Clay Water System

Clay particles have a platelike morphology commonly consisting of negatively charged faces
and positively charged edges when suspended in a polar solvent, such as water. These
particles readily undergo cation exchange reactions, swelling, adsorption, and even
intercalation of organic species to alter their surface charge, chemistry, and crystal
structure. Processing of clay-based ceramics is aided by their inherently plastic nature,
which provides excellent shape-forming capabilities. The plasticity of clay suspensions
results in part from their tendency to form the “house of cards” structure shown in Fig. 1.
This aggregated particle network results in an appreciable yield stress that allows
component shape to be maintained after the applied forming pressure is removed.
Through careful control of interparticle forces, suspensions can be prepared in the dispersed,
weakly flocculated, or strongly flocculated states, as shown schematically in Fig. 2.
In the dispersed state, discrete particles that exist
in the suspension repel one another on close
approach, provided the repulsive barrier is >>kbT.
In the weakly flocculated state, particles
aggregate in a shallow secondary minimum (well
depth ≈2–20 kbT), forming isolated clusters (or
flocs) in suspension at volume fractions below
the gel point (f < fgel) or a particle network at
higher volume fractions (f ≥ fgel). In this case, an
equilibrium separation distance exists between
aggregated particles. In contrast, particles
aggregate into a deep primary minimum in the
strongly flocculated (or coagulated) state,
forming either a touching particle network or
individual clusters in suspension, depending on
their concentration.
kb Boltzmann constant
f volume fraction of colloidal particles in
suspension
fgel volume fraction of colloidal particles in
suspension at the gel point
Colloidal stability is governed by the total interparticle potential energy, Vtotal, which can be
expressed as

VvdW is the attractive potential energy due to long-range van der Waals interactions
between particles, Velect the repulsive potential energy resulting from electrostatic
interactions between like-charged particle surfaces, Vsteric the repulsive potential energy
resulting from steric interactions between particle surfaces coated with adsorbed
polymeric species, and Vstructural the potential energy resulting from the presence of
nonadsorbed species in solution that may either increase or decrease suspension
stability.

The first two terms of Eq. (1) constitute the well-known DLVO theory developed by
Derjaguin and Landau and Verwey and Overbeek. This theory, which predicts the stability
of colloidal particles suspended in polar liquids.
 (1) van der Waals Forces
Long-range forces resulting from van der Waals (vdW) interactions are present and always
attractive between like particles. VvdW exhibits a power-law distance dependence whose
strength depends on the dielectric properties of the interacting colloidal particles and
intervening medium. For spherical particles of equal size, VvdW is given by the Hamaker
expression:

and where h is the minimum separation between the particle surfaces, a the particle radius,
and A the Hamaker constant ( Can be determined either under vacuum or across water)
To account for retardation effects, use of a distance dependent Hamaker function (A(h)) is
advised when interactions of interest exceed~ 5 nm.
Long-range, attractive vdW forces between particles must be satisfied during colloidal
processing to achieve the desired degree of suspension stability. One approach is to render
this force negligible by suspending particles in an index-matched solvent. However, this
approach is of limited practical importance because of the high index of refraction of most
ceramic powders. One must therefore rely on some type of interparticle repulsion, such as
electrostatic, steric, or depletion forces, to overcome the vdW attraction, as illustrated in
Fig. 3.
(2) Electrostatic Forces
The stability of aqueous colloidal systems can be controlled by generating like-charges of
sufficient magnitude on the surfaces of suspended ceramic particles. The resulting
repulsive Velect exhibits an exponential distance dependence whose strength depends on
the surface potential induced on the interacting colloidal particles and the dielectric
properties of the intervening medium.
For spherical particles of equal size that approach oneanother under conditions of constant
potential, Velect is given by

provided ka is sufficiently large (>10). In contrast, when the double layer around each
particle is extensive (< , 5), Velect is given by

where εr is the dielectric constant of the solvent, ε0 the permittivity of vacuum, 0 the
surface potential, and 1/k the Debye–Huckel screening length. k is given by
where Ni and zi are the number density and valence of the counterions of type i, and F the
Faraday constant. Ψ0 results from the dissociation of amphoteric hydroxyl groups present
on oxide surfaces and depends on pH and indifferent electrolyte concentration. It can be
estimated from the zeta potential (ζ), which measures the electrostatic potential at, or very
near to, the beginning of the diffuse double layer. By DLVO theory, dispersions can be
rendered unstable by either increasing ionic strength or adjusting pH toward the isoelectric
point (IEP). For multicomponent ceramic systems, it may be desirable to work in a pH range
where opposite charges are induced on different colloidal phases. This approach, termed
heteroflocculation, prevents unwanted phase segregation from occurring during processing.
Electrostatically stabilized suspensions are kinetically stable systems, where the rate of
doublet formation is controlled by the stability ratio, W:

