Volume 1 Number 10 Dennis R. Dinger 1 August 2003
An Update
Please forward any issues of the E-zine, or my web address, to any ceramists or materials engineers who might be interested. If this is the first issue you've seen, you can add your name to the mailing list by clicking HERE. All back issues can be accessed from the Publications page at the web site. Questions, suggestions, and/or requests for topics to be covered in future issues of this e-zine can be sent to QuestionsandComments@DingerCeramics.com .
Paperback and downloadable copies of Rheology for Ceramists and Particle Calculations for Ceramists are available at the Books and Downloads page of the web-site. Quantity discounts are available on the paperback books. If interested, please contact me for details.
I believe the topic in this issue will be applicable to many of you. It was suggested by questions from one individual, but it has broad application within ceramics.
Immobilize the Colloids
This subject applies to the rheology of suspensions and particularly to forming operations. When colloids are immobilized, it does not mean they are imbedded in concrete or anything like that. It means they are loosely bound into the gel structure; they are not free to move with the carrier fluids; and they are not free to segregate (under the influence of the flowing carrier fluid) by particle size. Carrier fluids exert major influence over the movement of colloidal particles in deflocculated suspensions, especially when processing speeds are high and the colloids are free to move as single particles (usually this happens under deflocculating conditions). Such phenomena are particularly important in those systems in which a 10 micrometer particle is considered to be a boulder (relatively speaking), and in any system which contains significant portions of colloidal particles.
Specific Surface Areas
The colloidal fractions in ceramic bodies are important to the rheological properties and the forming behaviors of bodies because they contribute large surface areas on which the chemical additives can function. Colloidal particles are frequently among the most expensive ingredients in a body. Depending on the reason for their inclusion in a body, only a small percentage may be required.
A single cube of a material 1 cm on edge has a surface area of only 6 cm2 (0.000006m2). But if it is chopped into sufficiently fine pieces, the single, original 1 cm3 cube of powder can become billions upon billions of particles with total surface areas of 10m2 (and higher). The masses of such colloid particles are so small that they are easily influenced by the Brownian motion of carrier fluid molecules. For this reason, colloidal particles generally don't settle easily due to gravity.
The combination of small particle masses and relatively huge specific surface areas allows colloidal particles to be heavily influenced by chemical additives in suspensions and forming bodies. Most additives in ceramic systems function by altering the surface properties of the particles. When electrostatic surface charges, lubricities, and hydrophilic/hydrophobic properties of surfaces are controlled by additives, the huge surface areas of the colloidal particles exert great influence over the rheological and forming properties of ceramic bodies and slips.
To maximize the contributions of colloids to the rheological and forming properties, one must first make sure they are all free to travel independently. This brings up the subject of high intensity dispersion (HID).
High Intensity Dispersion (HID)
By now, you must all realize that I am a fan of HID. Most ceramic bodies that have not been subjected to the intense agitation levels produced by HID are simply not well mixed. When bodies are not well mixed, the various ingredients will not be homogeneously distributed throughout the ware. Unfortunately, many ceramic bodies consist of white, white, white, tan, and white particles. It is impossible for a ceramic engineer, or a ceramic technician, or anybody else, to look at a body or slip and say, "Yup, it's homogeneous!" It does not matter how well calibrated the eye ball may be. It can't be done.
Another point to consider is that many chemically prepared materials are frequently highly agglomerated. Since it is extremely difficult to mill particles from bulk powder down to 100% less than 1 micrometer, in many cases when colloids are needed, the colloidal particles are chemically prepared. This means there is a high probability that they will be highly agglomerated. Unless the agglomerates are broken apart to release all particles to travel as individuals, and the free individual particles are then homogeneously distributed throughout the body, the properties expected from the addition of those colloids to the body may not be realized.
If such agglomerates are relatively strong, HID may not be sufficiently intense to break the agglomerate bonds and release the particles to travel as individuals. In such cases, stirred ball milling may be required. In all cases, for HID to properly deagglomerate, the particles in suspension must be slightly crowded and solids contents during HID must necessarily be somewhat high. Solids contents that are too high may produce extreme dilatancy, which is bad. But the mild dilatancy produced in slightly crowded suspensions is beneficial to the success of HID. A series of tests (using a milkshake mixer equipped with an HID blade) at a variety of solids contents should help to determine the required solids content that achieves "slightly crowded" conditions. If the milkshake mixer motor begins to smoke, the solids content is too high. If the temperature of the suspension doesn't increase much above room temperature after 30-60 minutes of HID, the solids content is too low. There will be a solids content, however, that increases suspension temperatures from room temperature to about 70oC in about 30 minutes of HID. This is an indication of the proper solids content range for HID.
