AaC

JJa

Vb

h

•

•

•

•

a

ARRAA

KPSHCF

1

s

h0

Colloids and Surfaces A: Physicochem. Eng. Aspects 458 (2014) 32–39

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

simple method to assess the hydrophilic lipophilic balance of foodnd cosmetic surfactants using the phase inversion temperature of10E4/n-octane/water emulsions

esús F. Ontiverosa,b, Christel Pierlota, Marianne Cattéa, Valérie Moliniera,ean-Louis Salagerb, Jean-Marie Aubrya,∗

Université Lille Nord de France, USTL & ENSCL, Cité Scientifique, EA-CMF 4478, Equipe, Oxydation et Physico-Chimie de la Formulation, F-59652illeneuve d’Ascq Cedex, FranceLaboratorio FIRP, Ingeniería Química, Universidad de Los Andes, Mérida 5101, Venezuela

i g h l i g h t s

Effect of an additional second sur-factant on the phase inversion tem-perature (PIT) of C10E4/octane/watersystem.dPIT/dC expresses the amphiphilicityof the second surfactant.dPIT/dC allows classifying surfactantsin a simple and reliable way.Application to food and cosmetic sur-factants.

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 21 October 2013eceived in revised form 18 February 2014ccepted 27 February 2014vailable online 12 March 2014

a b s t r a c t

The modification of the phase inversion temperature (PIT) of C10E4/n-octane/water emulsions wasstudied as a function of increasing amounts of additional second surfactants in order to rank themaccording to their hydrophilic lipophilic balance. Twenty five surfactants, selected from a wide rangeof chemical families, were studied. Well-defined polyethoxylated alkyl surfactants (CiEj) show a linearvariation of the PIT with their concentration C and can be used as standards to calibrate a scale in terms

eywords:hase inversion temperatureurfactant classificationLBosmetic surfactantood surfactant

of dPIT/dC. This parameter leads to a simple classification of surfactants with respect to C10E4. Positiveand negative values correspond to more or less hydrophilic surfactants compared to C10E4, respectively.Several industrial surfactants used in cosmetic and food industries (lecithins, sorbitan derivatives Spansand Tweens, sucrose esters, monoglycerides) were investigated and results are discussed with respectto the HLB scale of Griffin.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

The selection of an appropriate surfactant is a determinanttep to formulate emulsions or micro-emulsions. To achieve a

∗ Corresponding author. +33 320336364.E-mail address: Jean-Marie.Aubry@univ-lille1.fr (J.-M. Aubry).

ttp://dx.doi.org/10.1016/j.colsurfa.2014.02.058927-7757/© 2014 Elsevier B.V. All rights reserved.

good choice several tools can be used to compare the hydrophiliclipophilic balance of different surfactants. These methods will con-sider either the chemical structure of the surfactant molecule(HLB [1], packing parameter [2]), or the behavior of surfactantin water (cloud point), or the behavior of surfactant in the pres-

ence of water and oil (PIT [3], HLD [4], effective packing parameter[5]).

To classify non ionic surfactants, Griffin [1] introduced thehydrophilic lipophilic balance (HLB), an empirical scale arbitrary

A: Phy

d

H

wtmibskg

H

wheaatFt

ohthSltwHo

tttmbccpd

ttotaaa

ooCfisfs

J.F. Ontiveros et al. / Colloids and Surfaces

efined as follows:

LB = 15

(MH

MT× 100

)(1)

here MH is the molecular mass of the hydrophilic portion ofhe surfactant molecule, and MT the molecular mass of the whole

olecule, giving a result scaling from 1 to 20. Griffin’s HLB numbers widely used in particular for non ionic ethoxylated surfactantsut there is no proposed calculation for other kind of surfactantsuch as ionic ones. Davies [6] proposed a competitive coalescenceinetic model to explain the type of emulsions and suggested aroup contribution method to calculate the HLB values

LB =∑

Hh,i −∑

Hl,i + 7 (2)

here Hh,i and Hl,i are the contributions of hydrophilic andydrophobic parts, respectively. In this second expression, theffects of stronger and weaker hydrophilic groups are differenti-ted, however these empirical contributions are not defined forll kinds of polar heads, and there is for instance no possibilityo classify phospholipids, fluorinated or sucrose esters surfactants.urther, it must be noticed that there is not always a clear correla-ion between HLB values from Davies and Griffin.

The Phase Inversion Temperature (PIT) introduced by Shin-da et al. [3] corresponds to the temperature at which theydrophilic/lipophilic character of a nonionic ethoxylated surfac-ant switches by dehydration of the polyoxyethylene units duringeating, leading to the emulsion phase inversion from O/W to W/O.hinoda [3] studied the effect of surfactant polyoxyethylene chainength and hydrocarbon chain length on the PIT and found thathe emulsion phase inversion temperature was close correlatedith the cloud point of surfactant in solution. The PIT and theLB-number are also correlated, provided the distribution of thexyethylene chains of the emulsifiers is similar [7].

