Craig Martell on wed 30 jul 03
Hi:
One of the things that sort of pushed me over the edge with the porcelain
body thing was glaze fit and fired color but I was also looking at the zero
% absorbtion point relating to the type of potash feldspar used.
The body I use at present has 23.5 % custer spar and 27% added
quartz. When we are thinking about glaze fit from a crystalline body point
of view, this would be a logical step. Reduce the high expansion KNaO and
add some extra quartz for better fit. Problem is, that this recipe doesn't
fuse the body enough and the higher expansion crystalline body will cause
crazing in several types of glazes. I used a blue celadon with a moderate
high expansion as the control glaze. It crazed on my present body. When I
added more KNaO and lowered the quartz a bit the body was vitrified at the
Zero % level and the glaze fits. I did the freeze/ thaw tests on several
tiles and soaked them for a couple of weeks in H2O. No crazing. A good
thing to remember about porcelain is that the very low iron content of the
clays will allow the body to be fully fused without becoming weak and
brittle. Iron bearing stonewares will not take this kind of vitrification
unless they are fired in a full oxidizing atmosphere and I'm not even sure
that is OK.
Another thing I noticed with the fused bodies is the interface was more
fully developed and the glaze had a more intense fusion and color. This is
a "by eye" judgement.
Perhaps I should get to the point. I don't think Custer is a good potash
spar for porcelain. The Theoretical Seger Formula for potash spar is
basically 1.0 flux to 6.0 silica. Custer is 1.0 flux to 7.14 silica. So,
if you have a porcelain body with 25% Custer and you switch to G-200, which
is what I've done, you get more KNaO, less silica, and better
fusion. Custer is mined from a pegmatite mass that has a lot of silica and
they don't separate this out. They just grind the rock with the extra
silica. I'm not saying that Custer is an inferior spar. I'm just saying
that for some applications another spar might be better for your
purposes. G-200 is closer the the Theoretical potash spar. It has 1.0
flux to 6.3 silica. When I compare fusion buttons of Custer and G-200, I
see a difference. Custer has little or no deformation from the crucible
shape. It's glassy but very opaque and stiff. G-200 has a fair amount of
deformation and slumping and it's much more glassy and fused.
I do all this stuff for a couple of reasons. I'm interested in materials
and their behavior and I guess I enjoy doing the testing and learning. I
also have an end point in mind and I'm usually looking for results that
will improve my finished pots and get me closer to what I'm after. It
takes a little background work and late night weighing sometimes but you
most always get some useful information.
later, Craig Martell Hopewell, Oregon
Craig Martell on thu 31 jul 03
>In Body Building for Potters, one of the first steps in building a
>(stoneware) body is to add percentages of (KNaO) spar to the point
>where the lowest expansion glazes in the Glaze Test Series fit the body
>without shivering, meaning (I thought) that the thermal expansion of
>the"suite"
>of clays chosen were being lowered by the addition of spar. This does not
>square with the knowledge that (KNaO) spar is a very high expansion
>material, unless the thermal expansion of the "suite" of clays are radically
>altered to a state of low expansion, perhaps only when and if complete
>fusion occurs?
Hello David:
Adding the spar to stop shivering actually raises the expansion of the
claybody but nullifies cristobalite development thereby curing any
shivering. At this point, glaze fit is a problem until body expansion is
brought into balance with the addition of ground quartz.
So, you are looking for a good point of fusion for the claybody and also
the level of feldspar that will take the cristobalite into the body glass
and render it harmless.
It's hard to keep this stuff straight. I very often have to stop and think
long and hard and then I'm still not sure I've got it all in the right
place. At least we try.
regards, Craig Martell Hopewell, Oregon
Craig Martell on thu 31 jul 03
>Subbed Custer for G200 in the Ron/John Waterfall Brown and maturation
>went up to ^7+.
>Fit is fine on B-Mix 5-6, and more active rutile-like behaviour. Will
>add more boron frit to
>bring it back to ^6 and see if it still fits.
Hey Roger:
Boron is a low expander so it should help with glaze fit. Of course it
depends on the total frit composition.
regards, Craig Martell Hopewell, Oregon
Roger Korn on thu 31 jul 03
Craig Martell wrote:
> ...
>
> Perhaps I should get to the point. I don't think Custer is a good potash
> spar for porcelain. The Theoretical Seger Formula for potash spar is
> basically 1.0 flux to 6.0 silica. Custer is 1.0 flux to 7.14 silica.
> So,
> if you have a porcelain body with 25% Custer and you switch to G-200,
> which
> is what I've done, you get more KNaO, less silica, and better
> fusion. Custer is mined from a pegmatite mass that has a lot of
> silica and
> they don't separate this out. They just grind the rock with the extra
> silica. I'm not saying that Custer is an inferior spar. I'm just saying
> that for some applications another spar might be better for your
> purposes. G-200 is closer the the Theoretical potash spar. It has 1.0
> flux to 6.3 silica. When I compare fusion buttons of Custer and G-200, I
> see a difference. Custer has little or no deformation from the crucible
> shape. It's glassy but very opaque and stiff. G-200 has a fair
> amount of
> deformation and slumping and it's much more glassy and fused.
Subbed Custer for G200 in the Ron/John Waterfall Brown and maturation
went up to ^7+.
Fit is fine on B-Mix 5-6, and more active rutile-like behaviour. Will
add more boron frit to
bring it back to ^6 and see if it still fits.
Also, I think Custer brings more impurities (can't isolate them yet).