where Vmax is the maximum repulsive barrier height, k0 the rate constant for fast
irreversible flocculation (k0 = 4kbT/3η0), and k the rate constant of flocculation for the
system of interest. The stability ratio exhibits an exponential dependence on Vmax and a
linear dependence on the normalized electrostatic double-layer thickness (ka)–1, as
shown in Eq. (8).
For an aqueous suspension containing 60 vol% solids (500 nm in diameter), a characteristic
aggregation time (t) of ;0.02 s is predicted for rapid flocculation, where t = 1/Nk0 and N the
particle number density.
In the presence of a repulsive barrier, this
characteristic time is extended considerably.
This analysis assumes a system in which only
Brownian motion acts to bring particles
together. During colloidal processing, external
forces can “push” particles over the repulsive
barrier, further reducing suspension stability. In
practice, it may be difficult to effectively design
stable suspensions based only on electrostatic
stabilization. Particle solubility concerns may
limit the working pH range, whereas an
extended double-layer thickness may lead to
unacceptable drying shrinkage.
 (3) Steric Forces
Steric stabilization provides an alternate route of controlling colloidal stability that can be
used in aqueous and nonaqueous systems. In this approach, adsorbed organic molecules
(often polymeric in nature) are utilized to induce steric repulsion. To be effective, the
adsorbed layers must be of sufficient thickness and density to overcome the vdW attraction
between particles and to prevent bridging flocculation. Such species should be strongly
anchored to avoid desorption during particle collisions. The conformation of adsorbed layers
can vary dramatically, depending on solvent quality, molecular architecture, number of
anchoring groups, active surface site density, and colloid and organic concentrations in
solution. As an example, schematic illustrations of such layers adsorbed on ideal ceramic
surfaces are shown in Fig. 6 for varying molecular architectures, including homopolymers,
diblock copolymers, comblike copolymers, and functionalized short-chain dispersants.

Steric interactions occur when particles approach one another at a separation distance less
than twice the adlayer thickness (δ).

Their close approach can be divided into two domains:

1 the interpenetrational domain (δ < h < 2δ)2 and the interpenetrational-plus-
compressional domain (h < d).
When modeling homopolymer adsorption, a pseudohomopolymer model is used to
describe the mixing interactions that occur in the region δ<h <2δ, while the uniform
segment model describes the mixing and elastic interactions that occur at smaller
separations, h < δ. The pseudohomopolymer model accounts for chain conformations other
than tails (i.e., trains and loops) that are expected for such species. In the domain δ<h <2δ,
Vsteric is given by

where ρ2 is the density

and M2a the molecular
weight of the adsorbed
species
As predicted, such dispersions can be rendered unstable when solvent conditions become
poor (i.e., χ >0.5).
Reversible flocculation can be induced in nonaqueous ceramic suspensions by changing
temperature, which also decreased solvency.

Polyelectrolyte species are widely used additives that can impart electrostatic and steric
stabilization to a given colloidal dispersion. Such systems are often referred to as
electrosterically stabilized. Polyelectrolytes contain at least one type of ionizable group (e.g.,
carboxylic or sulfonic acid groups), with molecular architectures that range from
homopolymers, such as poly(acrylic acid), to block copolymers with one or more ionizable
segments. Polyelectrolyte adsorption is strongly influenced by the chemical and physical
properties of the solid surfaces and solvent medium. For example, adsorption is strongly
favored when polyelectrolyte species and the colloid surfaces of interest carry opposite
charges. At small adsorbed amounts, such species can promote flocculation either via surface
charge neutralization or bridging mechanisms. At higher adsorbed amounts, particle stability
increases because of long-range repulsive forces resulting from electrosteric interactions.
For a given system, the adsorption behavior and conformation of polyelectrolyte species can
be modulated by tailoring solvent conditions (e.g., pH and ionic strength). For anionic
polyelectrolytes, the degree of ionization (α) increases with increasing pH. Such species adopt
a compact coil configuration in solution at low pH (α 0) and adsorb in a dense layer
of large mass (Γads) and low adlayer thickness (δ), as shown in Fig. 7.
In contrast, when fully ionized (α 1), anionic polyelectrolytes adopt an open coil
configuration in solution because of intersegment repulsion. These highly charged species
would adsorb in an open layer of low Γads and high δ, also shown in Fig 7.

