Adsorption of TiO<sub>2</sub> Nanoparticles at the Surface of Micron-Sized Latex Particles. pH and Concentration Effects on Suspension Stability

(1)

Article

Reference

Adsorption of TiO

₂

Nanoparticles at the Surface of Micron-Sized Latex Particles. pH and Concentration Effects on Suspension Stability

LOOSLI, Frédéric, STOLL, Serge

Abstract

It is shown that although nanoparticles are now finding wide application in a variety of fields, a detailed understanding of their mode of action is important and has to be made on a detailed analysis of their physicochemical properties in solution. In this study, the interaction between micron-sized latex particles and TiO2 manufactured nanoparticles are reported and the importance of the electrostatic interactions is discussed. The surface charge variations of micron-sized latex particles and TiO2 nanoparticles are considered by adjusting the solution pH and three electrostatic scenarios are experimentally determined by performing electrophoretic measurements. In the first scenario,positively charged isolated TiO2 nanoparticles are rapidly adsorbed at the negative latex surfaces and the latex surface charge, latex aggregation and aggregation kinetic rates are found to be controlled by the TiO2 nanoparticle concentration. Fractal aggregates are obtained when latex surface charge is compensated by adsorption of the oppositely charged nanoparticles. Analysis of the variation of the kinetic rate with nanoparticle [...]

LOOSLI, Frédéric, STOLL, Serge. Adsorption of TiO

₂

Nanoparticles at the Surface of

Micron-Sized Latex Particles. pH and Concentration Effects on Suspension Stability. Journal of Colloid Science and Biotechnology , 2012, vol. 1, no. 1, p. 113-121

DOI : 10.1166/jcsb.2012.1003

Available at:

http://archive-ouverte.unige.ch/unige:26458

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

RESEARCH AR TICLE

Printed in the United States of America

Colloid Science and Biotechnology Vol. 1, 1–9, 2012

Adsorption of TiO ₂ Nanoparticles at the Surface of Micron-Sized Latex Particles. pH and Concentration

Effects on Suspension Stability

Frédéric Loosli and Serge Stoll

^∗

University of Geneva, F.-A. Forel Institute, Group of Environmental Physical Chemistry, 10 route de Suisse, 1290 Versoix, Switzerland

It is shown that although nanoparticles are now finding wide application in a variety of fields, a detailed understanding of their mode of action is important and has to be made on a detailed analysis of their physicochemical properties in solution. In this study, the interaction between micron-sized latex particles and TiO2 manufactured nanoparticles are reported and the importance of the electrostatic interactions is discussed. The surface charge variations of micron-sized latex particles and TiO2 nanoparticles are considered by adjusting the solution pH and three electrostatic scenarios are experimentally determined by performing electrophoretic measurements. In the first scenario, positively charged isolated TiO₂ nanoparticles are rapidly adsorbed at the negative latex surfaces and the latex surface charge, latex aggregation and aggregation kinetic rates are found to be controlled by the TiO₂nanoparticle concentration. Fractal aggregates are obtained when latex surface charge is compensated by adsorption of the oppositely charged nanoparticles. Analysis of the variation of the kinetic rate with nanoparticle concentrations indicates an uneven distribution of the TiO₂ nanoparticles at the latex surface leading to an extra attractive contribution to the interaction energy between the latex particles. In the second scenario, at the point of zero charge of the TiO2nanopar- ticles, and in presence of negatively charged latex particles, the suspension behavior is mainly controlled by the aggregation of the TiO2nanoparticles. The last scenario which is representative of the presence of negatively charged species is leading to stable suspensions.

Keywords: Nanoparticle Adsorption, Titanium Dioxide, Polystyrene Sulfate Latex, Nanoparticle Stability, Kinetic Rates, Fractal Dimension.

1. INTRODUCTION

Manufactured nanomaterials, in particular nanoparticles (NPs), are today widely produced and used due to their remarkable properties. Nanomaterials are defined as materials with one dimension less than 100 nm in at least one dimension, and NPs in at least two dimensions.¹ Nanoparticles cover a wide range of compounds such as, carbon based (fullerens, nanotubes), metal oxide (TiO₂, ZnO, Fe₂O₃, and metal (Au, Ag) NPs.² Due to huge surface/size ratio nanoparticles exhibit unique physical and chemical properties such as specific redox and catalytic properties and high UV filtering capacity. Depending on their sizes, shapes and chemical compositions NPs are today used in conductive and high-strength composite production, energy storage and conversion, sensors and in many domains of colloid science and biotechnology.^3–6

∗Author to whom correspondence should be addressed.

