STUDY OF THE EFFECT OF SMALL ADDITIONS OF HYDROPHILIC COLLOIDS ON THE GRANULATION PROCESS OF COARSE-DISPERSED MATERIALS

V. M. Vitgofin, I. N. Lanshman

(Presented by the scientific seminar of the Department of GCT)

At present, in world and domestic practice, a method has been mastered and is developing very progressively for preparing fluxed metallurgical raw material; in foreign practice this method has received the name pelletizing, while in our country it is called balling. This method consists in rolling moist mixtures of iron-ore concentrates and fluxes with a particle size of less than \(100\mu\) in mechanical drums or pan granulators into spherical granules 10–20 mm in size, followed by drying and firing operations to obtain very strong capillary-porous bodies. Practice has established that the effective implementation of the granulation process for such relatively coarse-dispersed materials in pan and drum balling units is possible only in the presence of small additions \((0.2–1.0\%)\) of hydrophilic colloids, such, for example, as bentonite clays. However, the mechanism of action of these small, in quantity, but decisive additives has at present been insufficiently studied, which naturally makes it difficult to optimize the technological regime of the process.

To a considerable extent this shortcoming is due to the insufficient clarity of the mechanism of granule formation in such complex systems.

The overwhelming majority of domestic and foreign researchers, considering the main factors that determine the process, have confined themselves to analyzing the structure and mechanism of bonding in the formed green granules and have ignored the essential significance of the dynamics

the very process of granule formation. At the same time, it is quite obvious that, over the course of this process, from the initial loose mass, which is a three-phase system with a significant proportion of the gas phase, to the compact final state with a predominance of the solid and liquid phases, the bonding mechanism changes continuously. It is precisely at the stage of granule formation that the effect of small additions of hydrophilic colloids is manifested most clearly. Indeed, the previously mentioned fact of the effective granulation of coarse-dispersed materials only in the presence of small amounts of these additives confirms this proposition. In fact, the strength of the finished granules changes only slightly when additives are introduced, while the effectiveness of the process as a whole is determined mainly by the presence of these additives.

As a rule, the main component of the charge being granulated—magnetite concentrate—is a wet, relatively coarse-dispersed material, practically deprived in the wet beneficiation process of the colloidal fraction (particle size 0.1–0, clay-fraction content from 1 to 3%). In addition to magnetite, the concentrate contains gangue minerals such as pyroxenes, amphiboles, garnet, calcite, sulfides, and others. The secondary charge components of the mixture, fluxes and hydrophilic colloids, are usually represented by limestone or dolomite ground to 100 μ in an amount of 5–15% and by bentonite clays (0.5–1.5%). Thus, in the initial state the mixture to be pelletized is a loose polymineral, polydisperse, heterophilic, polyaggregate three-phase system with an approximate volume ratio S:L:G = 40:20:40. The content of the fraction smaller than 10 μ in the main components, as a rule, does not exceed 1–3%. In bentonite clays the colloidal fraction amounts to 80–90%. The subsequent transformation of this system up to the formation of wet granules is determined to a considerable extent by the specificity of the aggregation process in disc and drum pelletizers. The principal feature is that granule formation, taking place in a centrifugal-gravitational field, occurs with a high degree of freedom of volumetric deformations of the growing, compacting primary aggregates. Naturally, in this case the effectiveness of the granule-formation process will be governed by the predominance of plastic deformations. In a coarse-dispersed system being pelletized, such plastic deformations can develop only if thixotropic coagulation structures are present in the working volume. The formation of such structures and

takes place in the pore space of the compacting aggregates.