To accurately model colloidal

interactions in the presence of
polyelectrolyte species, assignments
of the vdW plane, the plane of
charge (σ0), and the steric
interaction length (ϑ) are of critical
importance (refer to Fig. 7).
Biggs and Healy directly measured such interactions between zirconia surfaces with
adsorbed poly(acrylic acid) (MW ~ 2000 g/mol) using atomic force microscopy (AFM). At low
pH (α0), they observed that the steric interaction length and calculated plane of charge
(estimated from the normalized force versus separation distance curves) were coincident
and occurred ;1 nm away from the bare particle surfaces. As pH increased, they observed a
dramatic increase in the steric interaction length, with almost a 10-fold increase (δ ≈ 10 nm)
at pH 9. Simultaneously, they found a more modest shift for the calculated plane of charge
away from the particle surface, which doubled to ;2 nm under the same pH conditions. As
their results illustrate, the plane of charge is often located at some intermediate distance
between the solid–polyelectrolyte and polyelectrolyte–solution interfaces. One would
expect this location to depend strongly on the polyelectrolyte architecture and solution
properties of a given system.
 (5) Depletion Forces
Depletion forces occur between large colloidal particles suspended in a solution of
nonadsorbing, smaller species (e.g., polymers, polyelectrolytes, or fine colloidal particles).
Such species, known as depletants, may promote flocculation or stabilization of primary
colloidal particles. Depletion denotes the existence of a negative depletant concentration
gradient near primary particle surfaces. The concentration of rigid depletant species
decreases at bare particle surfaces and increases to its bulk solution value at some distance
away from these surfaces. This distance, known as the depletion layer thickness, is of the
order of the depletant diameter (2adep). For uncharged depletant species, Vdep is given by

where Φ dep is the depletant

volume fraction in solution and λ
=(h – 2adep)/2adep. This expression
accounts for second-order
concentration effects and
assumes that the depletant
species can be modeled as rigid,
uniform spheres
Otherwise stable dispersions are known to undergo transitions from stable depletion
flocculation  depletion restabilization with increasing depletant volume fraction.
Destabilization occurs when such species are excluded from the interparticle gap,resulting
in an osmotic pressure difference that promotes flocculation.
Ceramic slips are dispersions. These dispersions contain finely divided solids in a liquid
medium. Solids in ceramic slips consist of silicate ceramic, oxide ceramic, and non-oxide
ceramic raw materials. The dispersant will be assumed to be water in the following
explanations.
Gravity and electrostatic forces make the solids particles in these systems agglomerate and
sediment. In order to prevent agglomeration, i.e., to stabilize the slip, it is necessary to
arrange for optimal repulsive forces to act between the solids particles. Here, the addition of
suitable deflocculants and dispersants makes it possible to exert a controlled influence.

Definition of rheological terms

The viscosity of a ceramic slip is a measure of the internal friction between the molecules of
the dispersion. The constant of proportionality of the ratio of the shear stress and the shear
rate is the shear viscosity.

External forces, e.g., the shear forces produced on stirring a slip, influence the flow
behaviour. This leads to variation in the behaviour as a function of the strength and
duration of the external forces
Newtonian flow behaviour: Here the shear stress produced by a shearing force is directly
proportional to the rate of shear, hence this is ideally viscous or Newtonian flow behaviour. In
general, the technological systems, such as ceramic slips, deviate from ideal flow behaviour.

Pseudoplastic and dilatant flow behaviour: If a ceramic slip is subjected to increasing shear
stress it exhibits pseudoplastic (structural viscous) or dilatant flow behaviour. In contrast to
Newtonian flow behaviour, the ratio of shearing force to rate of shear is not constant in
pseudoplastic or dilatant flow behaviour. A single point measurement of the viscosity is
therefore not adequate to describe the flow behaviour of ceramic slips.
 In pseudoplastic flow behaviour the viscosity decreases with increasing shear force, in the
case of dilatant flow behaviour it increases. The flow behaviour depends on the type of raw
material used, its particle size distribution and particle shape, and the addition of
deflocculants.

Thixotropic and rheopexic behaviour

By thixotropic behaviour we understand the degradation of a suspension‘s structure during

a stress phase and its complete rebuilt during a subsequent relaxation phase, i.e. the
viscosity decreases under stress and will recover during relaxation. This process is reversible.
Rheopexic substances behave in precisely the opposite manner.

