Effect of Salts and Dissolved Gas on Optical Cavitation near Hydrophobic and Hydrophilic Surfaces
ABSTRACT The effect of four 1:1 electrolytes (KCl, KBr, NH4Cl, and CH3COONa) on optical (stimulated by laser pulse) cavitation in thin layers bounded by hydrophobic and hydrophilic surfaces has been explored. For water and all salts (up to 1 M) in the case of hydrophobic surfaces, the cavitation probability is enhanced as compared with the case of hydrophilic walls. The increased cavitation probability observed with hydrophobic surfaces can be linked to an enhanced concentration of gas-filled submicrocavities close to them. The phenomenon seems to depend strongly on dissolved gas. Variations in the probability of cavitation that occur with electrolyte are significant and depend on its concentration and type. The specific effect of electrolytes on optical cavitation in a thin layer likely makes sense only in terms of the previously neglected ionic dispersion interactions. The results obtained may have implications for the mechanisms of the long-range hydrophobic interactions between surfaces and hydrophobic slippage.
- SourceAvailable from: Jaroslaw Drelich
Dataset: 2007Enc Bubble-Solid[Show abstract] [Hide abstract]
ABSTRACT: Interaction of bubbles with solid surfaces is considered with respect to approach/contact, film rupture, and bubble attachment, with particle separation by flotation the classic example. During approach/contact bubble deformation may occur and the interaction involves momentum transfer as described by hydrodynamic forces. Subsequent interaction involving film rupture is governed by interfacial forces including van der Waals forces, electrostatic forces, hydration forces, and hydrophobic forces. The structure and stability of the film is considered with respect to interfacial water structure and the presence of surface stabilized, nanosized gas bubbles. Featured properties of the solid surface (roughness, heterogeneity) have a significant influence on water film stability and rupture. Finally the bubble attachment is examined in terms of the three-phase line of contact, its formation and relaxation.
- [Show abstract] [Hide abstract]
ABSTRACT: Interaction of bubbles with solid surfaces is considered with respect to approach/contact, film rupture, and bubble attachment, with particle separation by flotation the classic example. During approach/contact bubble deformation may occur and the interaction involves momentum transfer as described by hydrodynamic forces. Subsequent interaction involving film rupture is governed by interfacial forces including van der Waals forces, electrostatic forces, hydration forces, and hydrophobic forces. The structure and stability of the film is considered with respect to interfacial water structure and the presence of surface stabilized, nanosized gas bubbles. Featured properties of the solid surface (roughness, heterogeneity) have a significant influence on water film stability and rupture. Finally the bubble attachment is examined in terms of the three-phase line of contact, its formation and relaxation.Encyclopedia of Surface and Colloid Science, Edited by P. Somasundaran, 01/2007: chapter 1:1: pages 1-29; Taylor & Francis.
