Journal of Colloid and Interface Science 424 (2014) 56–61

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

www.elsevier .com/locate / jc is

Nanoscale adhesive forces between silica surfaces in aqueous solutions

http://dx.doi.org/10.1016/j.jcis.2014.03.0200021-9797/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding author. Fax: +56 41 2204691.E-mail address: petoledo@udec.cl (P.G. Toledo).

Paula Troncoso a, Jorge H. Saavedra a, Sergio M. Acuña b, Ricardo Jeldres a, Fernando Concha c,Pedro G. Toledo a,⇑a Chemical Engineering Department and Surface Analysis Laboratory (ASIF), University of Concepción, PO Box 160-C, Correo 3, Concepción, Chileb Department of Food Engineering, University of Bio-Bio, PO Box 447, Chillán, Chilec Metallurgical Engineering Department, University of Concepción, PO Box 160-C, Correo 3, Concepción, Chile

a r t i c l e i n f o a b s t r a c t

Article history:Received 2 May 2013Accepted 6 March 2014Available online 15 March 2014

Keywords:Nanoscale forcesAdhesionSilicaAFMCavitationContact surfacesDehydroxylation

Nanoscale adhesive forces between a colloidal silica probe and a flat silica substrate were measured withan atomic force microscope (AFM) in a range of aqueous NaCl, CaCl2, and AlCl3 solutions, with concentra-tions ranging from 10�6 to 10�2 M at pH �5.1. Notably, the measured force curves reveal large pull-offforces in water which increase in electrolyte solutions, with jump-off-contact occurring as a gradualdetachment of the probe from the flat substrate rather than as a sharp discontinuous jump. The measuredforce curves also show that the number and size of the steps increase with concentration and notablywith electrolyte valence. For the higher concentration and valence the steps become jumps. We proposethat these nanoscale adhesive forces between mineral surfaces in aqueous solutions may arise fromnewly born cavities or persistent subnanometer bubbles. Formation of cavities or nanobubbles cannotbe observed directly in our experiments; however, we cannot disregard them as responsible for thediscontinuities in the measured force data. A simple model based on several cavities bridging the twosurfaces we show that is able to capture all the features in the measured force curves. The silica surfacesused are clean but not intentionally hydroxylated, as contact angle measurements show, and as such maybe responsible for the cavities.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction

It is well known that silica suspensions exhibit remarkablestability at their isoelectric point [1] against coagulation and sedi-mentation [2,3] and low viscosity [1,4]. The short-range repulsionnot predicted by DLVO theory [5,6] but usually observed in directforce measurements between silica surfaces arises from asurface-induced water effect, from the creation of a hydrogen-bonding network at the surface level [4,7–24]. This short-rangerepulsion as well as the small attractive van der Waals forces forsilica in water and in aqueous electrolyte solutions are likelyresponsible for the unusual stability of silica suspensions particu-larly at their isoelectric point. In the past this issue has been ofinterest to a wide range of processes involving transport andprocessing of silica slurries and pulps. Interest today is even great-er considering that industry is moving fast toward higher solidsloadings, and thus an appropriate control of rheological behaviorand physicochemical stability of the suspensions is crucial toobtain fluidity and stability as desired. A great variety of

experimental techniques have been used to characterize silica–silica interactions, however for assessing interaction forcesbetween silica surfaces mediated by aqueous solutions the surfaceforce apparatus (SFA) and the atomic force microscope (AFM) arethe favorites. Here we are interested in the forces that arise inthe separation of two silica surfaces after reaching direct contact.Given the repulsive character of the interaction in the approachingof the surfaces, one might expect a separation virtually free of hys-teresis. This is true for the interaction in water [20,22,25,26] butnot in aqueous electrolyte solutions [18,27–31]. Although the sil-ica–silica system has been reinvestigated several times, interesthas focused on approach force curves rather than on separationforce curves, therefore adhesive forces have not always been ob-served [14,19,25–27,32–34]. It has been reported that adhesion be-tween silica surfaces in water occurs only when the surfacesremain in contact for a long time [19] although according toYaminsky et al. [25,26], after several days in water the strengthof the adhesion decreases. Similar behavior is reported by Chapel[18]; silica adhesion in 0.1 M NaCl at pH 5.5 disappears after fewminutes. Less common in the literature is to find trends followedby silica–silica adhesion with electrolyte concentration in themeasurement of forces. Yaminsky et al. [26] and Meagher [28]

Concentration (M)0 1 2 3 4 5 6

Con

tact

ang

le (°

)

20

30

40

50

60

70NaClCaCl2AlCl3

Fig. 1. Glass contact angles measured through the liquid phase for variouselectrolyte solutions and concentrations.

