Studies in Surface Science and Catalysis 132 Y. Iwasawa, N. Oyama and H. Kunieda (Editors) •c 2001 Elsevier Science B.V. Ail rights reserved.
Dynamic Mobility of Concentrated Suspensions in the Presence of Polyelectrolytes N. Tobori and T. Amari Graduate School of Science and Technology, Chiba University 1-33 Yayoi-cho, Chiba 263-8522, Japan The dynamic mobility of CaC03 suspensions in the presence of polyelectrolyte as a dispersing agent was measured with varying volume fraction of particles, and it was converted to zeta potential and size distribution by theoretical treatment of dynamic mobility spectra. The results of the suspension that the particles have cationic, anionic and nearly neutral charge by adsorption of polyelectrolytes fitted to the theoretical calculation derived by O'Brien. For the suspension stabilized by steric hindrance, the zeta potential and size distribution calculated by dynamic mobility spectra can be obtained up to 0.6 of volume fraction. 1. INTRODUCTION Dynamic mobility can be determined by measuring the magnitude and phase angle of the sound wave generated by electrophoretic oscillations of particles (electrokinetic sonic amplitude; ESA effect), if an ahemating electric field applied in aqueous suspension. The theoretical treatment of dynamic mobility was developed by O'Brien , and it was verified for various systems. Since the frequency dependence of magnitude and phase lag corresponds to the effect of particle inertia, we can obtain the particle size distribution of particles in concentrated suspensions without any dilution [1,2]. This technique is very useful for the suspensions in many industrial situations, especially for investigating the effect of dispersing agents that is used to attain high solid concentration with adequate fluidity . In this study, we attempted to apply the electroacoustic measurement for the suspensions with high solid concentration. To do this, the dynamic mobility of CaCO^ suspensions including a polyelectrolyte as a dispersing agent were measured with varying volume fraction of particles, and it was converted to zeta potential and size distribution by theoretical treatment of dynamic mobility spectra. Since we can obtain highly concentrated suspensions with volume fraction up to 0.6 by adding the polyelectrolytes, the results provide some useful information about electroacoustic characterization in highly concentrated suspensions. 2. EXPERIMENTAL 2.1. Materials In this study, a dry powder of heavy calcium carbonate was used. The average size of CaCOs particles(median diameter) was 1.8/z m and its size distribution was relatively broad that ranged from 0.3 to 10 M m. The specific surface area was 3.2mVg determined by BET
technique, and the density was 2.71g/ml. Two types of polyelectrolyte; sodium polyacrylate (PAA) and comblike graft copolymer based on methacrylate (PG) were used as dispersing agents. The PG copolymer consists of sodium methacrylate and methacrylate ester of methoxy-polyethylene glycol (polymerization degree: 22) with molar ratio of 0.75/0.25. These polymers were obtained by radical polymerization in our laboratory. The molecular weight of PAA and PG polymer were 9000 and 40000, respectively. Aqueous concentrated suspensions of CaCOj particles were prepared by slowly adding the required amount of dry powder of CaCOs to an aqueous solution of the polymer employed under a controlled stirring. Air bubbles in the suspension were removed by evacuation. The obtained suspensions were allowed to stand for 1 day and were re-stirred using a homogenizer before loading for each measurement. 2.2. Measurements The dynamic mobility was determined using an Acoustosizer instrument (Colloidal Dynamics Ltd.). This instrument applies a highfrequencyalternating voltage to a suspension, causing the particles to oscillate at a velocity that is dependent on their size and charge. The resulting pressure force that arise at the suspension boundaries produce pulses of sound waves in a phenomenon known as the electrokinetic sonic amplitude (ESA) effect. The ESA signal is related to the particle averaged dynamic mobility, < ju d>, by the equation: ESA( a))=A( CO,«)<<)A p/
where co istheangularfrequencyof applied field, <> / is volumefractionof particle, A is an instrument constant depending on co and * , A /o is the density difference between the particle and the medium density, io . For a spherical particle with a thin double layer, n ^ can be related to the C -potential by the relationship: /id=G(coa^/v) £ C / T ]
where f and TJ are the permittivity and viscosity of the medium respectively. G term of the equation is a measure of the inertia of particles, having the radius a and kinetic viscosity of medium v at the frequency co. In a polydisperse system, = J M(co,a)/7(a)da
where /7(a)da is the mass fraction of particles with radius between a+da/2 and a-da/2. Therefore, the zeta potential C and size distribution /7(a) can be determined by fitting measured dynamic mobility spectrum to theoretical one.
8 10 12 Frequency (MHz) Fig. 1. Dynamic mobility spectra of the CaCOs suspensions containing various polyelectrolytes. Filled symbols; magnitude, open symbols; phase lag and curves; theoretical fits to experimental data.
