2012_savary et al._impact of emollients on the spreading properties of cosmetic products a combined...

8/13/2019 2012_Savary Et Al._impact of Emollients on the Spreading Properties of Cosmetic Products a Combined Sensory a

1/8

Colloids and Surfaces B: Biointerfaces 102 (2013) 371378

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces

journal homepage: www.elsevier .com/ locate /colsur fb

Impact ofemollients on the spreading properties ofcosmetic products:A combined sensory and instrumental characterization

Graldine Savary, Michel Grisel, Cline Picard

Universit duHavre, URCOM, EA 3221, FRCNRS3038, 25 rue Philippe Lebon, B.P. 540, 76058 LeHavre cedex, France

a r t i c l e i n f o

Article history:

Received 15 March 2012

Received in revised form 19 June 2012Accepted 19 July 2012

Available online 9 August 2012

Keywords:

Spreading

Cosmetic emulsions

Sensory evaluation

Contact angle

Texture analysis

Correlation

a b s t r a c t

This study deals with the impact ofemollients on the spreading properties ofcosmetic products using a

combined sensory-instrumental approach. To that purpose, three esters and one silicone were selected

and incorporated separately into an oil phase. Different cosmetic o/w emulsions were then prepared

with these different oil phases. Both of them were analyzed by instrumental techniques and in vivo

sensory analyses. A significant effect of the emollient used was established in emulsions and in oil

phases as well. Concerning emulsions, results reveal a clear correlation between in vivo spreading

evaluation and friction coefficient parameters measured by texture analyzer, despite a fairly low cor-

relation coefficient (Pearson coefficient =0.78). Concerning oil phases, characterization of spreading

was done by monitoring the contact angle relaxation of a drop of solution after deposition on a flat

PMMA surface whereas sensory procedure was based on spontaneous spreading ofoil phases onto the

skin. Finally, good correlations between in vivo sensory analysis and instrumental measurements ofboth

oils and emulsions were found, thus promising the possible development ofpredictive tools to evaluate

spreadability.

2012 Elsevier B.V. All rights reserved.

1. Introduction

Sensory properties in skin care formulations mainly result from

ingredients such as emollients, rheology modifiers, emulsifiers and

humectants. Generallyspeaking, emollients, usedat levels between

3 and 20% (w/w)in emulsions, representthe second major ingredi-

ent after water [1]. Emollients can be of varied chemical structures

including esters and silicones. When incorporated in cosmetic

emulsions, esters and silicones are hydrophobic ingredients that

compose part of the oil phase. Despite their dubious reputation,

cyclic siliconesare widely found in skin care products fortheir spe-

cific properties when compared to the other emollients [2]. Esters

belong to a large family of compounds used either as emollient

or as emulsifier in cosmetic emulsions and can be used to replace

silicones [3]. Related to their physico-chemicalproperties, skin-feeleffects of theseemollients arecomplexand can be perceivedduring

and/or afterapplication on skin:gliding, sliding, moisturizing, plas-

ticizing, protecting, conditioning, softening, smoothing, etc. [1,4,5].

Concerning the properties during application on the skin, emol-

lients decrease the friction coefficient due to their lubricant prop-

erties and modify the spreading performance of the product [6].

Corresponding author. Tel.: +33 232 744394; fax: +33 232 744391.

E-mail address: [email protected](C. Picard).

To achieve adequate efficacy and user acceptance of a cos-

metic emulsion, spreading is an important property. Descriptive

sensory analysis of skin care products usually includes attributes

such as difficulty of spreading or slipperiness that are evalu-

ated during application of either pure ingredients [7] or creams

and lotions [8,9]. For example, Parente et al. [2] characterized

eight liquid emollients including esters and silicones (i.e. decyl

oleate, isopropyl myristate, dimethicone, cyclomethicone) for both

attributes. Authors compared properties of emollients alone but

they did not study their characteristics when included in cosmetic

emulsions.

However, sensory analyses are time-consuming and require an

available and well-trained panel of assessors. Therefore, different

studies were done inorderto establisha relationshipbetweentext-

ural attributes and structural or physical characteristics [1012].In the field of food sciences, many studies have been carried out

to develop an instrumental approach especially using texture pro-

file analysis [1315]. However, nowadays the number of studies

on spreading performed in the cosmetic domain remains limited.

Kusakari et al. [16] developed a measuring device to evaluate the

frictional force as an indicator of the spreading resistance whereas

DiMuzio et al. [17] established a correlation between spreadability

and rheological parameters obtained from stress sweep and creep

tests on o/w emulsions.