Kunieda [8–10] developed a mathematical model that defineshe hydrophilic–lipophilic–balanced plane (HLB plane) which is thehree-phase triangle where the surfactant (or mixture of surfac-ants S1 and S2) has the same affinity for oil and water. Using this

odel and by measuring the weight ratio S1/S2 that gives equili-rated three phase systems, weight ratio S1/S2 in the middle phasean be estimated. Thus, HLB for unknown S2 surfactant can be cal-ulated. However the experimental conditions to reach the threehase area are sometimes difficult to locate and measuring phaseiagrams is time-consuming.

For most industrial food or cosmetic surfactant mixtures allhese methods are difficult or impossible to use: molecular struc-ure not precisely known, no phase inversion between 5 and 90 ◦Cr experimental determinations very difficult to calculate surfac-ants parameters for HLD. The purpose of this work is to propose

new simple method able to assess the hydrophilic lipophilic bal-nce of all kinds of surfactant: well defined pure surfactants as wells commercial mixtures.

The principle of the method is to observe the modificationf the PIT of a reference SOW system with increasing amountsf various second surfactants. The reference system used is pure10E4/octane/water whose PIT is close to room temperature. In a

rst part, well defined CiEj will be studied to validate the new clas-ification method, and to discuss its limitations. Then 17 commonood and cosmetics surfactants will be classified, including tween,pan, monoglycerides, sucrose esters and lecithins.

sicochem. Eng. Aspects 458 (2014) 32–39 33

2. Experimental

2.1. Chemicals

Pure decyl tetraethyleneglycol ether (C10E4) was synthesizedaccording to a method described elsewhere [11,12]. Its puritywas assessed by NMR and GC analyses (>99%) and by com-paring the cloud point temperature with the reference value[13]. The compound was distilled twice under reduced pressure(174 ◦C/2.4 × 10−2 mbar) leading to a cloud point temperatureof 20.4 ◦C at 2.6 wt% in agreement with the published value(20.56 ◦C/2.6 wt%). Dodecyl tetraethyleneglycol ether (C12E2) wassynthesized in the same way, by reaction of dodecyl bromidewith diethyleneglycol. Potassium oleate was prepared by neu-tralization of oleic acid (SIGMA >99%) in ethanol with ethanolicKOH (SIGMA >90%) and then by evaporation of the solvent underreduced pressure.n-Octane (99%) was obtained for Sigma-Aldrich.Sodium chloride NaCl (≥99.5%) was supplied for Acros Organics.The commercial surfactants S2 studied in this work are shown inTable 1 and were used without further purification.

2.2. Sample preparation

10 mL of the C10E4/n-octane/water system were prepared in a20 mL vial (d = 2.5 cm, h = 5.5 cm), 24 h before the experiment bypouring in a reactor a mass mw (4.250 g) of NaCl 10−2 M aque-ous brine, a mass mo (4.250 g) of n-octane, and then an aliquot ofS1 = C10E4 (0.263 g), equivalent to 3 wt%. The water fraction (definedby Eq. (3)) is 0.5 in all systems.

fw = mw

mw + mo(3)

The system was hand-shaken during a few seconds and left topre-equilibrate at room temperature. A first heating–cooling cyclewas performed (see next paragraph) and then the second surfactantS2 was added. Eq. (4) defines S2 weight percentage in the system.

S2 wt% = mS2

mw + mo + mS1 + mS2

× 100 (4)

The amount of C10E4(

mS1

)was adjusted so that the proportion

of S1 surfactant remains constant at 3 wt% of the whole system inall experiments.

2.3. Heating–cooling cycles

During the whole experiment, the system was kept undercontinuous stirring using a square-cross magnetic stirrer (1.8 cmdiameter) at 500 rpm. It was maintained at the initial temperatureTo (about 12 ◦C below the PIT) for 10 min, before starting tempera-ture cycles. The sample was then subjected to two heating–coolingcycles at a linear rate of 1 ◦C/min. Temperature-controlled waterusing a HUBER Ministat 125 was circulated through the vesselin which the vial was introduced. The conductivity and temper-ature were simultaneously measured by a Radiometer AnalyticalCDM 210 conductimeter fitted with a CDC741 T platinized plat-inum probe. The software used was custom written in a Labview 7.1National Instruments platform. The program carries out real timeacquisition at ca. 2 data per second.