One of the more produced NPs type is titanium dioxide (TiO₂ with an annual production of about 4.3 million tons.⁷ TiO₂ is widely used as UV protective agent in painting and cosmetics, in photovoltaic, photocatalytic and sensor domains.^8–15 TiO₂ synthesis, properties, mod- iﬁcations and applications are summarized in Ref. [3].

Since nanomaterials risk evaluation is a subject of grow- ing interest,^16–18 the possible impact of TiO₂ NPs in the environment as well as their fate and transport are also under investigation.¹⁹ Inﬂuence of pH, ionic strength, salt valency, and surface potential on aggregation and dispersion of TiO₂ NPs have been reported,²⁰²¹ as well as TiO₂ NPs stability in presence of natural organic matter to get an insight into their possible behavior in aquatic systems.²²²³ Polymer latex based colloids are widely used in colloid science and applications such as the manufacture of water based coating, microscope calibration, packing material in chromatographic columns, biomedicine and biomedical especially in drug delivery systems.^24–27 They

(3)

RESEARCH AR TICLE

Adsorption of TiO₂ Nanoparticles at the Surface of Micron-Sized Latex Particles Loosli and Stoll are also known to be ideal models due to their monodis-

perse sizes and surface group functionalization possibili- ties. Latex particles or ﬁlms are also used as templates for NPs and biopolymer polyelectrolyte coating or for elec- trochemical deposition of metal to improve UV protecting properties, produce photonic material, high density mag- netic data storage devices and biosensors.^2428–34

Among these applications, polymer films made from colloidal aqueous dispersions, called latexes, are used more and more nowadays. To improve the long term stability of such films, TiO₂NPs are now considered because of their potential to improve the film UV protecting properties. One problem of such films is that their properties (optical, mechanical, etc.) may suffer from the inhomoge- neous distribution of the TiO₂ NPs in the final film from the drying mechanism itself but also the initial distribution of the TiO₂NPs adsorbed on the latex surface which will influence their final distribution in the dry film and consequently its properties. Prior film formation the initial distribution of the TiO₂NPs is expected to depend on the adsorption of TiO₂ NPs at the latex surface, aggregation formation processes, and parameters such as the extend of surface coating, surface charge modification of the latex particles, solution pH and ionic strength.

In this study, negatively charged micron-sized latex particles and TiO₂NPs mixtures are considered, and the inﬂu- ence of the TiO₂NPs concentration on the surface charge properties of the latex particles is investigated using electrophoretic mobility measurements. Since the TiO₂ NPs surface charge is pH dependant, the role of the pH on the TiO₂ NPs adsorption amount at the latex particles is also investigated. Three electrostatic scenarios are then investigated in details corresponding to negatively charged latex particles in presence of positive, negative and uncharged TiO₂NPs respectively. The aggregation of the latex particles resulting from the presence of the TiO₂ NPs is also evaluated and kinetic rates are determined by adjusting both pH and TiO₂ NPs concentration. Aggregation rates are also compared with salt induced particle destabilization. Based on a detailed analysis of electrophoretic measurements and kinetics rates, aggregation mechanisms are discussed and comparison is made with DLVO predictions.

Aggregate structures are then examined to obtain the fractal dimensions of the resulting structures and SEM analysis is made to get an insight into the distribution of TiO₂NPs on the latex particle surface.

2. MATERIALS AND METHODS 2.1. Materials

Monodisperse polystyrene sulfate latex spheres with a mean diameter of 099±003 m determined by TEM were obtained from IDC (Interfacials Dynamics Corpora- tion) as a 81 g/L latex suspension in water. Latex particles exhibit negative charges due to the presence of sulfate

groups (76×10⁵per particle) and have a specific surface area equal to 57×10⁴ cm²/g. TiO₂ anatase NPs with an average particle size of 15 nm were obtained from Nanos- tructured and Amorphous Material Inc as a 170 g/L TiO₂ suspension in water and with a specific surface area of 240 m²/g. Hydrochloric acid (1 M HCl, Titrisol^®, Merck) and sodium hydroxide (1 M NaOH, Titrisol^®, Merck) were used to adjust the solution pH. Sodium chloride (NaCl, 99.5%, Acros Organics) was employed to adjust the final ionic strength to 0.001 M. All the solutions were prepared to the target experimental concentrations with deionized Milli Q water (R >18 Mcm). All experiments were performed in 25 mL polypropylene (PP) jupe tubes 25× 90 mm (Milan) with a crosshead single 8×10 mm mag- netic stirrer (VWR).

