h..
CHINE-SE JOURNAL OF PHYSICS
VOL. 30, NO. 5
OmOBER 1992
Studies of Formation and Interface of Oil-Water Microemulsion
Hung-Jang Gi’, Shou-Nan Chen2, Jenn-Shyong Hwang3, C. Tien3,
and M. T. Kuo3
‘D epaltment of Chemisq, New York University, New York, hY10003, USA.
2Department of Chemisby, National Cheng Kimg University, Tainan, Taiwan 701, R.O.C.
3Departrnent of Physics, National Cheng fing University, Tainan, Taiwan 701, R.O.C.
(Received January 17,1992; revised manuscript received April 29,1992)
Oil-in-water microemulsions were prepared by the titration method. A mixture of fatty acid
and oil was added to a caustic solution to produce a microemulsion, which was then titrated with
a cosurfactant, an alcohol, until the system turned clear. It was found that as the chain length of
the surfactant increased, microemulsions with significant transmittances by visible spectrum could
be formed with oils of longer chain lengths. It was also found that different alcohols affected the
formation of microemulsions in different ways. The best results, in terms of the greatest percent
.
transmittance coupled with the widest range of oil (dispersed in water) concentration, were
obtained from short or branched alcohols.
The properties of the transparent system, n-cyclohexadecanone/margaric acid/4-methylcyclohexanol (4-methyl-1-pentanol)/water was investigated by photon correlation spectroscopy.
Droplet sizes varied from 14 nm to 46 nm in diameter, and the surfactant cross-sectional area was
calculated to be approximately 0.017 nm2 to 0.029 nm2 per molecule. This implied that surfactant
molecules at the interface were incorporated by the cosurfactant molecules.
I. INTRODUCTION
The term microemulsion introduced by Schulman and co-works’ may be one of those unfortunate words, all too common in science, which arise from the necessity to name a
phenomenon before it is completely understood and remain as a source of semantic confusion.
The spontaneously formed, transparent water-in-oil (W/O) systems studied by Schulman and his
collaborators2*3V4Ts76 contained spherical particles with diameters in the range of 10 nm to 100
nm, and were thought to be in a state of thermodynamic equilibrium. Since emulsions are normally considered to be a heterogeneous mixture of phases kinetically but not thermodynamically
stable, there has been a prevalent notion that microemulsions are emulsions with unique properties (e.g. small particle sizes transparent to light).7)8 Winsorgpl’ studied similar systems and
together with Palit et al.” characterized the clear phases obtained as solubilized systems consisting of swollen micellar units. Adamson12 prefers the term micellar emulsions to microemul665
.,
.
01992 THE PHYSICAL SOCIETY
OF THE REPUBLIC OF CHINA
666
STUDIES OF FORMATION AND INTERFACE OF OIL-WATER...
VOL. 30
sions to emphasize the micellar nature of these systems as well as their emulsion-like properties,
i.e., light scattering, similar composition, and the method of formulation. On the other hand,
Fowkes13 regards Schulman’s original definition appropriate. But, Ekwall et al. recommend that
Schulman’s microemulsions be considered as solubilized micellar solutions.14715 Nevertheless,
the opinion16Y’7 persists that microemulsions are true dispersions of one liquid within another
and that their formation is not the same process as the solubilization of the oil phase be the interior of the micelles. Extensive investigations of the phase equilibria and structures in ternary
system OS of amphiphiles 18719920 reveal that the phenomena of micelle formation and solubiliza-
.
tion may be have a common mechanism.
Interest in microemulsions has intensified because of their practical applications. Many
reviews of microemulsions have appeared. Among the earliest, Osipod’ pointed out the difficulty of distinguishing between microemulsions and solubilized systems. Reviews by Prince22
and Rosano expound the theories of the Schulman school while a recent review by Shinoda
and Friberg24 contrasts some of the differences in opinions with reference to their work on
solubilized phases.
The details of the various theories of microemulsions discussed above are amplified in this
paper.
II. EXPERIMENTAL
11-l. Microemulsion Preparation
The surfactants used in this research were sodium soaps of straight-chain primary fatty
acids. The fatty acids, cyclohexadecanone, cetyl alcohol and sodium hydroxide were purchased
from E. Merck Chemical Company (Tainan, Taiwan).
