Langmuir 2001, 17, 6999-7010 6999

Effect of Oily Additives on Foamability and Foam Stability.

1. Role of Interfacial Properties
Luben Arnaudov,† Nikolai D. Denkov,*,† Irena Surcheva,† Patrick Durbut,‡
Guy Broze,‡ and Ammanuel Mehreteab§
Laboratory of Chemical Physics Engineering (Formerly Laboratory of Thermodynamics and
Physicochemical Hydrodynamics), Faculty of Chemistry, Sofia University, 1 James Bourchier
Avenue, 1164 Sofia, Bulgaria, Colgate-Palmolive Research & Development, Inc.,
Avenue Du Parc Industriel, B-4041 Milmort (Herstal), Belgium, and Colgate-Palmolive
Technology Center, 909 River Road, Piscataway, New Jersey 08854-5596

Received April 24, 2001. In Final Form: July 23, 2001

Foam tests and model experiments with sodium dodecylbenzenesulfonate solutions are performed to
clarify how the foam stability and the foaminess are affected by several oils of different chemical structure.
The foam tests show that 2-butyloctanol (2BO, branched alkanol) and isohexyl-neopentanoate (IHNP,
branched ester) exhibit a significant antifoam activity at concentrations as low as 0.005 wt %. n-Heptanol
also acts as an antifoam, but at concentrations above 0.15 wt % due to its higher solubility in the surfactant
solution. The model experiments prove that the antifoam activity of pre-emulsified oils is determined
primarily by the barrier to drop entry, which controls the drop emergence on the solution surface. If the
entry barrier is high (e.g., n-dodecanol and silicone oil), the oil drops remain arrested in the Plateau borders
during the process of foam drainage, without being able to destroy the foam. Thus branched long-chain
alkanols (like 2BO) and esters (IHNP) behave as active antifoams, because they combine the advantages
of long-chain and medium-chain n-alkanolsslow solubility and low entry barrier, respectively. No direct
correlation between the spreading behavior of the oils and their foam breaking activity is observed. The
effect of these oils on the foamability of the solutions is far more complex. At low concentrations (below
and around their solubility limit) the oils reduce the dynamic surface tension of the solutions, facilitating
in this way the formation of fresh surface and enhancing the foamability. At higher oil concentrations,
however, the emulsified oil drops induce a coalescence of the foam bubbles during foaming and, as a result,
the foamability of the solutions decreases. That is why the foamability is a nonmonotonic function of the
oil concentration.

1. Introduction It has been established that normal linear alkanols (n-

alkanols) affect strongly the stability of foams produced
Two important characteristics of the surfactant solu-
from surfactant solutions. Kruglyakov and coauthors,11-13
tions are their foamability (foam produced by agitation)
systematically studied the effect of n-alkanol chain length
and foam stability (foam remaining after a certain period
on the stability of foams produced by several surfactants:
at rest), which determine the applications in areas such
sodium dodecyl sulfate (SDS); sodium dodecylbenzene-
as personal and house-hold care, washing, food industry,
sulfonate (SDDBS); nonylphenyl polyoxyethylene-10; sa-
fire fighting, ore flotation, and many others.1-4 Different
ponin. With all these systems, they found an optimum
additives, often called foam boosters, are used to improve
chain length of the alkanol that corresponded to most
the foaming properties of surfactant solutions.5,6 In other
pronounced antifoam effect. At very low surfactant
systems, additives are applied to prevent the formation
concentrations, long-chain alkanols (n-decanol, n-unde-
of an excessive foam and are termed antifoams.2-4,7-12
canol) were most efficient as foam breakers. At higher
surfactant concentrations, typical for most applications,
* To whom correspondence may be addressed. Phone: (+359)
2-962 5310. Fax: (+359) 2-962 5643. E-mail: ND@LTPH.BOL.BG. the medium-chain alkanols (n-hexanol to n-octanol) were
† Laboratory of Chemical Physics Engineering, Sofia University. more efficient. These results were explained11-13 by two
‡ Colgate-Palmolive Research & Development, Inc., Milmort competitive effects: (i) gradual decline in the alkanol
(Herstal). solubility and (ii) decrease of the oil polarity, with the
§ Colgate-Palmolive Technology Center, Piscataway.
alkanol chain length. The solubility is important, because
(1) Prud’homme, R. K., Khan, S. A., Eds. Foams: Theory, Measure- the antifoam effect is observed at alkanol concentrations
ments, and Applications; Surfactant Science Series; Marcel Dekker:
New York, 1996; Vol. 57. above the solubility limit, when dispersed oily drops appear
(2) Exerowa, D.; Kruglyakov, P. M. Foams and Foam Films; in the working solutions (see Figure 1). Therefore, at low
Elsevier: Amsterdam, 1998. surfactant concentrations more efficient as an antifoam
(3) Pugh, R. J. Adv. Colloid Interface Sci. 1996, 64, 67.
(4) Zocchi, G. In Handbook of Detergents, Part A: Properties; Broze, would be an alkanol of lower solubility which gives
G., Ed.; Surfactant Science Series; Marcel Dekker: New York, 1999;
Vol. 82, Chapter 10. (9) Wasan, D. T.; Christiano, S. P. In Handbook of Surface and Colloid
(5) Lomax, E. G., Ed. Amphoteric Surfactants; Surfactant Science Chemistry; Birdi, K. S., Ed.; CRC Press: Boca Raton, FL, 1997; Chapter
Series; Marcel Dekker: New York, 1996; Vol. 59. 6. Koczo, K.; Koczone, J. K.; Wasan, D. T. J. Colloid Interface Sci. 1994,
(6) Basheva, E.; Ganchev, D.; Denkov, N. D.; Kasuga, K.; Satoh, N.; 166, 225.
Tsujii, K. Langmuir 2000, 16, 1000. Basheva, E.; Stoyanov, S.; Denkov, (10) Denkov, N. D.; Cooper, P.; Martin, J.-Y. Langmuir 1999, 15,
N. D.; Kasuga, K.; Satoh, N.; Tsujii, K. Langmuir 2001, 17, 969. 8514. Denkov, N. D. Langmuir 1999, 15, 8530.
(7) Garrett, P. R. In Defoaming: Theory and Industrial Applications; (11) Kruglyakov, P. M. In Thin Liquid Films: Fundamentals and
Garrett, P. R., Ed.; Surfactant Science Series; Marcel Dekker: New Applications; Surfactant Science Series; Ivanov, I. B., Ed.; Marcel
York, 1993; Vol. 45, Chapter 1. Dekker: New York, 1988; Vol. 29, Chapter 11.
(8) Aveyard, R.; Binks, B. P.; Fletcher, P. D. I.; Peck, T. G.; Rutherford, (12) Chapter 9 in ref 2.
C. E. Adv. Colloid Interface Sci. 1994, 48, 93. (13) Kruglyakov, P. M.; Koretskaya, T. A. Kolloid. Zh. 1974, 36, 682.

10.1021/la010600r CCC: $20.00 © 2001 American Chemical Society

Published on Web 09/29/2001
7000 Langmuir, Vol. 17, No. 22, 2001 Arnaudov et al.