2.2. Zeta Potential and Initial Size Distribution Measurements

A Malvern Zetasizer Nano ZS was used to determine the zeta (potential values as well as size distributions of the latex particles, TiO₂NPs, and mixtures of latexes and NPs as a function of TiO₂concentration and pH. For each situation, triplicate measurements were performed and each sample was measured 3 times to determine the zeta potential values and the size distributions. For the zeta potential determination 15 sub-runs, with a delay of 5 s between them, to relax and stabilize the system, whereas 12 sub- runs of 10 s with a 5 s delay for the size distribution measurements were made. The Smoluchowski approximation model was used to determine the size distributions. The determination of TiO₂ and latex zeta potentials and size distributions were made using 50 mg/L solutions vigor- ously stirred for 24 hours and sonicated during 20 minutes before analysis.

2.3. Aggregation Kinetic Determination

The influence of TiO₂adsorption and concentration on the aggregation kinetic of the latex suspensions was investigated by monitoring the decrease of the number of non agglomerated 1 m latex monomer particles as a function of time using a Coulter Counter Multisizer^® II. This instrument determines the size distribution by recording through a 256 channels device the number of particles or agglomerates as well as their equivalent sizes (comprised between 1 and 30 m) when passing through a 50m aperture in an Isoton support electrolyte. The Iso- ton IIA electrolyte was previously filtered on a 0.22 m Millipore^® ISOPORE^™ GTTP filter to enhance the signal noise ratio. 100L of 500 mg/L latex with different TiO₂ concentrations mass reactant gently stirred were added to 100 mL ISOTON IIA in a 150 mL PP beaker (VWR) to perform the measurement.

The mixture was then gently agitated with a 28×6 mm mechanic axial propeller stirrer. Each Coulter Counter

2 J. Colloid Sci. Biotechnol. 1, 1–9,2012

(4)

RESEARCH AR TICLE

measurement was performed during 30 s, which corresponds to a total solution volume of 287L. Each sample was then measured three times and triplicates were made for each different latex-TiO₂ concentration ratio and the decrease with time of the free latex particles number was measured in the 1 m channel. The isolated latex particle number variation with time was then used to calculate kinetic aggregation rates.

2.4. Aggregate Fractal Dimension Determination Fractal dimension (D_fis an important geometric parameter for the characterization of aggregation processes and aggregate compactness.³⁵ It can be expressed as a scaling relationship between aggregate mass (M) and a character- istic aggregate length (l) according to,

M∼l_f^D (1) The fractal dimension indicates how the nominal volume occupied by an aggregate is ﬁlled with primary particles. Fractal dimension gives also important infor- mation for the calculation of sedimentation and diffusion coefﬁcients which are important transport and diffusion parameters. The aggregate fractal dimensions were determined using a Olympus BX61 microscope equipped with a 3.2 Mpx CCD camera. The samples were gently deposited on a glass plate and measurements were made immedi- ately to avoid any changes in aggregate geometries due to the drying process. The image analysis was made with SigmaScan Pro software. 12 pictures were taken with a 10×magnitude objective on three different solutions. The fractal dimensions were then determined by considering relative masses and aggregate dimensions and calculated by considering the slope of a log–log plot of the number of pixels per aggregate as a function of the major axis length of aggregates according to Eq. (1).

2.5. SEM Image Analysis

Sample morphology was observed using a scanning electron microscope (SEM, JEOL, JSM 7001F) operated at 15 kV. Samples (10L of solution), previously deposed on a silica wafer (<1 nm surface roughness, Agar scien- tiﬁc), were allowed to dry 1 day before the analysis. An ultra-thin coating of 3 nm gold was then deposited on the samples by low vacuum sputter coating prior to imaging.

3. RESULTS AND DISCUSSION

3.1. Material Characterization 3.1.1. Latex Particles

To get an insight on the surface charge variation of the latex particles in solution as a function of pH, the corresponding titration curve was determined with the measurement of zeta potential values. 50 mg/L latex solutions

with a ﬁnal ionic strength equal to 0.001 M (adjusted with NaCl) were used and the pH was adjusted from pH 11 to pH 2 with HCl at variable concentrations. As shown in Figure 1, due to the presence of surface sulfate groups, latex particles exhibit a negativepotential value which is decreasing from−70±1 mV at pH 2, to−109±1 mV at pH 11. No Point of Zero Charge (PZC) is observed. Owing to the low pKa value of sulfate groups (pKa<2) and the presence of negative surface charges, latex particles are thus stabilized against aggregation upon pH variations and at low ionic strength. It should be noted that aggregation could be promoted through screening effects by increasing the ionic strength. Previous work showed that the critical coagulation concentration of similar latex particles was equal to 0.25 M at pH 6 for a 500 mg/L latex solution with the use of monovalent salt.³⁶

Sizes were also measured in solution using a Zetasizer Nano ZS and the z-average diameter was found equal to 1199±26 nm. The latex solution was found stable and monodisperse in size in the full pH domain.