The first goal of the present study was to test the poten-

tial development of texture analysis to evaluate the spreading

0927-7765/$ seefrontmatter 2012 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.colsurfb.2012.07.028
http://localhost/var/www/apps/conversion/tmp/scratch_6/dx.doi.org/10.1016/j.colsurfb.2012.07.028http://localhost/var/www/apps/conversion/tmp/scratch_6/dx.doi.org/10.1016/j.colsurfb.2012.07.028http://www.sciencedirect.com/science/journal/09277765http://www.elsevier.com/locate/colsurfbmailto:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_6/dx.doi.org/10.1016/j.colsurfb.2012.07.028http://localhost/var/www/apps/conversion/tmp/scratch_6/dx.doi.org/10.1016/j.colsurfb.2012.07.028mailto:[email protected]://www.elsevier.com/locate/colsurfbhttp://www.sciencedirect.com/science/journal/09277765http://localhost/var/www/apps/conversion/tmp/scratch_6/dx.doi.org/10.1016/j.colsurfb.2012.07.028


2/8

372 G. Savary et al. / Colloids and Surfaces B: Biointerfaces 102 (2013) 371378

properties of cosmetic emulsions. The second target of this study

was to establish an instrumental measurement to characterize the

spreadingattributes of the various oil phases used to formulate the

o/w emulsions.

In their study, Gorcea and Laura [1] assessed the physicochemi-

cal properties of four cosmetic emollient esters in vitro to correlate

with in vivo sensory attributes. Spreading properties were charac-

terized by spreadingvalues on Vitro-Skin, surface tension,contact

angle and viscosity measurements. Indeed, surface wetting and

absorption dynamics are two key phenomena that govern the

spreading of a fluid on a given surface [18].

Wettability phenomena can be evaluated by studying the con-

tact angle formed at the air/solid/liquid contact point when a drop

of liquid is deposited on the solid surface [19]. When a liquid drop

is placedin contact with a flat, horizontal substrate, capillary forces

drive the interface spontaneouslytowards equilibrium. As the drop

spreads, the contact angle changes from its initial maximum value

at the very first instant of contact, towards its final equilibrium

angle 0 in the case of partial wetting, or 0 if the liquid com-

pletely spreads and wets the solid. The results more or less fit a

master curve showing relationships between contact angle versus

time [20]. Several parameters affect wettability [1922]: chemical

characteristics of the liquid and substrate, roughness and hetero-

geneity of the solid, viscosity, surface tension, density and volumeof the drop of the liquid and its potential evaporation.

In the present study, three esters and one silicone ingredients

chosen as major emollients used in the cosmetic industry were

studied as components of the oil phase of the given emulsions. The

impact of emollients on the spreading properties of the oil phases

and of the o/w emulsions formulated with these oil mixtures was

studied by a combined sensory-instrumental approach.

Spreading of emulsions was first investigated by in vivo tests

and in vitro methods including texture analysis. Then spreading

and penetration of the oil phases were characterized using sensory

evaluations and contact angle measurements. Results were then

analyzed and compared, and a correlation matrix was drawn up to

establish the relationships between both sensory and instrumental

measurements.

2. Material and methods

2.1. Preparation of the emulsions

Table 1 reports the name and physico-chemical properties of

the cosmetic grade esters and the silicone selected for this study.

Esters 1 and 2 are ester or diester from penta- or dipenta-

erythrityl alcohol respectively. Theyexhibita star-shaped structure

and a relatively high molar mass, so they are generally selected for

the consistency and texture they bring to the formulations. Ester 3

is a linear ester with a low molar mass. It exhibits properties very

close to those of silicone: improvement of spreading on skin, silkyafter-feel sensation on skin. Silicone belongs to the so-called group

of cyclic volatile dimethylsiloxanes or cyclomethicones. It is well-

known in cosmetics for producing a silky after-feel sensation and

improvingspreading on skin. Esters 1, 2 and3 were kindlygivenby

Starinerie Dubois (France) and silicone was obtained from Evonik

Goldschmidt (Germany).

A standard emulsion and four test emulsions were prepared by

varying the composition of the oil phase in order to analyze the

effectof each emollient by itself (Table 2). Ingredients were chosen

in order to respect two main criteria:

1. Those that produce a stable emulsion close to an industrial cos-

metic one in terms of complexity and number of ingredients

2. Those whose oil phase is a good solvent to solubilize the differ-

ent emollients chosen. Thus, at room temperature, a liquid oil

phase with or without emollient is obtained, making it easier to

characterize.

Forthe standardemulsion(200g),the oilphase (A) was firstpre-

pared by mixing the ingredients as listed in Table 2 and by heating

the mixture to 75C under stirring. In the meantime, the aque-

ous phase (B) was prepared using purified water (type 1). The mix

of water and glycerine was slowly sprinkled with carbomer pow-

der and left 20min without stirring. Then the aqueous phase was

heatedto 75 C.When bothphases were at75 C,the water loss due

to evaporation was compensated by adding water to the aqueous

phaseand theoilphasewas thengently addedto PhaseB understir-

ringfor 3 min with a D7801 homogenizer (Ystral GMBH, Germany).