2.4. Phase inversion temperature determination

Temperature and conductivity profiles for two heating–cooling

cycles were obtained as a function of time. The parallel tangentmethod described elsewhere [14] was used to calculate the phaseinversion temperature (PIT). The reported value is the average ofthe four temperatures, unless otherwise specified. In the two cycles,

34 J.F. Ontiveros et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 458 (2014) 32–39

Table 1Supplier and purity of surfactants.

No. Surfactant Supplier Purity and others specifications

1 Diethylene glycol monododecyl ether (C12E2) Synthesized >99%2 Pentaethylene glycol monododecyl ether (C12E5) Fluka >98%3 Hexaethylene glycol monododecyl ether (C12E6) SIGMA >98%4 Octaethylene glycol monododecyl ether (C12E8) TCI >95%5 Decaethylene glycol monododecyl ether (C12E10) SIGMA –6 Tetraethylene glycol monodecyl ether (C10E4) Synthesized >99%7 Octaethylene glycol monodecyl ether (C10E8) SIGMA >98%8 Oleic acid SIGMA >99%9 Potassium oleate Synthesized >99%10 SPAN 80 Fluka (60% C18:1; balance primarily C18:2, C18:3 and C16:0)11 SPAN 60/Sorbitan monostearate Alfa Aesar –12 SPAN 40/Sorbitan monopalmitate Sigma Aldrich –13 SPAN 20 Fluka (>44% C12:0; balance primarily C14:0; C16:0 and C18:3)14 Tween 80 Acros –15 Tween 60 Fluka (47–55% C18:0; 35–50% C16:0)16 Tween 40 Fluka (90% C16:0 balance primarily C18:0)17 Tween 20 Alfa Aesar –18 Glycerol monopalmitate TCI >95%19 Glycerol monolaurate TCI >98%20 Sucrose ester SE3P S. Dubois (66.8% C18:0; 20% monoester)21 Sucrose ester SE5S S. Dubois (65.7% C18:0; 30% monoester)22 Sucrose ester SE15P S. Dubois (78.2% C16:0; 70% monoester)

SigmCargCarg

i(

3

C3famabettos

FS

23 l-�-Phosphatidylcholine

24 Epikuron 200

25 Epikuron 130

ndividual values of PIT for heating and cooling are very similar±0.5 ◦C).

. Results and discussion

Pizzino et al. [15] reported the fish diagram for the10E4/octane/10−2 M NaCl system at fw = 0.5. From this diagram, at% C10E4, the equilibrium physicochemical behavior must changerom an aqueous microemulsion in equilibrium with oil (WIccording to Winsor nomenclature) at low temperatures to an oilicroemulsion in equilibrium with water (WII) at higher temper-

tures, passing through a three phase behavior (WIII) in a rangeetween 22 and 29 ◦C. Under stirring, the transition from an O/Wmulsion to a W/O occurs at 24.8 ◦C (the so-called phase inversion

emperature PIT) using mechanical agitation [16]. Using the pro-ocol defined in the experimental section, the samples containingnly C10E4 showed a phase inversion temperature of 23.9 ◦C with atandard deviation of 0.2 ◦C, thus in fairly good agreement with the

0

100

200

300

400

5 15 25 35 45

Cond

uc�

vity

(µS/

cm)

Tem peratur e (°C)

1.20%1.04%0.75%0.56%0.34%0.17%0% C12E2

3% C10E4/Octane /Water+ C12E2

A

ig. 1. Temperature-dependence of the conductivity of the emulsion formed with 3% C1

2 = C12E2. (B) S2 = C12E6.

a Aldrich >99%ill (93.5% L-�-phosphatidylcholine)ill (30% L-�-phosphatidylcholine)

value previously reported [16], which validates the choice of thesmall-dimension phase inversion set-up used in the present work.

3.1. Effect of the addition of a second well-defined CiEj on the PITof C10E4/n-octane/10−2 M NaCl system

Fig. 1 shows the conductivity vs. temperature profiles for twowell-defined polyethoxylated dodecyl surfactants (C12Ej) added tothe 3% C10E4/n-octane/10−2 M NaCl system. All systems exhibit theclassical shape of a temperature induced phase inversion. At lowtemperature, the high conductivity increasing with temperatureevidences an O/W emulsion up to a temperature range in whichthe conductivity drastically falls, indicating the inversion to a W/Oemulsion. In every conductivity profile of Fig. 1, a tiny conductivitybump appears just after the fall, which is usually attributed to the

temporary occurrence of liquid crystals [17–19].