- [Show abstract] [Hide abstract]
ABSTRACT: Aggregation of nanodispersed particles plays an essential role in the technology of colloidal systems. In this work the formation of clusters composed of long-living gas nanobubbles in electrolyte aqueous solutions has been studied experimentally by combining three distinct but complementary laser techniques: dynamic light scattering (DLS), laser phase microscopy, and polarimetric scatterometry. We propose a mathematical approach to modeling the structure of spherical particle clusters that is based on the solution of inverse problem of optical wave scattering with allowance for cluster−cluster aggregation. In this way we found the characteristic size of nanobubbles and fractal properties of their clusters in aqueous solutions of NaCl. The described method can be applied to the exploration of clustering in a wide class of disperse systems of spherical particles.Journal of Chemical & Engineering Data 09/2012; · 2.05 Impact Factor
E ffect of Salts and Dissolved Gas on Optical Cavitation
near Hydrophobic and Hydrophilic Surfaces
N. F. Bunkin,†O. A. Kiseleva,‡A. V. Lobeyev,†T. G. Movchan,‡
B. W. Ninham,§and O. I. Vinogradova*,‡
Department of Wave Phenomena, Institute of General Physics, Russian Academy of Sciences,
38 Vavilova Street, 117 942 Moscow, Russia, Laboratory of Physical Chemistry of Modified
Surfaces, Institute of Physical Chemistry, Russian Academy of Sciences, 31 Leninsky Prospect,
117 915 Moscow, Russia, and Department of Applied Mathematics, Australian National
University, Canberra 0200, Australia
Received March 19, 1996. In Final Form: J anuary 22, 1997X
The effect of four 1:1 electrolytes (KCl, KBr, NH4Cl, and CH3COONa) on optical (stimulated by laser
pulse) cavitation in thin layers bounded by hydrophobic and hydrophilic surfaces has been explored. For
water and all salts (up to1 M) in the case of hydrophobic surfaces, the cavitation probability is enhanced
as compared with the case of hydrophilic walls. The increased cavitation probability observed with
hydrophobic surfaces can be linked to an enhanced concentration of gas-filled submicrocavities close to
them. The phenomenon seems to depend strongly on dissolved gas. Variations in the probability of
cavitationthat occur withelectrolytearesignificant anddependonitsconcentrationandtype. Thespecific
effect of electrolytes on optical cavitation in a thin layer likely makes senseonly in terms of thepreviously
neglected ionic dispersion interactions. The results obtained may have implications for the mechanisms
of the long-range hydrophobic interactions between surfaces and hydrophobic slippage.
Hydrophobization of a solid surface plays a role in
phenomena like adhesion, wetting, film stability, cavita-
and the nature of liquids adjoining it are the subject of
much interest. Especially in the last decade with the
advent ofnewtechniques,it hasbeenintensively studied.
measurements of the force-distance profiles between
macroscopic hydrophobic bodies. These measurements
haverevealedthepresenceof strong attraction (orders of
magnitude larger than the van der Waals force) and
extremely long range (measurable to 100 nm).1-5The
existence of an interaction at such distances challenges
fundamental notions on liquid structure, and despite
considerabletheoretical efforts, theorigins of theselong-
rangeattractionsremain controversial. Therehavebeen
several theoretical attempts to explain the phenomena.
Thefirst approach is based on water structural effects.6,7
However, thesetheories areunlikely tobeabletoexplain
thelong-rangenatureof interaction. Other ideas invoke
electrostatic interactions to account for the range, and
electrostatictheories predict theinteraction should scale
with a range of twice the inverse Debye length of the
electrolyte. Sometimes,thissalt dependenceisobserved,
e.g. ref 11. However it is not valid in general.12A more
recent proposal, focusing on the observed cavitation of
the water when the hydrophobic surfaces are separated
from contact,13involves the metastability of the film due
toits confinement between hydrophobic walls,14or sepa-
ration-induced phase transition.15
recently argued that the presence of dissolved gas in the
aqueous mediumhas been implicatedas thesourceofthe
long ranged hydrophobic attractions.16-18
Again, in parallel and equally mysterious is the
phenomenon of slippage of water over a hydrophobic
surface.19,20This alsodepends on thedissolvedgas being
in a different form18,21,22and on thechangein viscosity of
the solvent near the hydrophobic wall.23
of emulsions, hydrophobic suspensions, filtration of fine
hydrophobic particles, latex polymerization gassed or
degassed, andion bindingtomicellesalsoimplicatea role
for dissolved gas.24-26
It has also been
* To whom correspondence should be addressed.
†Institute of General Physics, Russian Academy of Sciences.
‡InstituteofPhysical Chemistry, Russian Academy ofSciences.
§Australian National University.
XAbstract published in AdvanceACS Abstracts, April 15, 1997.
(1) Blake, T. D.; Kitchener, J . A. J . Chem. Soc., Faraday Trans. 1
1972, 68, 1435.