The effect of pH on the type of initial coarse emulsion was determined in the following
manner. A solution of 0.003 mole of fatty acid in cetyl alcohol or cyclohexadecanone was added
to 10.0 mL of distilled water with continuous stirring. This mixture was then titrated with 1.0 N
NaOH in water while the pH was monitored. The good coarse emulsions were obtained above
pH 11.8 and 11.5, corresponding to cyclohexadecanone and cetyl alcohol. The high pH were
necessary to insure the complete ionization of the fatty acid.2s From these preliminary experiments it was concluded that 0.150 N NaOH produced a sufficiently high pH to insure complete
ionization of the acid and formation of the suitable emulsions.
Microemulsions were prepared at 27°C by a titration method.“6 As prepared above, the
fatty acid was dissolved in the soil phase and then the aqueous NaOH solution was mixed with
it. The resulting coarse emulsion was then titrated with the cosurfactant to form the microemulsion.
VOL. 30
HUNG-JANG GI, SHOU-NAN CHEN. JENN-SHYONG HWANG...
467
X1-2. Photon Correlation Spectroscopy
Particle size measurements were performed using a photon correlation spectrometer. The
instrument was shown in the precious paper. 31j32 When the light beam traverses a path containing scattering centers which are undergoing Brownian motion, the intensity of light scattered at
an angle 90” from the forward direction fluctuates due to the motion of the particles. If the scattered light strikes the cathode of a PMT, the rate of photoelectric pulses produced also fluctuates
in time. The pulses are then counted electronically. This fluctuation is used to find the translational diffusion coefficient (DJ. In turn, the particle hydrodynamic radius (R) can be determined from Dr and used of the Stokes-Einstein equation,
(1)
where K is the Boltzmann constant, 7 is the viscosity of the continuous phase (0.81 CP was used),
and T is the absolute temperature. The measured dissymmetry was less than 6%, indicating that
the droplets were essentially spherical.
11-3. ‘k ansmittance Measurements
A Bausch and Lomb Spectronic-20-spectrometer was used to monitor the transmittance
of the emulsions at 520 nm (as chosen), during the titration with cosurfactants. Distilled water
was employed as the reference solution with a transmittance of 100%. The deviations of the
transmittance measurements were within + 5%.
III. RESULTS
For studying the emulsions, two basic models were used:
1.0/W (X mole of margaric acid was dissolved in 2.0 mL of oi1)/(20 mL of 0.150 N NaOH)
2. W/O (3.0 mL of 1.0 N NaOH)/(lO.O mL of oil solution containing y mole of margaric
acid)
We used the high concentration of NaOH in these systems to provide an excess of NaOH
which assured the complete ioniztion of the fatty acid.
For emulsions prepared using cyclohexadecanone and cetyl alcohol as the oil phase, transmittance is a function of the volume of cosurfactant added. As shown in Figures 1, 2, 3 and 4
each shown a maximum. At low soap concentrations the width of the curve is narrow. At higher
concentrations it becomes much broader, but more alcohol is needed to achieve maximum transmittance.
Dispersed phase volume also affected the optical properties of these systems (Figures 5,
6, 7 and 8). Solutions were prepared with 0.002 and 0.004 mole of margaric acid in
668
SI’UDIES OF FORMATION AND INTERFACE OF OIL-WATER...
\
\
.
VOL. 30
\
\
\
Volume of 4-Methylcyclohexanol (mL)
.
FIG. 1. Transmittake Curve for O/W Systems at 27’?Z as a Function of +Methylcyclohexanol and Soap
Content. 2.0 mL of Cyclohexadecanone, 20 mL of O.lSON NaOH, Moles of Margaric Acid: 0.0015
(-), 0.002 (. . * *), 0.003 (-.-), 0.004 (----).
l o o -
/‘\.
,:i‘
I!
l /-_
\ \\
l \ ‘\
-k.,
l
\ '\
i
i
50 -
i
'1,
\\
i
i
i
0'
I
2
\\
\
\
\
,
L
,
i
i.