Just trying different
potash spars, mostly local AZ found stuff, to see what happens.
Roger
> I do all this stuff for a couple of reasons. I'm interested in materials
> and their behavior and I guess I enjoy doing the testing and learning. I
> also have an end point in mind and I'm usually looking for results that
> will improve my finished pots and get me closer to what I'm after. It
> takes a little background work and late night weighing sometimes but you
> most always get some useful information.
>
> later, Craig Martell Hopewell, Oregon
>
> ______________________________________________________________________________
>
> Send postings to clayart@lsv.ceramics.org
>
> You may look at the archives for the list or change your subscription
> settings from http://www.ceramics.org/clayart/
>
> Moderator of the list is Mel Jacobson who may be reached at
> melpots@pclink.com.
>
David Beumee on thu 31 jul 03
7/30/03 10:46:00 PM, Craig Martell wrote:
The body I use at present has 23.5 % custer spar and 27% added
>quartz. When we are thinking about glaze fit from a crystalline body point
>of view, this would be a logical step. Reduce the high expansion KNaO and
>add some extra quartz for better fit. Problem is, that this recipe doesn't
>fuse the body enough and the higher expansion crystalline body will cause
>crazing in several types of glazes.
In Body Building for Potters, one of the first steps in building a
(stoneware) body is to add percentages of (KNaO) spar to the point
where the lowest expansion glazes in the Glaze Test Series fit the body
without shivering, meaning (I thought) that the thermal expansion of the"suite"
of clays chosen were being lowered by the addition of spar. This does not
square with the knowledge that (KNaO) spar is a very high expansion
material, unless the thermal expansion of the "suite" of clays are radically
altered to a state of low expansion, perhaps only when and if complete fusion occurs?
David Beumee
>Hi:
>
>One of the things that sort of pushed me over the edge with the porcelain
>body thing was glaze fit and fired color but I was also looking at the zero
>% absorbtion point relating to the type of potash feldspar used.
>
>The body I use at present has 23.5 % custer spar and 27% added
>quartz. When we are thinking about glaze fit from a crystalline body point
>of view, this would be a logical step. Reduce the high expansion KNaO and
>add some extra quartz for better fit. Problem is, that this recipe doesn't
>fuse the body enough and the higher expansion crystalline body will cause
>crazing in several types of glazes. I used a blue celadon with a moderate
>high expansion as the control glaze. It crazed on my present body. When I
>added more KNaO and lowered the quartz a bit the body was vitrified at the
>Zero % level and the glaze fits. I did the freeze/ thaw tests on several
>tiles and soaked them for a couple of weeks in H2O. No crazing. A good
>thing to remember about porcelain is that the very low iron content of the
>clays will allow the body to be fully fused without becoming weak and
>brittle. Iron bearing stonewares will not take this kind of vitrification
>unless they are fired in a full oxidizing atmosphere and I'm not even sure
>that is OK.
>
>Another thing I noticed with the fused bodies is the interface was more
>fully developed and the glaze had a more intense fusion and color. This is
>a "by eye" judgement.
>
>Perhaps I should get to the point. I don't think Custer is a good potash
>spar for porcelain. The Theoretical Seger Formula for potash spar is
>basically 1.0 flux to 6.0 silica. Custer is 1.0 flux to 7.14 silica. So,
>if you have a porcelain body with 25% Custer and you switch to G-200, which
>is what I've done, you get more KNaO, less silica, and better
>fusion. Custer is mined from a pegmatite mass that has a lot of silica and
>they don't separate this out. They just grind the rock with the extra
>silica. I'm not saying that Custer is an inferior spar. I'm just saying
>that for some applications another spar might be better for your
>purposes. G-200 is closer the the Theoretical potash spar. It has 1.0
>flux to 6.3 silica. When I compare fusion buttons of Custer and G-200, I
>see a difference. Custer has little or no deformation from the crucible
>shape. It's glassy but very opaque and stiff. G-200 has a fair amount of
>deformation and slumping and it's much more glassy and fused.
>
>I do all this stuff for a couple of reasons. I'm interested in materials
>and their behavior and I guess I enjoy doing the testing and learning. I
>also have an end point in mind and I'm usually looking for results that
>will improve my finished pots and get me closer to what I'm after. It
>takes a little background work and late night weighing sometimes but you
>most always get some useful information.
>
>later, Craig Martell Hopewell, Oregon
>
>______________________________________________________________________________
>Send postings to clayart@lsv.ceramics.org
>
>You may look at the archives for the list or change your subscription
>settings from http://www.ceramics.org/clayart/
>
>Moderator of the list is Mel Jacobson who may be reached at melpots@pclink.com.
>
David Beumee on fri 1 aug 03
8/1/03 11:14:51 AM, Jim Murphy wrote:
>Cristobalite from quartz is seldom observed because normally quartz added to
>bodies is not fine enough for cristobalite generation.
I thank you for your extensive notes from Tichane.
Peter Sohngen's article on cristobalite, confirmed by Ron's dilatometry,
proves that added amounts of silica to a high fire clay body can
form significant amounts of cristobalite. Sohngen concludes that
the solution is to add a combination of fine and coarse quartz silica,
and though it may require some work, true 125 mesh silica is available.
As for a typical 200 mesh silica, silica 295, Ottawa branch Sil-Co-Sil,
95% will pass through a 200 mesh screen. That means all the super fines
also pass through the screen, fines that are sometimes plenty fine
enough to contribute to considerable cristobalite growth if spar content
isn't of a high enough percentage in a body. My fusion button tests
of Custer and G-200 confirm what Craig has found. It takes less G-200
to get comparable fusion in a body than using Custer, meaning that a greater
proportion of plastic materials can be included, meaning more workable clay
with no cristobalite formation.
