Water-based muds and clay chemistry

A water-based mud is not just “water with clay”. It is a system of several groups of components, and each group has its own job (types of drilling mud).

Water-based muds are divided by what happens to the clay in them. Dispersed muds are clay muds with thinners (lignosulfonates and lignites: chemicals made from wood pulp and brown coal) and a high pH. The clay in them breaks into the finest particles. Such a mud tolerates a lot of solids and high temperatures. But it also disperses the drilled clay, so it quickly gets loaded with it. Non-dispersed muds are polymer or polymer–clay muds with little bentonite. The polymers do not let the cuttings fall apart, and they are removed at the surface whole. Inhibitive muds suppress the hydration of clays (their uptake of water) with potassium and calcium ions, with salts and with special additives. In Russian-language terminology, muds are often named by their composition: clay, polymer–clay, low-clay, clay-free biopolymer, lime, gypsum, potassium chloride and salt-saturated.

To see why the same additives sometimes thicken a mud and sometimes thin it, and why some clays turn into a slurry in water and others do not, we must look inside clay minerals.

Clay minerals are layered aluminosilicates. Their crystals are built of atomic sheets of two kinds. In a tetrahedral sheet each silicon atom is surrounded by four oxygen atoms. The tetrahedra are joined by their corners into a flat net with six-sided cells, and six-sided hollows remain on the surface of the sheet. In an octahedral sheet, atoms of aluminium, magnesium or iron are surrounded by six oxygen atoms and hydroxyl ions. If the octahedra hold trivalent aluminium, two positions out of three are filled. If they hold divalent magnesium, all three are filled.

Sheets are joined into layers by shared oxygen atoms. A layer of one tetrahedral and one octahedral sheet is called 1:1. A layer with an octahedral sheet between two tetrahedral ones is called 2:1. A clay crystal is a stack of layers, like a pack of paper sheets. Between the layers there are exchangeable cations and water, or nothing. The distance at which the layer repeats is the basal spacing. It is measured by X-ray diffraction, and it identifies the mineral.

What sets smectite apart is its huge surface. When it is fully separated into layers, it exposes about 750 m² per gram: both sides of each layer about 1 nm thick. Kaolinite has about 10–20 m²/g, and illite about 100 m²/g. So a few tens of kilograms of bentonite per cubic metre of water give the mud its viscosity, its gel and a thin, dense cake.

The chain-like clays stand apart: attapulgite (palygorskite) and sepiolite. Their particles are needles, not plates. They give viscosity by catching on each other mechanically. So they work in salt water, where bentonite is helpless, and sepiolite also resists high temperatures. But they build a poor cake, because needles do not lie tightly. The clay rocks of the section often hold mixed-layer minerals: illite–smectite, in which swelling layers alternate with non-swelling ones. With depth and rising temperature, roughly in the interval of 60–120 °C, smectite slowly turns into illite and gives up the water between its layers (Abnormal formation pressures: origin, prediction and recognition). So young, shallow clays are usually the most “active”, and deep claystones are denser and more brittle.

If every position of the lattice were occupied by the “proper” ions, Si⁴⁺ in the tetrahedra and Al³⁺ in the octahedra, the layers would be electrically neutral. But when a mineral forms, some ions are replaced by ions of similar size and lower charge: Al³⁺ in place of Si⁴⁺ in the tetrahedra, Mg²⁺ or Fe²⁺ in place of Al³⁺ in the octahedra. This isomorphous substitution leaves a permanent negative charge on the layer. It is balanced by cations that sit on the surface and between the layers. In water they can go into the solution and be replaced by others. These are the exchangeable cations, and their amount is the cation exchange capacity (CEC), in milliequivalents (meq, a unit that counts electric charge) per 100 g of dry clay. It is measured, among other ways, by the methylene blue test (tests of the mud).

Where the charge sits decides how the mineral behaves. In illite the substitutions are mostly in the tetrahedra. The charge is large and close to the surface, and the potassium between the layers is held so firmly that it hardly exchanges at all. In montmorillonite, the main smectite of bentonites, the charge is smaller and hidden in the middle of the layer, in the octahedra. The cations between the layers are held weakly, exchange easily and surround themselves with water.

The second source of charge is the edges of the plates, where the lattice is broken off. Groups there behave like weak acids or bases, and their charge depends on the pH. In acid and near-neutral water the edges are positively charged, and in alkaline water they are negatively charged. This is why muds are kept alkaline. With positive edges, the plates attract each other: an edge sticks to the negative face of a neighbour. Cations are held more firmly the larger their charge and the weaker their hydration (the water shell around them). Divalent (doubly charged) Ca²⁺ and Mg²⁺ push out monovalent (singly charged) Na⁺, and potassium holds more firmly than sodium. But the exchange is reversible and depends on concentration: in a strong solution of potassium chloride, potassium pushes out calcium too.