neglected, which makes the explanations in ref 14

incomplete.
Interestingly, in other studies the fast reduction of the
surface tension is considered to be the main factor for the
increased foamability of SDS solutions in the presence of
long-chain alkanols, that is, the same factor is used to
explain the opposite effect. For example, Patist et al.20
introduced n-dodecanol in SDS solutions at a molar ratio
of 1:20, when the additive is entirely solubilized within
the surfactant micelles and no foam destruction by oily
drops occurs. In these experiments lower DST cor-
responded to better foamability of the solutions. This result
was explained by a simple relation implying that the
energy introduced during foaming, W, is proportional to
the surface tension, σAW, and the change in the surface
area, ∆A: W ∝ σAW ∆A, that is, the lower dynamic surface
tension enhances the foamability of the solutions. Other
studies21-25 show that the incorporation of n-dodecanol
Figure 1. Mechanism of foam destruction by oil drops.6,9 The
leads to dense and “rigid” mixed adsorption layers with
drops are expelled from the foam films into the neighboring
Plateau borders in the early stages of foam evolution. The liquid SDS, which decelerate the foam film thinning and stabilize
drainage leads to a gradual narrowing of the Plateau borders the foams.
and to an increase of the capillary pressure compressing the The results described above suggest that branched,
trapped drops. When the compressing capillary pressure exceeds nonlinear long-chain alkanols (or other amphiphiles of
a certain critical value, which depends on the stability of the similar chemical structure) might have interesting and
asymmetric oil-water-air film, the oil drops enter the surface useful properties as additives to foaming solutions. The
of the Plateau border and rupture the neighboring foam films
(the detailed mechanism of film rupture is still unclear). The
low polarity of these substances ensures their low solubility
oil drop radius should be larger than the radius of the inscribed in water, while the branched hydrocarbon chains could
sphere (i.e., than the minimal radius of a trapped drop), RD, preclude the formation of dense mixed adsorption layers
which depends on the foam height, H. RPB is the radius of with the molecules of the main surfactant. Therefore, one
curvature of the Plateau border wall and 2θF is the contact may expect that these oils might be efficient antifoams at
angle foam film meniscus. a relatively low concentration, because they combine the
beneficial features of the medium-chain and the long-
numerous oily drops. However, at high surfactant con- chain linear n-alkanols. It is impossible to predict in
centration (around and above the critical micelle concen- advance what would be the effect of the branched long-
tration, cmc) the adsorption layers formed at the alkanol- chain alkanols on the foamability of the solutions.
water interface are possibly denser for long-chain alkanols, To check the above ideas, we performed a comparative
because the latter are less polar than the short-chain study of the properties of several oily additives of different
alkanols. Kruglyakov and coauthors11-13 hypothesize that chemical structure (one branched two-chain alkanol, one
these dense adsorption layers stabilize the asymmetric two-chain ester, two n-alkanols of different chain lengths,
oil-water-air films against rupture (Figure 1), which a long-chain saturated hydrocarbon, and silicone oilssee
makes the long-chain n-alkanols inefficient foam breakers. Figure 2) in SDDBS solutions. The major aims of the study
The effect of n-alkanols on the foamability was not studied are (1) to compare the effect of these oils on the foaminess
by these authors. and the foam stability of the surfactant solutions, (2) to
Abe and Matsumura14 studied the effect of alkanols and analyze the obtained results from the viewpoint of the
alkanediols on the foamability of SDDBS solutions. They dynamic and static interfacial properties, and (3) to explore
found that lower dynamic surface tension (DST) cor- the importance of the so-called “entry barrier”,6-13,26-30
responded to lower foamability of the solutions containing which is connected to the stability of the asymmetric oil-
n-alkanols and concluded that n-octanol should be most water-air films and to the possibility for emergence of
efficient as an antifoam. Their explanation was that the emulsified oil onto the foam film surface (Figure 1).
additives, which are able to reduce rapidly the surface The first paper in this series concerns primarily aims 1
tension, would eliminate the surface tension gradient (the and 2, while the second paper31 describes and discusses
Marangoni effect). The latter is considered to be among
the major factors ensuring the stability of the foam films (20) Patist, A.; Axelberd, T.; Shah, D. O. J. Colloid Interface Sci.
during their drainage.7,15-19 However, no correlation 1998, 208, 259.
between DST and foamability was found for alkanediols, (21) Brown, A. G.; Thuman, W. C.; McBain, J. W. J. Colloid Sci.
and the effect of the additives on the foam stability was 1953, 8, 491.
(22) Jones, M. N.; Reed, D. A. J. Colloid Interface Sci. 1969, 30, 577.
not explored at all (see also the comments in section III.C (23) Bergeron, V. Ph.D. Thesis, University of California, Berkeley,
of ref 7). The possibility that the oily drops or lenses, 1993.
floating on the air-water interface, could destroy the foam (24) Lu, J. R.; Purcell, I. P.; Lee, E. M.; Simister, E. A.; Thomas, R.
K.; Rennie, A. R.; Penfold, J. J. Colloid Interface Sci. 1995, 174, 441.
films by the mechanisms discussed in refs 6-13 was (25) Angarska, J. K.; Tachev, K. D.; Kralchevsky, P. A.; Mehreteab,
A.; Broze, G. J. Colloid Interface Sci. 1998, 200, 31.
(14) Abe, Y.; Matsumura, S. Tenside Deterg. 1983, 20, 218. (26) Bergeron, V.; Fagan, M. E.; Radke, C. J. Langmuir 1993, 9,
(15) Ivanov, I. B. Pure Appl. Chem. 1980, 52, 1241. 1704.
(16) Ivanov, I. B.; Dimitrov, D. S. In Thin Liquid Films: Fundamentals (27) Lobo, L.; Wasan, D. T. Langmuir 1993, 9, 1668.
and Applications; Surfactant Science Series; Ivanov, I. B., Ed.; Marcel (28) Kulkarni, R. D.; Goddard, E. D.; Kanner, B. J. Colloid Interface
Dekker: New York, 1988; Vol. 29, Chapter 7. Sci. 1977, 59, 468.
(17) Kralchevsky P. A.; Danov, K. D.; Ivanov, I. B.; Chapter 1 in (29) Koczo, K.; Lobo, L. A.; Wasan, D. T. J. Colloid Interface Sci.
ref 1. 1992, 150, 492.
(18) Danov, K. D.; Kralchevsky, P. A.; Ivanov, I. B. In Handbook of (30) Aveyard, R.; Binks, B. P.; Fletcher, P. D. I.; Peck, T.-G.; Garrett,
Detergents, Part A: Properties; Broze, G., Ed.; Surfactant Science Series; P. R. J. Chem. Soc., Faraday Trans. 1993, 89, 4313.
Marcel Dekker: New York, 1999; Chapter 9; Vol. 82. (31) Hadjiiski, A.; Tcholakova, S.; Denkov, N. D.; Ivanov, I. B.;
(19) Narsimhan, G.; Ruckenstein, E. Chapter 2 in ref 1. Durbut, P.; Broze, G.; Mehreteab, A. Langmuir 2001, 17, 7011-7021.
Effect of Oily Additives on Foams Langmuir, Vol. 17, No. 22, 2001 7001