1
L
\
I
1
4
,
6
Volume of 4-Methylcyclohexanol (mL)
FIG. 2. Transmittance Curve for O/W Systems at 27°C as a Function of -I-Methylcyclohcxanoi and Soap
Content. 2.0 mL of Cetyl Alcohol, 20 mL of O.lSON NaOtI, Pvloles of blsrgnric Acid: 0.0015 (-),
0 . 0 0 2 (. . .), 0.003 (-.--), o.oc-4 (----).
VOL. 30
HUNG-JAiiG GI, SHOU-NAN CHEN, JENN-SHYONG HWANG...
669
/__________________
100 /’
/
-*--**-•*-•.
: /”
/he
\
.
l \.
‘!
‘!
‘\
\
i.
‘\
'\
i.
;.
\
I
li.
4
6
Volume of 4-Methylcyclohexanol (mL)
.
FIG. 3. Transmittance Curve for W/O Systems at 27°C as a Function of 4-Methylcyclohexanol and Soap
Content. 10 mL of Cyclohexadecanone, 3.0 mL of 1.0 N NaOH, Moles of Margaric Acid: 0.0015
(-), 0.002 ( . . . . ), 0.003 (-.---), 0.0035 (-* -), 0.004 (- -- -)_
100 -
/
_____-__------___-___
.._..-..-. . -*a-
3
0
‘\
5
s
2
z
._
E
2
l\
i
50 -
i
i
i
i
s
@
i
0
i
.
L
2
4
6
Volume of 4-Methylcyclohexanol (mL)
FIG. 4. Transmittance Curve for W/O Systems at 27oC as a ITunction of I-~Icth~lcyclohcx~~n~~l anti Soap
Content. 10 mL of Cetyl Alcohol? 3.0 mL of 1.0 N NaOII, Moles of hd;irgaric Acid: 0.0015 (-),
0.002 (’ . . ‘), 0.003 (-.--,. 11.rms (-. .---), O.OM (--- -).
670
SHJDIES OF FORMATION AND INTERFACE OF OIL-WATER...
100
VOL. 30
5
4
3
60
2
1
n
2
4
6
6
10
12
Volume of Cyclohexadecanone (mL)
.
FIG. 5. Transmittance Curves for Cyclohexadecanone-in-Water System at 27°C with 4-Methylcyclohexanol
as Cosurfactant as a Function of Oil Content. 20 mL of 0.15N NaOH, pH 11.6. Moles of Margaric
Acid: 0.002 (o,A), 0.004 ( 0 ,m).
I
2
I
4
6
I
I
I
a
10
12
Volume of Cetyl Alcohol (mL)
FIG. 6. Transmittance Curves for Cetyl Alcohol-in-Water System at 27°C with -I-Methylcyclohexanol as
Cosurfactant as a Function of Oil Content. 20 mL of 0.15N NaOII, pI1 11.6. Moles of Margaric Acid:
0.002 (o,A), 0.004 (0 ,q.
VOL. 30
HUNG-JANG GI, SHOU-NAN CHEN, JENN-SHYONG HWANG...
671
‘J
100
4
3
50
2
1
0
2
4
5
Volume of Cyclohexadecanone (mL)
FIG. 7. Transmittance Curves for Cyclohexadecanol-in-water Systems with 4Methyl-l-pentanol as Cosurfactant as a Function of Oil Content. 20 mL of 0.15N NaOH, pH 11.6. at 27”C,
Moles of Margaric
Acid: 0.002 (o,A), 0.004 ( 0 ,B).
I
I
2
4
6
3
Volume of Cetyl Alcohol (mL)
FIG. 8.
Transmittance Curves for Cetyl Alcohol-in-Water System with -I-~lcthyl-l-pentanol as Cosurfxtant
as a Function of Oil Content. 20 mL of O.lSN NaOI-I, pI1 11.6, at 27”C, Molts of M:rr@c Acid: 0.02
(o,A), O.O@I (
l
P).
672
.
STUDIES OF FORMATION AND INTERFACE OF OIL-WATER...
VOL. 30
cyclohexadecanone and cetyl alcohol. These were dispersed in aqueous NaOH and titrated with
the alcohol indicated. The maximum transmittance and the volume of alcohol required are a
function of the volume of oil used. The results show that for a given oil, aqueous phase, surfactant, and cosurfactant combination there is an optimal amount of dispersed phase required to
produce maximum clarity.