In water, a cloud of cations, the counter-ions, gathers around a negatively charged plate. The electric attraction pulls them to the surface, and the thermal motion drags them back into the solution. Together with the charge of the surface, this cloud is called the electrical double layer. When two plates come closer and their clouds overlap, there are more ions in the gap than in the water around. Water tends to enter the gap, and the plates repel each other. The thickness of the cloud is described by the Debye length $\kappa^{-1}$: at this distance from the surface, its potential has fallen by a factor of about $e$ (2.718).

Water enters between the layers of smectite in two stages. The first is crystalline (surface) swelling. Water molecules hydrate the exchangeable cations and the surface of the layers, and they settle between the clay layers as films one, two, or three to four molecules thick. The repeat distance grows from about 1.0 to 1.9 nm. The energy of this stage is huge: to squeeze the last layers of water out of a clay, we need pressures of hundreds of megapascals. But the volume grows only to a limit. Then, if the cations between the layers are sodium and the water is fresh, osmotic swelling begins. The double layers of neighbouring surfaces repel each other, water is drawn into the gaps, and the layers move apart by tens of nanometres. In the end the crystal falls apart into separate plates about 1 nm thick and a few tenths of a micrometre across. This is how bentonite turns into a suspension.

The exchangeable cation decides whether the second stage happens:

• Sodium is monovalent and strongly hydrated, and it holds the layers weakly. In fresh water, sodium smectite swells without limit.
• Calcium and magnesium are divalent. One ion binds two neighbouring layers like a staple, and the swelling stops at the crystalline stage (repeat distance about 1.9 nm). Calcium smectite falls apart into stacks, not into separate plates.
• Potassium is weakly hydrated, and its size fits exactly into the six-sided hollows of the silicon–oxygen sheets. It “locks” the layers, as in illite, and smectite swells little. The ammonium ion acts in the same way.

In the well, the hydration of clay rocks leads to trouble. The rock takes up water, swells and loses strength. The walls cave in and fall, and caverns form. The soft cuttings of young clays stick to the bit and the bottom hole assembly and clog them (this is called bit balling). In the mud they fall apart into fines that can no longer be removed on the shale shakers (the vibrating screens). The mud gets thicker, and the plastic viscosity grows. Dense fractured claystones swell little, but the filtrate (the liquid that leaves the mud) enters them along microfractures and bedding planes and splits the rock. There is also a purely hydraulic mechanism. There is almost no filter cake on a clay, so the pressure of the mud is slowly passed on to the pores of the rock at the wall, and the wall loses its support. So the instability of clays grows with time, as long as the hole stays open.

To choose a mud for a particular clay, the clay is studied on cuttings and core. The mineral composition is found by X-ray diffraction analysis, and the cation exchange capacity by the methylene blue test. The reaction to the mud is found by a swelling test (an instrument records the linear expansion of a pressed sample in the mud) and a dispersion test (cuttings are rolled in hot mud, and we weigh how much of them stays on a sieve). By comparing candidate muds on the same clay, we choose the composition and the concentration of inhibitors.

The clay particles in a mud can be arranged in different ways, and the properties of the mud depend on this. There are two independent pairs of states:

• Aggregation and dispersion tell how many plates there are in one particle. In aggregation the plates are gathered face to face into stacks. In dispersion the stacks have fallen apart into separate plates.
• Flocculation and deflocculation tell whether the particles attract each other. In flocculation they stick edge to face and edge to edge into a loose framework (a “house of cards”). In deflocculation they repel each other and move independently.

Stacks can be flocculated or deflocculated, and so can separate plates. So the words “dispersed mud” say nothing yet about its YP.

Flocculation is caused by anything that weakens the repulsion or creates links between the particles:

• Salt squeezes the double layers. Bentonite hardly swells in water with a salinity of about one per cent or more, and bentonite that has already swelled flocculates.
• Divalent cations (Ca²⁺ from cement, gypsum and anhydrite, Mg²⁺ from sea water and magnesium salts) squeeze the layer more strongly and tie the plates together with bridges.
• A low pH makes the edges of the plates positive, and they stick to the negative faces.
• High temperature makes the particles collide more often and makes the hydration shells thinner. Clay muds thicken when heated (see the advanced section).
• Polymers with very long molecules (high-molecular-weight polymers) in a small concentration catch several particles with one molecule and pull them into flocs (loose clumps). This is used to settle out cuttings when drilling with water and no clay.