David Beumee
Earth Alchemy Pottery
Lafayette, CO
>Hello again to all,
>
>Craig's button test result of G-200 and Custer is interesting. I wonder
>though, what part of the G-200 being "more glassy" is due to its (G-200)
>KNaO and Silica levels ?
>Perhaps, by its inherent mineralogy, G-200 is already a more glassy feldspar
>material compared to Custer.
>
>For me, when trying to understanding porcelain clay body fusion and CTE when
>looking at Feldspar and body-recipe Quartz (Silica) levels, I need to
>consider Feldspar and Silica particle size and also consider the ceramic
>change(s) caused by these crystalline and glassy materials in the clay body
>recipe and keep in mind there is CTE of (a) the overall body; and (b)the
>"glass" formed during the firing process, i.e., like the formed "glass"
>surrounding a discrete Silica particle in porcelain, and specifically, I
>mean the larger free Silica grains added as a component of the body recipe
>of which some may never go into solution during a Cone 10 firing.
>
>Like Craig, I do not mean to debate which Feldspar is "better" in a
>porcelain body recipe. It's just that there's alot more going on with
>Feldspar (crystalline or glassy) and Silica during firing.
>
>Perhaps then, for better clay body fusion, we need to focus more attention
>on the mineralogy of the Feldspar component as well as grain size of both
>Feldspar and Quartz (Silica).
>
>For this topic, I offer some notes, below, I took while reading Robert
>Tichane's book "Clay Bodies". I highly recommend this book. Perhaps, there's
>some information which will help make sense.
>
>Notes from "Clay Bodies" by Robert Tichane:
>
>Silica by itself can either be crystalline or glassy. Crystalline (high
>expansion) forms are normally either quartz or cristobalite. The glassy (low
>expansion) form of pure silica will ordinarily not be found in a ceramic
>body because the temperatures reached will not be high enough to cause
>silica glass formation (which would require about 1450 C).
>
>Cristobalite will grow via two mechanisms. In the 1st case it can develop
>just through thermal effects on quartz, but, the quartz would need to be
>very fine particle size AND high temps (1250-1350 C) must be reached. The
>2nd mechanism for cristobalite formation is from thermal decomposition when
>clay (particularly kaolin) is heated above 1000 C it will decompose and
>eventually form both mullite and cristobalite.
>
>Cristobalite from quartz is seldom observed because normally quartz added to
>bodies is not fine enough for cristobalite generation. The elimination of
>cristobalite from kaolin occurs whenever there is more than 20% feldspar in
>a ceramic body.
>
>Quartz - Inexpensive filler that has high relative strength, does not
>contain water and thus undergo no shrinkage during firing. On the negative
>side, quartz is nonplastic and has a very high coefficient of thermal
>expansion (>100), thus, it is a mismatch from an expansion standpoint, for
>the clay-like materials and the glassy materials which occur in bodies. A
>residual particle of quartz observed in a completely fired average porcelain
>body will be surrounded by cracks due to its high expansion (and
>contraction). The quartz crystals will contract much more than the other
>parts of the body and there will be circumferential cracks associated with
>large quartz crystals after firing is completed. These cracks may be
>considered a source of weakness or may be viewed as a source of strength
>because if a crack starts in a body and propagates, the route of the crack
>will be from quartz particle to quartz particle and will not usually pass
>through those crystals. In this situation quartz functions as a crack
>terminator.
>
>Quartz and Feldspar - More silica will be dissolved by the feldspathic glass
>with: fine quartz and feldspar; thorough mixing; high temperatures; long
>firing times.
>
>For common triaxial body (clay, feldspar, quartz), when fired to a temp of
>1250 C or above, certain things happen:
>
>1. The feldspar will change from crystalline material to a glassy
>material (with different CTE). This glassy feldspar, especially with
>extended periods of firing, will dissolve some of the quartz and some of the
>clay. Naturally, the finer the feldspar, the more dispersed it will be and
>the greater will be its ability to dissolve quartz and clay. Since clays are
>usually of fine particle size, there will be no potential for
>size-solubility effects.
>
>2. The quartz crystals, however, are another matter. In a body, the finer
>the quartz crystals, the more easily they will be dissolved by the feldspar
>glass. Therefore, quartz particle size and size distribution will be very
>important as far as the CTE of the body glass is concerned.
>
>3. Clay will change from kaolin to mullite and cristobalite, plus glass
>(when it is dissolved).
>
>As quartz is dissolved by feldspar glass, it contributes a negative factor
>to the expansion of the glass and body. Quartz dissolving in feldspar will
>lower the total expansion of the resulting glass. Obviously, whether quartz
>is present in a body as very fine crystals or whether it is coarsely
>crystalline will determine how much it dissolves and how much it affects the
>CTE of the feldspar glass and the total body. With longer firing times, more
>silica and clay will dissolve in the feldspar glass.
>
>In general, crystalline materials in porcelain-type bodies will usually have
>higher CTE's than glassy materials (though glassy materials will obviously
>vary widely in composition and may have wide swings in CTE also).
>
>In glasses, as distinguished from ceramic bodies, one expects the expansion
>to increase as the quantity of alkali increases, and to decrease as the
>amount of silica in the glass increases.
