The Anisotropic Characteristics of Natural Fibrous Sepiolite As Revealed by Contact Angle, Surface Free Energy, AFM and Molecula

Home , Margarite

Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces A: Physicochemical and

Engineering Aspects

jou rnal homepage: www.elsevier.com/locate/colsurfa

The anisotropic characteristics of natural ﬁbrous sepiolite as revealed by contact

angle, surface free energy, AFM and molecular dynamics simulation

a,∗ b a

Birgul Benli , Hao Du , Mehmet Sabri Celik

Istanbul Technical University, Department of Mineral Processing Engineering, 34469 Maslak, Istanbul, Turkey

National Engineering Laboratory for Hydrometallurgical Cleaner Production Technology, 100190 Beijing, People’s Republic of China

a r t i c l e i n f o

a b s t r a c t

Article history: In this study, the anisotropic character of natural sepiolite was revealed by a combination of techniques

Received 1 December 2011

involving contact angle measurements, Atomic Force Microscopy (AFM), and molecular dynamics (MD)

Received in revised form 4 April 2012

simulations. Capillary rise technique was used to determine the change in contact angle values for several

Accepted 19 April 2012

wetting liquids and calculate the surface free energy and acid/base components of sepiolite. The surface

Available online 1 June 2012

heterogeneity of sepiolite for a single ﬁber surface corresponding to roughness parameters was calculated

from the AFM images. Finally, the interactions between water molecules and sepiolite mineral surfaces

Keywords:

were also computationally analyzed using MD simulation to understand the interfacial water structure

Sepiolite

and the conﬁguration of water molecules at the basal surface of sepiolite. It was found that due to the

Capillary rise

absence of hydrogen bonding sites, the hydrophobic sepiolite basal plane is not in close contact with

Contact angle

Surface energy components water molecules, thus leaving a 3 A void space at the basal plane.

1. Introduction determines the hydrophobicity and anisotropic character of sepio-

lite within the half-cell formula of Si12O30Mg8(OH,F)48H2O. While

Surface wettability, heterogeneity and hydrophobicity are hydrogen bonding sides of silanol groups (Si OH) are presented

important concepts in various ﬁelds including froth ﬂotation of on the external surface; owing to discontinuities and chain-layer

minerals [1], dewatering [2] and ﬁltration [3,4]. Hydrophobicity for molecular structure, Mg ions located at the edges of octa-

ﬂat surfaces is generally characterized by contact angle measure- hedral sheets exert more inﬂuence on the hydrophobicity of sepi-

ments between liquid, solid surface, and gaseous media. However, olite [7]. In other words; the breakage of Si O and Mg O bonds

the characterization of ﬁnely dispersed suspensions and the behav- provides many hydrogen bonding sites on the sepiolite edge sur-

ior of water conﬁned in porous materials such as sepiolite requires faces similar to the natural hydrophobic talc mineral whose edge

methods like capillary rise penetration which may help improve surfaces facilitate the formation of strong hydrogen bonds with

our understanding of bulk water/clay mineral interfaces. The use water dipoles [5]. The interest in sepiolite is more about its high

of molecular dynamics (MD) simulation for the study of mineral adsorptive capacity [8–10], catalytic performance [6,11], and rheol-

surface chemistry especially for characterizing hydrophobicity also ogy [12]. The extent of hydrophobicity or wettability is important in

provides an additional perspective on ﬂotation chemistry [5]. such applications. During water wettability measurements, water

Sepiolite, structurally similar to palygorskite, intersilite, amphi- molecules can enter into the porous structure, cause local swelling

bole, and raite, is one of the most important industrial and osmotic effects, upon which some macroscopic crack formation

magnesium-rich in 2:1 phyllosilicate clay minerals, i.e. octahe- occurs. This hinders the applicability of wettability measurement,

dral layer is bound above and below by a silica tetrahedral sheet i.e. sessile drop [13–15] and thin layer wicking method [16,17].

[6]. Since the discontinuous octahedral sheets extend only in one Generally, procedures to solve these type of problems are gene-

dimension, the tetrahedral sheets are divided into ribbons by a peri- rated from less direct and more complicated ones such as immer-

odic inversion of rows of tetrahedrons. The very large channels or sion calorimetry, inverse gas chromatography (IGC), dynamic vapor

tunnels are located between these ribbon strips and formed chain- sorption isotherms (DVS) [18], single particle contact angle using

layer molecular structure of sepiolite mineral as well as its unique colloidal probe technique by Atomic Force Microscopy [19] and

ﬁbrous structure (Fig. 1). Chain-layer molecular structure exactly recent developments of advanced analytical techniques which ana-

lyze the ﬁrst few atomic layers of the surface (X-ray Photoelectron

Spectroscopy (XPS) and Low Energy Ion Scattering Spectroscopy

∗ (LEISS), Time-of-Flight Secondary Ion Mass Spectrometry (ToF-

Corresponding author. Tel.: +90 212 285 6267; fax: +90 212 285 6323.

E-mail address: [email protected] (B. Benli). SIMS) and their combination [20]. However, Helmy and de Busetti

B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31 23

Fig. 1. Crystal structure of sepiolite at the (0 0 1) basal plane (a) and its edges orthogonal to the basal plane (b) and (c). The color representations are: red – oxygen atoms,

white – hydrogen atoms, yellow – silicon atoms and green – magnesium atoms. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to

the web version of the article.)

Table 1

[21] determined the surface properties of sepiolite from the exper-

Chemical analysis of the sepiolite sample.

imental data of spreading pressure using several minerals such as

palygorskite [22], talc [23], kaolinite [24] and montmorillonite [25]. Constituent Wt.%

Another alternative to direct contact angle measurements for sepi-

SiO2 49.85

olite is capillary rise technique based on the penetration of wetting Al2O3 2.38

liquids with its well established theoretical basis [26]. Fe2O3 0.87

MgO 20.15

MD simulation is an extremely efﬁcient tool that can be used

CaO 2.65

to explore water/water and water/mineral interactions, and to

Na2O 0.10

elucidate the structure of water at mineral surfaces, providing

K2O 0.36

detailed information and fundamental understanding on issues TiO2 0.13

LOI 23.5

such as mineral surface potential, dynamic behavior of nano-

conﬁned water, surfactant/macromolecule adsorption, and mineral

ﬂoatability [5,15,27]. In the past decade, considerable research

based on MD simulation has been reported on water structures,

Plasma Spectrometer) in ACME Analytical Lab., Canada. The miner-

dynamic and thermodynamic characteristics of water/mineral sys-

alogical analysis of the sample (Table 2) was also performed using

tems [5,28–34].

Shimadzu XRD-6000 equipped with Cu X-ray tube ( = 1.5405 A).˚

The aim of this study is to demonstrate the anisotropic character

The speciﬁc surface area of natural ﬁber sepiolite sample was

and the extent of hydrophobicity of natural sepiolite ﬁber surfaces 2

analyzed as 222 m /g using BET surface area by Quanto Chrome

with direct contact angle measurement using capillary rise method,

Monosorb Analyser (USA).

to calculate roughness parameters from AFM images and ﬁnally,

analyze the interactions between water molecules and sepiolite

surfaces using MD simulation.