Most of the used additives are partially soluble in the surfactant

solutions, mainly due to solubilization in the surfactant micelles.
To separate the effects created by the oily drops from the effects
due to molecularly solubilized oil, in some of the experiments we
use pre-equilibrated surfactant solutions, from which the drops
are removed as completely as possible. After 24 h of pre-
equilibration with 0.01 wt % of additive under continuous stirring,
the surfactant solution is kept at rest for 48 h in a separation
funnel for a gravity-driven separation of the oil drops. The lower
portion of the solution is drained from the funnel and centrifuged
at 5000g for at least 3 h. The upper layer of the liquid is removed
from the centrifugal vial by suction, and finally, the remaining
solution is filtered twice through a 220 nm membrane filter.
Even this complex procedure does not result in a complete removal
of the oil drops as evidenced by the scattering of light when a
laser beam is propagating through the precleaned solution.
However, the oil concentration in these solutions is much lower
as compared to that in the working solutions containing 0.01 wt
% of oil.
2.2. Methods. 2.2.1. Foam Formation and Foam Stability
Evaluation. The Ross-Miles test is used in most of the
Figure 2. Structural formulas of some of the studied oils: experiments to produce foams and to compare their stability. A
2-butyloctanol, 2BO; isohexyl neopentanoate, IHNP; silicone glass cylinder of 520 cm3 volume and 37 mm diameter is connected
oil, SO. The other studied oils are n-heptanol, n-C7OH; to a pump, which drives the circulation of the surfactant solution
n-dodecanol, n-C12OH; and n-hexadecane, n-C16. (300 mL). The liquid is pumped for 20 s, at a rate of 125 cm3/s
through an orifice, 7 mm in diameter, which is placed at 23 cm
results about the entry barriers for the studied oils (aim above the level of the liquid. The change of the foam volume with
3), as obtained by the recently developed film trapping time is monitored for a period of 15 min after liquid circulation
ceases. From the diameter of the cylinder of the Ross-Miles test,
technique (FTT).32,33 one can calculate that a reduction of the foam volume by 10 mL
corresponds to a decrease of the foam height by ≈1 cm. The
2. Experimental Details accuracy in the foam volume determination is (5 mL, whereas
the reproducibility is typically (10 mL.
2.1. Materials and Preparation Procedures. Sodium In another set of experiments, the foam is produced by five
dodecylbenzenesulfonate, SDDBS (C12H25C6H4SO3Na, 99% pu- rigorous hand shakes of 100 mL glass cylinders containing 20
rity, produced by Aldrich, Steinheim, Germany), is used as a mL of surfactant solution. Afterward, a 10 µL oil drop is placed
main surfactant at a concentration of 0.09 wt %, which is 2.6 mM on the wall of the cylinder (5 cm above the top of the foam column)
or approximately 13×cmc. The working solutions contain also and the foam destruction by the spreading oil is monitored for
12 mM NaCl (p.a. Merck, heated at 500 °C before utilization) 10 min.
and 0.01 wt % (unless another value is specified) of the oily 2.2.2. Size Distribution of Emulsion Droplets. The size
additive. As additives we studied 2-butyloctanol, 2BO (trade name distribution of the emulsion droplets is determined by video-
Isofol 12, Condea Chemie GmbH, Hamburg, Germany), n- microscopy observations (microscope Studar M, PZO, Warsaw,
dodecanol, n-C12OH (Sigma Chemical Co., St. Louis, MO), Poland; objectives 50× and 100×) in white transmitted light.
n-heptanol, n-C7OH (Sigma), isohexyl-neopentanoate, IHNP The image is recorded by a high-resolution CCD camera (Sony
(trade name Schercemol 65, Scher Chemicals, USA), silicone oil SSC-M370 CE) and digital memory VCR (Panasonic AG-7355).
SH200 of dynamic viscosity 5 mPa‚s (Kao Co., Tokyo, Japan), The recorded images are processed by homemade image analysis
SO, and n-hexadecane, n-C16 (Sigma). Hexadecane is refined by software for determining the drop size distribution. The diameter
passing it through a glass column filled with chromatographic of several hundreds drops is typically measured in each sample.
adsorbent Florisil.34 The other chemicals are used as received. The optical resolution of the used system is approximately 1 µm.
The solutions are prepared with deionized water from a Milli-Q 2.2.3. Surface and Interfacial Tension Measurements.
Organex system (Millipore). Ellipsometry. The surface tension of the surfactant solution is
The oily additives are emulsified by two different procedures measured by the Wilhelmy plate method on a Kruss K10T digital
to study the effect of the oil drop size: (1) 0.05 g oil is introduced tensiometer. After the surface tension of the solution is measured
into 150 mL of the surfactant solution and an intensive stirring in the absence of oil, an oil drop of 15-30 µL is gently deposited
for 1 h on a magnetic stirrer is employed. The obtained emulsion by a pipet on the solution surface away from the Wilhelmy plate.
is diluted to a volume of 500 mL to obtain the working The reduction of the surface tension, due to spreading of the oil
concentration of 0.01 wt %; the emulsion is additionally homog- over the surface, is monitored afterward. The thickness of the
enized by several hand-shakes before using it in the foam tests. spread oil layer is measured ellipsometrically by using the
(2) A 0.1 g portion of the oily additive is introduced into a 150 equipment from ref 35 and the procedure described in ref 36.
mL laboratory beaker containing 100 mL of surfactant solution These experiments are performed in closed containers to reduce
and an agitation by a rotor-stator homogenizer Ultra Turrax the evaporation of oil and water from the solution surface.
T25 (Janke&Kunkel Gmbh&Co, IKA-Labortechnik, Germany) The surface tension of bulk oils is measured by the Du Nouy
is applied for 20 s at 10 000 rpm. This emulsion is afterward ring technique on a Kruss K10T tensiometer. To check whether
diluted by surfactant solution to the desired oil concentration, the surfactant solution spreads over the oil surface, we first
and the obtained working solution is immediately used. For measure the surface tension of pure oil. Then a drop of the
brevity, we call hereafter the emulsions produced by magnetic surfactant solution is carefully placed in contact with the oil
stirring the coarse emulsions, while those obtained by the rotor- surface for at least 1 min, and the surface tension of the oil is
stator homogenizer are termed the fine emulsions (the respective measured again (no change of the oil surface tension was detected
drop size distributions are discussed in section 3.4). for all of the studied systems). The tension of the oil-solution
interface is measured by a homemade pendant drop method or
the spinning drop method (Kruss Site 04 tensiometer).
(32) Hadjiiski, A.; Dimova, R.; Denkov, N. D.; Ivanov, I. B.; Bor-
wankar, R. Langmuir 1996, 12, 6665. The entry, E, spreading, S, and bridging, B, coefficients are
(33) Hadjiiski, A.; Tcholakova, S.; Denkov, N. D., Ivanov, I. B. calculated from the interfacial tensions
Proceedings of the 13th Symposium on Surfactants in Solution (SIS
2000); Gainsville, FL, June 2000, in press. Hadjiiski, A.; Tcholakova, (35) Russev, S. C.; Argirov, T. V. Rev. Sci. Instrum. 1999, 70, 3077.
S.; Ivanov, I. B.; Gurkov, T. D.; Leonard, E. Langmuir, in press. (36) Denkov, N. D.; Marinova, K. G.; Christova, C.; Hadjiiski, A.;
(34) Gaonkar, A. J. Am. Oil Chem. Soc. 1989, 66, 1090. Cooper P. Langmuir 2000, 16, 2515.
7002 Langmuir, Vol. 17, No. 22, 2001 Arnaudov et al.

E ) σAW + σOW - σOA (1)

S ) σAW - σOW - σOA (2)

B ) σAW2 + σOW2 - σOA2 (3)

The subscripts AW, OW, and OA refer to air-water, oil-water,

and oil-air interfaces, respectively; σAW denotes either the surface
tension in the absence of oil (so-called initial EIN, SIN, and BIN
coefficients) or the tension after spreading of oil on the solution
surface (equilibrium EEQ, SEQ, and BEQ). The positive initial
spreading coefficient, SIN, indicates either a spreading of oil on
the solution surface as a multimolecular layer or a formation of
mixed adsorption layer of surfactant and oil molecules.30 The
equilibrium spreading coefficient, SEQ, could be either negative Figure 3. Experimental setup for determination of surface
(oil lenses are formed on the solution surface) or zero (complete dilatational elasticity, ESD, by the expanding drop method, EDM.
spreading of the oil as a duplex film).37 The role of these coefficients
in the antifoaming action of oils is discussed in refs 7-12. In by a dc motor (Newport 860A) and provides a constant liquid
general, positive values of E and B are considered to be a necessary flow. The two syringes, the working orifice of the pressure
(but insufficient) condition for having an active oil. Recent transducer, and the capillary are filled with surfactant solution.
studies7,38 show that positive values of S are not a necessary The electrical signal from the pressure transducer is amplified
condition for having an antifoaming activity of the oil, though and recorded on a computer via an analog-to-digital converter
the spreading might be helpful in some cases.10,31,39 (ADC). A drop of the surfactant solution is formed on the tip of
The dynamic surface tension of the solutions is measured by the capillary and is observed by an optical microscope. The whole
the maximum bubble pressure method (MBPM) on homemade system is mounted in a closed container, whose bottom is covered
equipment, which is described in ref 40. A hydrophobic capillary by the investigated solution, to suppress the evaporation of water
with a hydrophilic tip and the standard procedure for calculation and oil.
of the surface tension from the maximum pressure are used The experiment is carried out as follows. The experimental
(without a video-recording and image analysis of the bubble setup is filled with surfactant solution, and a drop is formed on
shape). the capillary tip by using syringe 2. The drop volume is kept
2.2.4. Stability of Foam Films. Horizontal foam films of constant for 1-4 h (depending on the particular system) to achieve
diameter ≈1 mm are formed and observed in a Scheludko cell.41,42 an equilibrium adsorption of surfactant on the drop surface. From
A foam film is formed from a biconcave drop of surfactant solution, the capillary pressure, PC, and the drop radius, R, measured by
placed in a vertical cylindrical glass capillary, by sucking out the pressure transducer and by the microscope, respectively,
liquid through an orifice in the capillary wall. The film radius one can calculate the equilibrium surface tension of the solution,
can be varied by a pressure control system. Special care is taken σAWEQ ) (PCR)/2. Afterward, a controlled expansion of the drop
to suppress the water evaporation. The films are observed from surface is accomplished for 1-25 s by squeezing a solution out
above (microscope Zeiss Axioplan, objectives LD Epiplan 10× of the capillary (syringe 1 is used at this stage). The changes of
and 50×) in reflected or transmitted white light. The observations the total stress τ(t) ) [PC(t)R(t) - PC(0)R(0)]/2, and of the drop
in reflected light provide a clear picture of the foam films, while area, A(t), are monitored during the expansion.
the transmitted light allows one to observe oil drops in the film The experimental data are processed by using a simple
or in the adjacent meniscus region.6 rheological model
In another set of experiments, the porous plate method of
Mysels2,23,43,44 is used to evaluate the critical capillary pressure τ ) ESD R (4)
inducing a rupture of the foam film. The liquid in the experimental
cell is in contact with a pressure transducer (Omega PC136G1), where ESD is the surface dilatational elasticity and R ) ln(A/A0)
which measures the pressure difference between the liquid phase is the relative change of the surface area. This model implies
and the ambient atmosphere (i.e., the capillary pressure). A that one can calculate the surface elasticity from the initial slope
correction for the hydrostatic pressure difference between the of the experimental curve (where τ is a linear function of R),
plane of the foam film and the level of the transducer membrane neglecting the effects of surfactant adsorption and viscous
is employed. The maximum capillary pressure attainable by the dissipation in the beginning of the drop expansion process. In
porous plate method is 5000 Pa, while the typical capillary the absence of viscous stresses, τ has the meaning of a deviation
pressure in the Scheludko cell is about 40 Pa. of the surface tension during expansion from its equilibrium
2.2.5. Measurement of the Elasticity of Water-Air In- value, τ ) σAW(t) - σAWEQ. Therefore, the obtained value of ESD
terface by the Expanding Drop Method (EDM). A home- corresponds to an effective surface elasticity, which does not
made equipment is used to determine the dilatational surface necessarily coincide with the thermodynamically defined Gibbs
elasticity, ESD, of the surfactant solution around the cmc (0.35 elasticity, EG. The experiments are performed around the cmc
mM SDDBS, 12 mM NaCl solution). The equipment consists of of SDDBS, because the surface stress created in the experiments
a glass capillary, which is connected to a piezoresistive pressure with the working solutions (2.6 mM) is too small to be reliably
transducer (163PC01D36, Omega) and two microsyringes (no. 1 detected and interpreted.
of 100 µL and no. 2 of 1 mL); see Figure 3. Syringe 1 is driven To study the effect of the spread layer of IHNP on the surface
elasticity of the surfactant solution, some of the experiments are
(37) Rowlinson, J. S., Widom, B. Molecular Theory of Capillarity; performed with drops whose surfaces are covered by a spread
Oxford University Press: Oxford, 1989; Chapter 8. layer of IHNP. Such drops are formed in the following way: First,
(38) Garrett, P. R.; Davis, J.; Rendall, H. M. Colloids Surf., A 1994, the experimental setup is filled with surfactant solution and an
85, 159.
(39) Denkov, N. D.; Marinova, K. Proceedings of the 3rd EuroCon- aqueous drop is formed from the solution on the tip of the capillary,
ference on Foams, Emulsions and Applications; MIT: Bremen, 2000; as described above. Next, IHNP is spread on the surface of the
p 199. bulk solution, which is placed in the container below the capillary;
(40) Horozov, T. S.; Dushkin, C. D.; Velev, O. D.; Danov, K. D.; see Figure 3. The capillary is moved downward until the drop
Arnaudov, L. N.; Mehreteab, A.; Broze, G. J. Colloid Interface Sci. 1996, coalesces with the bulk solution (i.e., an aqueous bridge is formed
113, 117. between the capillary and the bulk solution). Finally, the capillary
(41) Scheludko, A.; Exerowa, D. Kolloid Z. 1957, 155, 39.
(42) Scheludko, A. Adv. Colloid Interface Sci. 1967, 1, 391.
is detached from the surface of the bulk solution by moving it
(43) Mysels, K.; Jones, A. J. Phys. Chem. 1964, 68, 3441. upward. Thus a drop of the surfactant solution is formed again,
(44) Claesson, P. M.; Ederth, T.; Bergeron, V., Ruthland, M. W. Adv. with a spread layer of IHNP being transferred from the surface
Colloid Interface Sci. 1996, 67, 119. of the bulk solution onto the drop surface. After several hours
Effect of Oily Additives on Foams Langmuir, Vol. 17, No. 22, 2001 7003