It is interesting to note that when the aqueous phase and margaric acid are mixed, the system is turbid. The addition of cosurfactant (4-methyl-l-pentanol), increases the transmittance
to a maximum of 65-75%. Nevertheless, when the experiment is repeated in the presence of a
small amount of oil, almost 100% transmittance is obtained.
Figures 5, 6, 7 and 8 also show the sensitivity of the dispersions to the amount of the surfactant, and the type of the cosurfactant added. The volume of oil that can be dispersed at a
given maximum transmittance is much higher with 4-methylcyclohexanol than with 4-methyl-lpentanol. A larger number of other oils and surfactants were also studied using 4-methylcyclohexanol or 4-methyl-1-pantanol as a cosurfactant. The Tables I and II show that maximum
transmittance for each dispersion and the volume of alcohol required to produce it.
Photon correlation spectroscopy was used to measure particle size. This information was
then related to transmittance measurements and the amount of the surfactant. Figures 9 and
10 plot measured particle size, maximum transmittance, and the volume of alcohol required to
produce the maximum transmittance against the amount of surfactant used.
The results confirm the hypothesis, that the transmittance increases as the particle size
TABLE I. O/W Formulation - Relation of the oil and the Soap Chain Length for 4-Methylcyclohexanol at maximum Transmittance
K-Laurate
Vol.
Na-CPhenylcaprylate
Vol.
K-Palmitate
Vol.
Na-Lactarinate
Vol.
Oil
(mL)
%T
(mL)
%T
(mL)
%T
(mL)
%T
-c6
0.5
0.9
81
88
97
0.4
0.7
0.8
1.0
1.2
1.3
1.3
1.4
0.9
0.9
0.9
1.0
1.1
1.0
0.9
0.9
0.9
88
91
87
83
79
95
99
99
95
91
87
82
80
2.4
1.2
1.7
1.0
1.2
1.2
1.1
0.9
0.7
82
85
80
76
82
86
90
92
97
-c7
-CS
49
--Cl0
-Cl1
--Cl2
-Cl
-Cl6
1.1
1.4
1.5
1.6
1.8
75
64
56
49
69
89
87
85
HUNG-JANG GI, SHOU-NAN CHEN, JENN-SHYONG HWANG...
VOL. 30
673
TABLE II. O/W Formulation - Relation of the oil and the Soap Chain Length for 4-Methyl-lh
pentanol at maximum Transmittance
K-Lam-ate
Oil
-c6
47
-CS
49
-Cl0
-Cl1
-Cl2
-Cl
-Cl6
Na4-Phenylcaprylate
Vol.
K-PaImitate
Vol.
(mL)
%T
(mL)
%T
(mL)
0.5
0.6
0.9
0.9
0.8
1.0
1.3
1.7
2.1
95
98
100
96
91
90
89
92
78
0.6
0.6
0.8
1.1
1.3
1.2
1.0
1.3
1.6
92
95
98
97
95
93
94
92
90
0.8
0.9
1.4
1.0
1.2
1.3
1.5
1.6
1.9
Na-Lactarinate
%T
Vol.
(mL)
%T
90
85
92
89
87
82
77.
62
42
1.8
1.1
1.5
1.3
1.6
1.9
1.7
1.5
2.2
83
74
86
92
95
92
91
90
83
Vol.
.
decreases. That is, increased transmittance represents physical diminution of the dispersed
phase droplets.
Three other points are also evident. First, increasing the amount of surfactant in the system decreases the particle size. Second, a larger amount of cosurfactant is required to produce
maximum transparency. Third, a significant amount of surfactant must be added before the
transmittance increases appreciably.
IV. DISCUSSION
Several points can be seen in the results of the titration experiments. It has often been
observed27Y28 that a critical amount of cosurfactant is required before these dispersions become
transparent. This is reemphasized in Figures 1, 2, 3 and 4.
Furthermore, transmittance maxima are observed in most cases. These transmittance
mixima can be explained by the distribution of the cosurfactant. As the alcohol is added it partitions among the aqueous phase, oil phase and interface. This dynamic process provides some
of the energy to subdivide the dispersed phase. The distribution data show that the aqueous
phase is very quickly saturated with alcohols of six carbon atoms or greater. As the alcohol diffuses through the interphase the droplet size is made smaller and the transmittance increases.
