Cyrus - 21-5-2005 at 15:48
Well this post vanished last time I tried...
Hypotheses
The rate of shrinkage upon firing decreases linearly as the percentage of kaolin decreases.
The green strength (unfired strength) and fired strength of the rods decreases linearly as the percentage of kaolin decreased.
The strongest and most suitable formulation for crucibles is composed of clay and graphite.0
Experimental Procedure
1) Powder Mixing
-Various ceramic powders were weighed out using an Ohaus dial-o-gram scale and placed in a cup. (for example 6.00 g kaolin, 4.00 g silica; 10.00 grams
total material was used for most formulations)
-The contents were stirred for several minutes, placed in another cup, combined with enough water to make the mixture very plastic but not enough to
make the mixture fluid, and then mixed using a painter’s spatula for several more minutes.
2) Extrusion
-Using the spatula, the ceramic mixture was placed inside of a syringe with a 0.476 cm inside diameter extrusion orifice.
-The slurry was then extruded onto paper towels by gently pressing down on the syringe piston as the syringe was slowly drawn back toward the body at
a low angle relative to the paper towel. The bead diameter was kept as close to 0.476 cm as possible.
-The ceramic rods (approximately 5 per formula) were dried and organized.
3) Pre-firing measurements
-The lengths of 2 rods of each formula were measured in centimeters using a standard ruler accurate to the nearest millimeter, and estimating
to the nearest tenth of a millimeter.
-Several rods of each formula were placed on the 3 point flexural strength apparatus and the pressure on the rods was increased gradually until the
rod cracked. The force required to crack each rod was recorded.
4) Firing
-each ceramic rod was labeled using a glaze composed of black iron oxide, talc, and kaolin, placed in slip cast kaolin/alumina crucibles, and
preheated in the oven to 550 deg. F. This drives off any water remaining, preventing the rods from exploding from steam.
-the crucibles were then placed in a furnace, and fired for approximately 1 hr using an oil burner on a low setting, not going above approximately 900
deg. F.
- Wood kindling was then added to the furnace as air was blown through the oil burner for approximately 1.5 hrs. The ceramic rods reached
approximately an orange to yellow heat.
5) Post-firing measurements
-The previously measured rods (M1) were measured again (M2), and the shrinkage was determined by 100*(M1- M2)/M1.
-Several remaining rods of each formula were then placed on the flexural strength apparatus, and the force required to break each was recorded.
6) Variables
-The independent variables were the percentage of aggregate or additives and the
composition of those additives.
-The dependent variables were the percentage of shrinkage of the clay rods when fired, the green or unfired strength of the rods, and the fired
strength of the rods.
-The constants were the thickness of the rods, the methods of mixing, extruding, drying, and firing the rods.
Conclusion
The first hypothesis, that as the percentage of kaolin decreased the percentage of shrinkage would also decrease linearly, was found to be
approximately correct. This is due to the behavior of ceramic compounds at different temperatures. Kaolin is composed of many small flake-like
particles; when heated to high temperatures the molecules within these particles vibrate so rapidly that they begin to diffuse across the particles,
fusing the particles together. As the temperature further increases the molecules vibrate more rapidly and the particles behave more like a liquid;
surface tension draws the particles of kaolin together, causing the ceramic to shrink as a whole. If the temperature increases even further, the
kaolin will actually shrink into a puddle and become a liquid. Shrinkage depends greatly on the mobility of molecules, and their ability to contract,
which is determined mostly by temperature. Also, once clay has been fired to a maturing temperature, it will not shrink nearly as much when fired to
that temperature again; the particles have already fused into a mostly solid mass (matured) and cannot shrink much more. This is why grog decreases
the shrinkage of kaolin. While the kaolin in kaolin/grog ceramics does shrink when fired, the grog does not shrink, reducing the total amount of
shrinkage. Other additives such as alumina and graphite also decrease the shrinkage of ceramics because they do not shrink significantly when fired.
Their molecules are bound tightly together, as indicated by very high melting points, and thus cannot fuse together and shrink as easily as kaolin
does. Finally, the larger grained powders, grog and silica, caused the ceramics to shrink less than the smaller grained powders, talc and alumina.
The second hypothesis was that the green and fired strengths of kaolin based ceramics would decrease linearly as the percentage of kaolin decreased.
On the whole, the strength did decrease as the percentage of kaolin decreased, but not in a linear fashion. The reason for the reduction of strength
is simple in green or unfired ceramics. Kaolin has the unique property that when wet and dried, the particles adhere to one another significantly.
Other powders such as alumina and silica will not adhere to one another when wet and dried. Thus, as the percentage of kaolin decreases, the
percentage of particles that actually bind to other particles also decreases, causing a reduction in strength. All formulations showed a marked
increase in strength at 60% kaolin/ 40% additive. Since no chemical processes are taking place, this increase in strength is purely mechanical;
kaolin’s mostly flat particles have the most mechanical strength when mixed with 40% of other mostly rounded particles. A variation in sizes and
shapes of particles allows the particles to interlock more effectively, making them noticeably stronger. The fired strength of kaolin based ceramics
also decreases smoothly, except in the case of kaolin/talc and kaolin/alumina. For example, 50% kaolin/ 50% talc is much stronger than would be
expected (see graph). It could be that these rods were extruded improperly and were thicker than normal. In this case, though, the green strength
would also probably be noticeably higher, which it is not. It also could be that this ratio of kaolin and talc is near a eutectic point, the ratio of
2 chemicals at which their melting point is lowest. This would also cause the ceramic to shrink more; the shrinkage graph shows that 50% kaolin/ 50%
talc shrinks more than would be expected. In the case of alumina, it is not known what caused the peak in strength at 30% kaolin/ 70% alumina. It is
probably not a eutectic, because all combinations of kaolin/alumina have very high melting points, and so could merely be an error.