The parameter called “water”

The water used in ceramic technology generally contains dissolved salts as a natural
component. These influence the dissociation equilibrium that is set up, and, hence, the ions
present in the dispersion. The dissociation equilibrium is defined by the pH as well as by the
concentration of the dissolved ions. The following dissociation equilibrium is set up in
electrolyte-free water at neutral pH:
If the water contains soluble salts, a dissociation equilibrium is set up with the interaction of
other ions, e.g. Mg2+, Ca2+, Cl-. Depending on their type and their proportion the ions
present interact with the water and with the solids particles of a dispersion.
The bivalent cations, calcium and magnesium, contained in the dissolved salts, interfere with
deflocculation. The more of these cations present the greater is the degree of hardness of the
water. Here one degree = 10 mg CaO / 1 I water. The harder the water the more complexing
agent is required or a more effective one must be selected in order to bind these
cations.
If the pH ≤ 6, these cations are bound and have no effect on the deflocculation. If there is an
excess of OH- ions, that is a pH > 7, then it is necessary to complex the calcium or magnesium
ions in order to yield optimal slip properties with high solids content.
The parameter called “Soild”
The rheological properties of the dispersions and the starting conditions for an optimal
deflocculation depend on the state of the solids.
The sedimentation tendency, the specific surface and cation exchange capacity are all
influenced by the size and density of the particles. The sedimentation tendency increases with
increasing particle size and particle density. If the specific surface increases, the energy of the
whole system also increases, so that the tendency of smaller particles to agglomerate
is greater.
The surfaces of solids are interfaces in an unsaturated bonding state. The outermost layer is
composed of 02- ions. These ions originate either in the structure of the crystal itself (silicate
ceramic and oxide ceramic raw materials) or result from taking up oxygen from the
atmosphere (non-oxide ceramic raw materials). The mechanical destruction of the solids, e.g.
during milling or under great shear stress, also produces surfaces that are unsaturated. These
free bonds, which are created on fracture, also have a powerful tendency to saturate. The
compensation of these unsaturated bonding sites occurs by hydration - the addition of H2O
molecules - in aqueous systems.

Hydration
The molecules of water act as dipoles and are attracted by ions. The smaller the ion the
larger is its effect on the water molecules and the more strongly a certain number of water
molecules will be immobilized at the ion. This also applies when we consider the anionic
surfaces of ceramic raw material particles
pH-dependent interactions
The following pH-dependent interactions between ceramic particle surfaces (Me-OH surface
groups) and water follow as a result of hydration. Me represents, for instance, Si.
 From the scheme it is evident that the resulting surface charge on the raw material particle
is dependent on the pH of the water used as dispersion medium, i.e. it depends on whether
there is an excess of H30+ ions or of OH- ions.
The case where the surface charge, that results, is negative, i.e. pH ≥ 8, is relevant for the
deflocculation of ceramic raw materials. This ideal pH range for the optimal stability of a
slip can be deduced from arguments set out in Figure 5. In the case of silicate raw
materials this occurs because of the differing surface charge of the particles as a result of
diadochic replacement of ions in the silicate lattice. Silicate ceramic raw materials have
disturbed lattices, i.e. Si4+ is replaced by, for example, Al3+ or Al3+ by, for example, Mg2+,
so that excess negative charges occur at the edges of clay mineral platelets.
The variation in charge, which is also represented in Figure 5, results in a variation in the
spatial configuration of the particles in the water-solids system.
It can be seen from the figure that the spatial structure existing in alkaline medium is most
favourable for the long-term stability of a slip with a high solids content
Formation of the electrical double layer
An electrical double layer, with corresponding zeta potential (red curve), will form in the
water-solids-system, depending on the surface charge of the solids particles. This is shown in
Figure 6.
It is not only the resulting surface charge of the particles which is decisive for the
arrangement of the particles in the dispersion, it is also the charge density. Figure 7
schematically illustrates the differing thicknesses of the electrical double layer at varying
charge densities
The charge density is a function of the structure of the solids. The charge that results depends
on the degree of irregularity in the lattice structure of the raw material. The higher the charge
density the higher the thickness of the electrical double layer. Natural raw materials, as used
in silicate ceramics, have highly irregular lattice structures and, hence, a high charge density
on account of the impurities they contain. Therefore, there should be a large repulsion
potential.
On the other hand, the coating of the particle surfaces with bivalent cations in the water-
solids-system leads to the formation of a relatively thin electrical double layer, since these
cannot completely shield the negative surfaces of the clay minerals. Agglomerates of various
degrees of stability occur as a result of the effects of attractive forces.
The electrical double layer is also affected by shearing forces when the slip is stirred. It can
be assumed that this stress shears off a part of the diffuse layer. When the variation of the
zeta potential within the electrical double layer is considered, it becomes evident that
differing zeta potentials will occur depending on the depth to which this shearing occurs (see
Fig. 8). These varying potentials also necessarily require different saturations in order to set
up adequate repulsion forces between the raw material particles.
The water-solids-deflocculant-system
As described, the agglomeration, which interferes with high solids content and optimal
rheological function, is brought about by attractive electrostatic forces. The purpose of
deflocculants is to increase the distance between the raw material particles and, hence, to
inhibit agglomeration by influencing these electrostatic forces. Since the strength of these
attractive forces is influenced by the characteristic properties of water and solids described
above, it is necessary to add varying quantities and to combine the mechanisms of action of
the deflocculants, in order to bring about optimal deflocculation for every composition.