>
>Best wishes,
>
>Jim Murphy
>
>______________________________________________________________________________
>Send postings to clayart@lsv.ceramics.org
>
>You may look at the archives for the list or change your subscription
>settings from http://www.ceramics.org/clayart/
>
>Moderator of the list is Mel Jacobson who may be reached at melpots@pclink.com.
>
Jim Murphy on fri 1 aug 03
Hello again to all,
Craig's button test result of G-200 and Custer is interesting. I wonder
though, what part of the G-200 being "more glassy" is due to its (G-200)
KNaO and Silica levels ?
Perhaps, by its inherent mineralogy, G-200 is already a more glassy feldspar
material compared to Custer.
For me, when trying to understanding porcelain clay body fusion and CTE when
looking at Feldspar and body-recipe Quartz (Silica) levels, I need to
consider Feldspar and Silica particle size and also consider the ceramic
change(s) caused by these crystalline and glassy materials in the clay body
recipe and keep in mind there is CTE of (a) the overall body; and (b)the
"glass" formed during the firing process, i.e., like the formed "glass"
surrounding a discrete Silica particle in porcelain, and specifically, I
mean the larger free Silica grains added as a component of the body recipe
of which some may never go into solution during a Cone 10 firing.
Like Craig, I do not mean to debate which Feldspar is "better" in a
porcelain body recipe. It's just that there's alot more going on with
Feldspar (crystalline or glassy) and Silica during firing.
Perhaps then, for better clay body fusion, we need to focus more attention
on the mineralogy of the Feldspar component as well as grain size of both
Feldspar and Quartz (Silica).
For this topic, I offer some notes, below, I took while reading Robert
Tichane's book "Clay Bodies". I highly recommend this book. Perhaps, there's
some information which will help make sense.
Notes from "Clay Bodies" by Robert Tichane:
Silica by itself can either be crystalline or glassy. Crystalline (high
expansion) forms are normally either quartz or cristobalite. The glassy (low
expansion) form of pure silica will ordinarily not be found in a ceramic
body because the temperatures reached will not be high enough to cause
silica glass formation (which would require about 1450 C).
Cristobalite will grow via two mechanisms. In the 1st case it can develop
just through thermal effects on quartz, but, the quartz would need to be
very fine particle size AND high temps (1250-1350 C) must be reached. The
2nd mechanism for cristobalite formation is from thermal decomposition when
clay (particularly kaolin) is heated above 1000 C it will decompose and
eventually form both mullite and cristobalite.
Cristobalite from quartz is seldom observed because normally quartz added to
bodies is not fine enough for cristobalite generation. The elimination of
cristobalite from kaolin occurs whenever there is more than 20% feldspar in
a ceramic body.
Quartz - Inexpensive filler that has high relative strength, does not
contain water and thus undergo no shrinkage during firing. On the negative
side, quartz is nonplastic and has a very high coefficient of thermal
expansion (>100), thus, it is a mismatch from an expansion standpoint, for
the clay-like materials and the glassy materials which occur in bodies. A
residual particle of quartz observed in a completely fired average porcelain
body will be surrounded by cracks due to its high expansion (and
contraction). The quartz crystals will contract much more than the other
parts of the body and there will be circumferential cracks associated with
large quartz crystals after firing is completed. These cracks may be
considered a source of weakness or may be viewed as a source of strength
because if a crack starts in a body and propagates, the route of the crack
will be from quartz particle to quartz particle and will not usually pass
through those crystals. In this situation quartz functions as a crack
terminator.
Quartz and Feldspar - More silica will be dissolved by the feldspathic glass
with: fine quartz and feldspar; thorough mixing; high temperatures; long
firing times.
For common triaxial body (clay, feldspar, quartz), when fired to a temp of
1250 C or above, certain things happen:
1. The feldspar will change from crystalline material to a glassy
material (with different CTE). This glassy feldspar, especially with
extended periods of firing, will dissolve some of the quartz and some of the
clay. Naturally, the finer the feldspar, the more dispersed it will be and
the greater will be its ability to dissolve quartz and clay. Since clays are
usually of fine particle size, there will be no potential for
size-solubility effects.
2. The quartz crystals, however, are another matter. In a body, the finer
the quartz crystals, the more easily they will be dissolved by the feldspar
glass. Therefore, quartz particle size and size distribution will be very
important as far as the CTE of the body glass is concerned.
3. Clay will change from kaolin to mullite and cristobalite, plus glass
(when it is dissolved).
As quartz is dissolved by feldspar glass, it contributes a negative factor
to the expansion of the glass and body. Quartz dissolving in feldspar will
lower the total expansion of the resulting glass. Obviously, whether quartz
is present in a body as very fine crystals or whether it is coarsely
crystalline will determine how much it dissolves and how much it affects the
CTE of the feldspar glass and the total body. With longer firing times, more
silica and clay will dissolve in the feldspar glass.
In general, crystalline materials in porcelain-type bodies will usually have
higher CTE's than glassy materials (though glassy materials will obviously
vary widely in composition and may have wide swings in CTE also).
In glasses, as distinguished from ceramic bodies, one expects the expansion
to increase as the quantity of alkali increases, and to decrease as the
amount of silica in the glass increases.
Best wishes,
Jim Murphy
iandol on sat 2 aug 03
Dear Jim Murphy,
No doubt your assessment of the situation is true. But there are two =
points to ponder.
First. The Quartz-Crystobalite system is not about chemical changes but =
physical changes. These in essence are alterations in the angular =
dimensions between the Oxygen and the Silicon Atoms. These changes are =
not size related.