TiO₂ NPs. The TiO₂ NPs surface charge pH dependence and the resulting stability of the NPs in solution were determined through pH titration curves. As shown in Figure 2(a), TiO₂ NPs exhibit a stable and positive zeta potential from pH 2 to 4 (+50 mV). Then the potential value decreases to the PZC at pH 6.2±01 in good agreement with the values reported in the literature.³⁷³⁸ By increasing further the pH, the zeta potential becomes negative and is found to stabilize at −40 mV at pH 8.5.

Thez-average diameter evolution with pH was also determined and presented in Figure 2(b). The NPs size is found to increase to reach a maximum value at the PZC with values greater than 10 m indicating strong aggregation when the NPs surface charge is neutralized. By further increasing the pH, thez-average diameter is found to reach stable values again. As shown in the gray domains in

Fig. 1. Zeta potential variation of 1m latex particles as a function of pH; [latex]=50 mg/L;I=0001 M. Latex particles exhibit negative zeta potential values in the full pH range.

(5)

RESEARCH AR TICLE

Adsorption of TiO₂ Nanoparticles at the Surface of Micron-Sized Latex Particles Loosli and Stoll

Fig. 2. (a) Zeta potential variation of TiO₂NPs as a function of pH.

The PZC is found here to be equal to pH=62±01; (b) TiO2particles z-average diameter variation as a function of pH. TiO2NPs aggregation domain is found important and comprised between−30 mV and+30 mV (gray domain); [TiO2]=50 mg/L;I=0001 M.

Figures 2(a) and (b) TiO₂ NPs destabilization is found to occur in a zeta potential domain comprised between +30 and−30 mV.

3.2. Latex Surface Charge Modiﬁcations in Presence of TiO₂ NPs

According to the pH dependence of the zeta potential values for both the TiO₂ NPs and latex particles, different electrostatic scenarios of interaction processes can be deﬁned here. The ﬁrst one is considering the most favorable situation for the adsorption of the TiO₂ NPs at the surface of the latex particles i.e., positively charged NPs in presence of negatively charged latexes. In these conditions fast TiO₂ NPs adsorption at the latex surface is expected.

This will be achieved when pH<pH_PZC_TiO

2. The second one, when pH=pH_PZC_TiO

2, is related to the interaction of negatively charged latex particles with uncharged TiO₂ NPs. Such a condition is expected to rapidly promote the formation of TiO₂ aggregates in solution. The third one, when pH>pH_PZC_TiO

2, is representative of a system con- sisting in negatively charged latex particles and negatively charged TiO₂ NPs and thus represent unfavorable conditions for NPs adsorption at the latex surfaces.

The ﬁrst set of data was obtained at pH 3.7 by adjusting the TiO₂NPs concentration in the latex suspension. In this condition, the positively charged TiO₂NPs are rapidly adsorbed at the negatively latex surface and three distinct behaviors are obtained as shown in Figure 3(a):

(i) at low TiO₂concentration latex particles remain negatively charged ( from−70 mV to−60 mV).

(ii) then, by increasing TiO₂ concentration, the latex zeta potential is rapidly increasing to zero. The full charge neutralization of the latex particle and IsoElectric Point (IEP) value is achieved here when [TiO₂]=475 g/L for a TiO₂/latex particles ratio number equal to 741×10⁵.

Fig. 3. Variation of zeta potential values as a function of TiO₂NPs mass concentration for a 50 mg/L latex mass concentration (negatively charged latex particles) and I=0001 M at (a) pH=37: positively charged TiO₂NPs, (b) pH=pH_PZCTiO2=62: uncharged TiO2NPs, (c) pH=10:

negatively charged TiO₂NPs.

(6)

RESEARCH AR TICLE

(iii) then by increasing further the TiO₂ concentration charge inversion is obtained with latex particle exhibiting positive surface charges (= +60 mV).

For TiO₂ concentrations comprised between 5 g/L and 500 mg/L monomodal zeta potential as well as size distributions were obtained. The meanz-average diameter was found equal to around 1000 nm i.e., corresponding to the presence of isolated latex particles below and above the IEP. Higher values were found close to surface charge neutralization at the IEP. Such results indicate that the electrophoretic mobility signal was mainly related to the latex particles. For TiO₂concentrations ranging from 500 mg/L to 3 g/L (gray domain in Fig. 3(a)) bimodal zeta potential and size distributions were obtained. Then, for higher TiO₂ concentration, TiO₂ signal was found to be predominant and monomodal distributions were obtained withz-average sizes and potential values corresponding to the TiO₂ NPs.