Theemulsionwas then continuouslystirred at300 rpmusing a Tur-

botest motor generator with a 65mm turbine (VMI, France) until it

cooleddownto 50 C. Sufficient amountof 1 M aqueous NaOH solu-

tion (CarloErba, Italy) was then added to yield a final pH between

6.2 and 6.5 in the preparation. The stirring rate was then increased

to 500 rpm. Finally, the preservative (Phase C) was added at 40C

and the emulsion was kept under stirring for an additional period

of 5min until its temperature dropped to ambient temperature.

For the test emulsions with ester, the preparation was exactlythe same except for the composition of the oil phase (Table 2).

For the test emulsion with silicone, the emollient was added

with the preservative at 40 C in order to avoid its evaporation.

Samples were stored at 4 C during three months at the most.

2.2. Instrumental spreading characterization

2.2.1. Contact angle measurements

Spreading of emollients solubilized in the oil phase was inves-

tigated by the sessile drop method (Digidrop GBX goniometer). To

this purpose, a drop of product is formedat theendof a syringe until

it gently falls under its own weight on the solid support, and then

spreads onto it. A high speed camera (25 images/s) recorded the

evolution of the contact angle just after deposition and during theentire spreadingprocess. The solid support chosen for these exper-

iments was polymethyl methacrylate (PMMA) plates (Helioplate

HD2, Helioscreen, France). This support is commonly used in cos-

metics for the in vitro determination of SPF and offers several

advantages for our concerns, as compared to artificial skin: it is

less expensive, easier to handle in terms of surface preparation and

cleaning,it is reproducible in termsof surface chemistryand rough-

ness and it is reusable. For each ingredient, both the right and the

left contact angles were measured using the Windrop++ software,

and the corresponding average contact angle was followed over a

period of timeuntil stabilization. The final results correspond to the

mean of three reproducible kinetic experiments. All measurements

were realized at room temperature.

2.2.2. Texture analysis

Spreading of cosmetic creams was measured on a TA.XT

plus Texture Analyzer (Stable Microsystems, United Kingdom),

equipped with the A/FR Friction rig module at room temperature.

The principle of the test consists in displacing at a constant speed

of 3 mm s1 a PMMA plate (Helioplate HD6) surmounted by a

square weight (207.9g, 6.2cm length), on a static plate covered

with a polypropylene sheet. Prior to experiment, using a pipette

designed for creamy products (Microman, Gilson, France),4 drops

of cream of 40L each were deposited on the under-side of the

weight in order to design a 22 cm2. The force required to move

the weight over a length of 6 cm was monitored over time. Textu-

ral data related to friction andspreading of creams were calculated

from a graph obtained during experiments (Fig. 1). The positive


3/8

G. Savary et al. / Colloids and Surfaces B: Biointerfaces 102 (2013) 371378 373

Table 1

Names and physico-chemical properties of the emollients used.

Code INCI name Structure Mw logPa

Ester 1 Pentaerythrityltetraethylhex-

anoate

640.9 11.3

Ester 2 Dipentaerythrityl

pentaisononanoate

(content > 97%) 949.0 12.0

Ester 3 Propanediol

dicaprylate

328.5 6.4

Silic one Cyclopentasiloxane,

cyclohexasiloxane

(content > 95%) 370.8 5.2

a logPvalues were obtained or calculated from http://pubchem.ncbi.nlm.nih.gov.

area between 0 and 14s was determined for each cream and the

mean value of 3 reproducible experiments was reported and con-

sidered as an instrumental parameter characterizing the spreading

of creams (TEM).

2.2.3. Physico-chemical characterization of ingredients, oil

phases and creams

Density of pure ingredients and oil phases containing one

ingredient was characterized with a portable densimeter (Mettler

Toledo, France).

Surface tension of pure ingredients and ingredients in the oil

phase was determined by the pendant drop method (Digidrop GBX

goniometer). Surface tension value was the mean value obtained

when measuring ten drops of the same volume. Both measure-

ments were done at room temperature.

Rheological properties in flow mode of pure ingredients, oil

phases and creams were measured on a AR 2000 rheometer

(TA Instruments, USA) at 25C. Cone-and-plate measuring sys-

tem geometries were used for these characterizations, and chosen

depending on the viscosity range, respectively: 60mm diameter

aluminium coneplate with an angle of 01957, 60mm diameter

Table 2

Composition of thestandard and test emulsions.

