP1106,TD1204 and ES 1205.

A: Phy

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

C.D. Ampatzidis et al. / Colloids and Surfaces

eferences

[1] B.P. Binks, Modern Aspects of Emulsion Science, Royal Society of Chemistry,Information Services, 1998.

[2] H. Schubert, H. Armbruster, Principles of formation and stability of emulsions,International Chemical Engineering 32 (1992) 14–28.

[3] M.J. Rosen, Surfactants and Interfacial Phenomena, Wiley, 1978.[4] Z. Dong, M. Lin, H. Wang, M. Li, Influence of surfactants used in surfactant-

polymer flooding on the stability of Gudong crude oil emulsion, PetroleumScience 7 (2010) 263–267.

[5] M.D. Gernon, D. Alford, C.M. Dowling, G.P. Franco, Enhancing oil/water emul-sion stability: the use of capillary contact angle measurements to determineliquid/liquid interfacial tensions between aqueous alkanolamine solutions andoils, Tribology Transactions 52 (2009) 405–414.

[6] S. Hartland, S.A.K. Jeelani, Effect of interfacial tension gradients on emulsionstability, Colloids and Surfaces A: Physicochemical and Engineering Aspects 88(1994) 289–302.

[7] R. Nagarajan, D.T. Wasan, Measurement of dynamic interfacial tension byan expanding drop tensiometer, Journal of Colloid and Interface Science 159(1993) 164–173.

[8] P. Van der Meeren, J. Vanderdeelen, L. Baert, On the Relationship betweenDynamic Interfacial Tension and O/W Emulsion Stabilizing Properties of Soy-bean Lecithins, in: J. Appell, G. Porte (Eds.) Trends in Colloid and InterfaceScience IX, Steinkopff, 1995, pp. 136–139.

[9] M. Wulff-Pérez, A. Torcello-Gómez, A. Martín-Rodríguez, M.J. Gálvez-Ruiz, J. deVicente, Bulk and interfacial viscoelasticity in concentrated emulsions: the roleof the surfactant, Food Hydrocolloids 25 (2011) 677–686.

10] E. Varka, E. Coutouli-Argyropoulou, M. Infante, S. Pegiadou, Synthesis, charac-terization, and surface properties of phenylalanine-glycerol ether surfactants,Journal of Surfactants and Detergents 7 (2004) 409–414.

11] E.-M.A. Varka, M.G. Heli, E. Coutouli-Argyropoulou, S.A. Pegiadou, Synthe-sis and characterization of nonconventional surfactants of aromatic aminoacid–glycerol ethers: effect of the amino acid moiety on the orientation and sur-face properties of these soap-type amphiphiles, Chemistry–A European Journal12 (2006) 8305–8311.

12] C.D. Ampatzidis, E.M.A. Varka, T.D. Karapantsios, Dynamic surface propertiesof eco-friendly phenylalanine glycerol ether surfactants at the W/A interface,Colloids and Surfaces A: Physicochemical and Engineering Aspects (2012).

13] C.D. Ampatzidis, E.M.A. Varka, T.D. Karapantsios, Adsorption behavior of non-conventional eco-friendly tyrosine glycerol ether surfactants, Colloids andSurfaces A: Physicochemical and Engineering Aspects 438 (2013) 104–111.

14] G. Loglio, P. Pandolfini, R. Miller, A.V. Makievski, J. Krägel, F. Ravera, B.A. Noskov,Perturbation–response relationship in liquid interfacial systems: non-linearityassessment by frequency–domain analysis, Colloids and Surfaces A: Physico-chemical and Engineering Aspects 261 (2005) 57–63.

15] O. Malinowska, H. Szelag, The investigation of application possibilities and rhe-ological behavior of crude fruit seed oil in emulsion systems, InternationalJournal of Research in Cosmetic Science 3 (2013) 7–13.

16] J. Meyer, G. Polak, R. Scheuermann, Preparing PIC. Emulsions with a Very FineParticle Size, in: Cosmetics &Toiletries, Degussa GmbH, Germany, 2007, pp.61–70.

17] A.R. Rahate, J.M. Nagarkar, Emulsification of vegetable oils using a blend ofnonionic surfactants for cosmetic applications, Journal of Dispersion Scienceand Technology 28 (2007) 1077–1080.

18] H. Sagitani, Making homogeneous and fine droplet o/w emulsions using non-ionic surfactants, Journal of the American Oil Chemists’ Society 57 (1981)738–743.

19] S.L. McSweeney, R. Healy, D.M. Mulvihill, Effect of lecithin and monoglycerides

on the heat stability of a model infant formula emulsion, Food Hydrocolloids22 (2008) 888–898.

20] C.D. Ampatzidis, Rheological characterization of interfaces for the study ofemulsions stability using non-compatible surfactants, PhD Thesis, Departmentof Chemistry, Aristotle University of Thessaloniki, Thessaloniki, 2014, p. 250.

