THE ROLE OF chemical flocculants in the disposal of red mud
K.E. Bremmell, P.J. Scales, T.W. Healy
Advanced Mineral Products Special Research Centre
The School of Chemistry
The University of Melbourne
Parkville, VIC 3051
Flocculants are important chemical additives in the treatment of red mud. They are used to control particulate slurry rheology to achieve a high recovery of caustic and a high solids waste for disposal. This work investigates the mode of action of polymeric flocculants used in the clarification and washing stages of the Bayer process that feed the final thickening and waste disposal stages.
The Atomic Force Microscope (AFM) allows the interactions between approaching surfaces, similar to the interactions that occur between particles in a slurry, to be measured. In this study, a model red mud surface, developed specifically for use in the AFM, was used to investigate the role of the chemical composition of the polymers on flocculation under Bayer conditions. AFM and electrokinetic techniques have been successful in providing an understanding of the molecular role of additives in controlling floc structure. The results provide an understanding of the molecular role of flocculants in controlling the surface chemistry and subsequently, the floc structure and rheology of the red mud. An understanding of the mechanism of interaction of the flocculants with bauxite residue surfaces under process conditions should enable the solid-liquid separation to be optimized with subsequent improvements in washing performance.
flocculants, red mud, adsorption, atomic force microscope, particle interactions
THE ROLE OF chemical flocculants in the disposal of red mud
K.E. Bremmell, P.J. Scales, T.W. Healy
Interactions between particles play an important role in the rheological behaviour of solid/liquid dispersions. Inter-particle forces have been shown to be important in controlling the rheology of particulate fluids (Johnson, et al., 1998). The flocculation, washing and de-watering stages of red mud processing are all influenced by particle-particle interactions. To develop a more efficient process, it is thus important to understand how these interactions control the final flow and de-watering properties of the red mud.
The inter-particle force in the presence of polymers is assumed to be dominated by long-range attractive bridging forces that promote particle aggregation to form flocs. These polymers also increase floc robustness. An understanding of the action of various chemical flocculants used in the mud washing circuit will enable the rheology of the system to be better characterized. Conditions providing a suitable particle network structure for optimum caustic recovery and mud disposal should then be attainable. The purpose of this paper is to describe a method, namely Atomic Force Microscopy, to enable these particle interactions to be investigated.
1.1 Atomic Force Microscopy
Development of the Atomic Force Microscope (AFM) has allowed both imaging and the interactions between colloidal particles to be investigated in an aqueous environment. The instrument can be adapted to allow the force between two surfaces to be directly measured. Both long and short range forces responsible for the behaviour of dispersed systems have been identified (Israelachvili, 1992) and can be used to interpret the measured interactions. The AFM has been used to study the interaction between surfaces in the presence of ions, small chemical additives and polymers. The inter-relationship between these measured forces and macroscopic rheological properties such as shear and compressive yield stress of suspensions can then be determined. Measurement of the interaction forces in the presence of polyelectrolytes has also been investigated, providing an understanding of the molecular role of additives in controlling floc structure. Most studies have involved a surface and a polymer of opposite charge (Dahlgren et al., 1994; Bremmell et al., 1998). The system of interest to this investigation is a flocculant and surface both with a negative charge. The Bayer process provides an ideal opportunity to apply this technique to characterize the role of flocculants down the washing and disposal trains.
All AFM measurements were performed using a Digital Instruments Inc., Nanoscope III atomic force microscope. Figure 1 shows a schematic of the essential components of the instrument. As the cantilever tip is scanned over the surface, the force between the tip and the sample deflects the cantilever. The deflections are monitored by a laser beam reflected off the tip onto a photo-diode detector. Topographical images can then be collected.
Schematic of the Atomic Force Microscope (AFM)
Measurement of forces involves sticking a colloidal probe on the cantilever tip, and measuring the force of attraction or repulsion between the particle and a flat surface, which is moved toward and away from the probe with a piezo-electric crystal as demonstrated in Figure 2. The force between the two surfaces is then measured as a function of separation. Deflection of the cantilever spring, due to its interaction with the surface, will change the position of the laser beam on the detector shifts. The photodiode voltage versus displacement data can then be converted to a force versus separation curve as shown in Figure 2.