Thinners (deflocculants) are lignosulfonates, lignites, tannins, polyphosphates and low-molecular-weight polyacrylates: large anions (negatively charged particles). They sit on the positive edges of the plates, neutralise them and add a negative charge. The edges stop sticking to the faces, and the YP and the gel strength fall. Lignite and some other thinners are weak acids that really dissolve and work only in an alkaline medium. So they are added together with caustic soda. In Russian-language practice, lignite was long prepared together with alkali, as the “coal–alkali reagent” (in Russian, UShchR). Polyphosphates, when heated above about 65 °C, turn into inactive orthophosphates, and they are good only for shallow intervals.

Both pairs of states are easy to see if lime or gypsum is added to a fresh bentonite mud. The ionic strength rises, the double layers are squeezed, and first of all the clay flocculates. The viscosity and the gels jump. But calcium gradually takes the exchange positions, and the sodium clay becomes calcium clay and gathers into stacks. There are fewer particles, and the viscosity falls, often below the starting value. This transition is made on purpose when a lime mud or a gypsum mud is prepared. It is usually done with a thinner and caustic soda, so that the peak of thickening passes without trouble. The clay of such a mud is already “saturated” with calcium, so the calcium from cement or anhydrite hardly harms it. And stacks instead of separate plates let the mud carry more cuttings without thickening.

Bentonite is a clay that is 70–90% montmorillonite. It formed from volcanic ash. Natural sodium bentonites are rare, and most deposits are calcium bentonites. Calcium bentonite is activated with soda ash: the carbonate precipitates the calcium (takes it out of the water as a solid), and the clay turns into the sodium form. In Russian-language practice such a product is called modified clay powder. The quality of the clay powder is judged by its yield: the volume of mud of a given viscosity that we get from one tonne of clay. In English-language standards this is a suspension with an apparent viscosity of 15 mPa·s ($\theta_{600} = 30$), and Russian-language standards have their own criteria.

Polymers are long molecules that unfold in water and make it harder for the water to flow. In water-based muds they do four jobs. They thicken the mud, especially at low shear rates, and so hold the cuttings. They lower the filtration: they plug the pores of the cake and raise the viscosity of the filtrate. They coat the cuttings and the walls. And in another form they thin the clay or flocculate it. Their main advantage is that they give the needed properties with few solids: drilling with a low-clay polymer mud is faster than with a clay mud.

The pH of water-based muds is kept with caustic soda, NaOH (in potassium muds with caustic potash, KOH, so as not to bring in sodium), or with lime. A high pH makes the edges of clays negative, helps thinners and lignites to work, slows corrosion and keeps hydrogen sulfide in the mud as ions. A pH that is too high speeds up the hydrolysis and breakdown of some polymers and raises the use of chemicals. So the pH is kept within the limits set by the programme.

Many problems with the mud are contaminants: substances from the section or from cement that change the chemistry of the water. For most of them there is a chemical remedy, a reagent that turns the harmful ion into an insoluble precipitate. Contaminants are recognised from the analyses of the filtrate (tests of the mud).

Why soda ash against gypsum, and bicarbonate against cement? Both reagents precipitate calcium as calcium carbonate, but they act differently on the pH.

Shale inhibition is the suppression of the hydration, swelling and dispersion of clays by the mud. There are several mechanisms, and modern systems usually combine two or three of them:

• Exchange of cations. Potassium (less often ammonium) or calcium replaces sodium in the positions between the layers and does not let the layers move apart.
• Lower activity of water. Water leaves a concentrated brine for the clay only reluctantly. At a high enough salinity it can even be drawn out of the clay (see the advanced section).
• Encapsulation (coating). Long anionic polymers, first of all partially hydrolysed polyacrylamide, are adsorbed (held on the surface) on the cuttings and the wall. They slow the entry of water and the breaking up of particles.
• Plugging of micropores and fractures. Silicates, glycols above their cloud point (the temperature at which they separate out of the water as droplets), bitumen and asphalt additives and fine particles block the paths along which filtrate and pressure enter the clay.
• Building in organic cations. Amines sit between the layers of the clay and push out the water.

A mud is chosen for each interval. We compare the rocks, the pressures and the temperatures with the limits: protection of the productive formation, the requirements of logging, the environment, logistics and cost (types of mud). For a water-based mud the reasoning usually goes like this.

If no water-based mud can cope with the clays or the salt, and the environmental requirements allow it, we change to an oil-based mud (Oil-based muds). And the other way round: where cuttings with an oil base cannot be discharged and hauling them away is costly, more and more often the aim is to manage with highly inhibitive water-based muds.