Second. Commercial minerals are mined and may be mixtures of several =
minerals, with one predominating. Manufacturers blend to get a =
consistent product for the market. This situation allows potential for =
the emergence of Eutectic Reactions during the heating cycle, such as =
that which may happen when we heat Nepheline Syenite. Were people to use =
hand picked stone they might have samples which were closer in analysis =
to the pure mineral.
Best regards,
Ivor Lewis. Redhill, South Australia
Ron Roy on sat 2 aug 03
Just a note on all this - predicting expansion of clay bodies is not
something I would indudge myself in. Crystals are unpredictable and there
are other factors which have a profound effect on expansion - like quartz
and cristobalite.
Dilatometery is the way to go - if you are to see the trends.
We can eliminate cristobalite in spar fluxed bodies - there will not be any.
RR
>The body I use at present has 23.5 % custer spar and 27% added
>>quartz. When we are thinking about glaze fit from a crystalline body point
>>of view, this would be a logical step. Reduce the high expansion KNaO and
>>add some extra quartz for better fit. Problem is, that this recipe doesn't
>>fuse the body enough and the higher expansion crystalline body will cause
>>crazing in several types of glazes.
>
> In Body Building for Potters, one of the first steps in building a
>(stoneware) body is to add percentages of (KNaO) spar to the point
>where the lowest expansion glazes in the Glaze Test Series fit the body
>without shivering, meaning (I thought) that the thermal expansion of the"suite"
>of clays chosen were being lowered by the addition of spar. This does not
>square with the knowledge that (KNaO) spar is a very high expansion
>material, unless the thermal expansion of the "suite" of clays are radically
>altered to a state of low expansion, perhaps only when and if complete
>fusion occurs?
Ron Roy
RR#4
15084 Little Lake Road
Brighton, Ontario
Canada
K0K 1H0
Phone: 613-475-9544
Fax: 613-475-3513
Jim Murphy on sat 2 aug 03
on 8/1/03 11:54 PM, David Beumee at hotpots@BOULDER.NET wrote:
> Sohngen concludes that the solution is to add a combination of fine and coarse
quartz silica,...
> It takes less G-200
> to get comparable fusion in a body than using Custer, meaning that a greater
> proportion of plastic materials can be included, meaning more workable clay
> with no cristobalite formation.
Hi David,
I understand your rationale regarding a lower recipe amount of G-200 perhaps
allowing a greater proportion of plastic materials to be used.
However, as far as overall porcelain body FUSION is concerned, I'd be very
careful about how much large and small grain-size Silica is included in the
porcelain body.
Recent research suggests it's not so much what Silica particle size goes
into the body, but rather, what's left in the fired matrix. The use of too
much large grain Silica may leave a higher percentage of residual quartz in
the fired body matrix making it more likely for pyroplastic deformation,
weaker body strength and cracking on cooling to occur.
With large grain Silica, "Quartz dissolution rims" may develop. As
temperature progresses, there's a greater chance of forming ONLY "pockets of
glass" around larger grains of Silica, rather than a more fused body made up
of a Silica-saturated glass phase without the larger residual quartz
(silica) grains left in the fired matrix.
If a less viscous Spar is used in a porcelain body, perhaps even more
consideration needs to be given to Silica grain size.
What happens to a porcelain body when the Clay (plastics) component is held
constant, while Spar and Silica grain-size substitutions are made ? Would a
less viscous Spar, like G-200, require more large grain Silica for less
pyroplastic deformation ? Would a more viscous Spar, like Custer, require
the use of only smaller Silica grains for better body glass formation ?
Something to think about.
Best wishes,
Jim Murphy
Ron Roy on sun 3 aug 03
Hi Jim,
I find the Tichane notes somewhat confusing. Let me just add some comments
that will perhaps lead to a more clear understanding of this process.
First of all - I have never seen any cristobalite in porcelain - it would
show up on a dilatometer chart - in spite of the fact that we mostly use
very fine (micro fine) silica in our bodies. This is because the KNaO in
porcelain "eats" cristobalite as it is formed - because it is of such a
small size. This happens because of the greater amounts of Spar in
porcelain bodies compared to stoneware bodies.
As an aside - that is why Sohngen says - at least 10% feldspar in stoneware
bodies and keep the micro fine out or at a minimum.
In the old days - when you bought 200M silica there was not as much micro
fine silica - due to improved (?) grinding equipment these days - there is
much more micro fine silica in most of the silica we buy.
If you want your porcelain to have a lower expansion (harder to stop
crazing) then melt more silica because melted silica has a low expansion
rate.
If you want a higher expansion (and higher contraction on cooling) then
don't melt the silica so much and/or add bigger grains of silica to get a
bigger quartz inversion at 573C.
There are some interesting microphotographs in Ceramic Masterpieces by
Kingerly and Vandiver showing how the quartz grains - tear away from the
clay body as they decrease in size during the quartz inversion.
If any of you want to do some experiments - I will do the dilatometery - on
condition that the results will be published. I don't mind helping but I
don't want to be the one to write the article.
RR
>Craig's button test result of G-200 and Custer is interesting. I wonder
>though, what part of the G-200 being "more glassy" is due to its (G-200)
>KNaO and Silica levels ?
>Perhaps, by its inherent mineralogy, G-200 is already a more glassy feldspar
>material compared to Custer.