(2) Israelachvili, J . N.; Pashley, R. M. J . Colloid InterfaceSci. 1984,
(3) Pashley, R. M.; McGuiggan, P. M.; Ninham, B. W.; Evans, D. F.
Science 1985, 229, 1088.
(4) Rabinovich, Ya. I.; Derjaguin, B. V. Colloids Surf. 1988, 92, 243.
(5) Christenson, H. K. In Modern ApproachestoWettability: Theory
and Application; Schrader, M. E., Loeb, G., Eds.; Plenum Press: New
York, 1992; Chapter 2.
(6) Eriksson,J .C.;Ljunggren,C.;Claesson,P.J .Chem.Soc.,Faraday
Trans. 2 1989, 85, 163.
(7) Derjaguin, B. V.; Churaev, N. V. Langmuir 1987, 3, 607.
(8) Attard, P. J . Phys. Chem. 1989, 93, 6441.
(9) Podgornik, R. J . Chem. Phys. 1989, 91, 5840.
(10) Tsao, Y.-H.; Evans, D. F.; Wennerstro ¨m, H. Langmuir 1993, 9,
(11) Ke ´kicheff, P.; Spalla, O. Phys. Rev. Lett. 1995, 75, 1851.
(12) Christenson, H. K.; Fang, J .; Ninham, B. W.; Parker, J . L. J .
Phys. Chem. 1990, 94, 8004.
(13) Christenson, H. K.; Claesson, P. M. Science 1988, 239, 390.
(14) Be ´rard, D. R.; Attard, P.; Patey, G. N. J . Chem. Phys. 1993, 98,
(15) Parker, J . L.; Claesson, P. M.; Attard, P. J . Phys. Chem. 1994,
(16) Craig, V. S. J .; Ninham, B. W.; Pashley, R. M. J . Phys. Chem.
1993, 97, 10192.
(17) Meagher, L.; Craig, V. S. J . Langmuir 1994, 10, 2736.
(18) Vinogradova, O. I.; Bunkin, N. F.; Churaev, N. V.; Kiseleva, O.
A.; Lobeyev, A. V.; Ninham, B. W. J . Colloid Interface Sci. 1995, 173,
(19) Churaev, N. V.; Sobolev, V. D.; Somov, A. N. J . Colloid Interface
Sci. 1984, 97, 574.
(20) Blake, T. D. Colloids Surfaces 1990, 47, 135.
(21) Ruckenstein, E.; Rajora, P. J . Colloid Interface Sci. 1983, 96,
(22) Vinogradova, O. I. Langmuir 1995, 11, 2213.
(23) Vinogradova, O. I. J . Colloid Interface Sci. 1995, 169, 306.
(24) Karaman, M.; Ninham, B. W.; Pashley, R. M. J . Phys. Chem.
1996, 100, 15503.
3024 Langmuir 1997, 13, 3024-3028
S0743-7463(96)00265-X CCC: $14.00© 1997 American Chemical Society
Recently we have obtained some results that may add
a new dimension to the problem. Laser-induced optical
cavitation27,28suggested that thereexist in aqueous bulk
long-life charged submicrocavitiessbubstons.30
size appears to be of the order of nanometers.32Conse-
quently (although we donot, and cannot, claim universal-
ity), the measured probability of optical cavitation (break-
down)33can serve as an indirect method of investigation
of the relative change in concentration of gas-filled
submicrocavities (see ref 27). In previous work18we
explored optical cavitation in a thin water layer confined
walls. In the case of hydrophobic surfaces, we observed
theincreased cavitation probability that can belinked to
The aim of the present paper is to explore the effect of
electrolytes and dissolved gas on optical cavitation near
hydrophobic and hydrophilic surfaces. Here we limited
ourselves toaqueous solutions of several 1:1 electrolytes.