Table 1. Measured Interfacial Tensions and Calculated Entry, E, Spreading, S, and Bridging, B,
Coefficients for Different Oilsa
σAWEQ, σOW, σOA, EIN, EEQ, SIN, SEQ, BIN, BEQ,
additive mN/m mN/m mN/m mN/m mN/m mN/m mN/m (mN/m)2 (mN/m)2 PCCR, Pa
2BO 27.7 3.6 26.5 7.7 4.8 0.5 -2.4 247 78 44
IHNP 24.8 1.8 23.6 8.8 3.0 5.2 -0.6 383 61 73
n-C12OH 24.5 6.0 27.8 2.7 -9.3 -137 >1500
n-C7OH 27.9 4.6 25.6 6.9 -2.3 144 b
SO 23.5 5.6 18.7 17.5 10.4 6.3 -0.8 618 234 >3000
n-C16 30.0 2.8 27.2 6.2 5.6 0.6 0.0 204 168 ≈400
a σ IN is 30.6 mN/m in all cases except for n-C OH, where it is 30.2 mN/m (at 27 °C). The accuracy of the calculated values of E and
AW 12
S is (1 mN/m; for B it is (15 mN/m. b Not measured.

at rest (for reaching the equilibrium surface tension), this drop a relatively slow process) leads to establishment of the
is expanded as described above. equilibrium surface tension in these systems. Other oils,
All experiments are performed at an ambient temperature of however, spread as continuous layerssthick or thin,
25 ( 2 °C, except those with n-C12OH, which are performed at depending on the disjoining pressure acting across the oil
27 ( 0.2 °C to avoid the oil crystallization (the melting point of
n-C12OH is around 24 °C).
film.45 The main driving force for such spreading is the
initial spreading coefficient, SIN, which represents the
3. Results and Discussion difference in the interfacial tensions acting on the three-
phase contact line. As shown by Bergeron et al.46,47 with
First, results from model experiments with different several surfactants above their cmc, the oil spreading in
oils are presented. Next, the foaminess and the foam this case is very fast and obeys a simple hydrodynamic
stability of the surfactant solutions are discussed and law.
explained on the basis of the results from the model
Indeed, visual observations and measurements by
experiments.
ellipsometry show that IHNP spreads rapidly and com-
3.1. Entry (E), Spreading (S), and Bridging (B)
pletely as a thick oil layer (so-called duplex film7) on the
Coefficients, Entry Barriers. 3.1.1. Oil Spreading.
surface of SDDBS solutions. This spreading behavior is
The measured interfacial tensions and the calculated
in agreement with the measured equilibrium spreading
values of E, S, and B coefficients are presented in Table
coefficient for IHNP, which is virtually zero in the
1. The initial coefficients for n-C12OH and n-C7OH are not
framework of the experimental accuracy. In contrast, when
presented, because these two alkanols are more soluble
2BO is placed on the solution surface, several oil lenses
in the SDDBS solutions than the other studied oils. As a
are formed and the surface tension slowly decreases (by
result, an intensive mass transfer across the oil-water
about 2 mN/m for 20 min) but no spread layer is detected
interface is observed by optical microscopy when a drop
by ellipsometry. Even after 1 h, the ellipsometrical signal
of n-C12OH or n-C7OH is placed in contact with the
remains unchanged, which means that the thickness of
solution. Therefore, the interfacial tensions, σOW, which
the adsorption layer is not affected by the spreading oil
can be measured only with pre-equilibrated phases, cannot
(the resolution of the equipment is about 0.1 nm). These
be used to calculate the initial E, S, and B coefficients for
results show that a monomolecular, mixed adsorption
these two alkanols. Direct measurements showed that
layer of SDDBS and 2BO is formed on the solution surface,
the surface tension of the SDDBS solutions decreases
in coexistence with lenses of 2BO.
very rapidly by several mN/m when a drop of n-C12OH or
n-C7OH is placed on the solution surface. This observation One may speculate that the faster oil spreading should
proves that both alkanols rapidly spread, i.e., that SIN is lead to a more efficient foam destructionsthe results from
positive. There is no problem to determine the equilibrium our foam stability tests, however, do not support such a
coefficients for all oils, because the interfacial tensions general trend (section 3.6), which means that other factors
are measured with pre-equilibrated phases. are more important.
The results show that the initial spreading coefficients 3.1.2. Entry and Bridging Coefficients, Entry
of the remaining oils are all positive. It is worth noting Barriers. As seen from Table 1, the values of all entry
that the oils possessing a relatively large SIN > 5 mN/m and bridging coefficients are positive except the nega-
(IHNP and SO) spread very rapidlysthe surface tension tive equilibrium bridging coefficient determined with
reduction after placing the oil drop on the solution surface n-C12OH. The positive bridging coefficients indicate that
occurs for less than 1 s. On the contrary, 2BO and n-C16, the oil bridges would be unstable if they are formed in the
which possess SIN ≈ 0.5 mN/m (actually, this value is foam films.7,10 On the other side, the formation of oil bridges
smaller than the accuracy of our determination, (1 mN/ from pre-emulsified oil drops requires the rupture of the
m) exhibit much slower kinetics of surface tension asymmetric oil-water-air films, formed between the
reduction, on the order of several minutes. drops and the solution surface (Figure 1). As seen from
The observed qualitatively different spreading kinetics Table 1, the studied oils exhibit rather diverse entry
of the studied oils is probably related to different structure barriers: for 2BO and IHNP the barrier is below 100 Pa,
of the formed spread layers. As discussed by Binks et al.,45 for n-C16 it is 400 Pa, while for n-C12OH and SO it is much
some oils are only involved in a “chain mixing”sa process higher (above 1500 and 3000 Pa, respectively). The entry
that leads to formation of a mixed adsorption layer with barrier for n-C7OH could not be measured by the FTT for
the surfactant molecules, without an actual spreading of technical reasons, which are described in ref 31. However,
a continuous oil layer on the solution surface. The main the results obtained by Kruglyakov,11-13 and the poor foam
driving force for this process is the gradient of the surface stability in the presence of n-C7OH drops, suggest that
concentration of oil molecules in the mixed adsorption the entry barrier for this oil is low.
layer, which suggests that the surface diffusion (which is
(46) Bergeron, V.; Cooper, P.; Fischer, C.; Giermanska-Kahn, J.;
(45) Binks, B. P.; Crichton, D.; Fletcher, P. D. I.; MacNab, J. R.; Li, Langevin, D.; Pouchelon, A. Colloids Surf., A 1997, 122, 103.
Z. X.; Thomas, R. K.; Penfold, J. Colloids Surf., A 1999, 146, 299. (47) Bergeron, V.; Langevin, D. Phys. Rev. Lett. 1996, 76, 3152.
7004 Langmuir, Vol. 17, No. 22, 2001 Arnaudov et al.