>
>For me, when trying to understanding porcelain clay body fusion and CTE when
>looking at Feldspar and body-recipe Quartz (Silica) levels, I need to
>consider Feldspar and Silica particle size and also consider the ceramic
>change(s) caused by these crystalline and glassy materials in the clay body
>recipe and keep in mind there is CTE of (a) the overall body; and (b)the
>"glass" formed during the firing process, i.e., like the formed "glass"
>surrounding a discrete Silica particle in porcelain, and specifically, I
>mean the larger free Silica grains added as a component of the body recipe
>of which some may never go into solution during a Cone 10 firing.
>
>Like Craig, I do not mean to debate which Feldspar is "better" in a
>porcelain body recipe. It's just that there's alot more going on with
>Feldspar (crystalline or glassy) and Silica during firing.
>
>Perhaps then, for better clay body fusion, we need to focus more attention
>on the mineralogy of the Feldspar component as well as grain size of both
>Feldspar and Quartz (Silica).
>
>For this topic, I offer some notes, below, I took while reading Robert
>Tichane's book "Clay Bodies". I highly recommend this book. Perhaps, there's
>some information which will help make sense.
>
>Notes from "Clay Bodies" by Robert Tichane:
>
>Silica by itself can either be crystalline or glassy. Crystalline (high
>expansion) forms are normally either quartz or cristobalite. The glassy (low
>expansion) form of pure silica will ordinarily not be found in a ceramic
>body because the temperatures reached will not be high enough to cause
>silica glass formation (which would require about 1450 C).
>
>Cristobalite will grow via two mechanisms. In the 1st case it can develop
>just through thermal effects on quartz, but, the quartz would need to be
>very fine particle size AND high temps (1250-1350 C) must be reached. The
>2nd mechanism for cristobalite formation is from thermal decomposition when
>clay (particularly kaolin) is heated above 1000 C it will decompose and
>eventually form both mullite and cristobalite.
>
>Cristobalite from quartz is seldom observed because normally quartz added to
>bodies is not fine enough for cristobalite generation. The elimination of
>cristobalite from kaolin occurs whenever there is more than 20% feldspar in
>a ceramic body.
>
>Quartz - Inexpensive filler that has high relative strength, does not
>contain water and thus undergo no shrinkage during firing. On the negative
>side, quartz is nonplastic and has a very high coefficient of thermal
>expansion (>100), thus, it is a mismatch from an expansion standpoint, for
>the clay-like materials and the glassy materials which occur in bodies. A
>residual particle of quartz observed in a completely fired average porcelain
>body will be surrounded by cracks due to its high expansion (and
>contraction). The quartz crystals will contract much more than the other
>parts of the body and there will be circumferential cracks associated with
>large quartz crystals after firing is completed. These cracks may be
>considered a source of weakness or may be viewed as a source of strength
>because if a crack starts in a body and propagates, the route of the crack
>will be from quartz particle to quartz particle and will not usually pass
>through those crystals. In this situation quartz functions as a crack
>terminator.
>
>Quartz and Feldspar - More silica will be dissolved by the feldspathic glass
>with: fine quartz and feldspar; thorough mixing; high temperatures; long
>firing times.
>
>For common triaxial body (clay, feldspar, quartz), when fired to a temp of
>1250 C or above, certain things happen:
>
>1. The feldspar will change from crystalline material to a glassy
>material (with different CTE). This glassy feldspar, especially with
>extended periods of firing, will dissolve some of the quartz and some of the
>clay. Naturally, the finer the feldspar, the more dispersed it will be and
>the greater will be its ability to dissolve quartz and clay. Since clays are
>usually of fine particle size, there will be no potential for
>size-solubility effects.
>
>2. The quartz crystals, however, are another matter. In a body, the finer
>the quartz crystals, the more easily they will be dissolved by the feldspar
>glass. Therefore, quartz particle size and size distribution will be very
>important as far as the CTE of the body glass is concerned.
>
>3. Clay will change from kaolin to mullite and cristobalite, plus glass
>(when it is dissolved).
>
>As quartz is dissolved by feldspar glass, it contributes a negative factor
>to the expansion of the glass and body. Quartz dissolving in feldspar will
>lower the total expansion of the resulting glass. Obviously, whether quartz
>is present in a body as very fine crystals or whether it is coarsely
>crystalline will determine how much it dissolves and how much it affects the
>CTE of the feldspar glass and the total body. With longer firing times, more
>silica and clay will dissolve in the feldspar glass.
>
>In general, crystalline materials in porcelain-type bodies will usually have
>higher CTE's than glassy materials (though glassy materials will obviously
>vary widely in composition and may have wide swings in CTE also).
>
>In glasses, as distinguished from ceramic bodies, one expects the expansion
>to increase as the quantity of alkali increases, and to decrease as the
>amount of silica in the glass increases.
>
>Best wishes,
>
>Jim Murphy
>
>______________________________________________________________________________
>Send postings to clayart@lsv.ceramics.org
>
>You may look at the archives for the list or change your subscription
>settings from http://www.ceramics.org/clayart/
>
>Moderator of the list is Mel Jacobson who may be reached at melpots@pclink.com.
Ron Roy
RR#4
15084 Little Lake Road
Brighton, Ontario
Canada
K0K 1H0
Phone: 613-475-9544
Fax: 613-475-3513
David Beumee on sun 3 aug 03
8/2/03 11:56:49 AM, Jim Murphy wrote:
The use of too
>much large grain Silica may leave a higher percentage of residual quartz in
>the fired body matrix making it more likely for pyroplastic deformation,
>weaker body strength and cracking on cooling to occur.