P. Troncoso et al. / Journal of Colloid and Interface Science 424 (2014) 56–61 57

observed that adhesion between silica surfaces increases with saltconcentration, Freitas and Sharma [35] found no relationship be-tween the concentration of electrolyte and the adhesion values be-tween two silica surfaces, but Vakarelski et al. [36,37] and Fielden[38] did find such a dependency, adhesion values decrease withincreasing electrolyte concentration. These are three groups ofmanuscripts arriving to three different results. The experimentalevidence indicates that silica particle dispersion, aggregation andgelation depend on this elusive attractive force. The origin of sucha force is of great and broad scientific interest, particularly becausethere is as yet no single explanation of its cause. This force is in ex-cess to continuum van der Waals and electrical double-layer DLVOforces [5,6]. Here we use AFM to directly measure the adhesiveforce between silica surfaces in aqueous NaCl, CaCl2, and AlCl3

solutions, with concentrations ranging from 10�6 to 10�2 M atpH �5.1. Adhesive force measured between a colloidal silica probeand a silica substrate has unusual intensity and long range. Inrecent articles, Atkins et al. [29] and Guleryuz et al. [31] havereported long-range adhesive forces similar to ours explainingthem as originated in the compressive contact and elastic deforma-tion of the silica surfaces, contact forces however are too shortrange so as to be invoked to explain their measured forces. Ourforce curves reveal large pull-off forces in water and in electrolytesolutions, the jump-off-contact occurs as a gradual detachment ofthe probe from the flat substrate rather than as a sharp discontin-uous jump. Ladder type force curves have been measured severaltimes between hydrophobic surfaces (see for instance [38]) butnot between silica surfaces. Formation of cavities cannot beobserved directly in our experiments; however, we cannot discardthem as responsible for the discontinuities in the measured forcedata. We propose a simple model based on several cavitiesbridging the two surfaces to explain the measured force data.

2. Experimental

Glass microslides (B&C) typically 1 � 1 cm2 were used. The flatsilica substrates were glued to AFM stubs before use. Glass colloi-dal probes were prepared [39] by gluing a 20 lm in diametersphere (Duke Scientific Corporation) to the end of a tipless V-shaped, 100 lm long, 0.6 lm thick, Si3N4 cantilever (Veeco) withNorland Optical Adhesive 61 (Norland Products, USA). Spring con-stants of individual cantilevers were determined by the method ofstandards (standards provided by Park Scientific, USA) with theDimension 3100 AFM microscope and were typically 0.14 N/m.SEM and AFM images verified the quality of the modified cantile-vers. Interaction forces were measured in bi-distilled water, pHwas �5.1, and also in electrolyte solutions. NaCl, CaCl2, and AlCl3

(analytical chemical grade, Merck, Germany) were used in concen-trations ranging from 10�4 to 10�2 M. Experiments were carriedout without any buffering, pH was �5.1. All glassware used inthe preparation of solutions was detergent and alkali washed withfinal thorough rinsing in bi-distilled water. Prior to force measure-ment the mineral surfaces, substrate and sphere, were thoroughlyrinsed in high purity water (18.6 MO/cm), then with ethanol, andthen again with bidistilled water. No efforts were made to improvehydroxylation of the glass surface groups into silanol groups, thusperfect wetting should not be expected. Contact angles were mea-sured through the liquid phase for the various electrolyte solutionsused here with the Ramé-Hart contact angle goniometer. Surfaceroughness, assessed by AFM imaging with the Dimension 3100and measured as the root-mean-square roughness, was small forboth substrate, �2 nm, and microsphere, �1 nm. Forcemeasurements between AFM probes and substrates as functionof separation were conducted using an SPM-3 (Thermo, USA)multimode atomic force microscope equipped with a NanoscopeIIIa SPM control station, fluid cell (0.1 cm3), silicone pad for