Table 2

Mineralogical analysis of the sepiolite sample.

Mineral type Formula

2. Experimental

Sepiolite Mg4Si6O15(OH)2·6H2O

Dolomite CaMg(CO3)2

2.1. Materials

Albite (Na,Ca)Al(Si,Al)3O8

Minrecordite CaZn(CO3)2

The sepiolite consisting of 85 ± 3% sepiolite was obtained from Quartz SiO2

Calcite CaCO3

AEM Co., Sivrihisar region of Turkey. The chemical analysis of the

Montmorillonite NaO3(Al,Mg)2Si4O10(OH)2·4H2O

sample (Table 1) was accomplished with ICP (Inductively Coupled

24 B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31

Table 3

2.2. Contact angle measurements

Potential parameters for water/water, water/ion interactions [46,47].

Contact angle on sepiolite ﬁbers was measured with a Sigma 701 Species (Å) ε (kcal/mole) Charge (q)

tensiometer (KSV Instruments Ltd., Finland) using the capillary rise

Water hydrogen (HW) 0 0 0.41

␮

method. Sepiolite powder (size below 150 m) weighing 0.5 g was Water oxygen (OW) 3.169 0.1554 −0.82

−

placed in a standard tube of 9 mm internal diameter; the bottom Bridging oxygen (O) 3.169 0.1554 1.05

2 Silicon (Si) 3.706 1.84E−06 2.1

was replaced with a ﬁlter paper of 16,000 cm mesh density. The

Magnesium (Mg) 5.909 9.03E−07 1.36

procedure for obtaining a reproducible packing plays a key role in

the capillary rise method; therefore, a great care was taken when

preparing the packed bed. The assumption of the present study is

(AFM, XE-70E; Park Systems Corp., Suwon, Korea). As there is oscil-

that packed beds with a comparable bulk density have similar spa-

lation in cantilever frequency, it is typically chosen between 200

tial packing. In order to obtain a reproducible packing density, the

and 400 Hz in air [39]. The AFM measurements were performed

following packing procedure, which was the same for each exper- ◦

in the temperature (22 ± 2 C) and humidity controlled laboratory

iment, was performed; ﬁrst, the particles with a known mass of ∼

conditions during the winter season (RH 55%) at Istanbul Tech-

about 0.5 g were manually introduced into the column which then

nical University. Cantilevers were exposed to UV/O3 (UV Cleaner,

was tapped approximately 50 times and ﬁnally, compressed tightly

Bioforce Nanosciences) for 15 min prior to each experiment to

with a metal rod until the bed height no longer changed. Capillary

remove any possible contamination on each probe. In order to

rise experiments were carried out with the selection of wetting liq-

quantify the size of sepiolite ﬁbers on AFM images, all images were

uids for which properties were readily available in the literature;

processed by XEI software (Park Systems Corp., Suwon, Korea).

at least, three wetting liquids were necessary for the calculation

of surface free energy components. During the experiments, once

2.4. Molecular dynamics simulation

the gauze was in contact with the surface of the wetting liquid, the

weight of absorbed liquid was automatically measured during wet-

A 3-D crystal structure visualization of sepiolite was made

ting. The relation between the rate of capillary rise and penetration

using the interface program MERCURY [41,42] with access to The

time in such a system is given by the Washburn equation,

Cambridge Structural Database System (CSD System) and The Cam- r

h = cos t (1) bridge Crystallographic Data Centre (CCDC).

An MD simulation program DL POLY 214 [43] was used for the

analysis of sepiolite surfaces. A simple cubic cell containing water

where h is the height, m is the mass of liquid absorbed by the sample

molecules and the sepiolite mineral was deﬁned with periodic

in time t, r is the capillary tube radius, is the contact angle between

boundary conditions. The simple point charge (SPC) water model

the wetting liquid and the solid of the powder, is the liquid surface

[44], along with the recently developed CLAYFF force ﬁeld [45,46]

tension and is the liquid viscosity [35]. As the Washburn equation

was the basis for this simulation; the intermolecular potential

(Eq. (1)) is applicable to the porous media, the modiﬁed equations

parameters [46,47] are listed in Table 3.

are given below [36];

The initial conﬁguration of the sepiolite was constructed using

C 2

m2 = W lattice parameters provided by American Mineralogist Crystal

cos t (2a)

Structure Database [48,49]. Due to the lack of lattice information

of the water molecules within the sepiolite crystal, Simple Point

C 1 2

W = R(L · e · f ) (2b)

2 Charge (SPC) water molecules were added, and their structures

were analyzed in later sections. Initially, following the procedure

CW is known as Washburn or capillary constant, R is the average

reported elsewhere [29,45,50–53], sepiolite crystal was simulated

radius of the mean pore or effective pore diameter, f is the porous

in an isothermal–isobaric NPT ensemble, where moles (N), pressure

fraction, L is the length of column, and e is the column thickness.

(P) and temperature (T) are conserved with the pressure ﬁxed at

The surface free energy and surface free energy components of

0.1 MPa and the temperature ﬁxed at 300 K for 100 ps. After adding

natural sepiolite ﬁbers were derived from contact angle measure-

SPC water into the system, simulation was performed in a canon-

ments based on the Liftshitz–van der Waals interactions and van

ical, NVT ensemble that moles (N), volume (V) and temperature

Oss theory [37] as follows;

(T) are conserved, using Hoover’s thermostat [54]. The Leap-frog

LW LW + − − + method with a time step of 1 fs was used to integrate the particle

+ = + +

(1 cos ) L 2 S L S L S L (3)

motion. The Ewald sum has been used to account for the electro-

static interactions. A ﬁnal simulation time of 400 ps (4 × 10 steps

and

each of 1 fs) including a 200 ps equilibration period was performed.