The wetting moisture, concentrated at the initial moment of aggregation on the surface of the particles of the solid phase and at their few points of contact, is squeezed into the pore space as the system is compacted under the action of external forces, carrying with it suspended particles of the clay fraction of the main components and bentonite. As compaction proceeds, the volume concentration of the solid phase of the pore suspension increases, bringing it into an increasingly structured state. The content of the gas phase in the system is thereby reduced to a minimum determined by the phenomena of mechanical entrainment. In the formed granules, the volume ratio of the phases is on average S:L:G = 64:32:4. Thus, in the process of granule formation, the volume content of the solid and liquid phases increases by one and a half times, while the volume of the gas phase decreases by an order of magnitude. The change in the phase ratio and the restructuring of the dispersed-system structure during granule formation substantially affect the properties and the character of interaction of the phases.

The bulk of the solid-phase particles, representing the coarse-dispersed fraction, forms a rigid framework during compaction of the system. At the same time, at the final stage of the process, the initial wetting water is divided into water adsorbed by the framework grains and water of the pore suspension. Thus, in the process of granule formation the system is divided into two main parts. In this, the pore suspension plays the decisive role in the effectiveness of compaction, providing plastic deformations to one degree or another. Therefore, the amount and rheological properties of the pore suspensions formed in the pelletized material determine the effectiveness of the granule-formation process as a whole. The role of the framework part is reduced mainly to ensuring maximum compaction through a rational granulometric composition and sufficient adhesion capacity with respect to the structured pore suspension.

Proceeding from this working hypothesis of granule formation in coarse-dispersed materials with small additions of colloids, an experimental study was carried out of certain colloid-physical phenomena and structural-mechanical properties of industrial dispersions.

It is known that one of the indicators of the colloid-chemical properties and aggregate state of dispersed systems is the electrokinetic potential, which was determined by the electroosmotic method

with a correction for surface conductivity. Studies have established that the efficiency of the granulation process for dispersed materials is the higher, the smaller the values of the $\zeta$-potential.

A comparison of the values of the $\zeta$-potentials of diaphragms formed from the combined dispersed system (framework and pore suspension) showed that the introduction into the system of bentonite clays differing in quality has only a slight effect on the magnitude of the $\zeta$-potential of the system. At the same time, the values of the electrokinetic potential for different bentonites differ substantially. Thus, for example, the introduction of Azkamar bentonite ($\zeta = -17.1$ mV) and Sary-Tokh bentonite ($\zeta = -13.7$ mV) gives, respectively, values of the $\zeta$-potentials of the mixtures of 4.8 and 3.2 mV.

Similarly, the influence of the ionic composition of the dispersion medium (when distilled water is replaced by process water) on the magnitude of the electrokinetic potential for diaphragms of concentrate with bentonites is barely noticeable: the $\zeta$-potential of magnetite with an addition of Azkamar bentonite changed from 4.8 mV (for distilled water) to 4 mV (for process water). Conversely, this influence of the ionic composition of the dispersion medium on the electrokinetic potential for the bentonites themselves is quite substantial ($-17.1$ and 10.2 mV, respectively, for distilled and process water). The change in the $\zeta$-potential of magnetite concentrate as a function of the concentration of the introduced cations Ca, Na, and Al is shown in Fig. 1, from which it is seen that, with an increase in the concentration of cations in water, the $\zeta$-potential decreases; moreover, with increasing valence, the coagulating action of the cation increases in accordance with the Schulze–Hardy rule.

It is of interest to trace the change in the electrokinetic potential of magnetite concentrate upon addition of the humic-alkaline reagent (HAR), represented mainly by sodium humates. The initial increase in the $\zeta$-potential is caused by exchange of divalent cations for $\mathrm{Na}^{+}$ ions. With an increase in the concentration of HAR (0.1–0.3%), a decrease in the $\zeta$-potential occurs as a result of adsorption of humates, leading to shielding of the electrical double layer.

A further increase in the $\zeta$-potential indicates the formation of a layer of humate molecules with reverse orientation. This mechanism relating the action of HAR is also confirmed by sorption isotherms. Data on the change in the strength of the finished wet and dry granules indicate the maximum value of their strength with the addition

0.3% SHP, which corresponds to the minimum value of the \(\zeta\)-potential for this system.