vibration isolation, and acoustic enclosure. AFM allows continuousmeasurement of cantilever deflection vs. position as probe andsubstrate approach, commonly named extension, or separate, com-monly named retraction. Approaching force curves can be found inAcuña and Toledo [23], here we report on retraction force curves.Measurement of a typical force curve took less than 20 min; duringthis time the AFM roughness of the substrate remained unaltered.The force measurement protocol is well established [14,16] and thesample manipulation procedure is available [23]. Extension andretraction driving speeds were low, 280 nm/s, in order to minimizehydrodynamic contribution to the measured force. Force curveswere first verified to be independent of position on the substrate;measurements were always highly reproducible. Forces arereported normalized by the microsphere probe radius, that is, asinteraction energy between silica flat surfaces by virtue ofDerjaguins approximation, F(D) = 2pRE(D), where F is force,D distance, R probe radius, and E energy per unit area.

3. Results and discussion

Here we are interested in the forces that arise in the separationof two perfectly clean silica surfaces but not hydroxylated afterreaching direct contact. Fig. 1 shows glass contact angles measuredthrough the liquid phase for the various electrolyte solutions ofinterest here. Contact angle in pure water is 27� and in electrolytesolutions increases with concentration, very rapidly in the range<0.5 M, reaching 34� in NaCl, 36� in CaCl2, and 43� in AlCl3. Clearlyour surfaces are not perfectly wet by the liquid solutions and theirhydrophilic character decreases as electrolyte concentration andvalence increase.

Given the repulsive character of the silica–silica interaction inthe approaching of the surfaces, one might expect a separation vir-tually free of hysteresis. However, this is not so because the sepa-ration forces that we have measured for the silica–silica systemshow significant adhesive force with unusual intensity and longrange. It is important to mention that contact time between thesurfaces during the measurement is very short.

Fig. 2 shows approach-separation force curves measuredbetween a colloidal silica microsphere probe and a flat silicasubstrate in water and pH 5.1. Fig. 3 shows separation force curvesfor the same silica system immersed in aqueous NaCl, CaCl2 andAlCl3 solutions, with concentrations from 10�4 to 10�2 M and pHbetween 4.6 and 5.2. The curves clearly show the presence of anadhesive force. Force curves in Figs. 2 and 3 are representative ofat least six independent AFM measurements at two different pointson the substrate for each intervening liquid. Measurements were

D (nm)0 200 400 600 800 1000

F/R

(mN

/m)

-1.0

0.0

1.0

2.0ApproachSeparationvan der Waals

Fig. 2. AFM approach and separation force curves between a flat silica substrateand a silica-glass microsphere in bi-distilled water at 20 �C and pH �5.1. Approachforce data from Acuña et al. [23]. The curve is representative of several independentAFM measurements. The dashed line is the theoretical continuum van der Waalsforce with a Hamaker constant of 0.85 � 10�20 J.

58 P. Troncoso et al. / Journal of Colloid and Interface Science 424 (2014) 56–61

repeated three times per point. Clearly, the attractive componentof the force in Fig. 2, which is not van der Waals, should be evenstronger if we consider that is being shielded by a strongelectrostatic repulsive force, we have to note also that the surfacesare separated by a few hundred nm. The van der Waals force is alsoinsufficient as Fig. 2 shows to explain the extended ranges of theadhesive force that we have measured in aqueous electrolytesolutions.

In a recent article, Guleryuz et al. [31] have reported adhesiveforces similar to ours explaining them as originated in the com-pressive contact and elastic deformation of the silica surfaces. Asthey did we apply the DMT model [40] because the dimensionlessparameter of Tabor for the silica–silica system is much less than0.1 [41,42]. Table 1 shows equivalent calculations to [31] basedon the force curves measured here. Maximum adhesion force iscalculated as F = kDD where k is the elastic constant of the cantile-ver and DD is the surface-to-surface distance from the jump-off-contact distance to the equilibrium position of the cantilever(the zero line), which we read directly from the curves. Results cor-respond to averages over the six force curves measured here in twodifferent positions per electrolyte per concentration. The smallstandard deviations in Table 1 for each liquid medium suggest highreproducibility of the measurements. Assuming that the adhesionenergy totally originates in mechanical contact, as Guleryuz et al.[31] did, we also calculate the work of adhesion, the surface energyof the solid, and the total adhesion energy. Calculations areillustrated for the particular case of a suspension of particles with100 nm in radius. The total adhesion energy clearly increases withconcentration and electrolyte valence.

