Volume 2 Number 2 Dennis R. Dinger 1 December 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 .
If you are interested in purchasing paperback or downloadable copies of Rheology for Ceramists and Particle Calculations for Ceramists, they 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.
This being the holiday season, this issue will be relatively short. The topic was suggested by a reader. Keep those suggested topics coming!
Merry Christmas and Happy New Year!
The Use of Fibers in Ceramic Slurries
The Question:
When using fibers in ceramic slurries, how do size, aspect ratio, and loading effect slurry properties?
Some Stories First
Two stories apply: The first thing that came to mind was out of the dark ages of my past. A loooonnnnnngggggg time ago when I was in junior high school, my homeroom was the home economics classroom. We met there every morning; we had study halls there in the afternoon; and NO, I did not ever actually take the home ec course. This was the room in which they taught sewing. Some of the guys in my class liked to spend their time in this room dreaming about their future careers as place kickers on football teams. They would take any stray pins they could find (there were a lot of them around) and they'd prop them up on the desk with one finger (finger on top, point down, like the holder for a place kicker) and flick them up and across the room with a snap of the middle finger of their other hand. Everything (and everyone) in the room became potential goal posts (targets) for their games. Needless to say, most of these pins ended up on the floor and many were eventually found tangled in dust bunnies against the walls. A tangle of pins the size of a golf ball is a very prickly proposition to handle.
The second story that came to mind was something Jim Funk bragged about. According to him, he successfully made the first cermet when he was in grad school. It always sounded reasonable to me, so I'll repeat the story as I remember it. When he was a grad student at Alfred University, he worked in the old Air Force Lab building for Jim Tinklepaugh. He worked on a variety of projects during those years, but one that he was given had to do with the production of a ceramic/metal composite. One part was fibrous -- the other was not -- and I don't know which was which.
Others who had worked on the project had apparently been unsuccessful. The fibers would get tangled, like the pins in the first story, and that would prevent mixing and homogeneity. So they gave him the project. He decided that the problem was the aspect ratio of the fibers. He cut the fibers into relatively short lengths with a razor blade. Then he mixed them into the matrix material, formed them into the appropriate shapes, fired them, and Voilà! He had successfully made a cermet.
I'm sure I made that story a lot simpler and easier than it was at the time he was working on it, but the idea is clear. I believe this is the main trick to working with fibers in slurries -- any kinds of fibers in any kinds of slurries.
Aspect Ratios
The number that I remember is 20:1. If the fibers are much shorter than this, the benefits of the fibers will be lost. If the fibers are much longer than this, the fibers will tangle easily, mixing will be difficult, and uniformity of distribution throughout the body will be non-existent.
Each particular system of fibers in a slurry matrix will be different but I believe the 20:1 aspect ratio is a good starting point. A series of tests can determine how short the fibers can be to still provide the benefits of fiber content, how easy it is to successfully make and mix a fibrous slurry, and how well the fired properties of the composite stack up against the desired properties.
If the matrix adheres fairly well to the fibers, a lower aspect ratio may work well. If the matrix does not adhere well to the fibers, longer aspect ratios may be required, along with the added difficulty of mixing to achieve homogeneity. Some industrial burner tiles, for example, have imbedded metal fibers. Adherence of refractory to metal fibers is poor at best. In this application, randomly oriented fibers provide mechanical interlocking support like randomly oriented nails holding boards together. When the metal fibers are present, the burner refractories may crack, but the metal fibers will hold them together and prevent chunks of refractory from falling out from cracked areas. In this application, the metal fibers need to be fairly long, although their loading levels do not need to be particularly high. Mixing of such bodies, however, is difficult.
Loading Levels and Particle Size Distribution
As fiber loading levels increase, the fibers must be more highly oriented. When this occurs, the largest particle size of the matrix (DL) powder must necessarily decrease, and the remainder of the particles of matrix powders must be even smaller yet. For example, perfectly round #2 pencils stacked in a hexagonal arrangement in a box will fill more than 90vol% of the box (not counting edge effects). This shows that fibers, when well-oriented, can pack very densely. To pack the pores in such a box of pencils, however, the DL of the matrix material should be at most 1/20 of the pencil (the fiber) diameter. Smaller is better. Note that this does separate the pencils by putting slurry between them -- it merely fills pores. If such a system is to be slurried, the volume percentage of the fibers would need to be reduced. As the volume percentage of fibers decreases, the size of the largest matrix particle can gradually be increased.
A 3D stack of pencils with alternating layers oriented in X and Y directions, pierced with pencils in the Z direction can pack to about 59vol%. Again in this example, there is no space between fibers to allow any fluidity. To allow for fluidity, the volume loading must be reduced much further, and matrix powders must remain small relative to the fiber diameters.
