2d

Figure 4-15. Cations between the sheets of montmorillonite. (From Norrish26 Courtesy of Discussions of the Faraday Society.)

repels those of the same sign. The net result is a distribution of positive and negative ions, as shown schematically in Figure 4 16.-:" In the case of class, the surface charge is negative, as wc have seen, and the exchangeable cations act as counter ions.

The distribution of ions in the double layer results in a potential grading from a maximum at the clay surface to zero in the bulk solution, as shown in Figure 4 16.2 8

Particle

Solution

Figure 4-16a. Diffuse electric double layer model according to Gouy. (From van OlphenP Courtesy of John Wiley and Sons.)

Figure 4-16b. Diagram illustrating the zeta potential. (From Engelmann, Terichow, and Selim,2(i)

The layer of cations next to the surface of the particle, known as the Stern layer, is bound to and moves with the particle whereas the diffuse ions arc independently mobile. Thus, if a clay suspension is placed in a cataphoretic cell, the particle, the Stern layer, and some of the diffuse ions move toward the cathode. The potential difference from the plane of shear to the bulk of the solution is known as the zeta potential, and is a major factor controlling the behavior of the particle. Again, water flowing past stationary particles, as in the case of water flowing through the pores of a shale, removes the mobile ions, thereby generating a potential, which is known as the streaming potential.

The zeta potential is maximum, and the mobile layer is most diffuse when the bulk solution is pure water. Addition of electrolytes to the suspension compresses the diffuse layer, and reduces the zeta potential. The zeta potential decreases greatly with increase in valence of the added cations, especially if low valence ions are replaced by higher valence ones through base exchange, the ratio being approximately 1 to 10 to 500 for monovalent, divalent, and trivalent cations, respectively.28 The zeta potential is also reduced by the adsorption of certain long-chain organic cations, in some cases, it is possible to neutralize and reverse the zeta potential.

The potential difference between the surface of the particle and the bulk solution is known as the Nernstpotential. In a clay suspension this potential is independent of the electrolytes in solution.

Although ions are adsorbed mostly on the basal surfaces, they are also adsorbed at the crystal edge, and consequently a double layer develops there, also, it must be remembered, however, that the crystal structure is interrupted at the edge, so that, in addition to physical (electrostatic) adsorption, there may be specific chemical reactions with broken valences. Chemisorption, as it is called, is similar to ordinary chemical reactions, and only occurs under the appropriate electrochemical conditions. The edge charge is less than the basal surface charge, and may be positive or negative, largely depending on pH. For example, if kaolinite is treated with. HCI, it has a positive charge, and if treated with NaOH, it has a negative charge.-'1 The reason for this behavior is that aluminum atoms at the edge react with HCI to form AICI3, strong electrolyte which dissociates to Al + 3 + 3CI . whereas with NaOH, aluminum forms aluminum hydroxide, which is insoluble. (Remember that ion adsorption in kaolinite takes place almost entirely at the edge, so that the charge on the particle is determined by the charge on the edge).

The existence of positive sites on the edges of kaolinite has also been demonstrated by an experiment in which a negative gold sol was added to a kaolinite suspension.30 An electron micrograph showed the gold particles adsorbed only at the crystal edges (See Figure 4 17).

Figure 4-17. Electron micrograph of a mixture of kaolinite and a gold sot. (From van Olphen,?7 photographed by H. P. Studer. Courtesy of John Wiley and Sons.)

Particle Association Flocculation and Deflocculation

As mentioned in the beginning of this chapter, colloid particles remain indefinitely in suspension because of their extremely small size. Only if they agglomerate to larger units do they have finite sedimentation rates. When suspended in pure water, they cannot agglomerate, because of interference between the highly diffuse double layers. But if an electrolyte is added, the double layers are compressed, and if enough electrolyte is added, the particles can approach each other so closely that the attractive forces predominate, and the particles agglomerate. This phenomenon is known as flocculation. and the critical concentration of electrolyte at which it occurs is known its the fion ulation value.