LW AB LW + − The ﬁnal results were analyzed based on the production of 200 ps

= + = +

S S S S 2 S S (4)

simulation after the equilibration period.

where i is the Liftshitz–van der Waals component of the surface

+ −

free energy, is the electron acceptor and is the electron donor 3. Results and discussion

parameter of the Lewis acid–base component of the surface tension

of the material. Fig. 1a shows the crystal structure of sepiolite at the (0 0 1) basal

plane together with Fig. 1b and c which reveal the structures of

2.3. Atomic Force Microscopy (AFM) sepiolite at the edges orthogonal to the (0 0 1) plane. The red balls

represent oxygen atoms, the large light yellow balls are silicon

The main procedure for preparation and measuring the AFM atoms, the large green balls are magnesium atoms and the large

samples can be found in our previous paper on determining the white balls show hydrogen atoms. Atomic Force Microscopy (AFM)

size of sepiolite ﬁbers by AFM [38] and those previous articles on image was also undertaken to show porous and non-uniform sur-

montmorillonite and smectite type clay minerals [39,40]. The mea- face characteristics of sepiolite ﬁbers. Fig. 2 shows the AFM image

surements were performed in the non-contact mode using 910 of a ﬁber bundle, an assembly of multiple ﬁbers, with the length of

M-NSC14/Cr–Au-type cantilevers with 0.3–0.5 Hz scanning speed around 28 ␮m and approximately 4 ␮m in width. Wang et al. also

B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31 25

Fig. 2. AFM image of a single natural sepiolite bundle as an assembly of multiple ﬁbers (a), section analysis and the maximum slope of the surface feature (b).

obtained similar sepiolite nanoﬁbers ﬁbered by jet mill with their the appropriate wetting liquids. Evidently, polar and apolar liquids

␮

length greater than 9 m and average diameter of about 100 nm in our case were selected as hexane, octane, decane, water, ethy-

[55]. Using the corresponding AFM image in Fig. 2 and image anal- lene glycol, formamide and diiodomethane, as shown in Table 4.

ysis of line proﬁle by XEI software, the morphology of a single According to Holownia et al., evaluation of wetting properties of

sepiolite ﬁber surface was calculated by averaging the values of porous materials involving a bundle of capillaries requires a true

roughness parameters; the average roughness (Ra), the peak to val- selection of a reference liquid for which the contact angle is equal

ley (Rp-v) and the root mean square (rms) roughness (Rq) along two to zero. This is a big disadvantage of the capillary rise technique

different representative directions on the ﬁber surface – red and [57]. To eliminate this disadvantage, the selection of an n-alkane

green lines. The error in roughness was deducted by averaging high representing the highest penetration rate of n-alkane hydrocar-

precision profile measurements involving at least 5 profile lines. bons as a reference liquid is not sufficient; considering the complex

While the average Ra, Rq and Rp-v values were found as 68 ± 8, 83 ± 9 effect of viscosity, surface tension, and density, should be com-

and 293 9 nm along the red line, the same roughness parameters bined with another criteria for all wetting liquids [26]. Table 4

along the green line were 94 ± 8, 106 ± 9, 311 ± 9 nm, respectively. presents the change in penetration rate related to the slope of

As it is well-known, wettability is seriously affected by the pres- square root of liquid mass-penetration time (Eq. (2a)) and liquid

ence of micro- and/or nano-structures on solid surfaces [56]; the characteristics for all polar/apolar wetting liquids used; decane

dependence of roughness will be discussed in detail in the section was selected as the reference liquid as it has the highest material

−8

on contact angle studies. constant CW × cos = 50 × 10 and the corresponding liquid prop-

erties. Sepiolite ﬁbers exhibited anisotropic characteristics with

◦

72 of contact angle measured against water, similarly, Helmy and

3.1. Contact angle calculations from capillary rise tests

◦

de Busetti [21] obtained water contact angle value of 70.9 for a

natural sepiolite [21]. However, previous studies showed interest-

The relationships between the square of liquid mass and pene-

ing hydrophilic characteristics of some solids including glass beads

tration time for sepiolite bed (Eq. (2)) was used to determine the

using capillary rise method [26], thin layer wicking method (TLW)

contact angle values from the capillary rise curves in order to select

26 B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31

Table 4

◦

Properties of liquids used for capillary rise experiments (20 C) and obtained contact angle values.

Liquids Formula Character Density, Viscosity Surface tension MW (g/mol) Slope Calculation

3 2 2

g (cm ) (mPa s) (mJ/m ) (R = 0.99–1.0) parameter

CW × cos

(104)

Hexane C6H14 Apolar 0.655 0.294 18.43 86.18 0.0006 0.0024

Heptane C7H16 Apolar 0.684 0.386 20.14 100.21 –

Octane C8H18 Apolar 0.703 0.542 21.62 114.23 0.0008 0.0038

Decane C10H22 Apolar 0.727 0.907 23.83 142.29 0.0007 0.0050

Dodecane C12H26 Apolar 0.749 1.493 25.35 170.34 –

Water H2O Polar 0.999 1.002 72.80 18.00 0.0011 0.0016

Ethylene glycol HOCH2CH2OH Polar 1.110 19.900 47.70 62.08 0.0002 0.0006

1-Bromonaphthalene C10H7Br Polar 1.480 4.890 44.60 207.08 0.0000 –

Formamide HCONH2 Polar 1.129 3.302 58.20 45.04 0.0004 0.0019

Diiodomethane CH2I2 Polar 3.325 2.760 50.80 267.84 0.0070 0.0036

for glass [58], albite and feldspar [16]. Table 5 presents a number of analysis was found to increase the water contact angle signiﬁcantly

water contact angle data of clays and other minerals using different even at low surface roughnesses, thereby decreasing the wetta-

methods [59–66]. While for highly natural hydrophobic minerals, bility of the thin ﬁlm [70]. Similarly, moderately rough surfaces

◦

i.e. talc the water contact angle varied in the range of = 60–86.7 , obtained from the present AFM work, showed that a single natu-

the hydrophilic clays such as silica, mica, illite and kaolin the water ral sepiolite ﬁber has RMS roughness parameter of Rq = 83 ± 9 nm

◦ ◦ ◦ ◦

contact angles showed = 0 , 10 , 14 and 20 , respectively. along the ﬁber surface (red line in Fig. 2). Despite the high topo-

Interestingly, the natural hydrophobicity of sepiolite can be graphic resolution of AFM, it should be noted that the roughness

attributed to the surface roughness of single bundle and/or measurements via AFM are affected by the choice of scan range

ﬁbers whose AFM image is seen in Fig. 2. Young equation [67] and tip geometry. Depending on the quality of the tip, ﬁne details

deﬁnes the equilibrium contact angle ( E) on ﬂat and homoge- of the topography are more or less smeared out and Aimage can

nous surfaces. However, most surfaces are naturally rough and be generally smaller than Areal [72]. To eliminate the effects due

luckily the fundamental theories for rough and heterogeneous sur- to varying scan size, all images were taken at 1024 pixel size;

≈

faces proposed by Wenzel and Cassie-Baxter, and the theoretical at pixel numbers higher than 512, Aimage Areal [72]. Therefore,

models are available [68–70]. Among others, the Wenzel model the nanoscale corrugations could be captured consistently during

(cos measured = Rrough cos ﬂat) assumes that the liquid wets the the measurements. Here, measured is the experimentally measured

whole rough substrate to the roughness level usually being in apparent contact angle value of sepiolite ﬁbers using the capillary

␮ ◦

the range of 1 nm to 1 m [71]. The roughness of PTFE thin rise method (71.7 for water, Table 4). The Wenzel roughness fac-

ﬁlms (RMS roughness, Rq = 80 nm) as characterized by AFM fractal tor, Rrough, is the ratio between the real contact of a rough surface

to that of projected ﬂat surface [69,73]. Using the aspect ratio of

surface [74], Ar = height/bottom space = 8.3/27.94 = 0.297 ␮m, the

Table 5 height was deﬁned by the ratio of RMS roughness and Wenzel

A comparison of measured contact angles for water on several minerals.

roughness was calculated as Rrough = 1 + 2Ar = 1.594. Thus, the ﬂat

a (Wenzel) contact angle, , for the average contact angle of ﬂat

Mineral Technique Contact angle References ﬂat

◦

sepiolite surface was calculated as 78.66 . The calculated variation

Sepiolite CR 71.7 This study ◦ ◦

in the contact angles from 72 to 79 considering the effect of rough-

Sepiolite SS 70.9–71.4 [21]

Palygorskite SS 79 [22] ness demonstrates the anisotropic characteristics of sepiolite ﬁbers.