By adding increasing amounts of C12E2 to the3%C10E4/octane/water system, the phase inversion temperature(PIT) gradually decreases. C12E2 is a less hydrophilic surfactant than

0

100

200

300

400

5 15 25 35 45

Cond

uc�

vity

(S/

cm)

Tem perature (°C)

1.19%0.95%0.71%0.50%0.20%0% C12 E6

3% C10E4/Octane/Water+ C12E6

B

0E4/n-octane/10−2 M NaCl + S2 samples at fw = 0.5, on the second cooling cycle. (A)

J.F. Ontiveros et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 458 (2014) 32–39 35

0

10

20

30

40

50

0.0 0.5 1.0 1.5 2.0 2.5

Phas

e In

vers

ion

Tem

pera

ture

(°C

)

S2 concentration (wt. %)

3% C10E4/Octane /Water + S2 fw=0.5

FsC

CastthOCpTF

o

ccssTts

diofa

ttictapiosD

bps

0

2

4

6

8

10

12

14

16

18

20

0 20 40 60 80 10 0

dPIT

/dC

(°C

/wt.%

)

Cloud Point (°C)

C12E8

C12E6

C12E5C10E4

C10E8

ig. 2. Phase inversion temperature PIT vs. S2 weight concentration fraction for theystem 3%C10E4/octane/water + C12Ej at fw = 0.5. S2 = C12E5 (�); C12E6 (�); C12E8 (×);12E2 (�). Dotted lines indicate linear fitting.

10E4, a fact that is corroborated by the impossibility to measure cloud point (lower than 0 ◦C). Similarly, the C12E2/octane/waterystem cannot be inverted from a W/O to an O/W emulsion whenemperature diminishes until 5 ◦C. A less hydrophilic surfactanthan C10E4 diminishes the temperature required to change theydrophilic lipophilic balance of the entire system (C10E4 + S2).n the other hand, C12E6 is a more hydrophilic surfactant than10E4, a fact that can be demonstrated by comparing the cloudoints: 48 ◦C [20] and 20.4◦C for C12E6 and C10E4, respectively.he addition of C12E6 increases the PIT of the system as shown inig. 1B.

In Fig. 2, the PIT is represented as a function of the concentrationf the second surfactant S2 for several C12Ej.

Note that in this case each point on Fig. 2 has a different totaloncentration of surfactant, in order to maintain constant the C10E4oncentration in the system (3 wt%). For all polyethoxylated alkylurfactants studied, the PIT of the C10E4 + S2/octane/water systemhows a linear evolution with the second surfactant concentration.he size of the polar head in the ethoxylated surfactants impactshe slope (dPIT/dC) of each surfactant. The more hydrophilic theecond surfactant S2, the higher the slope dPIT/dC.

For linear tendencies of PIT vs. second surfactant concentration,PIT/dC can be interpreted as the theoretical change on the phase

nversion temperature for the C10E4/octane/water system when 1%f a second surfactant S2 is added. Table 2 shows the dPIT/dC valuesor each CiEj and their HLB values, calculated with Griffin and Daviespproaches [1,6].

Positive values of dPIT/dC indicate that S2 is more hydrophilichan C10E4, negative values indicate more lipophilic surfactantshan C10E4. The zero in this scale corresponds to C10E4, the non-onic surfactant used as reference. The dPIT/dC value thus allowsharacterizing surfactants and could represent a reliable parame-er for the quantification of the hydrophilic-lipophilic balance of

surfactant. The rise of dPIT/dC for the C12Ej surfactants when jasses from 2 to 8 indicates the increase in hydrophilicity. Compar-

ng dPIT/dC values for C12E8 and C10E8 (15.8 and 17.6), the influencef a shorter alkyl chain is evident and leads to a more hydrophilicurfactant, also indicated by HLB values reported by Griffin andavies.

To be used as reliable quantification of the hydrophilic-lipophilicalance, the dPIT/dC parameter should be correlated with an indis-utable experimental indicator of this balance for a given surfactantuch as the cloud point. In contrast, the HLB values may vary

Fig. 3. dPIT/dC for the system 3% C10E4/octane/water + CiEj at fw = 0.5 vs. Cloud pointof CiEj . Dotted line indicate linear fitting. (R2 = 0.99).

significantly depending on the used Griffin’s or Davies’ formulas,as shown in Table 2.

Differences in HLB values of Griffin and Davies for the CiEj sur-factants are remarkable. Davies pointed out that Eq. (2) is not validfor “very large molecules with for example 200 ethylene oxidesresidues”, nevertheless in our case the longer ethylene oxides chainis 8 (C12E8) and the differences are still observed. HLB values andcloud point values are both related to the number of ethylene oxidegroups however, the cloud points are undisputable and measurablephysico-chemical properties of the polyethoxylated surfactantsand therefore we find more accurate to look at the correlationbetween dPIT/dC and the cloud points for this homogeneous familyof non-ionic surfactants. Fig. 3 shows this correlation for the CiEjseries.