Mechanisms of action in the water-solids-deflocculant system

There are basically three mechanisms which act individually or in combination
depending on the deflocculant used:
• cation exchange and, thus, an effect on the thickness of the electrical
double layer of the raw material particles
• steric repulsion by the introduction of functional groups, which act as
spacers between the raw material particles
• binding of interfering cations by complexing
 Electrostatic repulsion by cation exchange

In the two substance water-solids-system the double layers are thin as they take up bivalent
cations from the mixing water. In order to enlarge the electrical double layer and, hence, to
increase the repulsion effect, it is necessary to replace the bivalent cations with monovalent
ones.
The occupation with bivalent cations (calcium, magnesium) is illustrated on the left-hand side
of the diagram. The negative surface charge of the clay minerals cannot be completely
screened off here. The attractive forces predominate at this moment.
An exchange of the multivalent cations against monovalent ones (e.g. sodium) leads to better
shielding and to an increase in the double layer as a result of the then externally neutral
behaviour of the clay mineral particles and the primarily repulsive forces associated with this.
Agglomeration of the particles is prevented and viscosity is reduced.

Steric repulsion
A steric repulsion is brought about by the use of dispersants with functional groups. These
functional groups act as spacers between the raw material particles.
This process is illustrated schematically in Figure 10.
On the left-hand side of the diagram the particles in the aqueous dispersion have sedimented on account
of gravity. The deposition of the dispersant on the surface of the raw material particles brings about a
repulsive effect that leads to a uniform distribution of the fine particles in the dispersion in which all
particles are ideally having the same distance from the nearest neighbouring particles. In the enlarged
cross section of the dispersion the effect of the deflocculant, a polymer with functional groups, is
illustrated schematically. The functional groups interact with the particle surface by charge compensation.
Here the positively charged "anchor" group of the additive attaches itself to negatively charged clay
particle surfaces. Since the polymer chains of the additive contain functional groups with positive and
negative charge, there are controlled attractive and repulsive effects. Hence, there is a targeted
compensation of agglomeration tendencies.
 Complexing
The presence of bivalent cations, from salts dissolved in the water, electrical double layers has
thin thickness these cations interfere with deflocculation. Exchanging them for monovalent
cations, is a means of having an effect here. Another way is to add inorganic complexing
agents, which bind free bivalent cations before they can become attached to the surface of
the raw material particle. The complexing of these free multivalent cations (e.g. Ca2+) is
illustrated schematically in Figure 11.

Complexing agents contain functional groups (e.g. amine groups), that have free valences in
their electronic shells. These valences are available for formation of electron pair bonds with
free multivalent cations. This reduces or eliminates the influence of these cations on the
electrical double layers of the clay minerals.
 Solids content of the slip and content of deflocculant
The mutual influence of the electrical attraction potentials of neighbouring particles must be
taken into account, particularly at high solids concentrations. Unsuitable deflocculant content
can lead to superimposition of the zeta potentials, thus preventing maximum repulsion of
neighbouring particles. The consequence can be undesired agglomeration.
On account of a low
density, the distance
between them is so large
that the zeta potentials
do not overlap.

If the density is greater,

there is superimposition of
the zeta potentials of
neighbouring raw material
particles.
At optimum electrolyte concentration, the zeta potentials of the neighbouring particles set up
an ideal equilibrium between attractive and repulsive forces. This leads to optimum repulsion
between the particles.
If more than the optimal quantity of deflocculant is added, the diffuse layer becomes thinner
with increasing electrolyte concentration. At too high electrolyte concentrations the attractive
forces predominate once again, the tendency to agglomerate increases because the diffuse
layers are too thin.
On account of the smaller distances between raw material particles, the agglomeration
tendency of slips, which have a too high content of deflocculant, increases with increasing
density.
Temperature of the slip
High slip temperatures lead to increased water evaporation. The loss of water necessarily
affects the density. The equilibrium between attractive and repulsive forces in the water-
solids-deflocculant-system is changed and, hence, the viscosity of the slip as well.