Albite TLW 43 [16] In another attempt to describe the inﬂuence of surface roughness

Orhoclase TLW 45 [16]

on the substrate wetting properties, the dependence of apparent

Calcite TLW 55.6 [59]

− ˛

contact angle ( measured = ﬂat ) on the shape of surface asperi-

Calcite TLW 77 [60]

Talc CR 83.9–86.7 [61] ties as described by (the maximum slope of the surface feature at

Talc TLW 78.3 [17] the liquid periphery) was studied [70]. Section analysis via AFM in

Talc TLW 84 [60]

Fig. 2b using the slope of the highest peak, ˛ was calculated as 2.27

Talc TLW 79.3 [59] ◦

and thus, ﬂat was obtained as 73.8 . Consequently, these results

Talc SS 60 [23]

are comparable. In this regard, further work is needed to develop

Talc SD 72 [15]

Kaolin SD 20 [62] new theoretical models to describe the wetting characteristics of

Calcite TLW and SD 55.59 and 6.2 [63] rough surfaces with nanometer size asperities and porosity.

Dolomite TLW and SD 76.46 and 51.7 [63]

Glass CR 47 [63]

Glass TLW 49.4 [59] 3.2. Water structure at sepiolite surfaces

Glass TLW and SD 49.41 and 9 [63]

Mica SD 10 [64]

Anisotropic characteristics of sepiolite ﬁbers were computa-

Silica CP 0 [65]

tionally revealed from molecular dynamics simulation of water

Methylated silica CP 90 [65]

Coarse silica TLW 34.9 [64] interactions at the sepiolite surface. The snapshot of water

Fine hydrophilic quartz CP 0 [66] molecules at sepiolite crystal surface is presented in Fig. 3. Inter-

Fine hydrophobic quartz CP 87.9–89.2 [66]

estingly, sepiolite exhibits unique crystallographic structures of

Hydroxyapatite TLW 68.1 [55]

talc-like ribbons linked to four identical ribbons through bridging

Octacalcium phosphate TLW 65.4 [55]

oxygen atoms. Sepiolite has complicated surface energy distri-

Fluroapatite TLW 75.2 [55]

Illite SD 14 [63] bution on the basal surface because of crystal structural defects

Pyrophyillite SD 38 [63]

(Fig. 3a), heterogeneity of cationic substitution (Fig. 3b) and sur-

SS, spreading pressure; TLW, thin layer wicking; CR, capillary rise; SD, sensile face heterogeneity of clay minerals [75]. Hydrophilic sites in close

drop. contact with water molecules on the basal surface of sepiolite can

B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31 27

Fig. 3. MD simulation snopshots of equilibrated water/sepiolite surface. The color representations are: red – oxygen atoms, white-hydrogen atoms, yellow – silicon atoms

2+ +

and green – magnesium atoms. Important structures are represented by (a) cationic substitution of Mg and formation Mg(OH) species, (b) crystal defects, (c) excluded

volume and hard wall effect, (d) water cluster bound to the surface, (e) water clusters bound to the edge of the channels, and (f) water clusters bound to the edge of the

tunnels. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of the article.)

be explained by the hydrolysis of Mg ions, as seen in Fig. 3b. Sim- angle values measured for diiodomethane (Table 6). The last two

ilar cationic substitutions of Mg ions are located on the basal terms on the right side of Eq. (3) drop out because all the com-

2+ LW

surface of talc, as at higher pH values, Mg begins to hydrolyze ponents are zero. The apolar part S was then calculated with

+ 2

and forms highly surface active Mg(OH) ions which decrease the the value of 33.5 mJ/m which represents an anisotropic character

negative charge of the surface [63]. The gaps between the water and nearly hydrophobic as advocated by van Oss and Giese [80];

phase and the talc-like ribbons in Fig. 3c reﬂect the dominance of the transition minerals exhibit both hydrophobic and hydrophilic

so-called “excluded volume” or “hard wall” effect observed with features with values up to 28 mJ/m .

typical hydrophobic surfaces [76,77]. In contrast, between talc-like

Using the calculated value for S , a system of equations for

ribbons, water molecules are tightly bonded to the edge surface water and ethylene glycol was solved to calculate the electron

of the ribbons, indicated by the close contact between the water acceptor and donor components of sepiolite surface as = 0.149

− S

molecules and these surfaces, as seen in Fig. 3d and e. and = 20.09. The sum of the two polar components was

S + −

The explanation for such water organization is that the pre- + = . >

positive as S S 20 24 0 causing a hydrophilic repulsion

sence of exposed magnesium atoms and bridging oxygen atoms or hydration pressure, this is in accord with van Oss and Giese’s

− 2

provides plenty of hydrogen bonding sites and facilitates the forma- < [80] classiﬁcation of S 28 mJ/m for hydrophobic materi-

− 2

tion of strong hydrogen bonds. Thus, water can wet these surfaces

≈

als, that is when S 28 mJ/m is the material to be neither

and the incompatibility between water and hydrophobic surface

hydrophobic nor hydrophilic. Among several clays and minerals,

accounts for the ﬂuctuating voids at the surface, as suggested in

the literature [78,79]. Sepiolite surface therefore is composed of

Table 6

hydrophobic/hydrophilic ribbons. Alternatively, it can be said that

Surface free energy components of polar liquids and contact angles of sepiolite

sepiolite ﬁbers show anisotropic surface characteristics. measured by capillary rise method.

LW + − AB

Liquid

3.3. Determination of surface free energy components of natural

Water 21.8 25.50 25.50 51.00 72.80 71.73

sepiolite ﬁbers

Ethyleneglycol 29.00 1.92 47.00 19.00 47.70 83.19

Bromonaphthaline 44.40 0.00 0.00 0.00 44.60 –

Formamide 39.00 39.6 2.28 19.00 58.20 67.84

The Liftshitz–van der Waals component of the surface free

LW Diiodomethane 50.80 0.00 0.00 0.00 50.80 75.64 energy, S is calculated from Eqs. (3) and (4) using the contact

28 B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31

Table 7

Values of the surface tension components and parameter in mJ/m for some clays and minerals [74–76].