The cloud point and the phase inversion temperature are closelylinked as evidenced by Shinoda et al. [3]. Fig. 3 evidences a linearitybetween the cloud point and dPIT/dC. The good fitting reinforcesour choice of the parameter dPIT/dC to construct a rating scale ofthe hydrophilic–lipophilic balance of surfactants.

3.2. Classification of food and cosmetic surfactants

Fig. 4 shows the conductivity vs. temperature profiles for otherS2 surfactants at various concentrations. Two sorbitan derivatives,Tween 20 and Span 40 were added to the 3% C10E4/n-octane/10−2 MNaCl system.

The presence of Sorbitan monopalmitate or Span 40 (Fig. 4A)as well as Tween 20 (Fig. 4B) impacts the conductivity profile of3%C10E4/octane/water system. As already observed in the case ofwell-defined surfactants, when S2 surfactant is less hydrophilicthan C10E4, the PIT decreases (Fig. 4A). Symmetrically, when S2is more hydrophilic than C10E4 (Fig. 4B), the PIT rises. Dependingon the nature and range of concentration for the second surfac-tant, the increase in S2 concentration modifies the verticality in theinversion curve, as can be seen in Fig. 4A. In this case, the phaseinversion zone rises, as already observed by Izquierdo et al. [22] inthe Brij30/hexadecane/water (0.01 M NaCl) system at fw = 0.8 andalso by Anton et al. [19] in the C18E6/mineral oil/water (0.1 M NaCl)system at WOR = 1.

As in the case of well-defined surfactants, the average phaseinversion temperature of the system varies linearly with theconcentration of S2 for most of the mixed surfactants over thestudied range of concentrations. This trend can be appreciated on

36 J.F. Ontiveros et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 458 (2014) 32–39

Table 2dPIT/dC, HLBGriffin, HLBDavies for well defined non ionic surfactants.

Surfactant Structure HLBGriffin HLBDavies dPIT/dCa (◦C/ wt%) Cloud point (◦C)

C12E2 (1) 6.4 4.3 −9.2 <0b

C12E5 (2) 10.8 5.3 1.1 26.5b

C12E6 (3) 11.7 5.7 6.1 48b

C12E8 (4) 13.1 6.3 15.8 77b

C10E4 (6) 10.5 5.9 0 20.4C10E8 (7) 13.8 7.3 17.6 85c

a Error <10%. The error is considerate as the maximal percentage of deviation of dPIT/dC reported values. It is calculated by ordinary least squares method fitting the PITand concentrations experimental data.

b Cloud points reported by Mitchell et al. [20].c Cloud point reported by Corti et al. [21].

0

100

200

300

400

5 15 25 35 45

Cond

uc�

vity

(µS/

cm)

Temperature (°C)

0.87%

0.58%

0.38%

0.20%

0% Span40

3% C10E4/Octane/Water + Span 40

A

0

100

200

300

400

5 15 25 35 45

Cond

uc�

vity

( µS/

cm)

Tem perature (°C)

0.60%0.52%0.41%0.23%0.16%0% T20

3% C10E4/Octane/Water+ Twee n 20

B

F 3% C1

S

Fo

wpt

FCS

ig. 4. Temperature-dependence of the conductivity of the emulsion formed with2 = Sorbitan monopalmitate (Span 40). (B) S2 = Tween 20.

ig. 5 for the Sorbitan monopalmitate (Span 40), the Tween 20 andthers sorbitan derivatives.

The ethoxylated derivates of sorbitan (Tween) increase the PIT

hile the non ethoxylated ones (Span series) diminishes it. Com-aring Fig. 5, the results for different non ethoxylated sorbitans, athe same concentration, the most important decrease is reached

0

10

20

30

40

0.0 0.5 1.0 1.5

Phas

e In

vers

ion

Tem

pera

ture

(°C

)

S2 concentra tion (wt. %)

3% C10E4/Octane /Water + S2 fw=0.5

ig. 5. Phase inversion temperature PIT vs. S2 weight percentage for the system 3%10E4/octane/water + S2 at fw = 0.5. S2 = Tween 20 (×); Tween 80(�); Span 20 (+);orbitan monopalmitate (�); Sorbitan monostearate (�).

0E4/n-octane/10−2 M NaCl + S2 samples at fw = 0.5, on the second cooling cycle. (A)

with sorbitan monostearate, then Span 40 (sorbitan monopalmi-tate) and finally Span 20 (sorbitan monolaurate). For the samepolar head, the size of the alkyl chain determines the slope: morecarbons on the average alkyl chain increase the hydrophobicity.Similar behavior is noted with the HLB values of these surfactants.Similar reasoning applies to ethoxylated sorbitan derivatives:Tween 80, having an oleate chain is less hydrophilic than Tween20, with laurate as average alkyl chain.