Slip ageing
With silicate ceramic bodies, an extended standing time leads to a better breakdown of the
clay mineral particles. An increase in the viscosity would be the result.
Changes can also take place in the water-solids-deflocculant-system during a prolonged
standing time. Since more time is available for cation exchange it is possible that the
viscosity may drop.
Natural clay, where the predominant exchangeable cation is calcium, are normally
flocculated. Salts of polyvalent cations are more effective flocculants than those of
monovalent cations. Thus, the deflocculation of a natural clay may be inhibited by the
presence of soluble salts, especially sulphates of calcium, magnesium or iron, Addition of
barium carbonate may effectively remove the sulphates:
 Plasticity of Clays

High plasticity : a high yield value and a small degree of “bottom curvature”
Montmorillonite > Ball clay > Fireclay > China clay
 Theory of Plasticity

Clay particles in plastic mass carry the normal negative charges and surrounding them is a
film of adsorbed water molecules, bonded by electrostatic forces.
• “bound” water present in the films
• “free” water in excess of that required to form films
• thickness of water film depends on the surface density of charge and on the
exchangeable cations present.
• The amount of bound water is greater, the greater the surface area on which it can be
adsorbed.
• small particle size high plasticity
• thin platy or fibrous particles high plasticity
Effect of exchangeable ions on the flow curve

• Ca-clay: strongly agglomerated

and packing is random and very
loose has a high yield value
and require a high water content
to flow.
• Na-clay: denser, low yield value
and required very little water to
flow.
 Suspension Rheology

Rheological measurements monitor changes in flow behavior in response to an applied stress

(or strain). Suspension, structure, and, hence, stability can be inferred from the observed
behavior. The critical parameters of interest include the apparent viscosity (h), the yield stress
under shear (ty) and compression (Py), and the viscoelastic properties (i.e., the loss
(G’) and elastic (G”) moduli) of the system. Such parameters must be tailored for the specific
forming method used during ceramics processing.

Most emerging forming routes require concentrated colloidal systems, with f (volume
fraction of colloidal particles in suspension) approaching 0.6.
Various types of flow behavior can be observed under steady shear depending on suspension
composition and stability, as shown in Fig. 8. Newtonian behavior is the simplest flow
response, where viscosity is independent of shear rate (see curve (a) in Fig. 8).
Pseudoplastic or shear-thinning behavior occurs when the viscosity decreases with shear rate
(see curve (b)). This response can be accompanied by a yield stress whose magnitude
depends on the strength of the particle network (see curves (d) and (e)). If the flow
curve is linear above ty, the system is referred to as Bingham plastic (curve (d)). Finally,
dilatent or shear-thickening behavior occurs when the viscosity increases with shear rate (see
curve (c)).
 Viscoelastic Behavior:

Concentrated colloidal suspension commonly display viscoelastic behavior, which can be

characterized by dynamic rheological measurements.

The complex shear modulus (G*) has a real and an imaginary component
frequency as v 0
 Compressive Flow Behavior

The stress acting on any position z in the network is given

 Effect of Interparticle Forces on Flow Behavior

(A) Hard-Sphere Systems: Hard-sphere colloidal suspensions do not experience interparticle

interactions until the point of contact (h =0), when the interaction is infinitely repulsive
 The relative viscosity (hrel = h/h0) of hard-sphere systems has been shown to scale with
the Peclet number (Pe) which is defined as
kb Boltzmann constant

a = particle radius

where h0 is the solution viscosity. Pe characterizes the relative importance of viscous and
Brownian contributions. Typically, viscous forces begin to dominate when Pe is
approximately unity, which correlates with the onset of shear-thinning behavior.

Newtonian flow at low solids loading (f< 0.3). When f> 0.3, low- and high-shear Newtonian
plateaus separated by a region of shear thinning are observed. Shear thinning occurs because
of the hydrodynamic interactions of rotating doublets, which eventually break up, thereby
reducing viscosity. Shear thickening also has been observed in such systems at elevated
volume fractions. This has been attributed to either an order– disorder structural transition
or cluster formation. Finally, as f = 0.6, a yield stress has been observed that results from the
structural disruption accompanying particle movement in dense suspensions.

For dilute systems, the Einstein relationship describes the relative viscosity dependence on
colloid volume fraction f:
At higher concentrations (f >05), hydrodynamic and Brownian many-body interactions affect
rheological behavior. In this regime, the Krieger–Dougherty model can be used to describe
this dependence
(B) Soft-Sphere Systems: Most ceramic suspensions can be classified as soft-sphere systems;
i.e., repulsive interactions occur some characteristic distance away from the particle surface.
For electrostatically and sterically stabilized systems, this distance is given by the electric
double-layer and adlayer thickness, respectively. Depending on the strength of these
interactions relative to vdW forces, such systems may exhibit dispersed to weakly attractive
behavior. Stable, soft-sphere systems have been shown to display flow behavior similar to
that described above for hard-sphere systems. Hard-sphere scaling has been successfully
applied in systems of particles with relatively thin layers (i.e., low d/a ratios), provided the
adsorbed layer has been accounted for in an effective volume fraction (feff). F eff is
enhanced by the volume occupied by the soft layer around each particle, according to

d adsorbed layer thickness

for irregularly shaped particles

where n = 2.5–3.
 Similarly, the shear moduli also exhibit a power-law dependence:

In flocculated systems, stress is transmitted through a spacefilling, aggregated particle

network formed when the colloid volume fraction exceeds the gel point.