The choice of the salts and of the range of their concen-
trations was partly motivated by some courious observa-
tion of salt dependence of coalescence of macrobubbles:16
It was found that some electrolytes reduce the mac-
robubble coalescense at high concentration (above ∼0.2
M), whereas others have no effect. These results could
indicatethat theions may beimportant in their effect on
water structure rather than for any further electrostatic
effect (which is consistent with the bubston model) and
that this ion-induced water structure could be connected
with hydrophobicattraction (in thecaseofref16it occurs
between macrobubbles). For some electrolytes the coa-
lescencephenomenon correlates with thesurfacetension
betweenmacroscopicbubblescouldplay animportant role
in thecoalescenceprocess.35Thus, herewetry toexplore
optical cavitation of just the salt pairs that affect mac-
robubble interactions differently. The differences ob-
served with solutions between hydrophobic and hydro-
philic surfaces seem to be quite dramatic and suggest a
possible connection between dissolved gas, electrolyte,
II. Materials and Methods
A. Optical Cavitation Measurements. Optical cavitation
was investigated with an experimental setup which has been
used in a few previous studies18,29(see Figure 1). A radiation
pulse in the single transverse mode of a pulse-repetitive YAG:
Nd3+laser (wavelength λ ) 1.06 µm and pulseduration τ ) 11.5
ns) with a fixed amount of energy was isolated from the train of
pulses and directed by a lens into a cuvette. Two designs of
cuvette were used.
The first design was the same as in previous studies;18,29i.e.,
the cuvette consisted of two polished glass plates separated by
a Teflon lining 25 µm thick with a confined drop of water or
water-electrolyte solution. The plates were fastened to each
other by theaid of metallic fixtures provided with water jackets
connected to a liquid thermostat.
thermostat was maintained at 20° with a precision of (0.1 °C.
It is clear that, for studying the cavitation in a thin cuvette, the
focal volume of the lens used should be located outside of it. In
our experiments the cuvette was mounted before the focus. As
a result, the diameter of the laser spot inside the cuvette was
approximately 1 mm; i.e., the focusing was not tight. This led
toa laser beamintensity that was sufficient toinduceformation
of cavitation (macro)bubbles. At the same time such a focusing
prohibited the formation of a laser spark (on cavitation there
was no breakdown proper) at the water-electrolyte solution/
glass interface, sothat there was nostrong thermal action. The
cuvetteareashot by laser pulseswaslocatedwithintheobjective
The temperature of the
(25) Zhou, Z. A.; Xu, Z.; Finch, J . A. J . Colloid Interface Sci. 1996,
(26) Zhou, Z. A.; Xu, Z.; Finch, J . A.; Liu, Q. Colloids Surf., A 1996,
(27) Bunkin, N. F.; Bunkin, F. V. Laser Phys. 1993, 3, 63.
(28) Bunkin, N. F.; Karpov, V. B. J ETP Lett. 1990, 52, 669.
(29) Bunkin, N. F.; Kochergin, A. V.; Lobeyev, A. V.; Ninham, B. W.;
Vinogradova, O. I. Colloids Surf., A 1996, 110, 207.
(30) In this approach the problem of the diffusion stability of
that in surrounding solution.31Generally speaking, thequestion of the
diffusion stability of a submicrobubble remains unresolved. However,
the charge of a submicrobubble can, in principle, provide a possibility
tension γ (that weknow must beless than theplanar value, 72 dyn/cm,
and greater than the limiting value γ ) 0 at zero radius) against an
opposing electrostatic pressure.32This is why such submicrobubbles
together with their ionic shells were termed bubstons (i.e. bubbles
stabilized by ions). Thepossibility of ion and ion pair adsorption on the
liquid/gas interfacewill becommentedon when discussing our results.
(31) Epstein, P. S.; Plesset, M. S. J . Chem. Phys. 1950, 18, 1505.
(32) Bunkin, N. F.; Bunkin, F. V. Sov. Phys. J ETP 1992, 101, 512.