The flocculation value of clays may be readily determined by adding increasing amounts of electrolyte to a series of dilute suspensions. The change from a deflocculated suspension to a flocculated one is very marked. Before flocculation, the coarser particles may sediment out, but the supernatant fluid always remains cloudy . Upon flocculation, clumps of particles big enough to be seen by the naked eye are formed: these sediment, leaving a clear supernatant liquid. The particles are very loosely associated in the floes, which enclose large amounts of water (see Fig. 4-18). and consequently form voluminous sediments.

Figure 4-18. Schematic representation of flocculated clay platelets (assuming negative edge potential).

The flocculation value depends on the species of clay mineral, the exchange cations thereon, and on the kind of salt added. The higher the valence of the cations (either on the clay or in the salt) the lower the flocculation value Thus, sodium montmorillonite is flocculated by about 15 meq/1 of sodium chloride, and calcium montmorillonite by about 0.2 meq/1 of calcium chloride. The situation is more complicated when the cation of the salt is different from the cation on the clay, because then base exchange occurs, but the flocculation value is always much lower whenever polyvalent cations are involved. For instance, the flocculation value of sodium montmorillonite by calcium chloride is about 5 meq/1, and of calcium montmorillonite by sodium chloride about 1.5 meq/1.

There is a slight difference in the flocculating power of monovalent salts, as follows. Cs > Rb > NH4 > K > Na > Li, This scries is known as the Huffmeister series, or as the lyotropic series.

If the concentration of clay in a suspension is high enough, flocculation will cause the formation of a continuous gel structure instead of individual floes. The gels commonly observed in aqueous drilling fluids are the result of flocculation by soluble salts, which are always present in sufficient concentrations to cause at least a mild degree of flocculation.

Gel structures build up slowly with time, as the particles orient themselves into positions of minimum free energy under the influence of Brownian motion of the water molecules (A position of minimum free energy would be obtained, for instance, by a positive edge on one particle moving towards the negative surface on another). The length of time required for a gel to attain maximum strength depends on the flocculation value for the system, and the concentration of the clay and of the salt. At very low concentration of both, it may take days for gelation even to be observable,31 whereas at high concentrations of salt, gelation may be almost instantaneous.

Flocculation may be prevented, or reversed, by the addition of the sodium salts of certain complex anions, notably polyphosphates, tannates, and iignosulfonates. For instance, if about 0.5% of sodium hexameta-phosphale is added to a dilute suspension of sodium montmorillonite, the flocculation value is raised from 15 meq/1 to about 400 meq/1 of sodium chloride, A similar amount of a polyphosphate will liquify a thick gelatinous mud. This action is known as peptization, or defiocculation, and the relevant salts are called deflocculanis or thinners in the drilling mud business.

There is little doubt that thinners are adsorbed at the crystal edges. The small amounts involved are comparable to the anion exchange capacity, and there is no increase in the c-spacing, such as one would expect if they were adsorbed on the basal surfaces. The mechanism is almost certainly chemisorption, because all the common thinners are known to form insoluble salts, or complexes, with the metals such as aluminum, magnesium, and iron, whose atoms are likely to be exposed at the crystal edges.,2 Furthermore,

—O— P—O—P—O— I1—0-— P - -i >— P—(>— P-4)— P—

-------------------- AI+ ■----------------------------------------- Al+ - - AP

Edge surface of particle

Figure 4-19. Schematic representation of a polyphosphate molecule adsorbed on clay crystal edge by bonding with exposed aluminum atoms. (From van Olphen.27 Courtesy of John Wiley and Sons.)