Geometrical constraints limit the evaluation of particle-particle interactions at small surface separations to smooth surfaces. In this study, the interaction between a flat hematite sample and a colloidal silica sphere attached to the cantilever are measured. Interaction forces as a function of solution pH and ionic strength will be shown, and flocculant addition will also be considered.
Measuring forces using the AFM. The force between the sample and colloidal probe is measured as a function of separation as the sample moves toward and away from the cantilever tip
2 Results and discussion
Bayer residues consist primarily of iron oxides. Hematite (a -Fe2O3) has been chosen as a model surface in this study as it is one of the major components of these residues (Yamada et al., 1980; Li et al., 1996). Prior to interaction force measurements in the AFM, the surfaces were characterized to ensure their suitability for use in this technique. AFM imaging and SEM surface analysis were performed.
Two, flat, smooth hematite surfaces were investigated. An image of a polished hematite sample is shown in Figure 3. This surface has a root mean square roughness (RMS) of 4 nm. The surface of the second sample, a specular hematite specimen, is illustrated in Figure 4, and was found to have an average surface roughness of 0.7 nm. The surface roughness results indicate both of these natural hematite samples to be suitable for study using the AFM. However, the polishing scratches visible in Figure 3 and the lower RMS roughness value for the specular hematite (Figure 4) suggest specular hematite would provide a more uniform substrate for this investigation.
Figure 3 Figure 4
AFM surface image of the polished hematite sample
image of specular hematite
Elemental analysis of the surface using SEM show both samples to be predominantly hematite, with small amounts of Cr and Ti present as impurities. The specular hematite surface was chosen for the interaction force experiments and a silica sphere was attached to the cantilever to act as the second surface. The interaction between the two surfaces was measured in aqueous solution as a function of pH and ionic strength and in the presence of flocculant (polyacrylic acid).
To investigate the effects of flocculant in a homogeneous iron oxide system, it would be beneficial to perform the force experiments between two hematite surfaces. A spray pyrolysis synthesis of hematite spheres has recently been developed in our laboratory. An example of one such sphere is shown in Figure 5. In future experiments we hope to arrange these spheres in the AFM to allow measurements in a homogeneous system to be performed.
Scanning electron micrograph of an iron oxide sphere
Direct interaction force measurements between specular hematite and a colloidal silica probe were measured in 1 mM NaNO3 solution. Figure 6 shows the interaction force between the approaching surfaces as a function of solution pH. The force between the two surfaces at pH 10.3 displays a repulsion at large surface separations and is well described by an electrostatic repulsive interaction force. The hematite and silica surfaces have a negative charge at this pH. As the pH is reduced to 7.8, the repulsive interaction between the surfaces is seen to decrease. It is known that silica has a negative charge in aqueous solution at pH values greater than approximately 2. Therefore, the reduction in the repulsive interaction suggests the charge on the hematite surface to be decreasing in magnitude. In a solution of pH 6, the interaction force shown in Figure 6 shows no repulsive interaction and is attractive at small surface separations. It is concluded that the charge at the hematite surface must be close to neutral. Therefore, the iso-electric point of the hematite occurs at a pH of approximately 6. When the solution pH is lowered to 5 an attraction is seen between the approaching surfaces. At pH values below the iso-electric point, the hematite surface has a positive charge and as the silica is negative at this pH, the surfaces experience an electrostatic attraction.
Direct interaction forces between silica and hematite as a function of pH. The iso-electric point occurs at a pH of approximately 6
An increase in the concentration of electrolyte between the surfaces alters the interaction forces between the solids. The interaction between the hematite surface and the colloidal silica probe, as a function of electrolyte (NaNO3) concentration at a solution pH of 10, is shown in Figure 7. In 0.001 M NaNO3 solution, the interaction is repulsive, as discussed above. As the NaNO3 concentration is increased to 0.1 M, the extent and the magnitude of the repulsion decreases. This behaviour is a result of the electrolyte ions screening the double-layer of charge present at the solid-liquid interface. The repulsive electrostatic interaction is therefore reduced.