The value of the electrokinetic potential for various systems, although it makes it possible to estimate the relative contribution of the pore suspension to the mechanism of particle bonding in aggregates, does not allow one to trace the influence of this active part of the system on the process of granule formation, since the determination of the \(\zeta\)-potential is tied to a diaphragm that is stable over the duration of the experiment. From this standpoint, the use of rheological indices of pore suspensions is far more productive; these indices depend on the properties and concentration of the solid phase, which changes during the granule-formation process.

Fig. 1. Effect of cation concentration on the \(\zeta\)-potential of magnetite

Investigation of the dependences of the limiting shear stress and plastic viscosity on the concentration of bentonite suspensions, and calculations of the corresponding plasticity, showed that the latter is an extremal function. In this case, the magnitude and position of the extremum are determined by the quality of the solid phase and the ionic composition of the pore solution. The dependences obtained make it possible to substantiate a method for evaluating the effectiveness of small additions of hydrophilic colloids, in particular bentonites, which industry urgently needs.

The essence of this method follows from the extremal dependence \(\psi - C\). The position of the maximum on the \(\psi - C\) curves can evidently serve as a qualitative–quantitative characteristic of the bentonite added to the material being granulated.

As the practice of pelletizing magnetite concentrates at the Sokolovsk-Sarbai concentrating combine shows, the optimum

the amount of bentonite in the charge corresponds to the maximum plasticity of the pore suspension. Thus, for example, the best results in granulating concentrate with Oglanly bentonite are obtained when the latter is added in an amount of 0.7%. The concentration of the pore suspension in this case is 22% and coincides with the concentration corresponding to the maximum plasticity of the pore suspension of the green pellets.

For selecting the type of bentonite and its consumption, one may use the index of specific maximum plasticity, which is the ratio of the maximum plasticity of the aqueous bentonite suspension to the bentonite concentration corresponding to the maximum plasticity. In accordance with the values of this index, the bentonites may be arranged in the following series: Sarigukh > Oglanly > Taganrog pink > Makharadze > Taganrog gray.

It is quite natural that, in the process of structure formation, the silty fraction of the principal components of the charge will also participate and exert an influence. As a rule, when iron-ore concentrates are granulated with fluxes, the silty fraction is represented by hydrophilic rock-forming minerals of the pyroxene, chlorite, calcite type, etc. These components of the pore suspension, while not themselves producing a structure-forming effect, significantly affect the rheology of bentonite clays (as a rule, increasing the yield shear stress). This indicates the need to take into account the quality and quantity of the principal components when selecting the optimum amount of hydrophilic colloidal additives to the pelletizing charge.

The extremal dependence of the plasticity of pore suspensions on the quality and quantity of the solid phase of the pore suspension makes it possible to determine one of the most important parameters of the granulation process—the optimum moisture content of the mixture before granulation. For this, evidently, only a correction is needed for the water adsorbed by the surface of the framework grains. The optimum pelletizing moisture \(\left(W_p^{\mathrm{opt}}\right)\), on this basis, can be determined from the following equation:

\[ W_p^{\mathrm{opt}} = W_{\mathrm{frame}}^{\mathrm{mg}} \cdot x + W_{\mathrm{p.s.}}^{\max}(100 - x) \]

\(W_{\mathrm{frame}}^{\mathrm{mg}}\)—maximum hygroscopicity of the framework portion of the charge;
\(x\)—content of the framework portion, %;
\(W_{\mathrm{p.s.}}^{\max}\)—moisture content of the pore suspension corresponding to the maximum plasticity.

Thus, studies of the colloid-chemical and structural-mechanical properties of disperse materials used in

in wet granulation processes make it possible to reveal the mechanism of action of small additions of hydrophilic colloids and to develop scientifically grounded methods for assessing the quality and quantity of these additives, as well as to predict the principal parameters of the technological regime of the process.