For NaCl and CaCl2 at concentrations 10�4 and 10�2 M, Table 1suggests that a suspension may remain indefinitely in sol state be-cause the total adhesive energy is not enough to overcome thethermal motion (�2kBT) for irreversible particle adhesion. Higherconcentrations of these electrolytes should lead to the gel state.For AlCl3 Table 1 suggests that the gel state may be attained forany of the concentrations considered here. Suspensions of particleslarger than 100 nm in radius should attained gel state for any of theelectrolyte conditions in Table 1.

Contact forces however are very short range and thereforeshould not be invoked to explain long range adhesive forces suchas those measured by Guleryuz et al. [31] at pH 12.5 and NaCl con-centrations of 100 and 1000 mN, which show separation jumps ofmore than 20 nm, nor to explain the adhesive forces measured byAtkins et al. [29], which show jumps of 50 nm at pH 5.5 and

600 nm at pH 4.1, nor to explain the adhesive forces measured inthe present work at pH �5 for a variety of electrolytes andconcentrations, which show separation jumps of more than300 nm. In our case the time of contact of the surfaces issufficiently short to discard any sintering mechanism of silicic pro-truding groups by the action of water. How to explain this adhesiveforce then?

Our force curves in Fig. 3 and Table 1 show clearly that the sur-face-to-surface distance from the jump-off-contact distance to theequilibrium position of the cantilever, and thus the adhesion force,increases with electrolyte concentration and particularly withelectrolyte valence. Notably, the force curves in Figs. 2 and 3 reveallarge pull-off forces in water and in electrolyte solutions and thejump-off-contact occurs as a gradual detachment of the probe fromthe flat substrate rather than as a sharp discontinuous jump. Suchladder type force curves have been measured several times but al-ways between hydrophobic surfaces (for instance see [43]). Theappearance of steps in the force curves between fluorinated glasssurfaces was anticipated by Parker et al. [43]. For our silica–silicasystem, Figs. 2 and 3 show that the number and size of the stepsincrease with concentration and notably with electrolyte valence(note the ordinate scale change for AlCl3 solutions). For the higherconcentration and valence the steps become jumps. We shouldnote that the expected electrostatic repulsive force between thesilica surfaces, which is large for water and diluted aqueous solu-tions, is greatly reduced in AlCl3 at 10�2 M, as a result the attrac-tion force in the latter case appears more fully and jumps can beseen more clearly. In Figs. 2 and 3 we see an adhesive force whichincreases with electrolyte concentration and valence, however thehidden hydrophobic force may be very similar in all cases. Thesecomplex features in the separation force curves and the complexsurface chemistry of our systems, particularly in AlCl3 [44], preventus from extracting the adhesive force as a simple subtraction of theclassical DLVO model from the measured AFM force data.

Formation of vapor cavities has been observed experimentally(for pioneering work see [45–47]) and tackled theoretically [48]always between hydrophobic surfaces. Our silica surfaces werenot intentionally hydroxylated to increase their hydrophilic char-acter; instead our surfaces were exposed to ambient air beforeuse in water. According to Atkins et al. [29] such exposure dehydr-oxylates the silica surfaces rendering them less hydrophilic whichis the case of our silica surfaces (Fig. 1). The hydrophilic characterof our silica surfaces decreases also as electrolyte concentrationand valence increases; this effect although present is small at theconcentrations used here. Thus, our silica surfaces behave as some-what hydrophobic instead of strongly hydrophilic as expected forhydroxylated silica surfaces. Under these conditions we proposethat these nanoscale adhesive forces between silica surfaces inaqueous solutions may arise from newly born cavities or persistentsubnanomenter bubbles. Cavities are driven by local high chemicalpotential of the intervening water in the vicinity of the contact ofthe surfaces. Water in such neighborhood is likely to be stressedat very small separations, possibly this water is unable to maxi-mize the formation of hydrogen bonds and thus would prefer toabandon such area thus creating a cavity. Persistent bubbles how-ever are not observed in the approach force curves [23], the ap-proach force curve in pure water is included in Fig. 2, no steps orjumps to the surface contact are observed, and the force in theapproach is always repulsive. Still a possibility is the existence ofsubnanometer sized bubbles, undetected in the approach, thatduring the contact between flat and microsphere probably hidein the nanometer sized roughness of these surfaces (less than1 nm). In the separation, these subnanometer bubbles cometogether to bridge the macroscopic surfaces. Here we propose thatcavities and or persistent bubbles, which may alternate with liquidbridges, have adhesive capabilities.