Phase Ingredients (INCI name) Suppliers Weight (%, w/w)

Standard emulsion Test emulsions

A Coco caprylate/caprate Starinerie Dubois (France) 14 8

Isohexadecane Croda (England) 8 3

Cetyl alcohol (and) glyceryl stearate (and) PEG-75

(and) ceteth-20 (and) steareth-20

Starinerie Dubois (France) 5 5

Emollient See Table 1 0 11

B Aqua 61.7 61.7

Glycerine CarloErba (Italy) 5 5

Carbomer Lubrizol (Belgium) 0.3 0.3

C Pr opylen e glycol (and) me th ylpr opan edio l( and)

potassium sorbate (and) methylisothiazolinone

Biophil (Italy) 1 1
http://pubchem.ncbi.nlm.nih.gov/http://pubchem.ncbi.nlm.nih.gov/


4/8


0

5

10

15

20

25

30

35

0 2 4 6 8 10 12 14

F

orce

(g)

Time (s)

Fig. 1. Texture profile obtained with the standard emulsion.

acrylic coneplate with an angle of 20153 and 40mm diameter

steel coneplate with an angle of 059. Flow curves were recorded

with a continuous ramp test in shear rate (0.0117,000s1) during

5min. Viscosity values for pure ingredients and oil phases (VOP)

were selected in the Newtonian region, while for the cosmetic

emulsions, viscosity values (VEM) at a shear rate of 1000s1 were

chosen.

2.3. Sensory analyses

Sensory analyses were carried out to evaluate spreading of the

five emulsions (SEM) on the one hand, and spreading and penetra-

tion of the five corresponding oil phases (SOPand POPrespectively)

on the other hand. Each of these three attributes was evaluated

separately. The sensory panel consisted of ten female volunteers

from 20 to 40 years old. The assessors were trained in two train-

ing sessions followed by two test sessions in which samples wereevaluated twice. Samples were labelled with three-digit random

numbers and were randomly presented at room temperature in

5 mLopaque vials. Sensory attributes were assessed on the inter-

nal side of the forearm according to well-defined procedures as

described below. Tests were carried out successively on the left

and rightarms.Prior toand between each sampletest,theassessors

cleaned the skin of theirforearm with skin cleanser, then waterand

finally dried it carefully. Testing took place at the sensory facilities

of theUniversityof Le Havre,France. Analysis conditions were con-

trolled before sessions, in particular the lighting in sensory booths.

Because ingredients and emulsions possess quite different

physicochemical properties, three different evaluation procedures

were defined, all with strict protocols as described below.

2.3.1. Evaluation of spreading for emulsions

Spreadability was defined as the ease of moving the cosmetic

emulsion over a given distance. The distance defined on the skin

was of 6c m. The assessor put down 50L of the sample at 6 cm

from the arm bend with a M250 positive displacement pipette

(Microman, Gilson, France). With the index finger, the sample is

spread only once towards the hand for a fixed distance of 6cm.

The assessor evaluated the force while applying the sample onto

the skin surface and indicated the score for each sample on a scale

from 0 to 9 with0.5 increments and verbal anchor points. To quan-

tify this attribute, assessors were trained to theresponsescalewith

internal referent samples (scores 1 and8). Duringthe test sessions,

assessors were authorized to re-test either samples or references

several times.

2.3.2. Evaluation of spreading for oil phases

Spreading of the oil phases was defined as the surface of skin

covered by the sample in one minute [23]. The larger the covered

surface, the easier the sample spreads. The assessors were asked to

put down 10L (10L syringe, SGE, Australia) of sample at 6 cm

from the arm bend and to keep the forearm horizontal, without

moving, for one minute. According to the anatomy of assessors

forearms,circular or ellipsoidal formswere coveredby the oils.Out-

lines of theform covered by thespread substance were then drawn

with a pen for skin and the lengths of the semi-major and semi-

minor axes of the ellipse were measured. The spreading of each oil

phase was then calculated according to the following equation:

SOP = a b

VS tS

where a and b are the lengths of the semi-major and semi-minor

axes respectively (in mm), VS is the volume of sample (in L) and

tS, the time (in min).

2.3.3. Evaluation of the oil phase penetration

Penetration of the oil phases was defined as the time necessary

to observe the total disappearance of the sample put down on skin.

0.5L of sample was put down at 6c m from the arm bend. Theassessor started the chronometer and kept her forearm still under

lighting. The assessor was asked to observe the deposit and indi-

cated when she could not observe the product any more on the

skin as theconsequence of its absorption into the skin. The time for

total penetration was recorded and the penetration rate was then

calculated according to the equation:

POP =tP

VP

where tPis the time for penetration (in min) and VP, the volume of

sample (inL).