Particle size distributions can also be measured to determine if HID is effective. In the cases when agglomerate sizes simply don't diminish with HID, milling must be used. Experience has shown that particle sizes decrease with HID, but specific surface areas (SSA) remain generally constant. Milling breaks particles, forms new surfaces, and surface areas increase. HID, however, doesn't break particles and form new surfaces. HID only releases and exposes the internal surfaces of agglomerates to the interparticle fluid environment. As agglomerates break apart, particle sizes will decrease to represent the newly released individual particles, but the total surface areas will generally not change.
Another reminder: deagglomeration is not a chemical process. If a dispersant is added to the fluid surrounding an agglomerate, the additive will not force its way into the middle of the agglomerate and push the particles apart. The additives will simply coat the outside of the agglomerates where they may then be unavailable if and when the agglomerates break apart and the internal surfaces are exposed to the interparticle fluid environment. Mechanical dispersion energy and impacts by mixing blades, by other particles, and/or by internal pipe and pump surfaces are required to break agglomerates. After particles are knocked apart by the mechanical energy, dispersants (if present and not already tied up) can then coat the particles, impart the desired surface properties, and hold them apart or cause them to flocculate (as desired).
Viscosity drift during aging is the result of shear and mechanical dispersion forces that expose new particle surfaces to the interparticle fluid environment in the absence of appropriate additive chemicals. If suspensions are sufficiently dispersed at the time of mixing, there should be no remaining particle surfaces that have not already been exposed to the interparticle environment. In the absence of time-dependent phenomena such as partial solubilities or evaporation of carrier fluids, when all particles travel as individuals, all individual particles are homogeneously distributed throughout the body, and all individual particles are coated uniformly with the additives, the viscosities of ceramic bodies should not drift.
Gel Structures
Immobilizing the colloids refers to the tying up of particles into the body structure by the gelation process. For ceramic wares to hold their shapes, the bodies must exhibit yield stresses. Yield stresses are produced by the gelation process. So when gelation processes are sufficiently strong to pull particles together to form continuous 3-D gel structures throughout the volume of a ceramic body, the colloids will be pulled quickly into the forming structures and immobilized.
The colloids will be pulled quickly into the gel structure because the particles' masses are small and they will be heavily influenced by the attractive interparticle forces which cause the gel structures to form. This occurs most readily in flocculated and partially-flocculated suspensions and bodies.
Flocculation/Deflocculation Characteristics
Deflocculation may be acceptable during mixing, but strong deflocculation and over-deflocculation are not generally desirable in ceramic forming bodies. Yield stresses will be minimized when suspensions are deflocculated, so wares will not easily hold their shapes. In some processes like slip casting, deflocculated suspensions can be cast, but they will produce hard, thin, brittle casts. Particles in such ware may be mechanically linked, but attractive interparticle forces will not be active to help build and/or strengthen the cast structures.
Deflocculation reduces fluid viscosities, so it can be beneficial during flow, but it is not helpful after a ware is formed. Deflocculated bodies and suspensions are characterized by repulsive interparticle surface forces. Rather than allow particles to come together into flocs to then grow into large gel structures, repulsive forces associated with deflocculation hold particles apart until they are forced into close proximity by other phenomena.
In highly deflocculated suspensions and bodies, colloidal particles are NOT immobilized. They are generally free to flow wherever the carrier fluid pulls them.
Rheology in Flocculated Systems
Colloids ARE immobilized when attractive interparticle forces pull them quickly into large 3-D gel structures. Immobilized colloids are not free to move with the fluids. They are free, however, to be released from the structures during shear. Under low shear, gel structures break down into relatively large flocs that move and are sheared within the suspension or body during flow. Under high shear, gel structures break down into smaller flocs that move and are sheared within the suspension or body during flow. Under extremely high shear, gel structures may break down into individual particles, but they will be individual particles with surfaces dominated by attractive forces. Individual colloidal particles will not freely flow with the fluids in flocculated systems because they will be strongly attracted to all particles in or near their paths.
Under shear, whether in high solids forming bodies or lower solids suspensions, flocculated colloids will be free to move and reposition relative to one another under shear, but they will generally be immobilized in the sense that they will generally be tied into and traveling with flocs of particles. The shear imposed during fluid flow can easily break through gel structures and cause all particles to move relative to one another. But colloidal particles will generally always be the first particles to reform into gel structures as shear rates decrease, and they will be the last particles released from such structures as shear rates increase.
Gelation breakdown during shear is associated with shear-thinning and thixotropic rheologies. Shear breaks gel structures and interparticle attractive forces rebuild gel structures. In each system, at each level of shear, a dynamic balance exists when a steady-state viscosity is achieved. Both destruction of the structure (from shear) and rebuilding of the structure (from gelation processes) occur simultaneously.
Rheology in Deflocculated Systems
Gelation processes generally do not occur in highly deflocculated systems. Electrostatic surface charges are either highly positive or highly negative, and in each case, the repulsion due to like-charges prevents particles from coming together into gel structures.