The zeta potential variation as a function of TiO₂ concentration at pH=pH_PZC_TiO

2 is now presented in Figure 3(b). The zeta potential values are found here to rapidly decrease to values close to zero. In such condition, TiO₂ NPs are expected to form large aggregates in solution. As a result at low TiO₂ concentration (from 50 to 500g/L), the obtained potential values correspond to latex particles (monomodal size distribution with a z- average diameter around 1000 nm). Then by increasing the TiO₂concentration, TiO₂aggregate formation is found predominant and to be in agreement with the expected z-average diameters and zeta potential values. At high TiO₂ NPs concentration, the potential values are close to zero due to presence of large aggregates composed of uncharged TiO₂ NPs.

Finally, experiments at pH 10 (pH>pH_PZC_TiO

2 were

realized, in which both latex and TiO₂ particles are negatively charged. Such conditions are not favorable to TiO₂ NPs aggregate formation, nor TiO₂ adsorption at the surface of the latex particles due to electrostatic repulsions.

As shown in Figure 3(c), the zeta potential is stable at

−105 mV in a relatively large domain of TiO₂ concentration. This value corresponds to the zeta potential of the isolated latex particles. By increasing further the TiO₂ concentration the potential value is found to increase up to −40 mV corresponding to the TiO₂ zeta potential value at this pH value. For a 500 mg/L or lower TiO₂ mass concentration, only the potential of the latex particles was measured due to its much larger size relative to TiO₂ NPs. On the other hand, only the TiO₂ zeta potential was observed for a TiO₂ concentration greater than 3000 mg/L. In between, two populations were observed, one at−105 mV and the other at−40 mV, hence giving an intermediary zeta potential value depending on the ratio between them. This was conﬁrmed by the analysis of the z-average diameters.

3.3. Aggregation Kinetic of Latex and TiO₂ NPs Mixtures

3.3.1. Aggregation Kinetic Rates

In Figure 4 are presented the variation of the inverse free latex particle concentration with time (cm³/particle) at different TiO₂NPs concentrations. The ﬁrst electrostatic scenario is mainly considered here i.e., negatively charged latex particles in presence of positively charged TiO₂NPs.

The kinetic aggregation rates were experimentally found to follow a 2nd order law according to

dN

dT = −KSN ² (2) whereN represents the isolated latex particle number and KS the kinetic aggregation rate (cm³/s). This formula is used to calculate KS values at early stages of coagulation when most of the latex particles are still as singlets.

The increase of TiO₂ NPs concentration is ﬁrst promot- ing the coagulation of latex suspensions until a maximum value. The kinetic aggregation rates varied from 492× 10⁻¹³ cm³/s for a 8 mg/L TiO₂ concentration to a maximum aggregation rate of 330×10⁻¹¹cm³/s at 4.75 mg/L corresponding to latex surface charge neutralization at the IEP. After this point, the increase of TiO₂ NPs concentration stabilizes the latex suspension. As shown in Figure 4, a signiﬁcant difference is found between the highestKS values obtained in presence of NaCl above the salt critical coagulation concentration and the one obtained here. The coagulation rate obtained with salt screening only (10×10⁻¹⁰ cm³/s) is 3 times greater than the value obtained at charge neutralization due to TiO₂ NPs adsorption.³⁶ The corresponding KS values are reported as a function of TiO₂ NPs concentration in Figure 5(a).

It is shown that the kinetic rate KS increases continu- ously until the negative latex negative surface charge is

Fig. 4. Determination of kinetic rates as a function of time for 500 mg/L latex mass concentration andI=0001 M at pH=37 for a TiO2mass concentration of: 3 mg/L, 1.5 mg/L, 4.75 mg/L,6 mg/L and 8 mg/L; — [latex]=500 mg/L, salt induced aggregation of latex particles (I=1 M) in the absence of TiO₂.