LW + − + −

Group Mineral S S S S S G

Kaolinite Kaolin 35.9 0.4 34.3 85.75

Brittle mica Margarite 38.1 0.4 57.6 96.0 +40.4

Mica Phlogopite 40.8 0.6 59.3 98.8 +39.4

Biotite 42.0 1.4 47.8 34.1 +22.3

Muscovite 40.6 1.8 51.5 28.6 +25.7

Vermiculite Vermiculite 38.4 1.0 59.7 54.6 +38.6

Illite LSH 40.2 1.3 42.6 32.9 +17.4

DP-10 37.5 1.6 46.6 29.1 +22.7

Smectite Montmorillonite 40.7 1.5 29.2 19.5 −0.4

Laponite 41.3 0.9 24.2 26.9

Saponite 41.2 2.6 28.1 10.8

Polygorskite and Sepiolite, Spain 30.5 0.1 22.5 225

sepiolite Sepiolite, NV 30.5 0.2 17.3 86.5

Sepiolite, Turkey 33.5 0.149 20.09 134 −13.1

Palygorskite, FL 29.5 0 28.7 –

Palygorskite, China 29.2 0 13.4 –

−

Zero layer charge Talc 31.5 2.4 2.7 1.1 40.4

Talc 30.7 1.8 5.9 3.3 −49.5

Pyrophyllite 33.9 1.7 4.9 2.9 −45.2

− 2

≈ . were produced [16]. Similar results were also seen for other amine

sepiolite with S 20 09 mJ/m value represents the boundary

between hydrophilic and hydrophobic materials (Table 7). The treated studies [17].

maximum hydrophobicity is seen for naturally hydrophobic min- Furthermore, Van Oss and Giese [83] suggest other crite-

erals like talc which shows a balance between electron donor and ria for testing of anistropic character of mineral surfaces using

− +

≤

+ surface free energy criteria. Accordingly, G 0 means hydropho-

acceptor sites and S S ratio = 1.1 approaches unity; conversely,

the bic and G > 0 refers to hydrophilic surfaces. The surface free

− +

+ ≈ energy of sepiolite ﬁbers was thus calculated using the following maximum hydrophilicity is S S ratio 20–30 for smectites

2 equation;

and 59.3 mJ/m for mica. The anisotropic character of sepiolite

− +

emerged from a moderate value of and very low 0 = − LW AB

G + S S 2( S S ) (5)

values.

AB 2 As expected, the surface free energy of natural sepiolite ﬁbers

Besides, S of 5.3 mJ/m , the surface free energy of sepiolite

2 was found to yield a hydrophobic character with the value of

was determined from Eq. (4) as = 36.96 mJ/m for natural sepio-

S 2

2 −13.14 mJ/m . This value suggests that sepiolite surfaces present

lite compared to the previous studies as 38 mJ/m for acid-puriﬁed

2 a hydrophobic character compared to other minerals given in

sepiolite and 60 mJ/m for deliberated sepiolite by jet mill [21].

2 Table 7. Comparison of this value with the naturally hydropho-

The similar value of 31 mJ/m was reported for synthetic hydrotalc,

2 2 2 bic talc (−40.4 mJ/m ) indicates that sepiolite surfaces present less

50 mJ/m for talc, 45 mJ/m for muscovite and 30 mJ/m for oxide

2 2 hydrophobicity due to the anisotropic character of rough ﬁbers.

minerals [81]; 31.1 mJ/m for glass [82], 24.8 mJ/m using TLW for

Finally, the Hamaker constant of sepiolite ﬁbers used for describing

talc [60]. For untreated feldspar minerals, the surface free energy

2 the van der Waals forces were calculated using surface free energy

of albite and orthoclase was reported as 56.66 and 52.2 mJ/m ,

2 components in Eq. (6) and that given for water in Eq. (7);

and these values respectively became 27.88 and 27.87 mJ/m , upon

2 LW

modiﬁcation with 0.8 mg/L amine, i.e. more hydrophobic minerals

A = H

SS 24 0 S (6)

Fig. 4. MD simulation snapshots of equilibrated water/sepiolite surface and water distribution along sepiolite crystal surface normal. The color representations are: red –

oxygen atoms, white – hydrogen atoms, yellow – silicon atoms, and green – magnesium atoms. (For interpretation of the references to color in this ﬁgure legend, the reader

is referred to the web version of the article.)

B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31 29

Fig. 5. MD simulation snapshots of equilibrated water/sepiolite surface and water dipole moment angle distribution along sepiolite crystal surface normal. The color

representations are: red – oxygen atoms, white – hydrogen atoms, yellow – silicon atoms, and green – magnesium atoms. (For interpretation of the references to color in

this ﬁgure legend, the reader is referred to the web version of the article.)

A 2 LW LW LW LW

= − H + −

SLS 24 0 2 S L S L (7)

and where ASS and ASLS are the Hamaker constants correspond-

ing to van der Waals interaction between two sepiolite surfaces

in the liquid medium and H0 is the minimum separation dis-

tance (for H < H0, strong Born repulsion would appear). The best

approximation of H0 is 1.58 A˚ for various systems [84]. This

constant can be determined if the LW component of the sur-

face free energy of each phase, Hamaker contact of sepiolite

−17 −19

ﬁbers was obtained as ASS = 6.3 × 10 J and ASLS = −1.24 × 10 J.

For different entities, the Hamaker constant can be positive

or negative resulting in attractive or repulsive van der Waals

force [85]. Similar results were obtained for the air–water–silica

interaction. The value of the Hamaker constant was reported

−20

negative as A = −0.8 × 10 J and therefore the van der Waals

force becomes repulsive [86]. However, while the Hamaker con-

stant for the silica–water–silica interaction was a positive value

−20

of 0.83 × 10 J, air–water–hydrocarbon was a negative value of

−20

−0.20 × 10 J and hydrocarbon–water–hydrocarbon was a pos-

× −20

itive value of 0.41 10 J. It is noted that a negative Hamaker

constant results in a repulsive van der Waals force [85].

Fig. 6. Water dipole moment angle distribution along sepiolite crystal surface nor-

mal.