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5NaCl 0.0001 M

(a)-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5NaCl 0.001 M

(b)

0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5NaCl 0.01 M

(c)

-1.5

-1.0

-0.5

0.0

0.5

1.0CaCl2 0.0001 M

(d)-1.5

-1.0

-0.5

0.0

0.5

1.0CaCl2 0.001 M

(e)

0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000-1.5

-1.0

-0.5

0.0

0.5

1.0CaCl2 0.01 M

(f)

-2.0

-1.5

-1.0

-0.5

0.0

0.5AlCl3 0.0001 M

(g)-2.0

-1.5

-1.0

-0.5

0.0

0.5AlCl3 0.001 M

(h)

D (nm)0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000

-2.0

-1.5

-1.0

-0.5

0.0

0.5AlCl3 0.01 M

(i)

D (nm)D (nm)

D (nm) D (nm) D (nm)

D (nm) D (nm) D (nm)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

F/R

(mN

/m)

Fig. 3. Separation force curves between a flat silica substrate and a silica-glass microsphere in a range of solutions at 20 �C and pH �5.1. Each curve is representative ofseveral independent AFM measurements. Electrolytes are NaCl, CaCl2, and AlCl3. Concentrations range from 10�4 to 10�2 M. AFM approach force curves can be found in Acuñaet al. [23].

P. Troncoso et al. / Journal of Colloid and Interface Science 424 (2014) 56–61 59

Fig. 4 illustrates a single axisymmetric vapor cavity between asphere and a plane in a liquid. For two planar or nearly planar sur-faces bridged by a single cavity, the force between them originatesin the axial component of the surface tension force plus the axialcomponent of the pressure force exerted by vapor and ambientliquid. In mechanical equilibrium, it is easy to show that whenthese contributions are evaluated in the mid section of the bridge,parallel to the substrate, the force is given by

F ¼ pR2c 1� R2

R1

� �ð1Þ

where c is surface tension and R1 and R2 are principal radii of cur-vature of the cavity surface. Eq. (1) correctly neglects here the axialcomponent force exerted by gravity. Stable cavities of 700 nm have

been calculated by Christenson and Claesson [46], and 500 nm to1 lm by Parker et al. [43], both in hydrophobic systems in water.According to Eq. (1), the attractive character of the force increaseswhen electrolyte concentration and valence increase because va-por–liquid surface tension and contact angle increase. Contact angledirectly affects the shape of the bridge; when such angle increasesthe curvature radius R1 increases too. The attractive character of theforce is not limited to hydrophobic surfaces but also to moderatelyhydrophilic surfaces provided R1 > R2. Force curves in Figs. 2 and 3,and data in Table 1, show this behavior even though the effect ofelectrolyte concentration and valence is small, althoughdistinguishable, over surface tension and contact angle.

Formation of cavities or persistent air bubbles cannot beobserved directly in our experiments; however, we cannot disre-gard them as responsible for the discontinuities in the measured

Table 1Adhesion force between silica surfaces measured in electrolyte solutions at 20 �C and pH �5.1. Electrolytes are NaCl, CaCl2, and AlCl3. Concentrations range from 10�4 to 10�2 M.F = kDD, where k = 0.14 N/m is the elastic constant of the cantilever and DD is the surface-to-surface distance from the jump-off-contact distance to the equilibrium position ofthe cantilever (the zero line), which we read directly from the curves. The work of adhesion for identical solid surfaces is Wadhesion = 2c, the surface energy of the solid c is obtainedfrom the DMT model through F/Rsphere = 4pc, here Rsphere = 10 lm. For example, for a suspension of silica particles with 0.1 lm in radius, according to DMT model the contactradius between the particles under zero external load is a0 = (4pR2c/K)1/3, where K is reduced elastic modulus of contacting particles. For silica K = 4.94 � 1010 N/m2. Totaladhesive energy W accumulated in the contact area between the particles is W ¼Wadhesionpa2

0 which is in units of kBT.