-1 >-E-2 .
• 0.039 o 0.084 A 0.136 A 0.197 • 0.269 o 0.355 • 0.462 L... 1
10 12 4 6 8 10 12 • 4 6 8 0 Frequency (MHz) Frequency (MHz) Fig. 2. Dynamic mobility spectra (a; magnitude, b; phase lag) for PAA suspensions at different volume fraction. Symbols indicate experimental results and curves indicate theoretical fits to experimental data. 0
3.RESULTS AND DISCUSSION The magnitude and phase lag of the dynamic mobility measured as a function of frequency for CaCOj suspensions at (t> =0.289 including various polyelectrolytes are shown in Fig. 1. In the case of additive free suspension, the sign of particle charge was positive. With addition of PAA or PG in the suspension, the surface of particles became high negative charge for PAA and little charge for PG. From the rheological measurements, the viscosity of the suspension decreased drastically when the adequate amount of PAA or PG was added into the suspension. Thus, it can be considered that the particles in suspension were stabilized by electrostatic force for PAA suspension and by steric force for PG suspension. In addition, it was found that the resuUs for three suspensions with different surface structure agreed with the theoretical calculation assuming a log-normal distribution using Eq. (l)-(3). The volume fraction dependence of dynamic mobility was also examined. Fig. 2 shows the dynamic mobility spectra for PAA suspensions at various volume fractions as indicated. The samples being measured are diluted to adequate volume fractions from original suspension of 0 =0.462. It was also found that the results of all suspension fitted to 60 [-0 o additive free theoretical calculation. The frequency 40 1 0 o 0 0 A PAA: 0.4% dependence was gradually decreased with 20 D PG: 0.4% increasing volume fraction, because of 0 preventing oscillation of particles and absorbing the generated sonic wave at high -20 volume fraction. The zeta potentials 1-40 calculated from dynamic mobility spectra are shown in Fig. 3. The zeta potential ^ - 6 0 [ A A . ^ A, A , A , , 1 -80 1 seemed to be constant irrespective of kinds of polyelectrolyte and volume fraction. In 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 particular, in the case of PG suspension, it is Volume fraction interesting that the zeta potential can be Fig. 3. Zeta potential calculated from dynamic measured and indicate constant value up to mobility spectra for various suspensions as a 0=0.6. function of volume fraction.
426 Figure 4 shows the particle size distributions (PSD) calculated by the dynamic mobility for these suspensions. In this figure, D15, D50 and D85 denote the particle size showing 15%, 50% and 85% of cumulative mass fraction assuming a log-normal distribution, respectively. For additive free suspension, the particle size was large as volume fraction decreased. This can be attributed to the change of pH in suspension because of dilution. At constant volume fraction, the PSD of PG suspension was smaller than that of PAA suspension at same polymer concentration. This means that the particles in PG suspension is more dispersed by steric hindrance. In addition, an interesting behavior was observed that the PSD becomes narrower with maintaining the median diameter as increasing volume fraction. This tendency was remarkable for weakly flocculated systems such as additive-free and PAA suspensions at high volume fraction. On the contrary, the PSD was kept constant as increasing volume fraction for PG suspension. From the viscoelastic measurements, it was found that the storage modulus, G' of additive free and PAA suspension was larger than that of PG suspension. Therefore, It can be deduced that the flocculated network structure in suspension affects the response of dynamic mobility, especially high frequency region. 100 a)additive free
• Disl OD50
• D15J b)PAA
0 0 0
• AA A A
OQO 0 0
•••• a a
OD50 A 0851
• D15 1
0 0.2 0.4 0.6 0 0.2 0.4 0.6 0.2 0.4 0.6 Volume fraction Volume fraction Volume fraction Fig. 4. Particle size distributions calculated from the dynamic mobility for various suspensions as a fiinction of volume fraction. a)additive free suspension, b)suspension with 0.4% PAA, c^susnension with 0 4% PG 0
4. CONCLUSIONS Dynamic mobility for highly concentrated CaCO^ suspensions can be measured. Since the zeta potential and size distribution calculated from dynamic mobility was mostly reasonable, the measurement of dynamic mobility is usefiil technique to analyze the dispersing state of particle in concentrated suspensions. The difference in the size distribution obtained by dynamic mobility can be explained by the difference of dispersing state between PAA suspension (electrostatically stabilized) and PG suspension (sterically stabilized). REFERENCES 1. R. W. O'Brien, D. W. Cannon and W. N. Rowlands, J. Colloid Interface Sci., 173 (1995) 406 2. R. J. Hunter, Colloids and Surfaces A, 141(1998) 37 3. H. Ohshima, and K. Furusawa (eds), Electrokinetic phenomena at interfaces. Marcel Dekker, 1998