I got off subject with my comments, which apply to stoneware bodies.
Ron has said more than once that cristobalite is not a problem with
porcelain because of the relative higher percentage of spar compared
to stoneware mixes.
It is very interesting what you say about increased slumping with the
use of larger grain size silica in porcelain bodies; a good reason to stay
with Sil-Co-Sil #90, although I would be very interested to find out if
Sil-Co-Sil #250 could be used to raise expansion for purposes of glaze
fit as a straight across substitution in a porcelain mixture.
All this discussion is a help and I am grateful for it, but I
understand why eyes glaze over when I speak of it, and why my
students simply buy what's available and hope for the best.
David Beumee
Earth Alchemy Pottery
Lafayette, CO
>on 8/1/03 11:54 PM, David Beumee at hotpots@BOULDER.NET wrote:
>
>> Sohngen concludes that the solution is to add a combination of fine and coarse
>quartz silica,...
>> It takes less G-200
>> to get comparable fusion in a body than using Custer, meaning that a greater
>> proportion of plastic materials can be included, meaning more workable clay
>> with no cristobalite formation.
>
>Hi David,
>
>I understand your rationale regarding a lower recipe amount of G-200 perhaps
>allowing a greater proportion of plastic materials to be used.
>
>However, as far as overall porcelain body FUSION is concerned, I'd be very
>careful about how much large and small grain-size Silica is included in the
>porcelain body.
>
>Recent research suggests it's not so much what Silica particle size goes
>into the body, but rather, what's left in the fired matrix. The use of too
>much large grain Silica may leave a higher percentage of residual quartz in
>the fired body matrix making it more likely for pyroplastic deformation,
>weaker body strength and cracking on cooling to occur.
>
>With large grain Silica, "Quartz dissolution rims" may develop. As
>temperature progresses, there's a greater chance of forming ONLY "pockets of
>glass" around larger grains of Silica, rather than a more fused body made up
>of a Silica-saturated glass phase without the larger residual quartz
>(silica) grains left in the fired matrix.
>
>If a less viscous Spar is used in a porcelain body, perhaps even more
>consideration needs to be given to Silica grain size.
>
>What happens to a porcelain body when the Clay (plastics) component is held
>constant, while Spar and Silica grain-size substitutions are made ? Would a
>less viscous Spar, like G-200, require more large grain Silica for less
>pyroplastic deformation ? Would a more viscous Spar, like Custer, require
>the use of only smaller Silica grains for better body glass formation ?
>
>Something to think about.
>
>Best wishes,
>
>Jim Murphy
>
>______________________________________________________________________________
>Send postings to clayart@lsv.ceramics.org
>
>You may look at the archives for the list or change your subscription
>settings from http://www.ceramics.org/clayart/
>
>Moderator of the list is Mel Jacobson who may be reached at melpots@pclink.com.
>
Jim Murphy on mon 4 aug 03
Hi Ron,
I'm with you Ron - and Tichane - there shouldn't be any cristobalite in
porcelain bodies fired to maturity with an adequate amount of spar. So, I
didn't and don't see cristobalite as an issue here.
I was focusing on better fusion of the porcelain body. G-200 was mentioned
as providing better fusion - compared to Custer - for a specific porcelain
body. Might be G-200 has a smaller particle size. Might be due to its
crystal structure. I think, though, the reason may have more to do with
G-200 being less viscous than Custer, rather than a spar chemical analysis
(KNaO/Silica) thing although I believe Craig was getting at G-200's higher
KNaO and lower Silica amount - in the porcelain body - as a reason why his
blue celadon "control glaze" crazed no more. I may have misunderstood though
if Craig also reduced the recipe amount of Silica.
G-200, by being less viscous (due to its finer particle size?) than Custer,
would flow more easily and have greater ability to dissolve quartz and clay
leading to better fusion.
Now, if I didn't misunderstand Craig's original post and assuming Craig's
sub of G-200 for Custer was the only body recipe change, i.e., no change in
Silica or Clay, then perhaps his porcelain body - with Custer - had too much
Silica to begin with, or lots of Silica did not go into solution, which is
why his Blue Celadon "control glaze" was crazing with Custer in the body.
I believe Craig used 200-mesh Silica in his porcelain body. If Craig were to
now sub in some larger grain Silica, I think his porcelain body expansion
would go back up and he'll also end up with more residual Quartz (Silica)
left in the fired matrix - less body fusion and more risk of deformation,
etc.
For a "better" fused porcelain body, I ask rhetorically, "How many residual
grains of Silica in the fired matrix are TOO many ?"
Ron, about those Kingery 573 C quartz inversion microphotographs you mention
showing how the quartz grains - tear away from the clay body as they
decrease in size during the quartz inversion. It can get worse at higher
temps.
It's my understanding, from research reported by W. Carty, mullite formation
ceases somewhere around 1150 C and as temperature increases "Quartz
dissolution rims" may form around Silica particles whereby the Quartz grain
shrinks away from the body glass & mullite needle matrix. Basically, this
leaves pockets of potentially low viscosity glass regions around discrete
grains of Silica.
Pyroplastic deformation/slumping is said to occur in these low viscosity
glass phase regions within a high viscosity matrix (clay relics and quartz
particles).
So, NO residual quartz particles in the fired matrix would be ideal as far
as pyroplastic deformation of porcelain is concerned.
Best wishes,
Jim
iandol on mon 4 aug 03
Dear Ron Roy,
<formed >>
What an image to play with!! Amoebic KayNayOs freely insinuating =
pseudopodia among and around sharp Mullite needles gobbling up little =
fragments of Cristobalite as they flee from decaying Kaolin.