Figure 4. Dynamic surface tension (DST) of solution containing

2.6 mM SDDBS, 12 mM NaCl, and 0.01 wt % of different oils
(coarse emulsion) as a function of t-1/2.

As discussed below, the entry barrier is among the most

important factors determining the activity of a given oil
as a foam breaker.
3.2. Dynamic Surface Tension (DST). In principle,
the DST of micellar surfactant solutions can be affected
by oils via two different mechanisms. The first one is
related to the molecular diffusion and adsorption of
surfactant and oil molecules on the solution surface. The
rate of surfactant adsorption in micellar solutions is
influenced by the demicellization rate.20,48-50 The presence
of solubilized oil in the micelles could strongly affect the
demicellization rate and the kinetics of surfactant ad-
sorption. In addition, part of the solubilized oil molecules
could also adsorb on the solution surface after their release
from the micellessthese oil molecules would participate
Figure 5. Dynamic surface tension of solutions containing 2.6
in the formation of mixed adsorption layer, reducing in mM SDDBS and 12 mM NaCl in the absence of oil, in the
this way the DST. The second mechanism of surface presence of 0.01 wt % oil (coarse emulsion), and after the oil
tension reduction, in the presence of oil, consists of drops have been removed: (A) 2BO, (B) IHNP solutions (see
coalescence of dispersed oil drops with the solution surface sections 2.1 and 3.2).
(drop entry), followed by oil spreading. Therefore, the
barrier to drop entry, the rate of oil spreading, and SIN as compared to the other oils, see Table 1, and (ii) a
(which determines the magnitude of the surface tension relatively low rate of demicellization in the presence of
reduction by the spread oil) are other important factors n-C12OH, which decelerates the surfactant adsorption at
for the DST of such systems. short times. Indeed, Patist et al.20 found that the addition
The DST of the working solutions, measured by the of n-C12OH to SDS solutions leads to reduced rates of
MBPM, is shown in Figure 4. Experiments with SDDBS demicelization and surfactant adsorptionsa similar effect
solutions in the absence and in the presence of 0.01 wt % can be expected in SDDBS solutions as well.
of oil (coarse emulsion) are performed. No experiments To separate the effect of the oil drops on the DST, we
with SO were made to avoid contamination of the made experiments with solutions that were pre-equili-
experimental equipment with this difficult-for-cleaning brated with 2BO and IHNP, and the oil drops were
oil. Remarkably, the curves for most of the oils lay very afterward removed as explained in section 2.1. The results
close to each other in the range of bubble lifetimes from (see Figure 5) show that the drop-deprived solutions of
0.2 to 4 s. One can notice that all of the oils reduce the 2BO show virtually the same DST as those of the working
DST by 4-6 mN/m in the interval between 0.5 and 2 s. solutions containing drops. Therefore, 2BO reduces the
This similarity in the DST curves is surprising (and DST mainly by the molecular mechanisms mentioned
probably fortuitous) having in mind the diverse chemical above, namely, by increasing the rate of demicellization
structure of the studied oils, which determines a rather and by molecular diffusion and adsorption of oil molecules
different spreading behavior and, probably, different rates onto the surface. On the contrary, solutions deprived of
of demicellization in the studied solutions. IHNP drops show virtually the same DST as that of the
The only curve that deviates significantly from the pure surfactant solution. Hence the surface tension
others is that of n-C12OH. This curve has a notably larger reduction in the IHNP-containing solutions occurs pri-
slope, which is caused by (i) the high surface activity of marily through a coalescence of oil drops with the solution
n-C12OH, which leads to lower equilibrium surface tension surface.
The importance of these results for the foamability of
(48) Joos, P. Dynamic Surface Phenomena; VSP: Zeist, The Neth- the surfactant solutions is discussed in section 3.6 below.
erlands, 1999.
(49) Dukhin, S. S.; Kretzschmar, G.; Miller, R. Dynamics of Adsorption (50) Dushkin, C. D.; Ivanov, I. B.; Kralchevsky, P. A. Colloids Surf.
at Liquid Interfaces; Elsevier: Amsterdam, 1995. 1991, 60, 235.
Effect of Oily Additives on Foams Langmuir, Vol. 17, No. 22, 2001 7005

Table 2. Surface Dilatational Elasticity, ESD, of 0.35 mM

SDDBS Solution in the Presence of 12 mM NaCl with
Coarse Emulsion and with a Spread Layer of Additive
(see the text for details)a
ESD, mN/m
additive emulsion spread layer
IHNP 18 ( 5 1.5 ( 0.5
2BO 12 ( 4 12 ( 5
n-C12OH 24 ( 4
a ESD ) 13 ( 1 mN/m in the absence of oil.

3.3. Dilatational Elasticity of the Water-Air In-

terface. The surface elasticity of the foaming solution is
another characteristic related to its foamability. Higher
elasticity corresponds to a more rapid increase of the
surface tension during surface expansion, which would
require more energy to create new bubbles during foam
generation at equivalent other conditions.
The surface elasticity of 0.35 mM SDDBS solutions in
the absence of oil additives was determined by EDM to
be ESD ) 13 ( 1 mN/m. The results obtained with 0.01 wt
% coarse emulsions of IHNP and 2BO were very similar
to those in the absence of oil (see Table 2). In the case of
n-C12OH emulsion, however, the obtained surface elastic-
ity was about twice higher, ESD ) 24 ( 4 mN/m, which is
due to formation of a particularly dense mixed adsorption
layer from the main surfactant and the alcohol.24 The large
value of ESD agrees qualitatively with the measured
steeper dependence of the DST on the bubble lifetime
(Figure 4), as well as with the relatively low foamability
of the solutions containing C12OH (Figure 9 below). On
the other side, as discussed in the literature,15-18 the higher
surface elasticity decelerates the thinning of the foam films
and stabilizes them against rupture.
Since the results described in section 3.2 show that the Figure 6. Experimental data obtained by EDM: (A) surface
reduction of DST of SDDBS solutions by IHNP is due to strain, R, vs time for emulsion of IHNP (circles) and for a spread
coalescence of oil drops with the solution surface, we layer of IHNP (boxes); (B) stress τ vs R for the same experiments.
The surface dilatational elasticity is calculated from the slope
performed EDM experiments with surfactant solutions of the lines drawn in (B). All solutions contain 0.35 mM SDDBS
whose surface was covered by a spread layer of IHNP. and 12 mM NaCl.
The presence of a spread IHNP drastically changes the
observed rheological behavior of the solution surface. In
Figure 6A the total strain, R, for two different experiments,
with emulsion of IHNP and with a spread layer of the
same oil, is plotted as a function of time. One sees that
the magnitude and the rate of surface expansion in these
two experiments are very similar, while the measured
surface stresses are very different, Figure 6B. The slope
of the surface stress in the case of spread IHNP layer is
an order of magnitude smaller (ESD ) 1.5 ( 0.5 mN/m)
Figure 7. A possible explanation of the measured low surface
than the one in the experiments with IHNP emulsion and elasticity of an aqueous drop in the presence of a spread IHNP
in the absence of oil (13-18 mN/m). layer. The expansion of the drop surface leads to rupture of the
In principle, one could try to explain this low value of oil layersthe subsequent surface expansion is accompanied
ESD by assuming that a thick uniform layer of IHNP is with a fast spreading of oil from the thicker oil regions (acting
present on the surface of the aqueous drop throughout its as reservoirs), so that the drop surface is always covered with
expansion. If this were the case, then the measured a thin oil layer, which maintains low surface stress.
elasticity should be equal to the elasticity of the oil-water
interface, which is formed between the spread layer and the period of surface expansion. Since the spreading of
the aqueous drop (or larger if the oil-air interface also the precursor oil film is a very fast process,46,47 the surface
contributes to the measured surface elasticity). However, of the expanding drop would be always covered with a
direct measurements of the oil-water interfacial elasticity thin oil layer, which decreases the surface tension and
with aqueous drops immersed in bulk IHNP showed that the measured surface stress. With respect to foaming,
it is too high, 5 ( 1 mN/m, to support the above explana- this corresponds to lower dynamic surface tension in the
tion. That is why we suggest another interpretation of the presence of spread oil, which might explain the enhanced
low surface elasticity in the presence of a spread IHNP foamability of IHNP-containing solutions in a certain
layer. We assume that the spread IHNP layer ruptures range of oil concentrations (see section 3.6.4 and Figure
almost immediately after the drop expansion has starteds 10A below), where presumably the coalescence of oil drops
one or several oil lenses are formed on the drop surface with the bubbles results in the formation of a spread IHNP
(Figure 7), which serve as a reservoir of oil throughout layer.
7006 Langmuir, Vol. 17, No. 22, 2001 Arnaudov et al.