When deflocculated systems are sheared, all particles will travel as individuals and they will stay as far apart as possible. Gel structures don't form because there are no attractive forces to cause them. As solids contents increase (for example during dewatering), particles come closer to one another but they still do not form gel structures. When solids contents are sufficiently high, particles may be forced into mechanical contact with one another, but this type of structure will be relatively brittle with little if any elasticity.
Dewatered flocculated bodies will have all particles in close proximity, but they will be attracted to one another by the interparticle attractive forces, and they will be able to shear relative to one another because they are not totally mechanically interlocked until solids contents are really high. But dewatered deflocculated bodies will be much more brittle because the only structures that will form are structures due to mechanical linkages and interferences between particles.
Since particles travel as individuals in deflocculated systems, there is no gel breakdown phenomenon to produce shear thinning behaviors. Particles flow as individuals at all shear rates and as shear increases, more and more particles will collide, which will cause viscosities to increase due to the phenomenon known as dilatancy.
Segregation of Particles by Size
When particles are not tied up in gel structures, they will move freely as individuals in deflocculated bodies and suspensions. In such systems, the smallest particles (the colloids), will be most easily influenced by the moving fluid, and they will tend to flow with the fluid. Wherever the fluid goes, the colloids will go there as well.
This may be okay during suspension flow, but during dewatering casting processes (filter pressing and slip casting, for example), the colloids will move with the water toward the mold surfaces. Depending on the relative sizes between the mold pores and the colloids, the colloids may travel with the water into the pores and block (ruin?) the molds. They may also travel with the water to the mold surface where they are then separated from the fluid. Neither case is desirable during ceramic processing. In the one case the molds can be ruined, and if they are not ruined, the molds may actually be insufficient to filter out the colloids. If this happens, colloids will be removed from the body. In the other case when the colloids are filtered by the mold surface, a dense, tightly packed layer will form at the mold surface. This will hinder all further casting because all fluid to be removed must pass through this tight layer -- and when the first layer of a cast is dense with extremely fine particles and pore channels that are even smaller diameters than the colloids, further casting will be extremely slow.
To prevent such behavior, colloidal particles must be immobilized into a gel structure by using (at a minimum) partially-flocculated bodies.
Process Speeds
As we try to process wares faster, faster, and faster and we want to use higher and higher solids contents because there is less water to remove to achieve target solids contents in the ware, we are moving towards more and more problems.
Higher solids contents usually require more deflocculation to achieve target process viscosities. More deflocculation makes suspensions and forming bodies more and more dilatant AND the onset of the dilatancy moves to lower and lower shear rates. This is not good.
Faster processing means higher shear rates and this increases the probability for dilatant behavior.
Here is a major myth which applies to slip casting operations: Bodies with higher solids contents dewater more quickly because there is less water to remove. Yes, the higher the solids content, the less water there is to remove. But the higher the solids content, the more deflocculated the system must be to achieve target viscosities. And the more deflocculated the system, the more the particles will segregate by size during dewatering, and the more likely a fine, tight layer will be produced at the surface which will seriously hinder all further dewatering.
The correct reasoning is this. The more flocculated a suspension, the faster it will dewater, cast, dry, etc. The more flocculated a suspension or forming body, the larger will be the channels within the gel structure, and the easier it will be for all fluid to pass through the cast or body. Even when solids contents are lower, it will be easier for the larger volumes of water to move through the gel channels in a flocculated cast, than the lower volumes of water to move through the dense, small pore channels in a deflocculated cast.
Also, the more flocculated a suspension or forming body, the more tightly the colloids will be bound into the gel structure. They will be effectively immobilized, and therefore, they will not be available to block or hinder further dewatering processes. But they will be free enough that shear can separate them from one floc so they can join another.
The faster a process speed, generally, the lower must be the solids content and the more flocculated must be the body for the ware to form successfully. This is generally backwards from the way many think. But it is the correct line of reasoning to speed up a process.
One other major benefit of more flocculated systems is that the more flocculated the body or suspension, the larger will be the pore channels through that body and the faster will be the casting and drying processes. If it casts quickly, it should also dry quickly.
So remember. Immobilize the colloids for excellent processing properties.
Miscellany
Short courses are completed for the year. If there is any interest in any other short courses for a particular topic, a particular month of the year, or a particular location, and whether they be courses of general interest or in-house courses for a specific company, please let me know.
Don't forget -- I'm still looking for suggested topics for future columns.
I look forward to hearing from you. See you next time. Thanks.
Thanks for signing up to receive Ceramic
Processing E-zine
To unsubscribe, click the "Remove Me" link below.
Copyright © 2003 Dennis R Dinger
103 Augusta Rd, Clemson, SC 29631 (864) 654-3155
All Rights Reserved.
All Rights Reserved.