Medium Concentration (M) DD (nm) F/Rsphere (mN/m) c (mN/m) Wadhesion (mJ/m2) W (kBT)

Water 1 � 10�6a 289.2 ± 21.5 4.05 ± 0.15 0.32 0.64 0.44

NaCl 1 � 10�4 370.8 ± 78.7 5.19 ± 0.55 0.41 0.83 0.661 � 10�3 568.3 ± 67.6 7.96 ± 0.48 0.63 1.27 1.351 � 10�2 649.2 ± 72.0 9.09 ± 0.51 0.72 1.45 1.69

CaCl2 1 � 10�4 485.0 ± 96.9 6.79 ± 0.68 0.54 1.08 1.041 � 10�3 536.7 ± 89.1 7.51 ± 0.63 0.60 1.20 1.231 � 10�2 666.7 ± 20.4 9.33 ± 0.15 0.74 1.49 1.76

AlCl3 1 � 10�4 732.5 ± 18.9 10.26 ± 0.13 0.82 1.63 2.061 � 10�3 741.7 ± 132.9 10.38 ± 0.93 0.83 1.65 2.111 � 10�2 764.2 ± 96.3 10.70 ± 0.68 0.85 1.70 2.21

a Water used is bi-distilled with conductivity of 18.6 MO/cm and concentration estimated in 10�6 M.

ϕ R

2θ1θ

1RD 2R

ϕ R

1

Fig. 4. Axisymmetric cavity/bubble between a sphere and a plane in a liquid. R isthe sphere radius, u the semi-angle subtended at the centre of the sphere by thecavity, h1 and h2 are the solid–liquid contact angles at the two triple lines, h1 = h2 = hwhen sphere and plane are made of the same material, R1 is the radius of curvaturedescriptive of the rotation about the vertical axis in the plane of the paper, R2 is theradius of curvature of the vapor–liquid surface in the plane perpendicular to thepaper, D is the length of the cavity.

(a)

(b)

iii

RF

D

i ii

Fig. 5. Schematic representation of several cavities/bubbles trapped between twosolid surfaces, some of the cavities are larger and more stable than others. Threestages of the separation are illustrated. (a) Contact angle is intermediate andremains the same in the three stages; the shape of the cavities is prolate spheroidalbut changes to almost a circular cylinder as separation of the surfaces increases.Cavities at the verge of failing are drawn with segmented lines. (b) Schematics ofthe corresponding forces in (a).

60 P. Troncoso et al. / Journal of Colloid and Interface Science 424 (2014) 56–61

force data. In a related work we have demonstrated by means of asimple macroscopic force balance (Eq. (1)) and molecular dynamicssimulation of a Lennard-Jones liquid that capillary bridges betweensurfaces of different nature are responsible for attractive forces notjust between solvophilic surfaces but also for moderately solvo-phophic surfaces; and thus for cavities or bubbles surrounded bythe same liquid the force is attractive even when the substratesare moderately solvophilic [49]. These results although for a LJfluid support the idea that for moderately hydrophilic surfaces itis also possible to have attractive forces due to cavities orpersistent bubbles bridging the silica surfaces. We propose asimple model to explain our force data based on several cavitiesand or nanobubbles bridging the two surfaces. Such cavities havea distribution of sizes and stability upon separation. The least sta-ble cavity fails first at small separation leading to a small step inaqueous solutions with low electrolyte concentration and valenceand to a jump in the presence of high electrolyte concentrationand valence; the second less stable cavity fails next and so on.

Fig. 5 shows a schematic representation of several vaporcavities trapped between two solid surfaces, some of the cavitiesare larger and more stable than others. Three stages of the separa-tion are illustrated. Frame (a) illustrates the array when contactangle is intermediate, proper of interaction in aqueous solutionswith high electrolyte concentration and valence, so the initialshape of the cavities is prolate spheroidal but changes to almosta circular cylinder as separation of the surfaces increases. Pressureinside the cavities is low and decreases as separation of thesurfaces increases. With increasing separation, R1 increases veryrapidly and R2 decreases slowly, the result according to Eq. (1) isan increasing net adhesive force between the surfaces. The thirdcavity from the left in Frame (a) is the weakest. The third stageshows the point at which this cavity fails (the cavity right before

the failure is drawn with segmented lines). After the failure, smallbubbles remain attached to the surfaces obeying the contact angleat the triple line. Frame (b) depicts a typical force curve for the sys-tem in (a). At point (i), the net adhesive force is given by Eq. (1) forR1 larger than R2. At point (ii) the net adhesive force increasesaccording to Eq. (1) because R1 increases more rapidly than R2

decreases, the third cavity in Frame (a) is about to fail. Contributionof this cavity to the net adhesive force is so significant that itsfailure drives a force jump to a much smaller adhesive force atpoint (iii). This event is repeated as many times as there are steps