3.4. Molecular dynamics simulation studies

Substitution of cations of different charge can create a nega-

− +

tive charge imbalance on the 2:1 layer. The ratio of S S where

water molecules hydrogen bonded with the sepiolite edge surfaces,

Mg OH is slightly basic and Si OH is acidic in character, can be

and the sixth peak is caused by interfacial water molecules.

illustrated with idealizing the ratio of Si OH/Mg OH in a single

Figs. 5 and 6 present the distribution of water dipole moments

layer 2:1 which theoretically results in predominantly acidic char-

direction relative to the surface normal. Figs. 5 and 6 reveal that

acter at the edges. This is also computationally proven by water

strong hydrogen bonding of water molecules within the sepiolite

density distribution along the sepiolite surface normal plotted in

crystal as well as water molecules located in the hydrophilic chan-

Fig. 4. Due to the presence of water molecules within the sepio-

nels all exhibit speciﬁc orientation as suggested by the distribution

lite crystal, water density distribution is quite complicated. There

peaks of sharp water dipole moment direction. Further examina-

are a total of six major oxygen peaks over a distance of 30 A.˚ The

tion of peaks located at 5 and 7.5 A˚ away from zero suggests that

ﬁrst three oxygen peaks are due to the water molecules which are

◦

these two peaks are symmetrical relative to 90 due to equal hydro-

tightly hydrogen bonded with the magnesium and oxygen atoms

gen bonding experience by these water molecules. The peak at 15 A˚

of sepiolite; these are called “crystallized water”. A close exami-

away from zero is the sharpest due to the strong hydrogen bonding

nation of the second and third oxygen density peaks suggests that

network originated from the interaction with the exposed magne-

there is another peak between these two peaks with a peak height of

sium and oxygen atoms at sepiolite edge surfaces. The peak at 20 A˚

approximately 3/4 of these two peaks; this small peak is ascribed to

away from zero is quite broad suggesting random distribution of

the presence of water molecules sandwiched between the “crystal-

interfacial water molecule dipole moments.

lized water”. The fourth and ﬁfth oxygen peaks are attributed to the

30 B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31

4. Conclusions [22] A. Helmy, S.G. de Busetti, E.A. Ferreiro, The surface energy of palygorskite,

Powder Technol. 171 (2007) 126–131.

[23] A. Helmy, E.A. Ferreiro, S.G. de Bussetti, The surface energy of talc, J. Colloid

The anisotropic characteristics of natural sepiolite ﬁbers were

Interface Sci. 285 (2005) 314–317.

studied by a multitude of techniques involving contact angle, AFM [24] A. Helmy, E.A. Ferreiro, S.G. de Bussetti, The surface energy of kaolinite, Colloid

Polym. Sci. 283 (2004) 225–228.

and MD simulation. The salient ﬁndings are given below.

[25] A. Helmy, E.A. Ferreiro, S.G. de Bussetti, The surface energy of montmorillonite,

Direct contact angle measurement using capillary rise method

J. Colloid Interface Sci. 268 (2003) 263–265.

was used to calculate the surface free energy components of ﬁbers [26] T. Dang-Vu, J. Hupka, Characterization of porous materials by capillary rise

method, Physicochem. Probl. Miner. Process. 39 (2005) 47–65.

from Liftshitz–van der Waals interactions and van Oss theory.

− 2 [27] N.W. Ockwig, J.A. Greathouse, J.S. Durkin, R.T. Cygan, L.L. Daemen, T.M. Nenoff,

The value of = 20.09 < 28 mJ/m represents the hydropho-

S Nanoconﬁned water in magnesium-rich 2:1 phyllosilicates, J. Am. Chem. Soc.

bic materials indicating that sepiolite behaves as an anisotropic 131 (2009) 8155–8162.

material. [28] P. Gallo, M. Rapinesi, M. Rovere, Conﬁned water in the low hydration regime,

J. Chem. Phys. 117 (2002) 369–375.

AFM images of a sepiolite bundle with the dimensions of

[29] A.G. Kalinichev, R.J. Kirkpatrick, Molecular dynamics modeling of chloride bind-

× ␮ × ␮

(length width) 28 m 4 m yielded the roughness parameters ing to the surfaces of Ca hydroxide, hydrated Ca-aluminate and Ca-silicate

phases, Chem. Mater. 14 (2002) 3539–3549.

of Ra, Rq and Rp-v of 68 ± 8, 83 ± 9 and 293 ± 9 nm, respectively;

[30] S.H. Lee, P.J. Rossky, A comparison of the structure and dynamics of liquid

this reveals the porous and non-uniform surface characteristics of

water at hydrophobic and hydrophilic surface-a molecular dynamics simula-

sepiolite. Interaction of water molecules with the sepiolite min- tion study, J. Chem. Phys. 100 (1994) 3334–3345.

eral surface was computationally analyzed using MD simulation to [31] J.R. Rustad, Molecular models of surface relaxation, hydroxylation, and surface

charging at oxide-water interfaces, Rev. Mineral. Geochem. 42 (2001) 169–197.

reveal the interfacial water structure and the conﬁguration of water

[32] J.R. Rustad, A.R. Felmy, E.J. Bylaska, Molecular simulation of the

molecules at the basal surface of sepiolite. Hard wall effect was

magnetite–water interface, Geochim. Cosmochim. Acta 67 (2003) 1001–1016.

observed at 3 A˚ distance from the surface and water molecules. Sim- [33] E. Spohr, C. Hartnig, P. Gallo, M. Rovere, Water in porous glasses. A computer

2+ + simulation study, J. Mol. Liq. 80 (1999) 165–178.

ilarly, cationic substitution of Mg and the presence of Mg(OH)

[34] J. Wang, A.G. Kalinichev, R.J. Kirkpatrick, Molecular modeling of water structure

ion at natural pH of sepiolite are important mechanisms for iden-

in nano-pores between brucite (0 0 1) surfaces, Geochim. Cosmochim. Acta 68

tifying the anisotropic characteristics of sepiolite ﬁbers. (2004) 3351–3365.

[35] E.W. Washburn, The dynamics of capillary ﬂow, Phys. Rev. 18 (1921) 273–283.

[36] V. Parry, E. Appert, J.C. Joud, Characterization of wettability in gas diffusion layer

for proton exchange membrane fuel cells, in: Fundamentals and Developments

References

of Fuel Cells Conference 2008 – FDFC2008, Nancy, France, 2008.

[37] C.J. Van Oss, R.F. Giese, Role of the properties and structure of liquid

[1] T.T. Chau, A review of techniques for measurement of contact angles and their water in colloidal and interfacial systems, J. Dispers. Sci. Technol. 25 (2004)

applicability on mineral surfaces, Miner. Eng. 22 (2009) 213–219. 631–655.

[2] G. Stroh, W. Stahl, Effect of surfactants on the filtration properties of fine par- [38] M.F. Can, M. Cınar, B. Benli, O. Özdemir, M.S. Celik, Determining the fiber size

ticles, Filtr. Sep. 27 (1990) 197–199. of nano structured sepiolite using Atomic Force Microscopy (AFM), Appl. Clay

[3] O. Ozcan, B. Benli, A.N. Bulutc¸ u, H. Manav, Correlations between the shear Sci. 47 (2010) 217–222.

strength of mineral ﬁlter cakes and particle size and surface tension, Colloids [39] M. Plaschke, T. Schafer, T. Bundschuh, T.N. Manh, R. Knopp, H. Geckeis, J.I. Kim,

Surf. A 187 (2001) 405–413. Size characterization of bentonite colloids by different methods, Anal. Chem.