(7)

RESEARCH AR TICLE

Fig. 5. Variation of the kinetic aggregation rate KS as a function of (a) TiO₂ mass concentration and (b) zeta potential; for [Latex]= 500 mg/L,I=0001 M and pH=37.

neutralized by the adsorption of the oppositely charged TiO₂ NPs. Then by increasing further the TiO₂ NPs concentration a rapid decrease of the KS values is observed owing to the charge reversal of the latex particles and the importance of the electrostatic repulsions between the positively charged TiO₂ coated latex particles. Similar trends are obtained when considering the kinetic rate variations as a function of the latex zeta potential (Fig. 5(b)). The optimum conditions are in good agreement with the opti- mal TiO₂ NPs dosage required for surface charge neutralization. Here also a rapid decrease of the KS values is observed in the charge reversal domain after the latex surface charge neutralization point. Two points are important to discuss here. The ﬁrst one is considering the asymmet- ric shape of the kinetic rate variations indicating a rapid restabilization of the latex particles after charge inversion of the latex particles. The second one is considering the relative broadness of the curves which indicates that even in presence of a small TiO₂ amount (1.5 mg/L) and a resulting latex particle potential value of −50 mV, the latex-TiO₂ system already starts to aggregate. In theory, for such zeta potential value the mixture should be stable and aggregation should be prevented. This denotes that charge neutralization is not the only mechanism of latex

destabilization. An alternative approach consists to con- sider the nature of the charge distribution on a latex particle with adsorbed TiO₂NPs. With TiO₂NPs of high charge density it is proposed an uneven charge distribution of the latex particle and that the patch mechanism would lead to an attractive contribution to the interaction energy between two latex particles.

The DLVO theory³⁹⁴⁰ was used to calculate the total interaction potential between two latex particles with variable surface charge potentials to help elucidate the possible aggregation mechanisms. The total interaction potentialVT

was calculated as the sum of long distance repulsive electrostatic double layer interactions V_RElect and short distance attractive van-der-Waals forcesV_AvdW according to:

V_TH=V_RElectH+V_AvdwH (3)

V_RElectH=20_ra²e^−kH (4)

V_AvdwH= A₁₃₁ 6

2a²

HH+4a + 2a²

H+2a²+ln

HH+4a

H+2a²

(5) where 0 and _r are the vacuum permittivity and the water relative permittivity respectively,athe latex particle radius, the latex surface potential, the Debye-Hückel parameter (with ⁻¹=961×10⁻⁹ m) and A₁₃₁ is the Hamaker constant. In these equations the linear superpo- sition approximation was used.⁴¹ In our calculations, the size modification of latexes due to the adsorption of TiO₂ was neglected and the surface particle potential was replaced, as a first approximation, by the experimental zeta potential values. The Hamaker constantA₁₃₁was defined by considering the latex intrinsic properties and fixed to 140×10⁻²⁰ J.⁴² In Figure 6, the total interaction potentials are then presented for different zeta potential values.

Fig. 6. DLVO interaction potential calculation between two latex particles as a function of distance between surfaces at different zeta potentials.

(8)

RESEARCH AR TICLE

It is shown that van-der-Waals attractive forces are fully predominant for particles with potential values less than 6 mV and that the absence of a secondary minimum indicates irreversible aggregation. For a 8 mV shear plane potential, even if a secondary minima is identiﬁed, the energetic barrier is almost equal to zerokBT and aggregation process should happen. In the case of a 10 mV zeta potential, a 10kBT potential is required to induce aggregation and it is thus expected to represent a limiting surface charge potential value. The theoretical calculation of the interaction potential between latex particles clearly indicates that out of a zeta potential region comprised between

−10 and +10 mV aggregation should not occur in our experimental conditions. This is not the case here owing to the fact that aggregation is observed even for relatively high zeta potential values as shown in Figure 5(b). Hence suggesting that, as already proposed, an uneven charge distribution of TiO₂ NPs on the latex surface leading to an extra attractive contribution energy.

3.4. Aggregate Morphologies and SEM Surface Analysis

Fractal analysis and fractal dimension calculations were performed on aggregates obtained in the ﬁrst scenario i.e., in presence of negatively charged latexes and positively charged TiO₂ NPs at the IEP corresponding to mixture at pH 3.7 with a latex and TiO₂ NPs concentration equal respectively to 500 mg/L and 4.7 mg/L.

Three samples representing 796 aggregates were ana- lyzed. In Figure 7 is presented a log–log plot of the variation of the aggregate mass (in number of pixels) as a function of the aggregate major axis length (represented here by the number of pixels). The linear distribution of

Fig. 7. Log–Log plot of the number of pixels per aggregate as a function of aggregate major axis length at the IEP with [latex]=500 mg/L, [TiO₂]=47 mg/L,I=0001 M and pH 3.7. Aggregates were obtained by the charge neutralization of negatively charged latex suspensions due to the adsorption of positively charged TiO₂NPs (10m corresponds to 29 pixels).