Loomis33 et al. obtained experimental evidence indicating chemisorption They treated clay suspensions with sodium tetraphosphate, centrifuged the suspensions, and analysed the supernatant liquor. They found that the phosphate was adsorbed, and that the amount required to effect maximum reduction in viscosity depended on the amount of phosphate per unit weight of clay, not on the phosphate concentration in the water phase, indicating chemisorption. In contrast, when suspensions were treated with sodium chloride, it was not adsorbed, and the amount required to cause maximum gelation depended on the concentration of the salt in the water. In addition, they found that when suspensions were treated with a series of complex phosphates of increasing molecular weight or with a tannate, the amount of each required to produce maximum reduction in viscosity was in proportion to the area that would be covered if the molecules were adsorbed on the clay.

van Olphen34 has postulated that the complex phosphate molecules are oriented on the clay edge surfaces by bonding with the exposed positive-charged aluminium atoms, as shown in Figure 4-19. Dissociation of the sodium ions then produces a negative edge surface, thus preventing the buildup of gel structures by positive edge to negative basal surface linkages. The creation of a negative edge is supported by the observation that the cataphoretic mobility increases after treatment with phosphates.

Although chemisorption appears to be the mechanism involved with small amounts of thinner, another mechanism must be responsible for the decrease in gel strength that is observed when larger quantities are added. In this case, the reduction in gel strength and the amount of thinner adsorbed, relative to the amounts added, are smaller. Probably the mechanism in this case is the exchange of simple anions in the double layer at the crystal edge with the large, multivalent anions of the thinner.

Some doubts have been raised about the action of ferrochromc lignosulfonate because it has been observed that base exchange occurs between Fe + 2 and Cr+3 from the lignosulfonate, and Na + and Ca + 2 on the clay,35 and this exchange would suggest that lignosulfonates were adsorbed on the basal surfaces. On the other hand, X-ray diffraction studies have shown no significant change in c-spacing. A possible explanation is that the lignosulfonates react with the aluminium at the crystal edges, but in doing so. release chromium and ferrous ions, which subsequently exchange with sodium and calcium ions from the basal surfaces.36

Aggregation and Dispersion

Although all forms of particle association are termed flocculation in classical colloid chemistry, in drilling fluid technology it is necesary to distinguish between two forms of association, because they have a profoundly different effect on the rheology of suspensions. The term flocculation is limited to the loose association of clay platelets which forms floes or gel structures, as discussed in the preceding section. The term aggregation, as used here, referes to the collapse of the diffuse double layers and the formation of aggregates of parallel platelets spaced 20 A or less, apart.37 Aggegation is the reverse of the sudden increase in c-spacing that Norrish20 observed when the layers in a flake of sodium montmorillonite overcame the attractive forces between them, and expanded to virtually individual units (See section on clay swelling mechanisms, earlier in this chapter). Thus, whereas flocculation causes an increase in gel strength, aggregation causes a decrease because it reduces (1) the number of units available to build gel structures and (2) the surface area available for particle interaction.

The lerm dispersion is commonly used to describe the subdivision of particle aggregates in a suspension, usually by mechanical means. Garrison38 proposed extending the term to the subdivision of clay platelet stacks, which is usually the result of electro-chemical effects, and thus to distinguish between the dispersion-aggregation process and the deflocculation-flocculation process. In the technical literature, the term dispersion is unfortunately still sometimes applied to the deflocculation process. The difference between the two processes (the floculation-defloculation process on the one hand, and the aggregation-dispersion process on the other) is illustrated schematically in Figure 4-20. The two left hand pictures show 1% suspensions of calcium bentonite and of sodium bentonite in distilled water. The calcium bentonite is aggregated and the sodium bentonite is dispersed, but both are deflocculated, as shown by the misty supernatant liquid after centrifuging. The picture on the lower right shows the calcium bentonite suspension after the addition of 0.0IN calcium chloride; the upper right-hand picture shows the sodium suspension after the addition of 0.1 N sodium chloride. Both are flocculated, as shown by the clear supernatant, but the calcium bentonite suspension is aggregated and the sodium bentonite suspension is dispersed, as shown by the much greater volume of sediment.