[4] C. Qing, W. Bigui, H. Yingdong, Capillary pressure method for measuring 73 (2001) 4338–4347.

lipophilic hydrophilic ratio of ﬁlter media, Chem. Eng. J. 150 (2009) 323–327. [40] B.R. Bickmore, M.F. Hochella, D. Bosbach, L. Charlet, Methods for performing

[5] H. Du, J.D. Miller, A molecular dynamics simulation study of water structure atomic force microscopy imaging of clay minerals in aqueous solutions, Clays

and adsorption states at talc surfaces, Int. J. Miner. Process. 84 (2007) 172–184. Clay Miner. 47 (1999) 573–581.

[6] J.E. Kogel, N.C. Trivedi, J.M. Barker, S.T. Krukowsk, Industrial Minerals & Rocks: [41] I.J. Bruno, J.C. Cole, P.R. Edgington, M.K. Kessler, C.F. Macrae, P. McCabe, J. Pear-

Commodities, Markets, and Uses, 7th ed., SME Inc., 2006. son, R. Taylor, New software for searching the Cambridge Structural Database

[7] E. Ruiz-Hitzky, Molecular access to intracrystalline tunnels of sepiolite, J. Mater. and visualising crystal structures, Acta Cryst. 58 (2002) 389–397.

Chem. 11 (2001) 86–91. [42] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock, L.R.

[8] E. Sabah, Adsorption mechanism of various amines onto sepiolite, Ph.D. thesis, Monge, R. Taylor, J. van de Streek, P.A. Wood, Mercury CSD 2.0 – new features

Osmangazi University, Eskis¸ ehir, Turkey, 1998. for the visualization and investigation of crystal structure, J. Appl. Cryst. 41

[9] M. Kara, H. Yuzer, E. Sabah, M.S. C¸ elik, Adsorption of cobalt from aqueous (2008) 466–470.

solution onto sepiolite, Water Res. 37 (2003) 224–232. [43] W. Smith, T.R. Forester, DL POLY user manual, J. Mol. Graphics 14 (2006)

[10] S. Lazarevic,´ I.J. Castvan, D. Jovanovic,´ S. Milonjic,´ D. Janackovi´ c,´ R. Petrovic,´ 136–141.

2+ 2+ 2+

Adsorption of Pb , Cd and Sr ions onto natural and acid-activated sepiolites, [44] H.J.C. Berendsen, W.F. Van Gunsteren, E. Egberts, J. De Vlieg, Dynamic simula-

Appl. Clay Sci. 37 (2007) 47–57. tion of complex molecular systems, ACS Symp. Ser. 353 (1987) 106–122.

[11] J. Liang, W. Lijuan, Effect of surface active agent on deﬁbering for sepiolite ﬁber [45] R.T. Cygan, Molecular models of hydroxide, oxyhydroxide, and clay phases and

bundles, Adv. Mater. Res. 96 (2010) 49–54. the development of a general force ﬁeld, Rev. Miner. Geochem. 42 (2001) 1–35.

[12] M. C¸ ınar, E. Sabah, M.S. Celik, Rheological properties of sepiolite ground in acid [46] R.T. Cygan, J.J. Liang, A.G. Kalinichev, Molecular models of hydroxide, oxyhy-

and alkaline media, Appl. Clay Sci. 42 (2009) 422–426. droxide, and clay phases and the development of a general force ﬁeld, J. Chem.

[13] B. Janczuk, E. Chibowski, T. Biaeopiptrowicz, L. Hoeysz, A. Kliszcz, Inﬂuence Phys. B. 108 (2004) 1255–1266.

of exchangeable cations on the surface free energy of kaolinite as determined [47] H.J.C. Berendsen, J.R. Grigera, T.P. Straatsma, The missing term in effective pair

from contact angles, Clays Clay Miner. 37 (1989) 269–272. potentials, J. Phys. Chem. 91 (1987) 6269–6271.

[14] J. Bachmanna, R. Hortonb, R.R. van der Ploega, S. Wochea, Modiﬁed sessile drop [48] J.W. Gruner, The crystal structures of talc and pyrophyllite, Z. Kristallogr. 88

method for assessing initial soil–water contact angle of sandy soil, Soil Sci. Soc. (1934) 412–419.

Am. J. 64 (2000) 564–567. [49] B. Perdikatsis, H. Burzlaff, Strukturverfeinerung am talk Mg3[(OH)2Si4O10], Z.

[15] J. Nalaskowski, B. Abdul, H. Du, J.D. Miller, Anisotropic character of talc surfaces Kristallogr. 156 (1981) 177–186.

as revealed by streaming potential measurements, atomic force microscopy, [50] H.J.C. Berendsen, J.P.M. Postma, W.F. Van Gunsteren, J. Hermans, Inter-

molecular dynamics simulations and contact angle measurements, Can. Metall. action models for water in relation to protein hydration, in: Jerusalem

Quart. 46 (2006) 73–87. Symposia on Quantum Chemistry and Biochemistry, vol.14, 1981,

[16] C. Karagüzel, M.F. Can, E. Sönmez, M.S. Celik, Effect of electrolyte on surface pp. 331–342.

free energy components of feldspar minerals using thin-layer wicking method, [51] R.J. Kirkpatrick, A.G. Kalinichev, J. Wang, Molecular dynamics modeling of

J. Colloid Interface Sci. 285 (2005) 192–200. hydrated mineral interlayers and the surfaces: structure and dynamics, Min-

[17] Z. Li, R.F. Giese, C.J. van Oss, J. Cases, The surface thermodynamic properties of eral. Mag. 69 (2005) 289–308.

talc treated with octadecylamine, J. Colloid Interface Sci. 156 (1993) 279–284. [52] R.J. Kirkpatrick, A.G. Kalinichev, J. Wang, X. Hou, J.E. Amonette, The application

[18] L. Galet, S. Patry, J. Dodds, Determination of the wettability of powders by the of vibrational spectroscopy to clay minerals and layered double hydroxides, in:

Washburn capillary rise method with bed preparation by a centrifugal packing CMS Workshop Lectures, vol. 13, 2005, pp. 239–285.

technique, J. Colloid Interface Sci. 346 (2010) 470–475. [53] J. Wang, A.G. Kalinichev, R.J. Kirkpatrick, Molecular dynamics modeling of the

[19] A.V. Nguyen, J. Nalaskowski, J.D. Miller, The dynamic nature of the contact angle 10-A phase at subduction zone conditions, Earth Planet. Sci. Lett. 222 (2004)

measured with atomic force microscopy, J. Colloid Interface Sci. 262 (2003) 517–527.

303–306. [54] S. Melchionna, G. Ciccotti, B.L. Holian, Hoover NPT dynamics for systems varying

[20] S. Abreu, C. Brien, W. Skinner, ToF-SIMS as a new method to determine the in shape and size, Mol. Phys. 78 (1993) 533–544.

contact angle of mineral surfaces, Langmuir 26 (2010) 8122–8130. [55] F. Wang, J. Liang, Q. Tang, L. Li, L. Han, Preparation and far infrared emission

[21] A. Helmy, S.G. de Busetti, The surface properties of sepiolite, Appl. Clay Sci. 255 properties of natural sepiolite nanoﬁbers, J. Nanosci. Nanotechnol. 10 (2010)

(2008) 2920–2924. 2017–2022.