One can notice in Figure 6B that there is some positive

intercept of the dependence τ(R). This intercept is due to RD(H) ) RPB(H)
3 [
31/2 π
6 ( ) π
6 (
sin - θF + cos - θF - 1 ≈ ) ]
[ ]
several phenomena accompanying the start of the drop
expansion process, such as the effect of liquid viscosity 2(31/2) 3 1/2
RPB(H) -1- θ 2 + O(θF3) ≈
(hydraulic shock), the initial nonlinear motion of the dc 3 3 F
motor, some mechanical disturbances, etc., that is, this RPB[0.155 - 0.577θF2] (6)
intercept is an artifact created by the used equipment.
3.4. Oil Solubility in the Surfactant Solution and
Drop Size Distribution. Microscope observations and where θF (expressed in radians) is the half of the contact
light-scattering experiments showed that all of the studied angle film-meniscus (Figure 1). Since the foam film
oils (except n-C7OH) were not completely dissolved in the thickness, h ∼ 10 nm, is much smaller than RPB, the film
surfactant solutions at the working concentration of 0.01 is considered infinitely thin in the derivation of eq 6. In
wt %sa fraction of the oil remained in the form of dispersed most cases, θF < π/30, which means that its contribution
drops. The solubility of n-C7OH was determined by can be neglected; i.e., a reasonable estimate of RD is given
measuring the intensity of light scattered from SDDBS by the expression6
solutions, in which different volumes of n-C7OH were
added. These measurements were made 5 days after the σAW 1
introduction of n-C7OH in the solutions. A sharp in- RD(H) ≈ 0.155 (7)
∆Fg H
crease of the intensity of the scattered light was observed
at n-C7OH concentrations around 0.15 wt %, which means For a foam column with height H ) 10 cm, eq 7 predicts
that its solubility is about 0.15 ( 0.05 wt %. We were RD ≈ 5 µm. This value is more than two times larger than
unable to determine the solubility of the other oils, because the mean drop radius in our samples, which means that
it was very low (below 0.005 wt %). most of the oil drops are too small to cause foam destruction
The drop size distribution in the samples containing (unless many oil drops are closely packed in a single
2BO and IHNP was determined to check the reproduc- Plateau border). However, as explained above, oil drops
ibility of the emulsification procedure and to compare the of radius above 5 µm, which can be compressed by the
drop size with the cross section of the Plateau borders narrowing walls of the GPB, are also present in the
(GPBs). Measurements of four independently prepared solutions. Moreover, a coalescence of the emulsion drops
samples of fine 2BO emulsion (0.1 wt %, emulsified by might occur in the GPBs during the liquid drainage, which
Ultra Turrax, see section 2.1) showed approximately log- would lead to an increase of the size of the drops captured
normal distribution with a geometrical mean drop radius in the foam.
of 1.9 µm and polydispersity σg ) 2.0; 200 drops were One can conclude that the foam destruction in our
measured in each sample. Most of the drops fell in the systems is accomplished by the biggest drops, i.e., not
range between 0.8 and 5 µm, but larger drops of radius only the mean size but also the polydispersity of the drops
up to 10 µm were occasionally observed in the samples. is important.
Virtually the same drop size distribution was determined 3.5. Foam films. 3.5.1. Foam Films in the Scheludko
in samples prepared at the working oil concentration of Cell. The experiments with small horizontal foam films
0.01 wt %; 100 drops were counted in this experiment. A in the Scheludko cell showed that the addition 0.01 wt %
very similar drop size distribution (mean radius 1.8 µm; of the studied oils did not change significantly either the
σg ) 2.2) was determined also in four independently film-thinning rate or the equilibrium film thickness. No
prepared samples containing fine emulsions of IHNP. multiple stepwise transitions in the foam film thickness
The oil drops in the coarse emulsions (prepared on (stratification) were observed, because the concentration
magnetic stirrer) were somewhat largersthe mean drop of micelles was too low. The final film thickness was 18
radii were 2.2 µm in 2BO and 2.9 µm in IHNP emulsions ( 1 nm, which corresponded to a common black film
and drops of radius as large as 20 µm were occasionally stabilized by electrostatic repulsion between the film
observed in these samples. The reproducibility of the size surfaces (and possibly containing one layer of micelles,
distribution in the coarse emulsions was lower as com- which does not leave the film spontaneously under these
pared to the one in the fine emulsions. conditions). The oil drops left the foam films during their
As discussed in section 3.5 below, the studied oils destroy thinning without causing a film rupture, similar to the
foam by rupturing the Gibbs-Plateau borders (not foam processes described in ref 6. Therefore, one can conclude
films). To compare the size of the oil drops with the cross that the drops are unable to rupture the thinning foam
section of the Gibbs-Plateau borders (Figure 1), one films, because the entry barriers are too high for drop
should estimate the radius of curvature, RPB, of the GPB entry to occur at this stage.6,39
wall. As explained in ref 6, RPB in the top layer of a foam 3.5.2. Foam Films in the Mysels Cell. The porous
column of height H can be estimated from the conditions plate method allows one to increase the suction capillary
for hydrostatic equilibrium pressure exerted on the foam film and to determine the
critical value, which induces a foam film rupture.23,26,44
RPB(H) ) σAW/PC ≈ σAW/(∆FgH) (5) Since the presence of spread oil on the film surfaces could,
in principle, lead to reduced stability of the foam films,
where PC(H) is the capillary pressure bubble-Plateau we studied surfactant solutions, which were pre-equili-
border, which is approximately equal to the hydrostatic brated with IHNP and 2BO, and the oil drops were
pressure, ∆FgH; ∆F is the difference between the mass afterward removed (section 2.1). The experiments revealed
densities of the aqueous and gaseous phases and g is the that the foam films from these solutions are extremely
acceleration of gravity. Equation 5 predicts that RPB ≈ 30 stable and the critical capillary pressure for their rupture
µm at equilibrium for a foam column of height H ) 10 cm, is above 5000 Pa. For comparison, the critical capillary
and RPB ≈ 300 µm for H ) 1 cm. A geometrical pressure for entry of the oil drops is about 2 orders of
consideration shows that the radius, RD, of a sphere magnitude lower; see Table 1. Therefore, one can conclude
inscribed in the GPB (i.e., the minimal radius of a trapped that the foam destruction in the presence of 2BO and IHNP
sphere) is given by the expression is due to dispersed oil droplets (which are the actual
Effect of Oily Additives on Foams Langmuir, Vol. 17, No. 22, 2001 7007

at these conditionssa very slight reduction (by 4-5 mL)