[

sicochem. Eng. Aspects 460 (2014) 176–183 183

21] R. Vijay, J. Singh, G. Baskar, R. Ranganathan, Amphiphilic lauryl ester derivativesfrom aromatic amino acids: significance of chemical architecture in aque-ous aggregation properties, The Journal of Physical Chemistry B 113 (2009)13959–13970.

22] K. Brunaldi, M.A. Miranda, F. Abdulkader, R. Curi, J. Procopio, Fatty acid flip–flopand proton transport determined by short-circuit current in planar bilayers,Journal of Lipid Research 46 (2005) 245–251.

23] W.F.D. Bennett, A.W. Chen, S. Donnini, G. Groenhof, D.P. Tieleman, Constant pHsimulations with the coarse-grained MARTINI model—Application to oleic acidaggregates, Canadian Journal of Chemistry 91 (2013) 839–846.

24] J.R. Kanicky, D.O. Shah, Effect of degree, type, and position of unsaturation onthe pKa of long-chain fatty acids, Journal of Colloid and Interface Science 256(2002) 201–207.

25] V.B. Fainerman, V.D. Mys, A.V. Makievski, J.T. Petkov, R. Miller, Dynamic surfacetension of micellar solutions in the millisecond and submillisecond time range,Journal of Colloid and Interface Science 302 (2006) 40–46.

26] S. Fu, W. Zhou, Z. Wang, B. Yin, J. Liu, D. Sun, X. Wei, Study on dynamic interfacialtension of 3-dodecyloxy-2-hydroxypropyl trimethyl ammonium bromide atliquid/liquid interface, Fluid Phase Equilibria 269 (2008) 93–97.

27] V.G. Babak, J. Desbrieres, V.E. Tikhonov, Dynamic surface tension and dilationalviscoelasticity of adsorption layers of a hydrophobically modified chitosan,Colloids and Surfaces A: Physicochemical and Engineering Aspects 255 (2005)119–130.

28] T. Dapcevic Hadna –dev, P. Dokic, V. Krstonosic, M. Hadna –dev, Influence of oilphase concentration on droplet size distribution and stability of oil-in-wateremulsions, European Journal of Lipid Science and Technology 115 (2013)313–321.

29] A. El-Hamouz, Effect of surfactant concentration and operating temperature onthe drop size distribution of silicon oil water dispersion, Journal of DispersionScience and Technology 28 (2007) 797–804.

30] C. Sun, S. Gunasekaran, Effects of protein concentration and oil-phase volumefraction on the stability and rheology of menhaden oil-in-water emulsionsstabilized by whey protein isolate with xanthan gum, Food Hydrocolloids 23(2009) 165–174.

31] P. Becher, Emulsions: Theory and Practice, second ed., Reinhold PublishingCorp, New York, 1965.

32] U. Klinkesorn, P. Sophanodora, P. Chinachoti, D.J. McClements, Stability and rhe-ology of corn oil-in-water emulsions containing maltodextrin, Food ResearchInternational 37 (2004) 851–859.

33] L.I. Osipow, Surface Chemistry: Theory and Industrial Applications, ReinholdPublishing Corp, New York, 1962.

34] P. Sherman, The Influence of Emulsifier Concentration on the Rheological Prop-erties of Emulsions, The MacMillan Co. (Pergamon Press Ltd.), New York, 1963.

35] K. Suzuki, T. Maeda, K. Matsuoka, K. Kubota, Effects of constituent concentra-tion on rheological properties of corn oil-in-water emulsions, Journal of FoodScience 56 (1991) 796–798.

36] R. Pal, Effect of droplet size on the rheology of emulsions, AIChE Journal 42(1996) 3181–3190.

37] R. Chanamai, D.J. McClements, Dependence of creaming and rheology ofmonodisperse oil-in-water emulsions on droplet size and concentration, Col-loids and Surfaces A: Physicochemical and Engineering Aspects 172 (2000)79–86.

38] C.D. Ampatzidis, E-M. A Varka, T.D. Karapantsios, Effect of the Different AminoAcid Hydrophilic Head in the Adsorption Behavior of Eco-Friendly GlycerolEther Surfactants at the W/A Interface, in: 27th ECIS Conference, Sofia, Bulgaria,2013, p. 1.

39] R. Vijay, A.B. Mandal, G. Baskar, 1H NMR spectroscopic investigations on theconformation of amphiphilic aromatic amino acid derivatives in solution: effect

of chemical architecture of amphiphiles and polarity of solvent medium, TheJournal of Physical Chemistry B 114 (2010) 13691–13702.

40] A.N. Ilia Anisa, H.N. Abdurahman, Affect of viscosity and droplet diameter onwater-in-oil (w/o) emulsions: an experimental study, World Academy of Sci-ence Engineering and Technology 38 (2010) 607–611.