Table 3 shows the values of the parameter dPIT/dC for fivefamilies of surfactants used both in food and in some cosmeticformulations and for the C12E10 industrial grade. Additionally,potassium oleate and oleic acid, the two references used by Griffinfor his HLB scale were included.

dPIT/dC values for non ethoxylated sorbitan derivatives areconsistent with chemical structure. The double bound in SPAN80 reduces the lipophilicity compared to SPAN 60, although theHLB is not able to reflect this phenomenon. However, lipophilicchains are variable in SPAN 80 and TWEEN 60 and the monooleateand monostearate are, respectively, the main components but notthe only ones. In the Tween series the grafting of ethoxylatedchains on sorbitan structure makes the analysis more difficult.The distribution of ethoxylate groups induces the occurrence ofpartition of the surfactant [8,23,24] between the interfacial filmand the oil. It is worth to notice that the dPIT/dC value expressonly the hydrophilic lipophilic balance of the surfactant located atthe W/O interphase. According a uniform chemical structure, theexpected decreasing hydrophilic order should be Tween 20 > Tween

40 > Tween 80 > Tween 60.

Glycerol monoesters derivatives exhibit more lipophilic behav-ior than C10E4; as in the case of sorbitan esters, the dPIT/dC valuesare negative. An increase in the carbon chain of 12 to 16 carbons

J.F. Ontiveros et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 458 (2014) 32–39 37

Table 3dPIT/dC, HLBGriffin, HLBDavies for cosmetic and food surfactants.

a /dC rcb

mt(Ti

fpcco

ld

Error <10%. The error is considerate as the maximal percentage of deviation of dPIToncentrations experimental data.HLB indicated by supplier are 3, 5 and 15 for sucrose ester SE3P, 5S and 15P.

akes dPIT/dC decrease of 4.2 ◦C/(wt%). This decrease is greaterhan observed between Span 20 and sorbitan monopalmitate1.4 ◦C/(wt%)) or Tween 20 and Tween 40 (3.2 ◦C/(wt%)), howeverween and Span are mixtures and therefore the carbon alkyl chains an average of all present chain lengths.

Sucrose esters follow the same behavior as indicated by the HLBrom manufacturer, the more hydrophilic (HLB = 15) SE15P has aositive value whereas SE3S (HLB = 3) has a negative value. In thisase the HLBGriffin in Table 3 corresponds to the accepted rule thatalculates the HLB of sucrose esters as 20 times the weight fractionf monoesters in the blend [25].

Lecithins, as well as monoesters, are widely used in food formu-ations with less restriction on the concentration level than sorbitanerivatives. Two selected commercial deoiled lecithins, with

eported values. It is calculated by ordinary least squares method fitting the PIT and

different phosphatidylcholine content were studied: Epikuron 200and Epikuron 130. A 99% pure sample of L-�-phosphatidylcholinewas also studied however there is no significant difference ondPIT/dC between the Epikuron 200 and that sample: both are morelipophilic than C10E4. Epikuron 130 presents a non linear behav-ior in the conductivity profiles: at low concentrations there is anincrease of the PIT and then a decrease. Epikuron 130 is a com-plex mixture of different polar groups: phosphadyethanolamine,phosphatic acid, phosphatidylinositol, phosphatidylcholine andlysophosphatidylcholine having probably different partition at theinterface at low concentrations. The dPIT/dC is evaluated in this

case as the derivative when S2 concentration tends to zero. Never-theless, this value cannot represent the behavior of the surfactantexcept if used at very low concentrations.

38 J.F. Ontiveros et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 458 (2014) 32–39

0 10 20-10-20

Twee

n 20

Twee

n 60

Twee

n 80

Twee

n 40

GlycerolMonopalmitate

Glyc erolMonolaurate

Span

20Sp

an40

Span

60

Span

80

SE3P

SE5S

SE

15P

Epi kuron 20 0 Epi kuron 130

Sucrose esters

LecithinsGlycerol Monoesters

C12

E 2

C12

E 5

C12

E 8C

10E 8

C10E4

Cosmetic and Food Surfactants

C12

E 6

More Hydrophilicthan C10 E4

More Lipo phili cthan C10E4

C12

E10

tech

115 120

Ole

icA

cid

Pot a

sium

Ole

ate

HLB Griffin Domain1 20

ation

cmoItClah

ffibtbrwtga

4

fCPtpto

aassNpres

Fig. 6. Surfactants Classific

The C12E10 used is a polydisperse mixture frequently used inosmetics. Its dPIT/dC value follows the C12Ej tendency; it is theost hydrophilic tested surfactant of this series. Fig. 6 is a scheme

f non ionic surfactants classification using the dPIT/dC method.n the top of the scheme are placed the cosmetic and food surfac-ants (industrial grade or not) and on the bottom the well definediEj surfactants. Pure glycerol monopalmitate has a slightly higher

ipophilicity than Span 80. Epikuron 200 has a similar balances Sucrose esters SE3P or SE5S, and polydisperse C12E10 is moreydrophilic than Tween 20.

