Of coagulating reagents with the sol

Coagulation has been analyzed so far as the result of two major reaction steps, the destabilization of the colloids and the transport and adherence of the destabilized dispersed solids to form permanent aggregates. And this model suffices to explain all observations as long as coagulation is studied in a laboratory context, i.e., small-volume batch reactions.

As soon as coagulation is explored in a technical context, under conditions of continuous flow and in larger reactor vessels, then the addition and mixing of the destabilizing reagent must be considered an additional and possibly rate-controlling step.

This may be illustrated by the following example: If one wants to effectively coagulate a suspension under the conditions of a large (hydraulic) flux, one will have to be careful not to use too large a volume of liquid (i.e., water) for the dissolution and transport of the chemical '). To mix such disparate streams will require additional reaction time. It may also lead, under unfavorable reaction conditions, to situations where this step will be rate-limiting.

If addition and mixing of chemicals is thought of as an independent reaction step, then the following configurations of physical and chemical boundary conditions could be envisioned:

1 In practical applications this can lead to situations where a suspension stream of 3000 1/h has to be mixed effectively with a chemicals stream of as little as 0.5 1/h, i.e. one thousands of the volume flow or less.

a) The coagulating chemical is formed in situ upon its entrance into the sol and its effectiveness depends upon the solution regime encountered, i.e., the homogeneous concentration of all solute constituents (e.g., hydrolyzing metal ions).

b) The destabilizing chemical is introduced already in its active form and the reaction is not dependent upon the mixing characteristics of the system in this instance (e.g., prepolymerized metal hydroxo-complexes).

c) The coagulant or flocculant is of such concentration and of such molecular size that its "interaction" with the colloid, for instance, its motion toward the colloid surface, is dependent upon turbulent transport as well as upon diffusion processes (high molecular weight organic polymers).

d) The interaction of the coagulating/flocculating chemical with the colloid is independent of turbulent transport phenomena and therefore cannot be improved by controlling the hydraulic regime (highly charged counterions).

e) The reaction of the destabilizing chemical with the colloidal surface is, from a point of view of available reaction time, irreversible (e.g., adsorption or organic polymers).

f) The surface-chemical reaction is to be considered reversible under conditions of practical application, i.e., non-homogeneous solution characteristics can be corrected or compensated later on (e.g., counterions).

For the most frequently encountered situations where hydrolyzed and hydroxo-complexing metal ions are used for the destabilization of aqueous sols, the situation, as it is to be envisioned, is shown in Fig. 1. The upper part describes schematically the theoretical situation: the first phase of the chemical mixing step leads to uniform hydroxilation of the metal ion. Subsequently, the metal-hydroxo-complex adsorbs - again uniformly — at the surface of the predominantly hydrophilic colloid (as encountered in aqueous systems).

The lower part of the figure describes the situation as it might be encountered under technical conditions where mixing and transport leads to non-homogeneous situations: First of all the hydroxo-complexation is nonuniform. This is "visualized" by showing mixed species of hydroxilated and hydrolyzed ions simultaneously. Second, the interaction of the colloid surface and the metal-hydroxo-complexes leads to insufficient destabilization in one case and re-stabilization in another, for reasons of less efficient transport. The overall effect is insufficient destabilization, even though the amount of chemicals added was sufficient on the basis of "theoretical" calculations. Such hpyotheses are confirmed by observations reported in the literature. Figure 2 illustrates with data by Masides et al. [2] the reversible or irreversible hydroxilation of aluminum as used in the coagulation of technical systems: depending upon the previous pH-regime, to

'IDEAL"

'IDEAL"

"REAL"

Fig. 1 Schematic depiction of the consequences for colloid de-stabilization if non-homogeneous mixing of chemicals (in this instance, trivalent metal ions) occurs

"REAL"

CHEMICALS ADDITION

CHEMICALS ADDITION

Colloid Destabilizing Effect

c high Ke c high Ke

Fig. 3 Mixing of chemicals in pipe flow, described by the change in concentration along the pipe axis and the pipe wall for two different hydraulic regimes

Fig. 1 Schematic depiction of the consequences for colloid de-stabilization if non-homogeneous mixing of chemicals (in this instance, trivalent metal ions) occurs

. fireein. aluni! hum h*dra*ide ' re-activateii

. fireein. aluni! hum h*dra*ide ' re-activateii

ig^imBmm

Fig. 2 Coagulation efficiency, i.e., destabilizing efficiency of aluminum prepared under different pH regimes, illustrating the irreversible hydroxilation or hydroxide formation (after [2])

pH^S.i ig^imBmm

Fig. 2 Coagulation efficiency, i.e., destabilizing efficiency of aluminum prepared under different pH regimes, illustrating the irreversible hydroxilation or hydroxide formation (after [2])

which the coagulant was exposed, the chemical is more or less effective in destabilizing water-borne colloids.

Observations on mixing in real systems, schematically described in Fig. 3 (after [3]) show that the mixing can be slower or faster, depending upon the hydraulic characteristics of the mixing reactor, denoted by lower or higher

Reynolds' numbers. Here, the point of addition of chemicals has been designed such that the highly concentrated stock solution is added into the central flow trajectory of the pipe flow. The flow regime is laminar or turbulent, depending upon the hydraulic load of the system. Measurements have shown that for conditions of "normal" loading of the system, i.e., lower flow velocity and therefore lower Reynolds numbers, the time required for complete mixing is very large. The state of mixing can be described by the difference in concentration, for instance, between the peripheral and the central flow trajectory. If the so-called through-put, i.e., the flow per unit cross-sectional area, is increased, this leads to an increase in the flow velocity and an increase in the Reynolds number. Within limits there is an increase in the degree of turbulence within the system. The result is a shortening of the mixing time, the time to reach homogeneous concentration of for instance, metal ions or hydroxilyzed metal species in the system.

The actual effect of such different hydraulic boundary conditions in the stage of chemicals addition, i.e., more or less successful chemicals' mixing, is shown by the data described in Fig. 4 (after [4]). Here, the coagulation efficiency is described for two different chemicals dosing systems. The difference between the two systems is a solely geometrical one, as the schematic indicates.

In the instance of the dosing fixture "A", one needs an amount of some 40 mg/1 prepolymerized metal-hydroxo-complexes coagulant to reach effective destabi-lization. This is recognized from measurements on elec-trophoretic mobility as well as from observations on the removal rate (measured as particle number, or turbidity turbidity (Formazin-Trûbungseinheiten)

turbidity (Formazin-Trûbungseinheiten)

Fig. 4 The effectivity of two slightly different designs of chemical dosing points expressed in terms of resulting electrophoretic mobility and the rate of the coagulation and separation process (measured by reduction in algae and turbidity after [4])

or chlorophyll alpha in the case of algal matter). If fixture "B" is used then for the same sol and the same chemical a (significantly) larger dosage is needed to attain values of the electrophoretic mobility that signal sufficient destabilization. Turbidity measurements and chlorophyl alpha recordings again support these electrophoresis data.

Clearly, the mixing situation with apparatus "A" leads to a more homogeneous or more efficient or faster mixing than the apparatus "B". And since the surface reaction is not readily reversible within the given detention time, this leads to satisfactory destabilization in one case and unsatisfactory destabilization in another.

0 0

Post a comment