of the foam volume is detected during the observation
period.
The solutions obtained after the above experiment were
shaken again to check how the emulsified oils affect the
foamability and the foam stability, Figure 8B. The ini-
tial foam volumes in the presence of oil were smaller than
VIN in the absence of oil. Most active as an antifoam is
n-C7OH (VIN ) 12 mL), while n-C16 has almost no effect
(VIN ) 70 mL). The remaining additives result in inter-
mediate VIN between 35 and 50 mL. Note that these
solutions contain 0.05 vol % oil, which means that the
foam generation is strongly affected by the dispersed oil
drops in the foaming medium (cf. with Figure 10 below).
For about 5 min, the produced foams are stable in all
systems. Afterward, a slow foam decay is observed in the
presence of 2BO and IHNP (not shown in Figure 8B) and
the foam volume in these systems reduces to 10-15 mL
after 1 h, while the other foams remain stable during this
period.
The differences in the defoaming activity of the oils
added on the top of the foam column and the antifoam
activity of the same oils when dispersed as drops in the
foaming medium, cf. Figure 8, are due primarily to the
decisive role of the entry barrier in antifoaming (when
the drops are pre-emulsified). If the oil is placed on the
top of the foam column (Figure 8A), several processes,
such as oil spreading, oil solubilization, and evaporation
(which could induce gradients in the surface tension and
destabilization of the foam films) possibly play a significant
role in the foam destruction. On the other hand, when the
oil drops are pre-emulsified, the drop entry appears to be
the critical step, which controls to a large extent the
antifoam efficiency of the oil.31 Indeed, the high entry
barrier of silicone oil and n-C12OH prevents the entry of
emulsified oil drops, which explains why these oils do not
Figure 8. (A) Foam destruction by spreading of different oils cause a foam destruction after emulsification (viz. Figure
on the top of a foam column formed from solution of 2.6 mM 8B). For n-C12OH, an additional problem is the negative
SDDBS and 12 mM NaCl. The oil emulsions (0.05 wt %) obtained equilibrium bridging coefficient. On the other side, the
after the end of experiment (A) are shaken again to produce drops of 2BO and IHNP, possessing low entry barriers
foams, whose evolution is shown in (B). Another shaking cycle
was made with these emulsions a week later, and the results and positive bridging coefficients, easily destroy the foam
were virtually the same as those presented in (B), except for by the mechanism illustrated in Figure 1.
heptanol which did not show any antifoam activity (see the The same solutions were tested again 1 week later to
respective curve in B), because it was already solubilized within see how the solubilization of the oils affects their antifoam
the surfactant micelles. activity. The results were almost the same with one
remarkable differencesthe foam produced from the solu-
antifoam entities) and is not related to reduced stability tion containing n-C7OH was large (70 mL) and remained
of the foam films. Similar conclusion for other systems stable for the observation period of 1 h (Figure 8B).
(foams in porous media in the presence of hydrocarbons) The microscope observations showed that n-C7OH was
was drawn previously by Bergeron et al.26 solubilized in this aged solution, which explained why
3.6. Foamability and Foam Stability. 3.6.1. De- n-C7OH had completely lost its antifoam activity. The other
foaming by Spreading of Oil over the Foam Column. oils are less soluble and dispersed oil drops were observed
The initial foam, VIN ) 79 mL, in these experiments is in all of the remaining solutions, which is why the antifoam
generated by five hand shakes of 20 mL of surfactant activity of these oils remained unaltered with time.
solution in the absence of oil. After 10 min at rest, the 3.6.2. Foam Destruction by Pre-emulsified Oil
foam volume decreases to about 72 mL due to liquid Drops. The experiments described in this and the
drainage. At that moment, 10 µL of the oily additive is following subsections are performed by the Ross-Miles
introduced on the top of the foam column, as described in test. In Figure 9 and Table 3 we compare the antifoam
section 2.2.1, and the foam decay is monitored for 10 min, activity of the oils at a concentration of 0.01 wt % (coarse
Figure 8A. The strongest destabilizing effect is observed emulsion). We first discuss the effect of the oils on the
with n-C7OH; the foam disappears almost completely foam stability. The effect on the foamability of the solutions
within 20 s. Though more slowly (for 2-3 min), IHNP also (which is far more complex for analysis) is discussed in
leads to an almost complete foam destruction; the final section 3.6.4.
foam volume of 6 mL is the same as with n-C7OH. The The results in Figure 9 show that only the drops of 2BO
deposition of 2BO and SO on the top of the foam column and IHNP are able to destroy the foam down to 40-50 mL
leads to similar decay profiles; after 2-3 min the foam for about 10 min. The foams produced in the presence of
volume levels off at about 20 mL. n-C12OH leads to a the other oils are very stable under these conditions. Note
moderate destruction of the foam down to about 35 mL. that the foam destruction pattern in the presence of 2BO
Only n-hexadecane is practically inactive as a defoamer and IHNP is very similar, though these two oils have very
7008 Langmuir, Vol. 17, No. 22, 2001 Arnaudov et al.

one of the drops trapped in the upper layer of the foam

column, the subsequent drop entry triggers an avalanche
of film ruptures and drop entry events, so that the top
portion of the foam column is destroyed. The visual
observations show that the foam layer, which is destroyed
in one step, roughly corresponds to the upper layer of dry
foam (it appears more transparent in transmitted light
than the remaining, wet part of the foam), which is formed
as a result of the water drainage.
The residual foam, observed with 2BO- and IHNP-
containing solutions (≈40 mL, corresponding to ≈4 cm
height of the foam column), can be explained by the
increased equilibrium cross section of the Plateau borders
at low foam height, H. Equation 7 predicts that the
minimal radius of the trapped sphere is RD ≈ 11 µm for
H ) 4 cm. As explained in section 3.3, the main fraction
of oil drops is of radius below 5 µm, and the larger drops
are relatively rare. Hence, the probability for trapping oil
Figure 9. Comparison of the foamability and foam stability
of 2.6 mM SDDBS solutions in the presence of 0.01 wt % of oily drops, which are large enough to cause foam destruction,
additive (coarse emulsion, Ross-Miles test). becomes rather small at low H. Furthermore, one sees
from Figure 9 that the last steps in the foam decay curves
Table 3. Initial, VIN, and Final, VF, Foam Volume in the become very small, i.e., the entry events do not lead to a
Presence of 0.01 wt % of Oil (coarse emulsion)a massive destruction of bubbles in the upper layer of foam.
additive VIN, mL VF, mL PCCR, Pa The visual observations show that the last 40-50 mL of
no oil 210 200
foam remains wet during the entire observation period,
2BO 200 42 44 so that this foam is less vulnerable to destruction.
IHNP 240 52 73 3.6.3. Experiments with Pre-equilibrated Solu-
n-C12OH 150 140 >1500 tions after Removal of the Oil Drops. As clarified by
n-C7OH 210 195 b Kruglyakov et al.,11-13 the antifoam activity of n-alkanols
SO 210 180 >3000 is primarily due to the emulsified oily drops. To check
n-C16 260 245 ≈400 whether the same is true for 2BO and IHNP, we performed
a For comparison, the critical capillary pressure for drop entry, foam tests with pre-equilibrated solutions, from which
PCCR, is also shown. b Not measured. the oil drops had been almost completely removed as
explained in section 2.1. The tests showed that the removal
different spreading behavior. This is another indication of IHNP and 2BO drops lead to foams that were stable for
that the oil spreading has a secondary importance for the a longer time (8-10 min). Afterward, a significant foam
foam stability in these systems, while the bridging destruction was typically observed in one or two steps
coefficients (should be positive) and the entry barrier (down to VF ≈ 80 mL).
(should be low) are more important. The lack of foam This result can be explained by taking into account the
destruction by the drops of silicone oil is certainly due to fact that some oil drops had remained in these solutions
its high entry barriers (Table 1). For n-C12OH, the entry (it was impossible for us to remove completely all of the
barrier and the equilibrium bridging coefficient are both oil drops). Therefore, if a single big oil drop is trapped
unfavorable with respect to its antifoam activity. The somewhere within the foam column, the entry of this drop
heptanol is completely solubilized, so that no oil drops are could induce a mechanical shock that is powerful enough
present in the solution. Hexadecane also has no antifoam to destroy a big fraction of the foam (other oil drops were
activity in the time scale of interest (15 min), though its certainly captured in the foam as well).
bridging coefficient is positive. Additional experiments It is worth noting that the foamability of solutions pre-
showed that the hexadecane indeed breaks the SDDBS equilibrated with 2BO was higher than that of pure
foam, but the destruction process starts only after ca. 1 SDDBS, while the foamability of the solutions pre-
h (for comparison, the SDDBS foam in the absence of oil equilibrated with IHNP was lower. This observation is
is very stable and no destruction is observed for more explained and discussed in the following subsection.
than 2 h). This delay is certainly due to the higher entry 3.6.4. Effect of the Additive Concentration. The
barrier of hexadecane (400 Pa)smore water should drain effect of the oil concentration was studied with fine
from the foam before the hexadecane drops can enter the emulsions of IHNP, 2BO, and n-C7OH (Figure 10), because
walls of the GPBs and become active as antifoam entities.6 the reproducibility of the drop-size distribution was better
One interesting feature of the results shown in Figure when a rotor-stator homogenizer was used to disperse the
9 is the stepwise destruction of the foam column observed oils.
with IHNP- and 2BO-containing solutions. The foam is In the absence of oil, the initial foam volume, VIN, is
destroyed in steps of 10-20 mL (each step including about 205-210 mL and the final foam volume, VF, is about
hundreds of bubbles), separated by a relatively long 195-200 mL. The experiments with different oils show
periods, 1-2 min, without visible foam destruction. No that VF, which is determined mainly by the antifoam
such steps are seen in Figure 8A, where the oil is spread activity of the oil drops, is a monotonic function of the oil
on the top of the foam column. Such a collective destruction concentration; see Figure 10. For IHNP-containing solu-
of many bubbles in a single step can be explained by the tions, VF sharply decreases with the oil concentration down
mechanism illustrated in Figure 1.6 The drainage of water to 60 mL at 0.005 wt % and remains practically constant,
from the foam column leads to a slow but steady narrowing VF ≈ 40-50 mL, at higher concentrations (0.01 to 0.05 wt
of the Plateau borders and to an increase of the pressure %). Similarly, for 2BO solutions VF drops to ≈60 mL at
that compresses the trapped oil drops. When the capillary 0.005 wt % and remains almost constant at higher
pressure exceeds the critical value (the entry barrier) for concentrations (Figure 10B). The final foam volume for
Effect of Oily Additives on Foams Langmuir, Vol. 17, No. 22, 2001 7009