The described method has the advantage of being a simple andast way to compare surfactants; nevertheless dPIT/dC of a sur-actant is not an absolute value but a comparative criterion thatndicates how the surfactant S2 impacts the 3% C10E4/octane/waterehavior, if it increases or decreases the hydrophobicity of the sys-em. This method can be used as a simple tool to classify surfactantseing more reliable than HLB. Potassium oleate and oleic acid, theeferences for the HLB scale from 1 to 20, are placed in our scaleith dPIT/dC of −17.7 and 118, respectively. One of the advan-

ages of this method is the inclusion of both, pure and industrialrade surfactants in the same classification with no ambiguities:ny surfactant can be analyzed using the described protocol.

. Conclusion

We developed a new simple and fast method to classify sur-actants. This method is based on the variation of the PIT of the 3%10E4/octane/water system by addition of a surfactant S2. Since thisIT varies linearly with surfactant concentration, dPIT/dC parame-er was defined to characterize the hydrophilic lipophilic balanceower. For well defined alkyl polyethoxylated compounds (CiEj)he good correlation between dPIT/dC and cloud point validatesur method.

Surfactant that causes an increase in the PIT will be considereds more hydrophilic than C10E4 and vice versa, so dPIT/dC is not anbsolute value but a comparative criterion that indicates how theurfactant S2 impacts the reference 3% C10E4/octane/0.01 M NaClystem. These experimental conditions (3% C10E4, fw = 0.5, 0.05 MaCl and octane as oil) must be kept constant in order to get com-

arable values. However, if the addition of a compound on theeference system does not cause any change in PIT (dPIT/dC = 0),ither the molecule has the same amphiphilicity as C10E4 or is noturface active (completely in the oil or water phase).

using the dPIT/dC method.

Increasing the number of ethylene oxide of pure CiEj extendsdPIT/dC values over a wide range. Thus lecithin is less hydropho-bic than C12E2 and Tween 20 is less hydrophilic than C12E8. UsingdPIT/dC values, we have obtained the following order for the testedfood and cosmetics surfactants (from more to less hydrophilic):Tween (20, 40, 60, 80) > sucrose esters dPIT/dC lecithin > span (20,40, 80, 60) dPIT/dC glycerol esters (Gly-C14, Gly-C10).

This method can be used not only with cosmetic or food sur-factants but with any other pure or crude amphiphilic compoundsand thus can be used to evaluate the amphiphilicity of new surfac-tants. Other parameters as oil and salinity influence on dPIT/dC forwell defined ionic and nonionic surfactants will be discussed in aforthcoming paper.

Acknowledgments

The authors thank the University of The Andes ScholarshipProgram for financing J.F.O. doctoral studies and the Postgradu-ate Cooperation Program PCP no. 20100000305 Green emulsions(FONACIT Venezuela and MAE-France) for sponsoring professorand graduate student exchanges.

References

[1] W.C. Griffin, Classification of surface-active agents by HLB, J. Soc. Cosmet. Chem.1 (1949) 311–326.

[2] J.N. Israelachvili, D.J. Mitchell, B.W. Ninham, Theory of self-assembly of hydro-carbon amphiphiles into micelles and bilayers, J. Chem. Soc. Faraday Trans. 2(72) (1976) 1525.

[3] K. Shinoda, H. Arai, The correlation between phase inversion temperature inemulsion and cloud point in solution of nonionic emulsifier, J. Phys. Chem. 68(1964) 3485–3490.

[4] J.-L. Salager, N. Marquez, A. Graciaa, J. Lachaise, Partitioning of ethoxylatedoctylphenol surfactants in microemulsion–oil–water systems: influence oftemperature and relation between partitioning coefficient and physicochemi-cal formulation, Langmuir 16 (2000) 5534–5539.

[5] D.J. Mitchell, B.W. Ninham, Micelles, vesicles and microemulsions, J. Chem. Soc.Faraday Trans. 2 (77) (1981) 601.

[6] J.T. Davies, A quantitative kinetic theory of emulsion type, I. Physical chem-istry of the emulsifying agent, in: Proceedings of Second International CongressSurface Activity, 1957, p. 426.

[7] K. Shinoda, H. Saito, H. Arai, The effect of the size and the distribution of the

oxyethylene chain lengths of nonionic emulsifiers on the stability of emulsions,J. Colloid Interface Sci. 35 (1971) 624–630.