P. Troncoso et al. / Journal of Colloid and Interface Science 424 (2014) 56–61 61

in the force curve. The enhanced wetting of the cavities in this caseincreases their attachment to the surfaces and so does the pull-offforce at the failure. This very simple model captures all the featuresin the force curves in Figs. 2 and 3.

4. Conclusion

Measured long range adhesive forces clearly increase with elec-trolyte concentration and valence. Contact forces may be too shortrange to be invoked to explain them. Likewise the attractive forceof van der Waals. In our experiments the time of contact of the sur-faces is sufficiently short to discard any sintering mechanism bythe action of water. Notably, our force curves reveal large pull-offforces in water and in electrolyte solutions, the jump-off-contactoccurs as a gradual detachment of the probe from the flat substraterather than as a sharp discontinuous jump. The measured forcecurves also show that the number and size of the steps increasewith concentration and notably with electrolyte valence. For thehigher concentration and valence the steps become jumps. Wepropose that these nanoscale adhesive forces between silica sur-faces in aqueous solutions may arise from newly born cavities orpersistent subnanometer bubbles. Formation of cavities and ornanobubbles cannot be observed directly in our experiments; how-ever, we cannot discard them as responsible for the discontinuitiesin the measured force data. We offer a simple model based on sev-eral cavities and/or nanobubbles bridging the two surfaces that isable to capture all the features in the measured force curves. Inthe model the least stable cavity fails first at small separation lead-ing to a small step in aqueous solutions with low electrolyte con-centration and valence and to a jump in the presence of highelectrolyte concentration and valence, the second less stable cavityfails next and so on. Dehydroxylation of our silica surfaces may beresponsible for the cavities/bubbles; however, cavities/bubblesneed to be considered in the analysis of measured force curves be-tween silica surfaces because they might never be totally absent.

Acknowledgments

We thank Project Conicyt/Fondecyt 1101023, Centro CRHIAM-Project Conicyt/Fondap-15130015 and Red Doctoral REDOC.CTA,MINEDUC Grant #UCO1202 for financial support. J.H.S. and R.J.thank CONICYT-Chile for graduate student fellowships.

References

[1] R.K. Iler, The Chemistry of Silica, Wiley Interscience, NY, 1979.

[2] L.H. Allen, E. Matijevíc, J. Colloid Interface Sci. 31 (1969) 287.[3] J. Depasse, A. Watillon, J. Colloid Interface Sci. 33 (1970) 430.[4] M. Colic, M.L. Fisher, G.V. Franks, Langmuir 14 (1998) 6107.[5] B.V. Derjaguin, L. Landau, Acta Physicochim. URSS 14 (1941).[6] E.J.W. Verwey, J.Th.G. Overbeek, Theory of the Stability of Lyophobic Colloids,

Elsevier, Amsterdam, 1948.[7] D.M. LeNeveu, R.P. Rand, V.A. Parsegian, Nature 259 (1976) 601.[8] D.M. LeNeveu, R.P. Rand, V.A. Parsegian, D. Gingell, Biophys. J. 18 (1977) 209.[9] J.N. Israelachvili, G.E. Adams, J. Chem. Soc. Faraday Trans. I 74 (1978) 975.