Fig. 8. Scanning electron microscope images for 50 mg/L latex at pH 3.7 in presence of (a) 0g/L TiO2,= −80 mV, (b) 200 g/L TiO₂, = −30 mV; (c) 470g/L TiO2,= −0 mV; (d) 1000g/L TiO2,= +30 mV; scale bar represent 100 nm and expanded area is a 2×zoom.

the data demonstrates the fractal character of the aggregates and the fractal dimensionDf is found equal to 182± 003 in good agreement with the cluster–cluster aggregation model and diffusion limited aggregation model fractal

(9)

RESEARCH AR TICLE

Fig. 9. Scanning electron microscope image for 50 mg/L latex and 1 mg/L TiO₂ NPs at pH=pH_PZCTiO2=62. SEM image indicates the formation of large uncharged and isolated TiO₂NPs aggregates which are not adsorbed at the negatively charged surface of latex particles. Scale bar represent 100 nm.

dimension as well as computer simulations (Df =180) and experimental latex suspension results (Df =183–

1.86).³⁵⁴³

SEM picture analysis was also made by considering latex particles in presence of TiO₂ NPs at variable concentrations. The SEM samples were prepared at pH 3.7, pH 6.2 and 9.7, considering a 50 mg/L latex and various TiO₂ mass concentrations, but without the presence of salt to minimize artifact effects due to the presence of salt during the drying processes. Distribution of TiO₂ on polystyrene sulfate latex surface clearly shows an uneven NPs distribution which is concentration dependent. For better comparison, in Figure 8(a) are represented bare latex particles. The presence of 15 nm TiO₂ NPs is clearly shown on the latex surface in Figures 8(b)–(d) which are representative of various TiO₂ concentration distributions for latex particles having apotential values around

−30 mV (Fig. 8(b)), 0 mV (Fig. 8(c)) and +30 mV (Fig. 8(d)). The latex surface coverage is found to increase with the TiO₂ NPs concentration. TiO₂ NPs are mainly adsorbed as individual particles even if, in some cases, small TiO₂ aggregates of less than 50 nm in size are observed on the latex surface. Additional SEM analysis was performed at the pH corresponding to the PZC of the TiO₂NPs. The corresponding picture which is presented in Figure 9 indicate that TiO₂ NPs are forming large aggregate which are not adsorbed at the latex surface in good agreement with our previous zeta potential measurements.

Finally, at pH 9.7, as expected, latex particles were found uncoated and TiO₂ NPs dispersed as singlets due to electrostatic repulsions.

4. CONCLUSIONS

Electrophoretic measurements, SEM image analysis and particle counting techniques were used to investigate in

a systematic way the interaction (adsorption and aggregation) processes, resulting structures, and solution stability of solutions containing micron-sized latex particles and TiO₂nanoparticles. It is show that the “functionalization”

of the latex particles by nanoparticles is achieved in very speciﬁc electrostatic conditions so as

(i) to promote the nanoparticles adsorption at the latex surface and

(ii) avoid the formation of large aggregates in solution.

Such conditions are obtained here below the point of zero charge of the TiO₂nanoparticles and for NPs concentration below the surface charge neutralization of the latex particle so as to avoid latex aggregation. Using image analysis, considering DLVO calculations and analyzing the kinetic aggregation rates we also concluded to an uneven nanoparticle distribution on the surface of the latex particles resulting to latex destabilization below the latex surface charge neutralization point.

Acknowledgments: The authors are grateful to Daniel Palomino and Philippe Le Coustumer for stimulating dis- cussions and Agathe Martignier and Apolline Lefort for the SEM analysis. We also acknowledge the ﬁnancial support received from the Swiss National Foundation (project 200021_135240 and 206021_133771).

References and Notes

1. American Society for Testing and Materials, Standard Termi- nology Relating to Nanotechnology, West Conshohocken (2006), E 2456-06.

2. Y. Ju-Nam and J. R. Lead,Sci. Total Environ.400, 396(2008).

3. X. Chen and S. S. Mao,Chem. Rev.107, 2891(2007).

4. Q. A. Pankhurst, J. Connolly, S. K. Jones, and J. Dobson,J. Phys.

D: Appl. Phys.36, R167(2003).

5. R. W. J. Scott, O. M. Wilson, and R. M. Crooks,J. Phys. Chem. B 109, 692(2005).

6. X. L. Luo, A. Morrin, A. J. Killard, and M. R. Smyth,Electroanal.

18, 319(2006).

7. M. A. Kiser, P. Westerhoff, T. Benn, Y. Wang, J. Perez-Rivera, and K. Hristovski,Environ. Sci. Technol.43, 6757(2009).