Why are we so willing to accept solvation in the case of Clay bodies but =
apparently deny that it should be taken into account when speaking of =
the maturing processes which process glazes?
By the way, do you have access to Differential Thermal Analysis =
Equipment?
Best regards,
Ivor Lewis. Redhill, South Australia
Ron Roy on tue 5 aug 03
I should have explained this earlier in this discussion - It appears to me
that there is still the wide spread belief that - when the bag says 200M -
many still believe that most - or even a significant amount of that quartz
is 200M
When we tested (sieved the silica) we found about 95% passed through - not
200M but a 400M sieve - it is that silica which is the problem.
The 400 mesh silica is about the same - a little more stays on 400M maybe 1
or 2% - it is virtually the same.
If you don't test your silica - or ask for a sieve analysis - you have no
idea what is in the bag.
I encourage all of you who are interested to read Peters article - he had a
difficult time finding properly separated silica - but did - and detailed
the results when he used different sized silica in stoneware bodies.
Beware - sieving silica - take the right precautions - it is one of the
materials you should not be breathing - especially the ultra fine stuff.
RR
>8/1/03 11:14:51 AM, Jim Murphy wrote:
>
>>Cristobalite from quartz is seldom observed because normally quartz added to
>>bodies is not fine enough for cristobalite generation.
>
> I thank you for your extensive notes from Tichane.
>Peter Sohngen's article on cristobalite, confirmed by Ron's dilatometry,
>proves that added amounts of silica to a high fire clay body can
>form significant amounts of cristobalite. Sohngen concludes that
>the solution is to add a combination of fine and coarse quartz silica,
>and though it may require some work, true 125 mesh silica is available.
>As for a typical 200 mesh silica, silica 295, Ottawa branch Sil-Co-Sil,
> 95% will pass through a 200 mesh screen. That means all the super fines
>also pass through the screen, fines that are sometimes plenty fine
>enough to contribute to considerable cristobalite growth if spar content
>isn't of a high enough percentage in a body.
Ron Roy
RR#4
15084 Little Lake Road
Brighton, Ontario
Canada
K0K 1H0
Phone: 613-475-9544
Fax: 613-475-3513
Ron Roy on wed 6 aug 03
Hi Ivor,
I did not invent "eats" but thought it was a good way to describe the
process to potters - who are much better at understanding when they can see
a picture.
How about - KNaO lurking about - waiting for those tiny quartz crystals to
come close.
I accept solvation in glazes - what sintering is all about for one thing.
I have a dilatometer - if that is what you mean - otherwise I am without.
RR
><>
>
>What an image to play with!! Amoebic KayNayOs freely insinuating
>pseudopodia among and around sharp Mullite needles gobbling up little
>fragments of Cristobalite as they flee from decaying Kaolin.
>
>Why are we so willing to accept solvation in the case of Clay bodies but
>apparently deny that it should be taken into account when speaking of the
>maturing processes which process glazes?
>
>By the way, do you have access to Differential Thermal Analysis Equipment?
>
>Best regards,
>
>Ivor Lewis. Redhill, South Australia
Ron Roy
RR#4
15084 Little Lake Road
Brighton, Ontario
Canada
K0K 1H0
Phone: 613-475-9544
Fax: 613-475-3513
Ron Roy on wed 6 aug 03
Hi Jim,
I wrote about this yesterday but it needs mentioning again - because the
bag says 200M or 400 - whatever - is no indication of what is in the bag -
a particle analysis would tell you though.
The gaps around the quartz grains are described as happened during the
quartz inversion - look like cracks - not completely around the crystal -
looks like they happen at 573C
As I have repeatedly said - I have never dilatometered any porcelain that
did not have a healthy inversion hump at 573C. I suspect - if you melt
enough so there is none - what you will have is more of a glaze than a pot
- and it would be very difficult to stop glazes from crazing over it -
remember - we rely on that quartz inversion to help keep glazes in
compression.
RR
>I believe Craig used 200-mesh Silica in his porcelain body. If Craig were to
>now sub in some larger grain Silica, I think his porcelain body expansion
>would go back up and he'll also end up with more residual Quartz (Silica)
>left in the fired matrix - less body fusion and more risk of deformation,
>etc.
>
>For a "better" fused porcelain body, I ask rhetorically, "How many residual
>grains of Silica in the fired matrix are TOO many ?"
>
>Ron, about those Kingery 573 C quartz inversion microphotographs you mention
>showing how the quartz grains - tear away from the clay body as they
>decrease in size during the quartz inversion. It can get worse at higher
>temps.
>
>It's my understanding, from research reported by W. Carty, mullite formation
>ceases somewhere around 1150 C and as temperature increases "Quartz
>dissolution rims" may form around Silica particles whereby the Quartz grain
>shrinks away from the body glass & mullite needle matrix. Basically, this
>leaves pockets of potentially low viscosity glass regions around discrete
>grains of Silica.
>
>Pyroplastic deformation/slumping is said to occur in these low viscosity
>glass phase regions within a high viscosity matrix (clay relics and quartz
>particles).
>
>So, NO residual quartz particles in the fired matrix would be ideal as far
>as pyroplastic deformation of porcelain is concerned.
Ron Roy
RR#4
15084 Little Lake Road
Brighton, Ontario
Canada
K0K 1H0
Phone: 613-475-9544
Fax: 613-475-3513
| |
|