[10] R.M. Pashley, J. Colloid Interface Sci. 80 (1981) 153.[11] Ya.I. Rabinovich, B.V. Derjaguin, N.V. Churaev, Adv. Colloid Interface Sci. 16

(1982) 63.[12] J.N. Israelachvili, R.M. Pashley, Nature 306 (1983) 249.[13] R.G. Horn, D.T. Smith, W. Haller, Chem. Phys. Lett. 162 (1989) 404.[14] W.A. Ducker, T.J. Senden, R.M. Pashley, Nature 353 (1991) 239.[15] J.N. Israelachvili, Intermolecular and Surface Forces, Academic Press, Second

Edition, London, 1992.[16] W.A. Ducker, T.J. Senden, R.M. Pashley, Langmuir 8 (1992) 1831.[17] A. Grabbe, R.G. Horn, J. Colloid Interface Sci. 157 (1993) 375.[18] J.-P. Chapel, Langmuir 10 (1994) 4237.[19] G. Vigil, Z. Xu, S. Steinberg, J.N. Israelachvili, J. Colloid Interface Sci. 165 (1994)

367.[20] J.N. Israelachvili, H. Wennerstrom, Nature 379 (1996) 219.[21] J.C. Fröberg, O.J. Rojas, P.M. Claesson, Int. J. Miner. Process. 56 (1999) 1.[22] H.-J. Butt, B. Cappella, M. Kappl, Surf. Sci. Rep. 59 (2005) 1.[23] S.M. Acuña, P.G. Toledo, Langmuir 24 (2008) 4881.[24] S.M. Acuña, P.G. Toledo, J. Colloid Interface Sci. 361 (2011) 397.[25] V.V. Yaminsky, R.K. Yusupov, E.A. Amelina, V.A. Pchelin, E.D. Shchukin, Colloid

J. USSR 37 (1975) 918.[26] V.V. Yaminsky, B.W. Ninham, R.M. Pashley, Langmuir 14 (1998) 3223.[27] W.R. Bowen, N. Hilal, R.W. Lovitt, C.J. Wright, Colloid Surf. A 157 (1999) 117.[28] L. Meagher, J. Colloid Interface Sci. 152 (1992) 293.[29] D.T. Atkins, P. Kékicheff, O. Spalla, J. Colloid Interface Sci. 188 (1997) 234.[30] J.J. Adler, Ya.I. Rabinovich, B.M. Moudgil, J. Colloid Interface Sci. 237 (2001)

249.[31] H. Guleryuz, A.K. Royset, I. Kaus, C. Filiàtre, M.-A. Einarsrud, J. Sol-Gel Sci.

Technol. 62 (2012) 460.[32] Ya I. Rabinovich, Colloid J. USSR 39 (1977) 1094.[33] G. Peschel, P. Belouschek, M.M. Müller, R.M. Müller, R. König, Colloid Polym.

Sci. 260 (1982) 444.[34] R.G. Horn, D.T. Smith, J. Non-Cryst. Solids 120 (1990) 72.[35] A.M. Freitas, M.M. Sharma, J. Colloid Interface Sci. 233 (2001) 73.[36] I.U. Vakarelski, K. Ishimura, K. Higashitani, J. Colloid Interface Sci. 227 (2001)

111.[37] I.U. Vakarelski, K. Higashitani, J. Colloid Interface Sci. 242 (2001) 110.[38] M.L. Fielden, R.A. Hayes, J. Ralston, Phys. Chem. Chem Phys. 2 (2000) 2623.[39] S. Huntington, S. Nespolo, Microscopy Today 1–3 (2001) 32.[40] B.V. Derjaguin, V.M. Muller, Y.P. Toporov, J. Colloid Interface Sci. 53 (1975) 314.[41] D. Tabor, J. Colloid Interface Sci. 58 (1977) 2.[42] Y.-P. Zhao, L.S. Wang, T.X. Yu, J. Adhes. Sci. Technol. 17 (2003) 519.[43] J.L. Parker, P.M. Claesson, J. Phys. Chem. 98 (1994) 8468.[44] Ch. Li, P. Somasundaran, J. Colloid Interface Sci. 148 (1992) 587.[45] V.S. Yushchenko, V.V. Yaminsky, E.D. Shchukin, J Colloid Interface Sci. 96

(1983) 307.[46] H.K. Christenson, P.M. Claesson, R.M. Pashley, Proc. Indian Acad. Sci. (Chem.

Sci.) 98 (1987) 379.[47] H.K. Christenson, P.M. Claesson, Science 239 (1988) 390.[48] V.V. Yaminsky, V.S. Yushchenko, E.A. Amelina, E.D. Shchukin, J. Colloid

Interface Sci. 96 (1983) 301.[49] J.H. Saavedra, R.E. Rozas, P.G. Toledo, J. Colloid Interface Sci. 13 (2014) 2686.