(33) Theoptical cavitation is theformation of a macroscopic bubble,
while the breakdown is the formation of laser plasma under a laser
pulse. The optical cavitation should be considered as the first stage of
the optical breakdown, because the spark itself is formed only inside
this macroscopic bubble.27For plasma formation the proper fluency of
laser radiation is required, sothat optical breakdown may beobserved
only insidethefocal volumeofthelens, providedthat focusing is rather
(34) Weissenborn, P.; Pugh, R. Langmuir 1995, 11, 1422.
(35) Weissenborn, P.; Pugh, R. J . Colloid Interface Sci. 1996, 184,
F igure1. (top)Schematicdiagramofthesetupforinvestigation
of optical cavitation: (1) YAG:Nd3+laser; (2) lens; (3) cuvette;
(4) thermostat; (5) microscope; (6) two-coordinate table; (7)
computer; (8) radiation energy measuring device. (bottom)
Cuvettesfor thecavitationstudy: (1)thincuvette(first design);
(2) thick cuvette (second design).
Optical CavitationLangmuir, Vol. 13, No. 11, 1997 3025
field of an optical microscope, through which the presence of
cavitation from a laser pulse was determined. We alsochecked
with the microscope that the windows were not damaged in the
process, and the solutions were not spoiled. The cuvette itself
was mounted on a program-controlled, movable two-coordinate
table, serving tomove the cuvette intothe plane perpendicular
tothelaser beam. After emission ofeach laser pulsethecuvette
was shifted to a new site, and a new shot was made.
The second design of the cuvette was designed for the
investigation of the relative change in the optical breakdown
probability close to the liquid/air interface as compared to the
bulk case. In this case the cuvette was of a larger size, and the
focal volume was situated inside the cuvette. We remark here
that, in spiteof thefact that thefocusing was rather tight, there
was no influence of the laser beam on the cuvette walls. The
focal volumeherewas located at a distanceabout 1 cm from the
cuvettewalls, whilethefront of high pressureand temperature
diminishes steeply at a distanceof several millimeters from the
center of the laser plasma flash. In this case, the program-
controlled movable table also served to shift the cuvette in the
vertical direction, sothat we were able tostudy the breakdown
probability at different levelsofthecuvette, includingtheregion
near tothe interface with air. We remark here that, due tothe
concave form of the meniscus in the curvette, we were able to
direct the laser beam at a tangent to the liquid/air interface.
Unfortunately, we did not succeed in studying breakdown near
the solid (hydrophobic or hydrophilic) surfaces. To show that
objections on the grounds of possible contribution of contamina-
tions tothe breakdown probability can be ruled out, this design
of the cuvette was also used to study breakdown in water and
aqueous solutions containing added particles that absorb laser
radiation, as well as in degassed water.
probability w)n0/n,wheren0isthenumber ofpulsesthat caused
cavitationandnisthetotal numberoflasershots. Theprocedure
guaranteed that a sporadic character for laser stimulation of
cavitation (breakdown) would be retained. In all cases w was
100 shots at a fixed temperature (i.e. n ) 100).
B. Solution and Surface Preparation. Stock solutions (2
M) offour 1:1inorganicelectrolytes (KCl, KBr, NH4Cl, andCH3-
COONa) were prepared in water. The salts were of analytical
reagent quality, and of purity greater than 99 or 99.5%. Before
measurements, the stock electrolyte solutions were diluted to
prepare solutions of five concentrations in the range from 0.05
to1M. Waterwasdoublydistilled,andtheelectrical conductivity
of the bulk water was 2 × 10-6Ω-1cm-1, pH 6.5. The water-
electrolyte solutions were always in equilibrium with the
atmosphere, and hence they contained dissolved gas.
All glassware was cleaned in chromic acid and hydrogen
glassplatesfor theoptical cavitation cuvettewerecleanedusing
thesametechniqueandconsideredtobehydrophilic. For these,
practically completewettingoftheplatesby water wasachieved.
by reaction with dimethyldichlorosilane. Theglass plates were
exposedtothesilanevapor for aperiodof48hinadry desiccator.