This stress, defined as the compressive yield stress (Py(f)), increases rapidly with f

compressive yield stress of flocculated systems

b = proportionality constant

where the gel point varies inversely with the strength of the interparticle attractions
 FLOW PROPERTIES
The flow properties of kaolinite-water systems must be divided into two groups: the pastes,
and the slips. The intermediate range is so sticky that at present we have no satisfactory way
of measuring its properties.
Pastes
This region goes all the way from dry powder to a mass so wet that it will just support itself
under gravity forces. The flow properties may be measured by twisting a bar in torsion, of
deforming it in shear.
Slips
As this consistency has such a low yield point that it cannot resist gravity forces, the
specimen must be held in a container. Usually, the test is carried out in a concentric cup
viscometer, where the rate of shear may be varied and the yield point determined under
static conditions.
PASTES
The flow properties of pastes may be varied by changing the water content, the particle size,
the rate of deformation, the surface tension of the water, and the adsorbed ions.
Water Content
Increasing water content, as everyone knows, decreases the yield point and increases the
maximum extension. This is shown clearly in Figure 1. If the product of the yield point and
the maximum extension is plotted against water content, a maximum will occur where the
mass actually feels as if in the most workable condition. (Figure 2).
 In general, the finer fractions give
greater workability and higher
yield points.

high deformation rates give

increased yield points and
greater extensions.
In other words, clay pastes are
more workable when rapidly
deformed.
When a clay paste is deformed,
there is at once a turning of the
clay particles closely into
alignment along a plane parallel
to the direction of shear.

The flow properties of pastes are greatly influenced by the adsorbed ions on the clay.
Everyone is familiar with the effect of deflocculents which supply mono-valent cations and
OH ions for adsorption. With a pure hydrogen clay, little change in flow properties occurs
until all of the ions are exchanged, and there are free OH ions available. Then the properties
change with great suddenness. For this reason, a paste will change into a slip with only a
small addition of deflocculent
The surface tension of water, which normally has a value of 72 dynes-cm., may be depressed
by small amounts of non-ionic wetting agents to around 16 dynes-cm. This lowering of the
surface tension causes a corresponding lowering of the workability, as in Figure 3. With the
lower surface tension liquids, the workability is so low that it is very difficult to form even the
simpler specimens.
 Theory of Plasticity Stretched membrane theory (Norton, 1948).