8. B. Mahltig, H. Bottcher, K. Rauch, U. Dieckmann, R. Nitsche, and T. Fritz,Thin Solid Films485, 108(2005).

9. F. E. A. Guo,Colloids Surf., Adoi:10.1016/j.colsurfa.2011.12.007 (2011).

10. M. Gratzel,J. Photochem. Photobiol., A168, 235(2004).

11. P. D. Cozzoli, R. Comparelli, E. Fanizza, M. L. Curri, and A. Agostiano,Mater. Sci. Eng. C23, 707(2003).

12. A.-P. Zhang and Y.-P. Sun, World J. Gastroentero. 10, 3191 (2004).

13. P. Wongkalasin, S. Chavadej, and T. Sreethawong,Colloids Surf., A 384, 519(2011).

14. X. Jimin, J. Deli, C. Min, L. Di, Z. Jianjun, L. Xiaomeng, and Y. Changhao,Colloids Surf., A372(2010).

15. R. Khan and M. Dhayal,Electrochem. Commun.10, 263(2008).

16. A. Nel, T. Xia, L. Madler, and N. Li,Science311, 622(2006).

17. M. R. Wiesner, G. V. Lowry, P. Alvarez, D. Dionysiou, and P. Biswas,Environ. Sci. Technol.40, 4336(2006).

18. V. L. Colvin,Nat. Biotechnol.21, 1166(2003).

19. B. Nowack and T. D. Bucheli,Environ. Pollut.150, 5(2007).

(10)

RESEARCH AR TICLE

20. R. A. French, A. R. Jacobson, B. Kim, S. L. Isley, R. L. Penn, and P. C. Baveye,Environ. Sci. Technol.43, 1354(2009).

21. K. A. D. Guzman, M. P. Finnegan, and J. F. Banﬁeld,Environ. Sci.

Technol.40, 7688(2006).

22. R. F. Domingos, N. Tufenkji, and K. J. Wilkinson, Environ. Sci.

Technol.43, 1282(2009).

23. A. A. Keller, H. Wang, D. Zhou, H. S. Lenihan, G. Cherr, B. J.

Cardinale, R. Miller, and Z. Ji, Environ. Sci. Technol. 44, 1962 (2010).

24. C. Pichot,Curr. Opin. Colloid Interface Sci.9, 213(2004).

25. D. Polpanich, P. Tangboriboonrat, A. Elaissari, and R. Udomsangpetch,Anal. Chem.79, 4690(2007).

26. A. Arkhis, A. Elaissari, T. Delair, B. Verrier, and B. Mandrand, J. Biomed. Nanotechnol.6, 28(2010).

27. A. Elaissari and H. Fessi,Macromol. Symp.288, 115(2010).

28. C. J. McDonald and M. J. Devon,Adv. Colloid Interface Sci.99, 181 (2002).

29. A. Imhof,Langmuir17, 3579(2001).

30. S. Ugur, S. Sunay, A. Elaissari, F. Tepehan, and O. Pekcan,Polym.

Compos.27, 651(2006).

31. S. H. Cho, W. Y. Kim, G. K. Jeong, and Y. S. Lee,Colloids Surf., A255, 79(2005).

32. P. N. Bartlett, P. R. Birkin, and M. A. Ghanem, Chem. Commun.

1671(2000).

33. A. S. Susha, F. Caruso, A. L. Rogach, G. B. Sukhorukov, A. Kornowski, H. Mohwald, M. Giersig, A. Eychmuller, and H. Weller,Colloids Surf., A163, 39(2000).

34. A. Rogach, A. Susha, F. Caruso, G. Sukhorukov, A. Kornowski, S. Kershaw, H. Mohwald, A. Eychmuller, and H. Weller,Adv. Mater.

12, 333(2000).

35. P. Meakin,Adv. Colloid Interfac.28, 249(1988).

36. D. Palomino, D. Hunkeler, and S. Stoll,Polymer52, 1019(2011).

37. T. Preocanin and N. Kallay,Croat. Chem. Acta79, 95(2006).

38. G. D. Parﬁtt,Pure Appl. Chem.48, 415(1976).

39. B. V. D. A. L. Landau,Acta Physicochim. URSS14, 633(1941).

40. E. J. W. Verwey,J. Phys. Colloid Chem.51, 631(1947).

41. J. Lopez, Electrostatic Heteroaggregation Process Arising in Two- Component Colloidal Dispersion(1996).

42. F. J. Rubiohernandez,Colloids Surf., A88, 141(1994).

43. S. Tang,Colloids Surf., A157, 185(1999).

Received: 2 March 2012. Accepted: 12 March 2012.

Adsorption of TiO<sub>2</sub> Nanoparticles at the Surface of Micron-Sized Latex Particles. pH and Concentration Effects on Suspension Stability

Article

Reference