Once the capacity for alcohol of the aqueous phase and interface is exceeded, the excess
dissolves in the oil phase. For O/W system this means an increase in droplet size and a concomitant decrease in transmittance. For W/O systems the excess of alcohol in the continuous
I
674
STUDIES OF FORMATION AND INTERFACE OF OIL-WAER..
Amount of Margaric Acid (0.001
VOL. 30
mole)
FIG. 9. Three of the Variables Dependent on the Amount of Surfactant.
phase will extract the soap from the interface. With less surfactant at the interface to stabilize
the droplets, they will grow in size and the transmittance will fall.
The same effect is seen when the volume of the disperse phase is increased and the amount
of surfactant is held constant. When the amount of the dispersed phase exceeds the capacity
of the surfactant to stabilize it, the transmittance falls (Figures 5,6,7 and 8). A given dispersed
phase volume requires a critical amount of surfactant to form a transparent dispersion (Figures
9c and 10~).
All of the titration results (Tables I, II and Figures l-10) show the specificity of the various
components. For a given oil-surfactant pair, the steric requirements of the cosurfactant deter-
VOL. 30
HUNGJANG GI, SHOU-NAN CHEN, JENN-SHYONG HWANG...
675
Amount of Margaric Acid (0.001 mole)
FIG. 10. Three of the Variables Dependent on the Amount of Surfactant.
mine the volume of dispersed that can be stabilized (Figures 5-8).
The surfactant type also play a significant role in determining whether a transparent system
will form.27 As shown in Tables I and II, as the chain length of the surfactant increases, dispersions with significant transmittance can be formed with oils of longer chain length.
Surfactant type, cosurfactant type, and the nature of the oil phase are three of the interacting variables that determine the size of the dispersed phase droplets. The aqueous phase was
kept constant throughout this study, so that such factors as a change in ionic strength would not
affect the resultant dispersion. 12
When preparing microemulsions by the titration method, various physical changes are
often noted. Sometimes, there are dramatic increases in viscosity just before the mixture be-
676
.
VOL. 30
STUDIES OF FORMATION AND INTERFACE OF OIL-WATER...
come transparent. Also, the mixture may progress from lactescent to clear rapidly or slowly as
drops of cosurfactant are added. These effects are probably related to the rate of alcohol diffusion between phases.
While the alcohol does diffuse through the interface,7 some remains there29 in weak association with the surfactant molecules. The amount that remains at the interface can be estimated. It is different for each system and also changes with the manner of preparation.%
Thus, it is path dependent. For the system depicted in Figure 9b, it was determined from the
slope that about two molecules of 4-methylcyclohexanol remain at the interface for every
molecule of surfactant. To calculate this, the assumption is made that all of the surfactant is at
the interface. In addition, the incremental amount of alcohol required for each additional increment of surfactant (slope of 9b) is also at the interface. The similiar condition can be also
obtained from Figure lob.
Surfactant and cosurfactant maintain a definite configuration at the interface with a given
cross-sectional area. This area can be calculated from the measured droplet size if it is assumd
that all bf the surfactant resides at the interface and the droplet are spherical. With these assumptions the cross-sectional area per surfactant molecule (a) is calculated b3’ u = 3V/nr,
where V is the droplet volume, n is the number of surfactant molecules, and r is the radius of
the droplet. The results are shown in Figure 11.
I
I
I
0.030
0
0.026
/
0.020 -
-,
/
0.015 -
10
I
1
I
20
30
40
50
Diameter (nm)
FIG. 11. Calculated Cross-Sectional Area as a Function of the Dispersed Phase Droplet Diameter for the System in Figure 9.
VOL. 30
HUNG-JANG GI, SHOU-NAN CHEN, JENN-SHYONG HWANG...
677
More explicitly, parameters V was taken as the sum of the volume of oil, surfactant, and
alcohol in the dispersed phase. The amount of alcohol in the dispersed phase was calculated
from the titration curves in Figures 9b and lob. That is, the intercept gives the amount in the
continuous phase. The difference from total alcohol gives the amount in the dispersed phase.