B. Benli et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 408 (2012) 22–31 31

[56] Y. Liu, W. Wu, G. Sethuraman, G.H. Nancollas, Intergrowth of calcium [72] H. Kamusewitz, W. Possart, Wetting and scanning force microscopy on rough

phosphates:an interfacial energy approach, J. Cryst. Growth 174 (1997) polymer surfaces: Wenzel’s roughness factor and the thermodynamic contact

386–392. angle, Appl. Phys. A 76 (2003) 899–902.

[57] D. Hołownia, I. Kwiatkowska, J. Hupka, An investigation on wetting of porous [73] A. Munoz-Noval,˜ M. Hernando Pérez, V. Torres Costa, R.J. Martín Palma, P.J. de

materials, Physicochem. Probl. Miner. Process. 42 (2008) 251–262. Pablo, M. Manso Silván, High surface water interaction in superhydrophobic

[58] E. Chibowski, L. Hołysz, Use of the Washburn equation for surface free energy nanostructured silicon surfaces: convergence between nanoscopic and macro-

determination, Langmuir 8 (1992) 710–716. scopic scale phenomena, Langmuir 28 (2012) 1909–1913.

[59] W. Wu, R.F. Giese, C.J. van Oss, Change in surface. Properties of solids caused [74] H. Nakae, M. Yoshida, M. Yokota, Effects of roughness pitch of surfaces on their

by grinding, Powder Technol. 89 (1996) 129–132. wettability, J. Mater. Sci. 40 (2005) 2287–2293.

[60] I. Yıldırım, Surface free energy characterization of powders, Ph.D. thesis, Mining [75] F. Villieras, L.J. Michota, J.M. Casesa, I. Berenda, F. Bardota, M. Franc¸ oisa, G.

and Minerals Engineering, Virginia Polytechnic Institute and State University, Gérarda, J. Yvona, Stud. Surf. Sci. Catal. 104 (1997) 573–623.

Blacksburg, VA, 2001. [76] F.F. Abraham, The interfacial density proﬁle of a Lennard–Jones ﬂuid in

[61] H. Casanova, J.A. Orrego, J. Zapata, Oil absorption of talc minerals and dispersant contact with a (1 0 0) Lennard–Jones wall and its relationship to idealized

demand of talc mineral non-aqueous dispersions as a function of talc content: ﬂuid/wall systems: a Monte Carlo simulation, J. Chem. Phys. 68 (1978)

a surface chemistry approach, Colloids Surf. A 299 (2007) 38–44. 3713–3716.

[62] Y. Hu, X. Liu, Z. Xu, Role of crystal structure in ﬂotation separation of dispore [77] C.J. Yu, A.G. Richter, A. Datta, M.K. Durbin, P. Dutta, Observation of molecular

from koalinite, pyrophyllite and illite, Miner. Eng. 16 (2003) 219–227. layering in thin liquid ﬁlms using X-ray reﬂectivity, Phys. Rev. Lett. 82 (1999)

[63] P.M. Costanzo, W. Wu, R.F. Giese, C.J. van Oss, Comparison between direct con- 2326–2329.

tact angle measurements and thin layer wicking on synthetic monosized cuboid [78] M. Colic, J.D. Miller, The signiﬁcance of interfacial water structure in colloidal

hematite particles, Langmuir 11 (1995) 1827–1830. systems–dynamic aspects, Surfact. Sci. Ser. 88 (2000) 35–82.

[64] E.M. Bryant, R.S. Bowman, J.S. Buckley, Wetting alteration of mica surfaces with [79] E. Ruckenstein, N.V. Churaev, A possible hydrodynamic origin of the forces of

polyethoxylated amine surfactants, J. Pet. Sci. Eng. 52 (2006) 244–252. hydrophobic attraction, J. Colloid Interface Sci. 147 (1991) 535–538.

[65] P.T.L. Koh, F.P. Hao, L.K. Smith, T.T. Chau, W.J. Bruckard, The effect of par- [80] C.J. van Oss, R.F. Giese, The hydrophilicity and hydrophobicity of clay minerals,

ticle shape and hydrophobicity in ﬂotation, Int. J. Miner. Process. 93 (2009) Clays Clay Miner. 43 (1995) 474–477.

128–134. [81] D.J.G. Durán, M.M. Ramos-Tejada, F.J. Arroyo, F. González-Caballero, Rheolog-

[66] C. Subrahmanyam, N.K. Thakur, R.T. Gangadhara, R. Khanna, M.V. Ramana, V. ical and electrokinetic properties of sodium montmorillonite suspensions, J.

Subrahmanyam, Tectonics of the Bay of Bengal: new insights from satellite- Colloid Interface Sci. 229 (2000) 107–117.

gravity and ship-borne geophysical data, Earth Planet Sci. Lett. 171 (1999) [82] R.F. Giese, W. Wu, C.J. van Oss, Surface and electrokinetic properties of clays

237–251. and other mineral particles untreated and treated with organic and inorganic

[67] T. Young, An essay on the cohesion of ﬂuids, Philos. Trans. R. Soc. Lond. A (1805) cations, J. Dispers. Sci. Technol. 17 (1996) 527–547.

65–87. [83] M. Lewin, A. Mey-Marom, R. Frank, Surface free energies of polymeric materials,

[68] R.D. Narhe, D.A. Beysens, Growth dynamics of water drops on a square-pattern additives and minerals, Polym. Adv. Technol. 16 (2005) 429–441.

rough hydrophobic surface, Europhys. Lett. 75 (2006) 98–104. [84] Z. Li, R.F. Giese, C.J. van Oss, The surface thermodynamic properties of synthetic

[69] G. Palasantzas, J.T.M. de Hossen, Wetting on rough surfaces, Acta Mater. 49 hydrotalcite compounds, Langmuir 10 (1994) 330–333.

(2001) 3533–3538. [85] N. Schreithofe, Investigation of particle-bubble interactions with a new

[70] J.D. Miller, S. Veeramasuneni, J. Drelich, M.R. Yalamanchili, Effect of roughness experimental setup, Ph.D. thesis, Helsinki, University of Technology,

as determined by atomic force microscopy on the wetting properties of PTFE Finland, 2003.

thin ﬁlms, Polym. Eng. Sci. 36 (1996) 1849–1855. [86] J. Laskowski, The relationship between ﬂoatability and hydrophobicity, in: P.

[71] D.L. Sedin, K.L. Rowlen, Inﬂuence of tip size on AFM roughness measurements, Somasundaran (Ed.), Advances in Mineral Processing, Littleton, Co, SME, USA,

Appl. Surf. Sci. 182 (2001) 40–48. 1986, pp. 189–192.