2.4. Data analysis

Analyses of variance (ANOVA) were conducted to check dif-

ferences between properties of each emollient. One-way ANOVAs

were performed on each instrumental measurement and two-way

ANOVAs were computed on each sensory attribute with samples

and assessors as variables. In case of significant difference at 95%,

Tukeys HSDpost hoccomparisons were then carried out to inves-

tigate the sample effect.

The correlation matrix of the mean values was used in order

to establish and interpret the relationship between sensory and

instrumental measurements.

All mathematical and statistical analyses were performed using

XLSTAT software.

3. Results and discussion

3.1. Sensory spreading of emulsions

Spreading of a cosmetic cream is a determinant textural

attribute that governs the performances of the product during its

application on the skin. In order to evaluate the impact of the

emollient on this attribute, a sensory analysis was carried out

on the five creams formulated as described above, by a trained

panel. The precise protocol of testing was established with physi-

cal references to distinguish and evaluate specifically this textural

attribute. Spreading of emulsions (SEM) scored between 0 and 9 is

indicated in Table 3. We observed that all emulsions were eval-

uated to be significantly different on this attribute with scores

ranging from 2.50 to 8.05. Spreading was higher with Ester 1 and


5/8


Table 3

Sensory properties and instrumental measurementsfor thedifferent emollients incorporatedeither in theoil phase (OP) or in theemulsion (EM).

Standard Ester 1 Ester 2 Ester 3 Silicone

Sensory properties

Oil phase penetration POP (min/L) 5.36c 10.1b 14.3a 5.85c 5.83c

Oil phase spreading SOP (mm2/min/L) 49.9b 29.0c 18.6d 43.0b 78.4a

Emulsion spreading SEM(score from 0 to 9) 3.55d 8.05a 2.50e 6.05b 4.95c

Instrumental measurements

Oil phase viscosity VOP (Pa s) 0.0058c 0.0152b 0.0344a 0.0063c 0.0046c

Contact angle at 0s A0s () 38.3bc 45.5b 55.6a 39.2bc 32.0c

Contact angle at 4s A4s () 12.3ab 16.2a 17.1a 8.97b 5.70b

Emulsion viscosity VEM(Pa s) 0.08c 0.16bc 0.61a 0.11c 0.24b

Texture of emulsion TEM(g s) 242c 112e 416a 324b 180d

Differentletters in thesame row indicate a significant difference between samples atp


6/8


Table 5

Physico-chemical propertiesof ingredients either pureor dispersedin the oil phase.

Densi ty Su rface tension ( mN/m) Viscosity (Pa s ) a t 1 0 s1

Pure ingredients

Ester 1 0.963 28.090.59 124103

Ester 2 0.976 Nd 3630103

Ester 3 0.932 20.160.09 10103

Silicone 0.957 17.370.07 4103

Oil phases

Ester 1 0.894 26.360.59 15103

Ester 2 0.900 25.550.91 34103

Ester 3 0.880 27.961.19 6103

Silicone 0.894 19.650.34 5103

Standard 0.827 25.030.91 6103

experiments, respectively, that could be different and induce dif-

ferent deformations of the products.

3.3. Characterization of the oil phases

To establish correlations between instrumental techniques and

in vivo sensory analysis in the case of the oil phases, some

experiments were carried out on the oil mixtures alone. Vari-

ous characterizations were performed including sensory spreadingand penetration evaluations, viscosity, and contact angle measure-

ments.

A previous preliminary study [25] has shown that the spread-

ing of pure ingredients, characterized by sensory and instrumental

analyses, was mainly influenced by the physical state of the ingre-

dients. Thus, surface tension and dynamic contact angle were not

measurable for highly viscous ingredients like Ester 2 and sensory

evaluation became particularly difficult as products could be liq-

uid as well as close to solid state. Table 5 sums up the interest

of studying the emollients solubilized in the oil phase (Phase A,

Table 2).

Incorporation of esters or silicone into the oil phase both

increases the oil phase density and leads to a lower density dif-

ference between the different emollients.Due to solubilization, and taking into account standard devia-

tion values, surface tensions of esters alone and oil phases fell into

the same range. Surface tension of silicone remained the lowest,

nevertheless gaps between emollients were strongly reduced.

Lastly, all emollients, either pure or included in the oil solution,

showed a Newtonian behaviour (curves not shown here). Com-

pared to pure ingredients, solubilization in the oil phase shifted

viscosity values (VOP) of Esters 1 and 2 to muchlower values. They

still remained significantly higher than that of Ester 3.