Under these conditions, the chemisorption of silane36-38on the
glass plates leads to the formation of strongly hydrophobic
of92-97°. Experimentally, therewasnomeasurabledifference
inthecontact angleonmethylatedglassofadroplet ofwater and
a droplet of a 1 M solution of KCl, KBr, or NH4 Cl. A small
amount of CH3COONa leads to a drastic decrease in contact
angle on methylated glass down to 65-70°. Futher change in
concentration has no influence on contact angle. However, at
about 1 M the contact angle appears to increase. This may be
due to the second-layer adsorption of CH3COO-.
on optical breakdown, we used carbon powder. The main size
of the carbon particles was 1 µ m. Suspensions of several
in water or electrolyte solutions.
Doubly distilled water for some experiments on breakdown
was degassed using a vacuum pump and a liquid N2cold trap.
Before degassing, water was kept at the boiling point for a
minimum of 1.5 h; then it was cooled quickly and submitted to
3 consecutive operations of freezing and melting at continous
degassing. With this arrangement, a vacuum of 10-3Torr or
better was obtained. Water was then sealed and injected into
the cuvette. Care was taken not to introduce any air into the
III. R esults
A. E ffect of E lectrolytes. With the cuvette of the
first design, experiments were carried out following the
above scheme. The laser pulse intensity was an easily
variableparameter intheexperiments. It wasprescribed
sothat the probability of cavitation for pure water at 20
°C in thecuvettewith thehydrophilicsurfaces, w0, would
amounttoabout0.02. Inotherwordsabaselevel situation
wasrealized,inwhichabout twolaser pulsesinahundred
would induce cavitation. This was attained when the
intensity of the laser beam was about I ∼ 109W/cm2. At
the same energy input, and under the same conditions,
the probability of cavitation, w1, was then measured for
hydrophobicsurfaces. At thetemperature20°C theratio
in the value of this ratio may be attributed to a number
of difficult-to-control factors in each new series (e.g.,
of the energy of the laser beam, etc.). We remark here
that in a previous report18we found that for pure water
the value w1/w0differs slightly from that in the present
paper. That is because earlier we used slightly different
conditions, i.e. a different laser pulse intensity, so that
the value of w0for pure water was alsodifferent. Recall
it being important only that both w1 and w0 for all
experimental conditions should be less than unity; i.e.,
for the laser stimulation is retained. Then, at the same
energy input, and under the same focusing conditions,
theprobability ofcavitationwasmeasuredfor electrolytes
of different concentration confined between hydrophobic
or hydrophilic surfaces. The results for different 1:1
electrolytes at 20 °C are shown in Figure 2.
Tofinish this subsection, wehavetonotesomefurther
observations. The macroscopic air cavities had a size of
the order of the cuvette thickness, i.e. about 10-20 µm.
(36) Tripp, C.; Hair, M. Langmuir 1991, 7, 923.
(37) Tripp, C.; Hair, M. Langmuir 1992, 8, 1120.
(38) Tripp, C.; Hair, M. J . Phys. Chem. 1993, 97, 5693.
F igure2. Opticalcavitationprobabilitywversusconcentration
of salt C for a thin layer of CH3COONa (a), KBr (b), KCl (c),
or NH4Cl (d) confined between two hydrophobic (+) or two
hydrophilic(O) surfaces. Theleftmost points correspondtothe
case of pure water.
3026Langmuir, Vol. 13, No. 11, 1997Bunkin et al.
line of the macroscopic bubbles has the shape of an ideal
circle. If thesizeof a bubblethus formed is smaller than
theinterlayer thickness, then thebubblerapidly tearsoff
fromthewall andfloats. Forthecuvettewithhydrophobic
walls, the macroscopic cavity, independently of its size,
continues to“stick” tothecuvettewalls. Thecontact line
slightly differs from a circular form. The shape of such
a cavity depends on surface methylation nonuniformity.