[8] H. Kunieda, K. Shinoda, Evaluation of the hydrophile–lipophile balance (HLB)of nonionic surfactants. I. Multisurfactant systems, J. Colloid Interface Sci. 107(1985) 107–121.

A: Phy

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

J.F. Ontiveros et al. / Colloids and Surfaces

[9] H. Kunieda, N. Ishikawa, Evaluation of the hydrophile–lipophile balance (HLB)of nonionic surfactants. II. Commercial-surfactant systems, J. Colloid InterfaceSci. 107 (1985) 122–128.

10] H. Kunieda, K. Hanno, S. Yamaguchi, K. Shinoda, The three-phase behaviorof a brine/ionic surfactant/nonionic surfactant/oil system: evaluation of thehydrophile-lipophile balance (HLB) of ionic surfactant, J. Colloid Interface Sci.107 (1985) 129–137.

11] T. Gibson, Phase-transfer synthesis of monoalkyl ethers of oligoethylene gly-cols, J. Org. Chem. 45 (1980) 1095–1098.

12] J.C. Lang, R.D. Morgan, Nonionic surfactant mixtures. I. Phase equilibria inCE–HO and closed-loop coexistence, J. Chem. Phys. 73 (1980) 5849.

13] J. Schlarmann, C. Stubenrauch, R. Strey, Correlation between film propertiesand the purity of surfactants, Phys. Chem. Chem. Phys. 5 (2003) 184–191.

14] A. Pizzino, Inversion de Phase des Emulsions: Relation avec le Comportement àl’Equilibre et Détection par Rétrodiffusion de Lumière, Université Lille 1, 2008.

15] A. Pizzino, V. Molinier, M. Catté, J.-L. Salager, J.-M. Aubry, Bidimensionalanalysis of the phase behavior of a well-defined surfactant (C10E4)/oil(n-octane)/water–temperature system, J. Phys. Chem. B 113 (2009) 16142–16150.

16] A. Pizzino, V. Molinier, M. Catté, J.F. Ontiveros, J.-L. Salager, J.-M. Aubry, Rela-tionship between phase behavior and emulsion inversion for a well-defined

surfactant (C10E4)/n-octane/water ternary system at different temperaturesand water/oil ratios, Ind. Eng. Chem. Res. 52 (2013) 4527–4538.

17] H. Kunieda, Y. Fukui, H. Uchiyama, C. Solans, Spontaneous formation of highlyconcentrated water-in-oil emulsions (gel-emulsions), Langmuir 12 (1996)2136–2140.

[

sicochem. Eng. Aspects 458 (2014) 32–39 39

18] B. Heurtault, P. Saulnier, B. Pech, J.-E. Proust, J.-P. Benoit, A novel phaseinversion-based process for the preparation of lipid nanocarriers, Pharm. Res.19 (2002) 875–880.

19] N. Anton, J.-P. Benoit, P. Saulnier, Particular conductive behaviors of emulsionphase inverting, J. Drug Delivery Sci. Technol. 18 (2008) 95–99.

20] D.J. Mitchell, G.J. Tiddy, L. Waring, T. Bostock, M.P. McDonald, Phase behaviourof polyoxyethylene surfactants with water: mesophase structures and par-tial miscibility (cloud points), J. Chem. Soc., Faraday Trans. 1 79 (1983)975–1000.

21] M. Corti, C. Minero, V. Degiorgio, Cloud point transition in nonionic micellarsolutions, J. Phys. Chem. 88 (1984) 309–317.

22] P. Izquierdo, J. Esquena, T.F. Tadros, C. Dederen, M.J. Garcia, N. Azemar, et al.,Formation and stability of nano-emulsions prepared using the phase inversiontemperature method, Langmuir 18 (2002) 26–30.

23] A. Graciaa, J. Lachaise, J.G. Sayous, P. Grenier, S. Yiv, R.S. Schechter, et al., Thepartitioning of complex surfactant mixtures between oil/water/microemulsionphases at high surfactant concentrations, J. Colloid Interface Sci. 93 (1983)474–486.

24] A. Graciaa, J. Andérez, C. Bracho, J. Lachaise, J.-L. Salager, L. Tolosa, et al., Theselective partitioning of the oligomers of polyethoxylated surfactant mixtures

between interface and oil and water bulk phases, Adv. Colloid Interface Sci.123–126 (2006) 63–73.

25] A.-S. Muller, J. Gagnaire, Y. Queneau, M. Karaoglanian, J.-P. Maitre, A. Bouchu,Winsor behaviour of sucrose fatty acid esters: choice of the cosurfactant andeffect of the surfactant composition, Colloids Surf., A 203 (2002) 55–66.