3.3.1. Sensory characterization of oil phases

Spreading of the five different oil phases (SOP) was evaluated by

the assessors according to a different procedure from the one used

for spreading evaluation of the emulsions. When deposited on theskin surface, the oil phases exhibited differentbehaviours from the

emulsions as the consequence of the effect of consistency. Emul-

sions formed a mound that assessors spread on the skin, whereas

the oil phases spread spontaneously without requiring any action

for the assessors. Therefore, oil phases promptly formed an oval

stain when deposited on the skin. Each sample was characterized

by a shiny appearance. The assessors were then able to visually

evaluate the surface of skin covered by the product by measuring

the corresponding area one minute after deposition.

During this time, dermal penetration of emollients may hap-

pen. As a result, we decided to ask the panellists to assess also skin

penetration of the five oil phases (POP). This property was evalu-

ated as the time necessary to observe the total disappearance of

the shiny stain formed by the sample deposited on the skin. In

this study, we assume that there was no difference in terms of

volatility between samples. This supposition was made since the

emollient represented only 11% (w/w) of the composition of the

oil phase. Among ingredients, silicone was the most volatile with a

vapour pressure of 0.2 mmHg at25 C and a boiling temperature of

210 C against for instance, Ester 3, 5.93106 mmHg and 352 C,

respectively. Measurements of skin penetrationPOPin Table 3 indi-

cate times above 5.36min. As a consequence, we assumed that the

duration of the spreading evaluation (1min) was short enough to

consider that there was no significant penetration occurrence of

samples into the stratum corneum.

Generally, the penetration ability of the emollients into the

lipophilic stratum corneum depends on both the polarity and the

molecular size of the compound [26]. This is confirmed here, pen-

etration variation among oil phases is mainly related to both the

polarity and the molecular size of ingredients (Table 1); the higher

the values of logPandMw, the longer the time needed to penetrate

into the stratum corneum as illustrated for Esters 1 and 2.

Concerning spreading, results reveal a large effect of the com-

position of the oil phase (SOP in Table 3). Values ranged from

18.6 to 78.4mm2/min/L, for Ester 2 and silicone respectively.

The standard oil phase displays an intermediary spreading of

49.9mm2/min/L. As discussed by other authors [5], the ingredi-

ents with higher molecular weight exhibit lower spreading values,as observed among esters (Table 1).

3.3.2. Instrumental characterization of oil phases spreading

Contact angle measurements for standard or emollient-oil mix-

tures were performed from initial time, as the drop of liquid was

deposited on PMMA surface, until it reached a measurable con-

stant value. At t= 0, initial contact angle values are respectively:

Ester 2 (55.63.9), Ester1 (45.52.4),Ester3(39.2 1.7), standard

(38.31.9), silicone (32.00.3). This emphasizes different affinity

of ingredients with the support. The curves of contact angle versus

time then exhibit a rapid exponential decrease, corresponding to

the rapid spontaneous spreading of each solution on the support,

until an asymptotic contact angle value is reached (after 4 s for sili-

cone(approx.6) and Ester 3 (approx.9),after14 s for Ester 2,Ester1 (approx. 12) and standard (approx. 11)). Selection of the data

for contact angles were thus restricted to the shortest time period

necessary to obtain one of these asymptotic values, that is 4s in

the case of silicone. Fig.2 shows the relaxation of the contact angle,

over a time period of 4 s, for each emollientoil mixture as well as

the profile for the corresponding drop of the five systems at t= 0 s

(immediately after deposition) and t=4s. Table 3 also displays the

values for the contact angle just after deposition (A0s) and at t= 4 s

(A4s).

Study of spontaneous spreading of a liquid on a solid surface

involves different key parameters that are: surface tension, vis-

cosity and density of liquid, surface free energy, roughness and

heterogeneity of solid support [19]. Due to solubilization of the

emollients in the oil phase, the impact of density, viscosity andsurface tension on spreading is actually reduced. When consider-

ing these last parameters for the different mixtures tested, only

differences in terms of surface tension and viscosity may explain

the contact angle results. Therefore, surface tension and volatility

differences between silicone and Ester 3 can explain the different

spreading that is observed. In the case of the ester molecules dis-

persed in oil and also oil phase, the main parameter governing the

different spreading behaviour is undoubtedly the viscosity.

It is also necessary to keep in mind that a droplet can evaporate

somewhat,depending on the wettingbehaviour and the hydropho-

bicity/hydrophilicity of the substrate [21]. In all cases, evaporation

phenomena remain negligible during the time of experiments

[21,22]. This appears evident as Orejon et al. [21] have demon-

strated that a typical evaporation time for ethanol droplets was


7/8


Fig. 2. Contact angle evolution of oil phasesdeposited on PMMA surface.

between 90s and 300s for hydrophilicand hydrophobic substrates

respectively.Initial and equilibrium contact angles values show an almost

complete wetting of the liquids on the PMMA substrate, this being

in good agreement with the roughness of PMMA plates (2m)

on the one hand, and surface free energy of the substrate on the

other hand. Alteraifi and Sasa [22] reported a critical surface ten-

sionof39mN/mforasmoothPMMA surface,which is characteristic

of intermediary properties (between PTFE around 20mN/m and

soda-lime glass 70mN/m). This is also in good agreement with the

relatively lowpolarity of theoilyliquids testedin the present study.