Generally speaking, the higher the desired level of fiber loading, the finer must be the coarsest matrix size. In this case, let's consider a dense pack of 20μm fibers. If the coarsest matrix particles are 1/20 of the fiber size, the largest matrix particles must be no larger than 1μm. And it is nigh unto impossible to obtain excellent packing slurries with particles 1μm or less. There simply are not enough ultra-fine particle sizes available to produce a broad, dense packing suspension when the DL of the particles is 1μm or smaller. But in a dense loading of 20μm fibers, this would be the requirement.
There are no simple formulas to relate fiber loadings to matrix particle DL sizes, but the general trend is: the higher the fiber loading required, the finer the matrix particles must be. As fiber aspect ratios and fiber loadings decrease, matrix particle sizes can be somewhat larger.
Mixing
Even if fiber aspect ratios are 20:1, mixing properties of such slurries will be poor. Dilatancy can be expected, and if it occurs, it can create the prickly fiber tangles mentioned in the first story above. To prevent dilatancy, mixing shear rates must remain low.
There are two ways to obtain excellent mixing: (1) high intensity dispersion (HID) which produces high shear rates at relatively low solids contents, and (2) high shear stress mixing at relatively high solids contents. The distinction applies to the difference, for example, between a slurry and an extrusion body. HID works well on slurries, and the application of high shear stresses works well on extrusion bodies. When dealing with fiber containing slurries, the lower shear rates will minimize dilatant effects, while the high shear stresses will help mixing. Unfortunately, there are no simple mixers that can easily apply this second mode to slips and slurries.
Everyone who has been following these E-zines has heard about HID. So I won't say more about it here. But I have said little about the other method -- low shear rates and high shear stresses. A variety of mixers are available which can provide general homogeneity to high solids, high viscosity bodies. But the best way to achieve the final mixing in such bodies is to extrude them several times. The high shear stress, low shear rate environment as the body passes through the extrusion die will perform this. A pelletizing die, for example, can be used to perform the final mixing of an extrusion body before it is extruding through the production die. Allow the body to extrude through a pelletizing die several times prior to its final pass through the production die. We have used this method on very dilatant bodies. After the first pass through the die, the body exited a round die in the shape of a horizontal pagoda. On the 5th pass through the same die, the column was smooth and shiny with no feathered edges whatsoever. The amount of mixing (in a lab mixer) prior to the first pass through the die had little if any effect on the quality of the column. The number of passes through the die, however, had remarkable effects.
This second mode of mixing (low shear rates at high shear stresses) should work well for fiber/matrix slurries. One might pump such slurries under pressure through relatively tight orifices to do the same job. If fiber loading levels are low, HID might be possible. If the matrix slurry alone (before the fibers are added) can be subjected to HID to produce uniformity, then a final, lower intensity mixing step might be sufficient to distribute the fibers uniformly throughout the body.
The higher the aspect ratio of particles, the more they will tend to produce dilatancy. Keep this in mind. Cutting the length of the fibers in half, might be all that is required to solve a mixing and formation problem, while having little effect on properties contributed by the fibers.
Particle Packing
There are two ways to consider the packing issues in fiber-containing suspensions.
A Two-Component System
When fiber loading levels are high, the fibers essentially form the boxes into which the matrix particles must pack. This effectively produces a two-component system in which (1) fibers are suspended in a (2) matrix suspension. In such systems, all matrix particles must be considerably smaller than the fiber diameters to achieve dense packing and reasonable suspension viscosities. If fiber loadings are high and dense packing (overall) is required, the particle size distribution of the matrix particles must define a dense packing system that can pack around the fibers.
A Single-Component System
As fiber loading levels and fiber aspect ratios decrease, the fibers can be considered part of the overall particle size distribution which must pack well to produce dense packing and/or low viscosities. In this case, the particle size distribution of the whole body (fibers plus matrix particles) will determine packing potentials and rheological properties.
Conclusions
The single most important point to successfully produce fiber-containing slurries is to keep the aspect ratio of the fibers to a minimum. This will require experimentation with each particular body to determine how low is low enough.
Then, the next most important point to remember is that shear rates must remain low to achieve mixing. High stress levels may be required to force matrix particles around and between the fibers. This should be achieved at low shear rates.
The higher the volume percentage of fibers required, the more mixing problems you will have.
The higher the volume percentage of fibers required, the finer must be the matrix powders that form the slurry. If the fiber diameters and the matrix particle diameters are similar, fiber volume percentages must necessarily remain low.
When volume percentages of fibers are high, the particle size distribution of the matrix suspension alone must pack well to produce low viscosities. As volume percentages of fibers decrease, the fiber sizes can be considered to be part of the overall particle size distribution of the suspension. They will then contribute to both the packing potential and viscous properties of the system.
Miscellany
Please continue to send suggested topics for future columns. Suggestions to date have been quite good and they are greatly appreciated. Keep them coming. Thanks!
Until next time ...
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.