B. E ffect of Dissolved Gas. An observation made
with cuvettes of the first design points strongly to a key
This was true for all salts and all concentrations used.
Namely, and we emphasize this, when the cavitation
immediately after flotation of the macrobubble could not
The effect of dissolved gas was studied with a cuvette
of the second design. Remember that in this case we
studied the probability of cavitation as a function of a
distance from the air/water interface. The intensity of
the laser beam in this experiment was about I ∼ 1010
W/cm2. We found that changes in the breakdown prob-
ability ascomparedwiththebulk casewereobservedonly
in a region that was less than 1 mm distant from the
boundary with air. Unfortunately, we cannot define the
value of this interval more precisely.
probability of breakdown in bulk water, then the prob-
ability w1near theinterfacewith air turns out tobemore
than10timeshigher. Thephenomenoncannot berelated
to a possible increase in concentration of dust particles
(from the atmosphere) near the boundary with air. The
influence of such absorbing radiation particles was
simulated by adding carbon powder to the water and
electrolyte solutions. In all the cases the breakdown
probability was found to decrease abruptly with an
increase in the suspension concentration. Clearly, the
laser beam intensity becomes insufficient to stimulate
breakdown. This is apparently connected with the
absorption of radiation before the focal region, as well as
with the thermal defocusing inside the focal neck.
A drastic decrease in breakdown probability was also
observed for degassed water.
tothebreakdowninwater incontact withtheatmosphere.
To stimulate the first flashes in degassed water, it was
as compared with that corresponding to the sporadic
regimein thegassedwater. In this casethepulseenergy
was about a 0.5 J ! These observations are crucial. They
confirm that the phenomenon is not due to impurities.
If w0 is the
It was found that no
IV. Concluding R emarks
Thus, the results presented extend our earlier work.18
They show there is significant dependence of optical
cavitation probability on electrolyte type, concentration,
substrate, and presence of dissolved gas.
previously interpreted the optical cavitation probability
as a method for comparison of concentrations of submi-
crocavities.27If the original model for optical cavitation
of submicrocavity structure. So, let us now list our main
observationsandtry togivesome(very tentative)analysis
from the point of view of such a structure.
There is a conclusion, made first for the case of pure
water, that it follows from our results can generalized at
least for 1:1 electrolytes. Namely, we found that for all
salts and all concentrations the probability of cavitation
than that in the case of hydrophilic surfaces. According
to our model,18,27this implies an increase in the concen-
tration of gas-filled submicrocavities close to the hydro-
phobic wall or directly associated with it as compared
with the hydrophilic case. We stress that in the present
study weobtained additional evidencetofavor theroleof
dissolved gas only, not impurities, in the cavitation
phenomenon. The existence of such submicrocavities is
in agreement with some observations made with some
hydrophobic force measurements5,13,15,17and slippage
Thereisalsoanimportant newobservationthat several
1:1 electrolytes givedifferent values for theprobability of
iscurrently notheory that explainstheadsorption ofions
at the liquid/gas interface, or indeed at the oil/water or
other (uncharged) interfaces.39The simple Onsager-
Samarasapproach40isalinear theory inwhichdesorption
of ions always occurs andis only very weakly ion specific.
Nonetheless, surface charge vs pH observations,41elec-
trokinetic measurements,42-45and thin-film studies46-49
allow one to conclude that the air/water interface is
charged and that thechargeis negativeand likely dueto
charging process.50Whether these carbonic acid anions
actually adsorb or simply enhance OH-adsorption
remains unclear.) Moreover, the measured values of
changes in interfacial tension with concentration at air/
water interfaces34,35dependcritically on ion pairs. Thus,
even for a planar interface, the Onsager theory often
fails,34,35,41-44,46-49a failure that can be traced to the
gasinterfacearepurely electrostaticinorigin, apart from
the linearization assumption and the very existence of
theHofmeister effect.39A completely neglectedeffect that
gives rise to specificity of adsorption of ion pairs is
dispersion interactions that dominate beyond 0.2 M
concentration.52If bubstons doexist, their very different
geometry and scale can affect this ionic adsorption in a
different way again. Henceit is not surprising that even
1:1 electrolytes, adsorption differences can occur. These
in turn could lead tothe expectation of different bubston
ultrastructure of solution. This, which we believe, was
reflected by our cavitation experiment (Figure 2).