These measurements appear particularly relevant to anticipate

the spreading ability of a sample, as evidenced by the correla-

tion coefficients between SOP and A0s or A4s (Table 4). All these

coefficients are negative since the contact angle was inversely pro-

portional to the surface of PMMA covered by the product. Thanksto this correlation, it is possible to envisage predicting the spread-

ing of different oil phases using an instrumental method instead

of sensory evaluation. Moreover, present contact angle measure-

ments were conducted with a PMMA surface, a fairly reproducible,

easy-to-use and low cost material when compared to real as well

as synthetic skin.

Atthis step,it is possible tocompare surface properties ofPMMA

and skin surface, in terms of surface tension and chemical compo-

sition, thanks to data from the literature. A surface free energy is

reported for human skin varying from about 27 to 38.2mJm2 in

the dorsal region of skin index [27] but also on the volar forearm

or forehead skin [28]. Lowest values are obtained for a dry skin

submitted to a cleaning with delipidizing solvent and highest ones

were obtained on skin washed with soap and rinsed thoroughlyunder tap water or unwashed skin during 16h. Concerning PMMA

surfaces, values are similar or higher than skin, data in the litera-

ture ranged from 39 to 48mJ m2 depending on methods used and

cleaning of the surface [22,27,29]. Taylor et al. [30] reported even

valuesrangingfrom46to54mJ m2 related to roughness andtreat-

ment of PMMA surfaces. Both types of surface exhibit also a high

dispersive component close to total surface free energy, an acidic

component close to zero, a basic component varying from about

5 to 20 mJ m2 for PMMA and from about 0 to 28 for human skin.

This can be related to a similar chemical surface composition: epi-

dermal lipids on the one hand and ester methyl groups [31] on the

other handfor PMMAsurface.Even if PMMAsurface doesnot mimic

stratum corneum in terms of chemical composition and porosity,

contact angle measurement is a pertinentmethod to measure pure

spreading properties of oils. However, it is important to take into

account that such physical characterizations are not easy or pos-sible with any kind of sample as for example with highly viscous

liquids or the cosmetic emulsions of the present study.

3.4. Relationships betweenoil phase and emulsion properties

When formulating a cosmetic product, a major point is to know

how the properties of the end product can be governed by the

intrinsic characteristics of ingredients. Our study shows that the

relationship between the properties of emollients and the cor-

responding emulsions is not necessarily obvious. Concerning the

viscosity of emulsions (VEM), we observed a significant correlation

with VOP(Table 4). In this case, it is possible to anticipate the viscos-

ityof thecreamaccording to thecomposition of the oilphase. As an

example,Ester2 induced the highest viscosityfor both the oilphase

and the corresponding emulsion. However, concerning spreading,

thereis no linear correlation between SEMand SOP(Table 4) asan oil

phase characterizedby a highspreading performance does not nec-

essarily generate a cosmetic emulsion with the same performance;

this is illustrated with the silicone ingredient.

Emulsions are complex structures that exhibit original

behaviours influenced by many parameters including composition

and proportion of the dispersed and continuous phases, droplet

size, emulsifier type, composition and structure of the interface,

and interactions between ingredients [32]. Concerning spreading

of emulsion,our experiments shouldbe completedwith other data

to develop a predictive model with multiple parameters to be con-

sidered.

3.5. Relationship between sensory and instrumental

measurements

In conclusion, as shown in Table 4, penetration of the oil phase

(POP) was best correlated with viscosity of the oil phase (VOP) fol-

lowed by contact angle at 0 s (A0s). Spreading of the oil phase

was correlated with both A0s and A4s contact angles among the

instrumental measurements. The sensory spreading of emulsion

correlated best with the spreading of emulsion instrumentally

measured in terms of work required to move a weight on a sur-

face. Consequently, all these results made it possible to implement

physical measurements of sensory evaluations of either oil phases

or emulsions to avoid time-consuming in vivo methods.


8/8


Acknowledgements

The authors thank different contributors. Authors sincerely

acknowledge Clmentine Lachaud for instrumental measurements

and Nathalie Loubat-Bouleuc (Starinerie Dubois, France) for help-

ful discussions.