Direct interaction forces between silica and hematite as a function of ionic strength (NaNO3)
When the solution is replaced with one of 2 M NaNO3, the interaction is attractive at small surface separations despite the negative charge present on both surfaces at pH 10. In this case, the Debye length is greatly reduced (k -1 is 9.7 nm in 10-3 M NaNO3, and 0.22 nm in 2 M NaNO3). Therefore, the electrostatic charge is screened to such an extent that the attractive van der Waals interaction between the surfaces at the small separations can dominate.
Interaction forces between silica and hematite with adsorbed polyacrylic acid flocculant
The effect of a high molecular weight fully charged anionic flocculant (polyacrylic acid) on the interaction is shown in Figure 8. The interaction force curve between the surfaces in the presence of 20 ppm flocculant at a pH of 10, is well described by electrostatic forces. If present, the steric barrier is small. These preliminary results suggest the polymer has adsorbed in a thin layer on the surface. It is known that for a system where the polymer and surface are opposite in charge that the polymer adsorbs in a flat conformation as a result of strong electrostatic attraction. In the present case, the polymer must therefore have an affinity for the surface in order for the flocculant molecule to lie flat. If the polyelectrolyte is adsorbed in an extended conformation on the surface, bridging between particles would be possible. Rinsing the system with a 2M NaNO3 solution decreases the extent of interaction to that when no polymer is present, suggesting the polyelectrolyte is adsorbed only weakly to the surface.
In the Bayer residue circuit, addition of this family of flocculants would initially promote bridging between particles. As the Bayer residue solids pass through the thickening and washer trains, the flocs may be easily broken not only as a result of shear, but also due to partial removal of the flocculant from the bridging configuration within the aggregates. Thus, further addition of flocculant is required in each unit to re-aggregate the particles.
This preliminary study has shown the Atomic Force Microscope to be a valuable tool in probing the interactions between particles representative of red mud dispersions. Interaction measurements of this scale have not been performed for the Bayer process, and should provide a greater understanding of the washing and de-watering process of red mud, leading to an increase in the efficiency of the process.
This work is supported through the Australian Research Council and the Australian Minerals Industries Research Association (with contributions from Alcoa, Comalco, QAL, Worsely, Allied Colloids and Cytec), through funding to the Advanced Mineral Products Special Research Centre (AMPC).
Bremmell, K.E., Jameson, G.J., Biggs, S. (1998). Polyelectrolyte adsorption at the solid/liquid interface. Interaction forces and stability. Colloids and Surfaces, 139, 199-211.
Dahlgren, M.A.G., Claesson, P.M., Audebert, R. (1994). Highly charged cationic polyelectrolytes on mica: Influence of polyelectrolyte concentration on surface forces. J. Coll. Int. Sci., 166, 343.
Hartley, P., Larson, I., Scales, P.J. (1997). Electrokinetic and direct force measurements between silica and mica surfaces in dilute electrolyte solutions. Langmuir, 13, 2207-2214.
Hogg, R., Healy, T.W., Fuerstenau, D.W. (1966). Mutual coagulation of colloidal dispersions. Trans. Faraday Soc., 62, 1638.
Israelachvili, J.N. (1992). Introduction to Molecular and Surface Forces, 2nd Edition, Academic Press, NY.
Johnson, S.B., Russell, A.S., Scales, P.J. (1998). Volume fraction effects in shear rheology and electroacoustic studies of concentrated alumina and kaolin suspensions. Colloids and Surfaces, 141, 119-130.
Larson, I, Drummon, C.J., Chan, D.Y.C., Grieser, F. (1995). Direct force measurements between dissimilar metal oxides.J.Phys, Chem., 99, 2114-2118.
Yamada, K., Harato, Y., Shiozaki, Y. (1980). Flocculation and sedimentation of red mud, Light Metals, 39-50.
Li, LY; Rutherford, GK, (1996). Effect of bauxite properties on the settling of red mud. International Journal of Mineral Processing, 48, 169-82.