The third hypothesis, that kaolin/graphite formulations would be the best, was completely incorrect. The graphite was burned away by oxygen in the
furnace, leaving the ceramics very porous and weak, the green strength was below average, the shrinkage was merely average, and graphite is one of the
harder to obtain chemicals used, making it the least practical. This research indicates that the best formulation for mechanical green strength was
determined to be 60% kaolin/ 40% additive, but the green strength of a ceramic is not as significant as its fired strength; crucibles would only be
used in their fired state. The best formulation for fired strength was determined to be 80% kaolin/ 20% alumina or talc. Fired strength, though, must
be balanced with low shrinkage. Several tests have indicated that crucibles with high rates of shrinkage will crack when fired. The best additive to
reduce shrinkage is about 70% kaolin/30% grog, and its fired strengths are not much lower than alumina. A solution that may meet all of these
requirements would be to fire pieces made of 80% kaolin/ 20% alumina, crush and powder them for use as grog, and then mix that with kaolin and alumina
in order to obtain a ceramic crucible with the formula 80% kaolin, 20% alumina, which is the strongest, but also comprised of 70% unfired kaolin and
alumina/ 30% fired kaolin and alumina grog, which would have low shrinkage.
This study had several errors, which could easily be fixed in future research. First, using only a simple syringe it was impossible to extrude rods
of the same diameter every time, causing some variation. This could be solved by using a proper clay extrusion device, and extruding square cross
section rods, instead of round cross section rods, which could be tested for flexural strength more accurately. Second, different amounts of water
were added to each formula in order to make the formula easily mixable; this may have affected the green strengths, and could be solved by using a
pipette to accurately deliver water. Third, the firing of the ceramic rods was inexact. Neither the exact length nor the exact temperature reached
was measured, and different parts of the furnace may have been at different temperatures. Using a large pottery kiln and pyrometric cones to measure
the temperature would solve this problem. Although the methods used in this experiment were not always precise, the data itself shows that there are
significant and quantifiable differences between the effects of various additives on kaolin based ceramics.
Attachment: Data and Graphs.xls (42kB)
This file has been downloaded 721 times
12AX7 - 21-5-2005 at 23:07
Good work 
But um... what the hell is 0.476cm? Sounds like 4.8mm to me, but 3/16" is a whole lot more convienient. (WTF is it about metric being used for
non-metric items, anyway?)
Don't suppose you have any leftover to fire to yellow-white heat?
Tim
Cyrus - 22-5-2005 at 12:18
I saved everything, so I can do some further tests to see which ones are fluxed at a white heat, however they are mostly too small to do another
strength test, so I will not be able to see how much the strength increases when fired again.
Yes, 0.476 cm is actually 3/16". 
And I did use a 3/16" drill bit, but
it just seemed aesthetically unacceptable to have an x cm long rod y inches thick being broken by z grams of force.
zoomer - 22-5-2005 at 20:22
Cyrus, interesting stuff, well done! 
You noted that the kaolin/talc mixture was unexpectedly strong. Is it possible that the strength resulted from talc having the same platy structure
as kaolinite (the major component of kaolin)? Both minerals are in the clay group of the phyllosilicate subclass (Strunz classification), and both
are often described as having a “fish scale” type of crystal habit. It seems a kaolin-talc “plate-plate” structure would be stronger in
several dimensions than the “plate-sphere” structure of your other mixtures.
Z
12AX7 - 22-5-2005 at 23:52
Probably a combination of that and the heavy fluxing action of the combination; temperature was hot enough to sinter and possibly fuse the eutectic
between particles (a tertiary between clay's AlSi and talc's MgSi mixtures, somewhere around cordierite). Basically, it matured more than
any.
I'd like to see what temp the alumina-containing rods melt at.
Tim
Cyrus - 24-5-2005 at 16:07
I doubt I'll get them to melt at all!
Zoomer, it could be that the plate-plate structure is stronger than the plate-sphere structure (as seen by talc's high green strength) but I
think the fired strength was influenced more by fluxing than anything else...
Incidentally, I fired some other rods that same day, and the strength of raku clay was off the scale. Literally. It was more than 510 grams, which was the most I could measure. That is because of
the various blends of clays in raku clay, some of which mature very early, some of which mature later, giving the clay a broad firing range, but also
a lower softening point.
Porcelain
MadHatter - 24-5-2005 at 20:17
This topic is interesting. My kiln will fire to Cone 10(2500 F) but the porcelain I use never
melts. It's made for high fire and is great for making crucibles that withstand high temps.
Excellent research BTW !