Figures 11 shows that the apparent cross-sectional area increase as the droplet diameter
of the dispersed phase increases. The area per molecule changes considerably at small
diameters. As the droplet sizes varied from 14 nm to 46 nm in diameter, the surfactant crosssectional area was calculated to be approximately 0.017 nm2 to 0.029 nm2 per molecule. This
could imply that the cosurfactant molecules were incorporated with surfactant molecules at the
interface between oil phase and aqueous phase.
REFERENCES
.
J. H. Schulman and T. S. McRoberts, Discuss Faraday Sot. 8,165 (1946).
J. A. Friend, Ibid. 4,497 (1949).
J. H. Schulman and T. S. McRoberts, Z. Electrochem. 107,15 (1948).
L. M. Prince, in “Emulsion and Emulsion Technology”; J. Ed. Lissant and Marcel Dekker, N. Y.,
1947, Part 1, pp. 164-165.
5. T. P. Hoar and J. H. Schulman, Nature 152,102 (1943).
6. J. H. Schulman and H. A. Friend, J. Colloid Interface Sci. 4,497 (1949).
7. W. E. Gerbacia and H. L. Rosano, J. Colloid Interface Sci. 44,242 (1973).
8. E. Elbing and A. G. Parts, J. Colloid Interface Sci. 37,635 (1971).
9. P. A. Winsor, Trans. Faraday Sot. 44,376 (1948).
10. P. A. Winsor, Trans. Faraday Sot. 46,762 (1950).
11. S. R. Palit, V. A. Moghe, and B. Biswas, Trans Faraday Sot. 55,463 (1959).
12. A. W. Adamson, J. Colloid Interface Sci. 29,261 (1969).
13. F. M. Fowkes, in “Solvent Properties of Surfactant Solutions”; K. Ed. Shinoda and M. Dekker,
1.
2.
3.
4.
New York, 1967, pp. 112.
14. G. Gellberg, H. Lehtinen, and S. Friberg, J. Colloid Interface Sci. 33,40 (1970).
15. S. I. Ahmad, K. Shinoda, and S. Friberg, J. Colloid Interface Sci. 47,52 (1974).
16. D. 0. Shah, A. Tamjedi, J. W. Falco, and R. D. Walder, Jr. AICHE J. 181116 (1972).
17. D. 0. Shah, Reprints 48th Nat’1 Colloid Symposium, Austin, Texas, 173 (1974).
18. S. Friberg, J. Am. Oil. Chem. Sot. 48,578 (1971).
19. P. Ekwall, L. Mandell, and K. Fontell, Mol. Cryst. Liq. Cryst. 8, 157 (1969).
20. P. A. Winsor, Chem. Rev. 68,1(1968).
21. L. I. Osipow, J. Sot. Cosmet. Chem. 14,277 (1963).
22. L. M. Prince, J. Sot. Cosmet. Chem. 21, 193 (1970).
23. H. L. Rosano, J. Sot. Cosmet. Chem. 25,609 (1974).
678
SlTJDIE3 OF FORMATION AND INTERFACE OF OIL-WATER..
VOL. 30
24. K. Shinoda and S. Friberg, Adv. Colloid Interface Sci. 4,281(1975).
25. H.Z. cummin s, N. Knable, and Y. Yeh, Phys. Rev. Letters 12,150 (1964).
26. H. L. Rosano, A. Weiss, and W. E. Gerbacia, in “Proceedings of the 12th Internation Congress on
Surface Active Substances” Moscow, USSR, September, 1,453 (1967).
27. J. Ziochower and J. H. Schulman, J. CoIIoid Interface Sci. 24,115 (1967).
28. W. Stoeckenius, J. H. Schuhnan, and L. M. Prince, Kolloid Z. 169,170 (1960).
29. W. E. Gerbacia and H. L. Rosano, in “Colloid and Interface Science” M. Ed. Kerker, Academic
Press, New York, Vol. 2, pp. 245.
30. H. L. Rosano, T. Lan, and A. Weiss, J. Colloid Interface Sci. 72,156 (1979).
31. J. S. Hwang, H. J. Gi, and S. N. Chen, Chin. J. Phys. 27,50 (1989).
32. J. S. Hwang, H. J. Gi, and S. N. Chen, J. Chin. Chem. Sot. 37,157 (1990).