tion, VIN ) 225 mL at 0.02 wt %. The further increase of

the oil concentration leads to reduction of the initial foam
volume, VIN ) 195 mL at 0.05 wt %. These variations of
VIN are beyond the experimental error ((10 mL) and have
been reproduced in several series of experiments.
Such a complex dependence of VIN on the oil concentra-
tion can be explained by assuming that the presence of
oil in the foaming solution leads to at least two opposite
effects that are competing with each other: (1) a reduced
DST, which facilitates the foam generation in the presence
of oil, Figures 4 and 5, and (2) an antifoam effect of the
oil droplets, which induce a coalescence of the newly
created bubbles, and thereby suppress the foam genera-
tion.
On the basis of the DST results, one can suggest the
following explanation of the observed dependence of VIN
on the IHNP concentration. At a very low oil concentration,
the DST of the solutions is practically the same as that
of the pure surfactant solution (Figure 5B). Therefore,
the oil acts only as an antifoam (effect 2), decreasing the
volume of the initially generated foam. At intermediate
oil concentration (0.01-0.02 wt %), the coalescence of the
oil drops with the solution surface during foaming leads
to reduced DST (Figure 4) and to larger VIN (effect 1
prevails). With the further increase of the oil concentration
(0.05 wt %), the DST remains the same, while the bubble
coalescence induced by oil drops becomes more intensive
(effect 2 prevails) and less foam is generated.
The dependence of VIN on the 2BO concentration seems
simplersVIN decreases from 220 mL at 0.002 wt % down
to 180 mL at 0.02 wt %. However, the foam tests with
solutions, pre-equilibrated with 2BO and removed drops,
have revealed that the drop removal results in a relatively
high VIN ≈ 230 mL. These changes in VIN are also larger
than the experimental error and have been reproduced in
several experimental series. These results can be explained
by DST and by the antifoaming action of oil drops. The
pre-equilibrated with 2BO solutions exhibit a reduced DST
(Figure 5A) even in the absence of oil drops. Therefore,
one may expect higher foamability of these solutions just
as observed in the experiment (effect 1). The increase of
the concentration of 2BO drops leads to enhanced coa-
lescence of the foam bubbles during foaming and, hence,
to a gradual decline of VIN (effect 2 takes over).
As discussed above, heptanol is totally inactive as a
foam breaker at 0.01 wt %, because it is rapidly solubilized
in the surfactant micelles. The initial foam volume is
increased in the presence of n-C7OH at low concentrations,
due to the reduced dynamic surface tension. One needs
concentrations well above the solubility limit of 0.15 wt
% to observe a substantial decrease of both VIN and VF.
These results confirm the conclusion of Kruglyakov et
al.11-13 that the antifoam activity of n-C7OH is primarily
Figure 10. Effect of the oil concentration on the initial foam due to the presence of oily drops.
volume, VIN, and the final foam volume, VF: (A) IHNP; (B)
2BO; (C) n-C7OH. The solutions contain 2.6 mM SDDBS, 12 The comparison of the results from the foam tests with
mM NaCl, and oil dispersed by rotor-stator homogenizer (fine fine and coarse emulsions (cf. Figures 9 and 10) shows no
emulsion, Ross-Miles test). big difference, in contrast to the large drop size effect
observed in other systems.6 This slight dependence of the
n-C7OH containing solutions is also a monotonically foam stability on the size of IHNP and 2BO drops can be
decreasing function of the alcohol concentration (Figure explained in the following way. In both types of emulsion
10C). The comparison of the respective curves for these (fine and coarse) we observe a significant number of large
oils shows that 2BO and IHNP are active antifoams even drops, which are able to destroy the foam after being
at concentrations as low as 0.005 wt %, whereas more compressed by the walls of the PBs (the entry barrier is
than 0.6 wt % of heptanol is needed for observing a very low for 2BO and IHNP). The stepwise transitions
significant antifoam activity. (presumably each of them induced by the entry of a single
Remarkably, the initial foam volume, VIN, is a non- large drop), observed in the foam decay curves, show that
monotonic function of the oil concentration: At IHNP several large drops are sufficient to trigger the observed
concentration of 0.002 wt %, VIN is substantially smaller process of foam destruction. In contrast, the systems
(160 mL) but gradually increases with the oil concentra- studied in ref 6 did not show a stepwise foam destruction
7010 Langmuir, Vol. 17, No. 22, 2001 Arnaudov et al.

pattern, which means that many drop entry events should hand, the emulsified oil drops induce a coalescence of the
take place for a significant foam destruction to occur. That foam bubbles during the foam generation process and, as
is why the stability of the foams studied in ref 6 was a result, the foamability of the solutions decreases at
strongly dependent on the mean drop size, while the upper higher oil concentrations. Therefore, the dependence of
boundary of the drop size distribution seems more the foamability of the surfactant solutions on the oil
important in the current systems. concentration is a nonmonotonic function and a careful
examination of this dependence (in relation to the oil
4. Concluding Remarks solubility limit) is needed before making any general
Foam tests and model experiments with several oils of conclusion about the effect of oil on the foaminess.
different chemical structure are performed to clarify how The foam tests confirmed the significant antifoam
these oils affect the foam stability and the foaminess of activity of n-C7OH, but at a relatively high concentration
SDDBS solutions. The results show that the activity of (above 0.2 wt %) due to its high solubility in the surfactant
these oils as foam breakers is determined mainly by their solution. For comparison, 2BO (branched alkanol) and
entry barrier (viz. by the stability of the asymmetric oil- IHNP (branched ester) demonstrate a significant antifoam
water-air films) which should be lowsotherwise the oil activity at concentrations as low as 0.005 wt %. Note that
drops remain arrested in the Plateau borders during the the n-C12OH (linear alkanol of similar molecular mass)
process of foam drainage, without being able to destroy has no antifoam activity at this concentration. Therefore,
the foam. This is the main obstacle for silicone oil to be the branched long-chain alkanols combine the advantages
an active antifoam in SDDBS solution. For n-C12OH the of long-chain n-alkanols (low solubility in the surfactant
high entry barrier and the negative bridging coefficient solutions) and medium-chain n-alkanols (low entry bar-
are both unfavorable with respect to its antifoam activity. rier). This makes the branched alkanols and other
On the other side, oils having low entry barrier and positive amphiphiles of similar molecular structure potentially
entry coefficient (2BO and IHNP) demonstrate a signifi- applicable as antifoam agents.
cant antifoam activity. Finally, let us note that the studied additives fall in the
No direct correlation between the spreading behavior category of the so-called “slow antifoams”,6,39 which destroy
of the oil (e.g., whether it spreads as a thin or thick layer the foam in the Plateau borders for a relatively long period
on the solution surface, rate of oil spreading, or value of of time (minutes or dozens of minutes). It has been well
the spreading coefficient) and its foam breaking activity established7,10,39 that the introduction of hydrophobic solid
is observed. Two oils of rather different spreading behavior particles in the oils leads to formation of mixed oil-solid
(2BO and IHNP) show similar antifoam activity, while compounds, which often act as “fast antifoams”, viz., they
n-C16 (whose spreading behavior is similar to that of 2BO) are much more active and destroy completely the foam
has no activity in the time scale of interest. On the other within seconds. Recent experiments showed10,39 that the
side, experiments described in refs 10 and 31 show that fast antifoams break the foam films almost immediately
the presence of a spread oil layer (even a very thin one) after their formation. The main physicochemical char-
on the surface of the foam film might reduce the entry acteristic, which determines whether a given antifoam
barrier. Therefore, the oil spreading might facilitate to would behave as a slow or fast one, is the entry barrier
some extend the drop entry and the foam destruction, (for fast antifoams it is below ca. 20 Pa).
without being a crucial factor, however.
Acknowledgment. The support of this study by
The effect of the oils on the foamability of the solutions
Colgate-Palmolive is gratefully acknowledged. The au-
is far more complex. The spreading oils are able to reduce
thors are indebted to Mrs. M. Temelska for the Ross-
the dynamic surface tension of the solutions, facilitating
Miles tests, to Dr. S. Roussev and Mr. T. Argirov for the
in this way the formation of a fresh surface. Hence, most
ellipsometrical measurements, and to Professor I. B.
of the oily additives enhance the foamability of the solu-
Ivanov and Professor P. A. Kralchevsky for the helpful
tions at concentrations below and around the oil solubility
discussions.
limit (long-chain linear alcohols such as n-C12OH might
be exceptions for reasons discussed in ref 20). On the other LA010600R