This theory may be illustrated by the two kaolinite plates surrounded by water films shown in
Figure 4. The particles are held together by the surface tension forces in the surrounding
water surface and held apart by an equal repulsive force, caused perhaps by like charges on
the two plates (electro-kinetic).
The fact that the kaolinite plates are
thin and flat allows one to slide over
the next a small amount without
breaking the water envelope, thus
giving extensibility to the mass. On the
other hand, should the extension
become too great, the envelope is
broken and the piece ruptures.
Liquid limit: The water content at which the soil changes from the liquid state to plastic state
is known as liquid limit. In other words, the liquid limit is the water content at which the soil
ceases to be liquid.
Plastic Limit: The water content at which the soil becomes semi-solid is known as plastic limit.
Shrinkage Limit: The water content at which the soil changes from semi-solid to solid state is
known as the shrinkage limit.
Adsorbed water layer is considered as an integral part of clay particle. As shown in Figure
4.2, when clay particles contain enough water, adsorbed water layers are not at all in contact
with each other, and thus, there is no frictional resistance. It is at a liquid stage (lean soup).
Now, if water is removed to a certain level at which all the adsorbed water layers are just in
contact, frictional resistance will be developed at the contact points. This is considered to be
the stage of LL. When it is further dried, overlapping of adsorbed water layer will take place.
The limiting stage of this overlapping is the level at which all particles themselves touch each
other and no further overlapping possible. This stage is considered as the SL (cheese). PL
may have some degree of overlapping of adsorbed water layers (soft butter).
Shear thinning can occur for
many reasons, e.g.,:
• alignment of rod-like particles
in the flow direction;
• loss of junctions in polymer
solutions;
• rearrangement of
microstructure in suspension
and emulsion flow and
• breakdown of flocs.
Shear thinning can occur for many reasons, e.g.,:
• alignment of rod-like particles in the flow direction;
• loss of junctions in polymer solutions;
• rearrangement of microstructure in suspension and emulsion flow and
• breakdown of flocs.
 SLIPS
This consistency comprises suspensions of kaolinite in water with a low enough yield point to
flow under gravity forces. The flow properties are influenced by the concentration of solids,
the particle size, and the adsorbed ions. Suspensions of kaolinite particles are not truly
viscous but have an apparent viscosity which varies with the rate of shear. In other words,
they are thixotropic in nature.
slip control
Viscometer consists of a wire, clamped
and suspended from its top end, with a
metal cylinder attached to the lower end
which is free to rotate with the wire (see
Figure ). When the cylinder is turned
through 360 degrees and then allowed to
unwind in a container of casting slip, the
extent of its rotation is governed by the
viscosity of the slip. A flywheel attached to
the wire above the cylinder gives a more
stable rotation. Hence the fluidity of the
slip can be readily expressed in terms of
degrees of rotation.
Before taking a measurement of fluidity, the
viscometer cylinder must have been wound through
360 degrees in a clockwise direction. With the
viscometer beaker in position under the viscometer,
the flywheel pin is released hereby enabling the
flywheel to rotate in an anticlockwise direction. The
overswing indicated by the pointer on the
flywheelagainst the graduated scale is recorded as a
measurement of fluidity
 measurement of thixotropy
Thixotropy, or the tendency of an undisturbed slip to thicken with time, is an important
property which has a marked effect on casting performance. It is measured by rewinding the
viscometer flywheel immediately after taking the fluidity reading, allowing the slip to stand
undisturbed for either 60 or 90 seconds, and then re-measuring the overswing. Thixotropy is
defined as the difference between the two readings. In some instances, a 5 or 6 minute
thixotropy reading is also useful. It is important that the temperature of the above
measurement is recorded as temperature can severely affect viscosity values.
As a guide, most casting slips perform satisfactorily when set to a fluidity of between 290 -
310 degrees overswing and a one-minute thixotropy of between 20-50 degrees, at a density
of between 1.800 to 1.825 kg/l. To achieve a fluid casting slip, at a suitable solids loading, the
clays have to be deflocculated. When possible, it is advisable to use a mixture of sodium
silicate and sodium carbonate, as this system is not too difficult to adjust. In some cases
when the clays used are particularly difficult to deflocculate, other reagents may be
necessary; for example, ‘Dispex’, a polyacrylate, and ‘Dolaflux’, a mixture of sodium humate
and sodium silicate, have proven successful. The use of phosphates should be avoided in
ceramic casting slips as these may give problems with poor ageing characteristics and can
also give rise to mould damage.
The casting slip control system is most easily understood through the construction of a graph
which illustrates how fluid properties vary with the addition of deflocculant. When sodium
silicate or sodium carbonate is added to casting slip the fluidity increases and the thixotropy
decreases. If the deflocculant is added to excess then a point will be reached where fluidity
decreases and thixotropy increases. This generally serves as an indication of over
deflocculation of the system.
Figure 7 illustrates typical deflocculation curves for a casting slip using in one case, sodium
silicate and in the other, sodium carbonate. Sodium silicate tends to give high fluidity slips
with low thixotropy, while sodium carbonate will give lower fluidity slips with higher
thixotropy. It is normal practice to use these two reagents together as this offers the
possibility of obtaining high fluidity slips with an appropriate degree of thixotropy.
The reagents are mixed in a ratio which will give the most effective control over fluid
properties for the given casting slip system. This ratio may vary from 1:1 silicate:carbonate
up to 4:1, but 3:1 is usually considered to be normal. Processed clays and clay bodies have
been developed that include deflocculant in the formulation so that only small additions of
sodium silicate are required to the slip prior to casting.
 the control of casting slips
When a silicate/carbonate mixture is used, a progressive addition of deflocculant will
increase fluidity to a maximum and further additions will have little effect on the level of
fluidity. Thixotropy, however, will decrease steadily as deflocculant is added and will
continue to do so even after fluidity has levelled off. In this way thixotropy can beadjusted
while fluidity is maintained. If the maximum value of fluidity is too low, then the density of
the slip may be reduced by the addition of water. This will have a marked effect on fluidity,
but only a slight effect on thixotropy. The density of the slip will not need to be drastically
reduced as a small decrease in density results in a large increase in fluidity. The aim of slip
control is to keep the density as high as possible (say 1.800 kg/l for tableware and 1.825
kg/l for sanitaryware) whilst maintaining acceptable fluid properties for effective casting.
casting faults