Although low concentrations of sodium chloride cause only flocculation, high concentrations cause aggregation as well. This was shown by some experiments in which increasing amounts of sodium chloride were added to a

chapter, which showed that the critical change in c-spacing occurred at a concentration of 300 meq/1 of sodium chloride. In addition, X-ray scattering studies have shown independent clay platelets in 0.1 N sodium chloride, and aggregates of 6 to 8 layers in 1 N solutions.13

The addition of polyvalent salts to sodium bentonite suspensions show flocculation at first, and then aggregation as the concentration increases (see Figures 4 22 and 23). Note that the critical concentrations decrease with increase in valency of the cation. The mechanism is complicated by base exchange reactions. Other studies have shown that the maximum gel strength occurs when the amount of calcium added is 60"(1 of the base exchange capacity, and the minimum is reached when 85% has been added.40

Many clays encountered in drilling are predominately calcium and magnesium clays, and hence are aggregated. When treated with thinner, both deflocculation and dispersion occur simultaneously—deflocculation because of the action of the anion, and dispersion because of the conversion of the clay to the sodium form. Dispersion is undesirable because it increases the plastic viscosity. Dispersion may be avoided by the simultaneous addition of a polyvalent salt or hydroxide with the thinner (See "Drilling Fluid Selection" in Chapter 1).

meq/liter NaCI

Figure 4-21. Flocculation and aggregation of sodium bentonite by sodium chloride. {From Dar/ey.39 Copyright 1957bySPE-AIME.)

meq/liter NaCI

Figure 4-21. Flocculation and aggregation of sodium bentonite by sodium chloride. {From Dar/ey.39 Copyright 1957bySPE-AIME.)

meq/liter CaCL

Figure 4-22. Flocculation and aggregation of sodium bentonite by calcium chloride. (From Darley.™ Copyright 1957 by SPE-AiME.)

meq/liter CaCL

Figure 4-22. Flocculation and aggregation of sodium bentonite by calcium chloride. (From Darley.™ Copyright 1957 by SPE-AiME.)

Up to this point, we have been concerned with the gelation of comparatively dilute (3%) bentonite suspensions, in which gel structures are not apparent unless sufficient salt is present to cause flocculation. Gelation does occur, however, at salt concentrations below the flocculation point, if the concentration of clay is high enough. The reason is that at high concentrations, the clay platelets are spaced so close together that their diffuse double layers interfere, and they must orient themselves in positions of minimum free energy. Thus, X-rays scattering curves showed no preferred orientation in 2% suspensions of sodium montmorillonite,13 but did show parallel orientation in 10% suspensions. As shown in Figure 4 24, the spacing increased with increase in water content.

The Mechanism of Gelation

Various linkages and plate orientations proposed to account for gel structure may be summarized as follows:

1 Cross-linking between parallel plates, through positive edge to negative surface linkages, to form a house-of-cards structure.29 4 ]

2 Edge-to-edge association, to form intersecting ribbons.42 The basis for this theory is, briefly, that because of the relatively high repulsive potential between the basal surfaces, the preferred platelet orientation will be parallel with edge-to-edge association.

3 Parallel association of plates, held together by the quasi-crystalline water between them.43

meq/liter ALC!,

Figure 4-23. Flocculation and aggregation of sodium bentonite by aluminum chloride. (From Darley.39 Copyright 1957bySPE-AIME.)

meq/liter ALC!,

Figure 4-23. Flocculation and aggregation of sodium bentonite by aluminum chloride. (From Darley.39 Copyright 1957bySPE-AIME.)

It may well be that all of these mechanisms are operative, and that their relative significance depends on such factors as clay concentration and the strength and sign of the double layer potentials on the platelet edges and surfaces.

In trying to visualize gel structures, one should bear in mind that in a crowded system of platelets up to 10,000 A wide, spaced somewhere around 300 A apart, orientation of the particles is restric ted by spatial considerations. Furthermore, the platelets are not the rigid little rectangles that we like to draw in schematic diagrams, but flexible films of di verse shapes and sizes, as shown in Figure 4-7.

In a concentrated suspension, one would expect to see local groups of platelets aligned roughly parallel under the influence of the basal repulsive

Getting Started With Solar

Getting Started With Solar

Do we really want the one thing that gives us its resources unconditionally to suffer even more than it is suffering now? Nature, is a part of our being from the earliest human days. We respect Nature and it gives us its bounty, but in the recent past greedy money hungry corporations have made us all so destructive, so wasteful.

Get My Free Ebook


Post a comment