One of our starting points was an experiment on
explicit correlation between our observations of optical
(39) Collins, K. D.; Washabaugh, M. W. Q. Rev. Biophys. 1995, 18,
(40) Onsager, L.; Samaras, N. J . Chem. Phys. 1934, 2, 528.
(41) Bergeron,V.;Waltermo,Å.;Claesson,P.M.Langmuir 1996,12,
(42) Li, C.; Somasundaran, P. J . Colloid Interface Sci. 1991, 146,
(43) Li, C.; Somasundaran, P. Colloids Surf., A 1993, 81, 13.
(44) Yoon, R. H.; Yordan, J . L. J . Colloid Interface Sci. 1986, 113,
(45) Graciaa, A.; Morel, G.; Saulner, P.; Lachaise, J .; Schechter, R.
S. J . Colloid Interface Sci. 1995, 72, 131.
(46) Manev, E. D.; Pugh, R. H. Langmuir 1992, 7, 2253.
(47) Waltermo, Å.; Manev, E.; Pugh, R. J . Dispersion Sci. Technol.
1994, 15, 273.
(48) Cohen, R.; Exerova, D. Colloids Surf., A. 1994, 85, 271.
(49) Radke, C. Private communication.
(50) It was recently well established that thechargeof theoil/water
interfaceis duetoadsorption of OH-ions, whiletheCO32-and HCO3-
ions are not potential-determining.51However, the hydrophobic air/
water interfacecouldbevery different from theoil/water interface(see
(51) Marinova, K. G.; Alargova, R. G.; Denkov, N. D.; Velev, O. D.;
Petsev, D. N.; Ivanov, I. B.; Borwankar, R. P. Langmuir 1996, 12, 2045.
(52) Ninham, B. W.; Kurihara, K.; Vinogradova, O. I. Colloids Surf.,
A, in press.
Optical Cavitation Langmuir, Vol. 13, No. 11, 19973027
cavitation and the experiment on macrobubble coales-
cence. This is likely due to the fact that the methylated
glasssurfacesusedinthepresent work arefundamentally
different in nature from the macroscopic bubbles.16
Presumably submicrobubbles may experiencea different
influence of salts as compared with the case of mac-
robubbles because of their much smaller size.
The above conclusion, however, does not exclude the
fact that the methylated glass surface is similar to the
liquid/gas interface. Actually, in both cases we found an
enhancedprobability ofcavitation. Infact,other physical
to consider it as a very hydrophobic surface.
an important qualitative role for dissolved gas and salt
in determing the ultrastructure of aqueous solutions, in
bulk and at interfaces. There is evidently a complex
interplay between dissolved gas, electrolyte type, elec-
gas in liquid and solution structure has hardly begun to
be considered theoretically in any depth and raises
questions of some interest. In solid state physics the
presence of impurity atoms or molecules and their
association is a central issue. In liquids, dissolved gas
molecules are present as such impurities at around 5 ×
10-3M, and it might be expected that they ought tohave
a similar role to play in setting ultrastructure.
Acknowledgment. This reseach was supported in
part by a grant from the J apan-Former Soviet Union
Scientist Collaboration Program of the J apan Society for
the Promotion of Science (J SPS) and in part by Grant
Nos. 96-03-32147 and 96-02-17236 from the Russian
Foundation for Basic Reseach (RFBR). The authors are
which led to this work.
3028 Langmuir, Vol. 13, No. 11, 1997Bunkin et al.