References

[1] M. Gorcea, D. Laura, Cosmet. Toilet. 125 (12) (2010) 26.[2] M.E. Parente, A. Gambaro, G. Solana, J. Cosmet. Sci. 56 (2005) 175.[3] N. Loubat-Bouleuc, OCL 11 (2004) 454.[4] E.Kim, G.W. Nam, S.Kim, H. Lee, S.Moon,I. Chang,SkinRes. Technol. 13 (2007)

417.[5] B.A. Salka, Cosmet. Toilet. 112 (1997) 101.[6] S. Nacht, J.A. Close, D. Yeung, E.H. Gans, J. Soc. Cosmet. Chem. 32 (1981) 55.[7] M.E. Parente, A. Gambaro, G. Ares, J. Sens. Stud. 23 (2008) 149.[8] I.S. Lee, H.M. Yang, J.W. Kim, Y.J. Maeng,C.W.Lee,Y.S. Kang, M.J. Rang, H.Y. Kim,

J. Sens. Stud. 20 (2005) 421.[9] G.V. Civille, C.A. Dus, Cosmet. Toilet. 106 (1991) 83.

[10] R.L. Goldemberg, C.P. De La Rosa, J. Soc. Cosmet. Chem. 22 (1971) 635.[11] R. Brummer, S. Godersky, Colloids Surf. A: Physicochem. Eng. Aspects 152

(1999) 89.[12] T.F. Tadros, S. Lonard, C. Verboom, V. Wortel, M.C. Taelman, F. Roschzttardtz,

in:T.F. Tadros(Ed.), Colloids in Cosmeticsand Personal Care, vol. 4, Wiley-VCHVerlag GmbH & Co. KGaA, Weinheim, 2008 (Chapter 8).

[13] G.Hough, A.N. Califaro,N.C. Bertola, A.E. Bevilacqua, E.Martinez, M.J. Vega, N.E.

Zaritsky, Food Qual. Pref. 7 (1996) 47.

[14] J.F. Meullenet, B.G. Lyon, J.A. Carpentier, C.E. Lyon, J. Sens. Stud. 13 (1998) 77.[15] R.A. de Wijk, L.J. van Gemert, M.E.J. Terpstra, C.L. Wilkinson, Food. Qual. Pref.

14 (2002) 305.[16] K. Kusakari, M. Yoshida, F. Matsuzaki, T. Yanaki, H. Fukui, M. Date, J. Cosmet.

Sci. 54 (2003) 321.[17] A.M. DiMuzio, E.S. Abrutyn, M.Y. Cantwell, J. Cosmet. Sci. 56 (2005)

356.[18] R.C. Daniel, J.C. Berg, Adv. Colloid Interface Sci. 123126 (2006)

439.[19] G. Kumar, K.N. Prabhu, Adv. Colloid Interface Sci. 133 (2007) 61.[20] M. de Ruijter, T.D. Blake, A. Clarke, J. De Coninck, J. Petrol. Sci. Eng. 24 (1999)

189.[21] D. Orejon, K. Sefiane, M.R. Shanahan,Langmuir 27 (2011) 12834.[22] A.M. Alteraifi,B.J. Sasa, J. Adhes.Sci. Technol. 20 (12) (2006) 1333.[23] U. Zeidler, Fette Seifen Anstrichmittel 87 (1985) 403.[24] H.A. Barnes,in: D. Petsev (Ed.), Emulsions: Structure,Stabilityand Interactions,

Elsevier, Albuquerque, 2004 (Chapter 18).[25] L. Gilbert, C. Picard, G. Savary, M. Grisel, Paper 0266. Proceedings of 5th World

Congress on Emulsion, 1214 October 2010, Lyon, 2010.[26] Y. Schiemann, M. Wegmann, P. Lersch, E. Heisler, M. Farwick, Colloids Surf. A:

Physicochem. Eng. Aspects 331 (2008) 103.[27] A. El-Shimi, E.D. Goddard, J. Colloid Interface Sci. 48 (2) (1974) 242.[28] A. Mavon, H. Zahouani, D. Redoules, P. Agache, Y. Gall, Ph. Humbert, Colloids

Surf. B: Biointerfaces 8 (1997) 147.[29] J.K. Chen, S.W. Kuo, H.C. Kao,F.C.Chang, Polymer 46 (2005) 2354.[30] R.L. Taylor, J. Verran, G.C. Lees, A.J.P. Ward, J. Mater. Sci.: Mater. Med. 9 (1998)

17.[31] J. Wang, C. Chen, S.M. Buck, Z.Chen, J. Phys. Chem. B 105 (2001) 12118.[32] P.E. Miner,in: D. Laba (Ed.),RheologicalProperties of Cosmeticsand Toiletries,

Marcel Dekker, New York, 1993 (Chapter 13).

2012_savary et al._impact of emollients on the spreading properties of cosmetic products a combined...

Documents