Earl Krueger on thu 16 oct 03
A while back Ron Roy, in somewhat of an addendum to a post, made the
comment that he was surprised there weren't more glaze calculation
questions on Clayart. I thought about that. I understand the
concepts. I understand the math. So, what don't I understand?
Having studied chemistry many years ago, the one thing that has been
like a thorn in the finger was why potters use oxides in their
calculations when using just the cations (Na, K, Ca, Mg, SI, Al) would
really be simpler and perhaps be more meaningful. This led me to my
original post with this subject line. Let me expound on the subject,
but first I must say that I am NOT a trained ceramic chemist.
If I think about Calcium Oxide (CaO) as a raw material of glazes I
visualize a regular crystalline structure, not composed of CaO units
like ...CaO-CaO-CaO... but of Ca and O as separate units, like
...Ca-O-Ca-O-Ca-O... Of course I extend this into three dimensions so
that a Calcium atom is surrounded by, and linked to, multiple Oxygen
atoms. Or if you look at it a little differently, each O is surrounded
by, and linked to, multiple Ca. You cannot say that a single Ca is
bound to one and only one O as CaO might be interpreted.
The only thing the formula of CaO tells us is the ratio of Calcium to
Oxygen atoms (i.e. 1:1). The same can be said for all the other oxides
in a glaze (K2O=2:1, Na2O= 2:1, MgO=1:1, SiO2=1:2, Al2O3=2:3, etc.) A
book I have lists the theoretical composition of Potassium Feldspar as
K2O.Al2O3.6SiO2. This does mean the feldspar consists of three
different entities K2O, Al2O3 and SiO2. The formula could just as
correctly be written as KAlSi3O8. Both of these formulae result in the
same ratio of atoms.
End of Part 1. Part 2 will discuss a glaze in terms of atomic ratios.
Bothell, WA, USA
John Hesselberth on fri 17 oct 03
On Friday, October 17, 2003, at 02:28 AM, Earl Krueger wrote:
> Having studied chemistry many years ago, the one thing that has been
> like a thorn in the finger was why potters use oxides in their
> calculations when using just the cations (Na, K, Ca, Mg, SI, Al) would
> really be simpler and perhaps be more meaningful. This led me to my
> original post with this subject line. Let me expound on the subject,
> but first I must say that I am NOT a trained ceramic chemist.
I think where a person comes out on tackling a subject like this
depends where they are on the spectrum between pure theoretician and
pure experimentalist. I would submit everyone is at a different place.
For example, the pure theoretician would not rest until the theory
perfectly described reality and would develop and test ever more
complex theories. She would probably not actually use the theories to
solve problems because she would be more concerned with perfecting the
theory. Most of those folks end up in universities focusing on their
On the other hand the pure experimentalist might say 'theory be
damned--I'm just going to mix glazes until I get what I want'. Some
potters are probably in that category. They are the ones Ron and I and
others keep trying to tug a little bit toward the theoretican side of
the ledger because we think they will be able to solve problems better
and faster if they were to do so.
Most of us (including, I think, you and me for certain) are somewhere
in between, but I suspect I am a step or two closer to the
experimentalist end of the scale than you are. To exemplify, while I
can agree that your points might well be valid, I have little interest
in pursuing them myself. Why? Because there is a well developed system
(oxides) that works pretty well and I can use it today to help people
solve problems. And there are so many problems to be solved. Further,
there is a huge body of literature references from which I can learn
that are oxide-based. For example, I can find (and simultaneously see
the logic behind) how pyrometric cones were developed by Herman Seger
in the late 1800s. And that same logic is extremely useful to me today
in solving glaze problems. Could I do it better if I made the switch
to an ionic model? Perhaps, perhaps not. That would take years of work
to demonstrate and take me away from solving problems. Therefore I
leave that work to people who are farther toward the theoretician end
of the scale than I am. I'd rather be solving problems using the oxide
model today--it more than meets my needs.
But if you are close enough to the theoretician end of the scale to
want to focus on developing and proving that an ionic model is superior
to an oxide model, then I say go for it. The world needs people at all
positions on that spectrum. Just don't be surprised when there aren't
more than a tiny number of people here on the Clayart forum who are
interested in pursuing it with you. As a group, we are heavily toward
the experimentalist end of the scale.
John Hesselberth on fri 17 oct 03
Earl, another thought on the subject. I just picked up "The Chemistry
of Glasses" by A. Paul, second edition, 1990. While it may be 13 years
out of date, I found his presentation of 4 theories of glass formation
interesting. These four researchers present a fairly complex set of
"rules" (really hypotheses) of the conditions necessary to form
glasses. Just to give you the titles to give you an idea of their
different thinking there is 1) Goldschmidt's radius ratio criterion for
glass formation, 2) Zachariasen's random network hypothesis, 3)
Smekal's mixed bonding hypothesis, and 4) Sun's bond-strength criterion
for glass formation.
Paul closes out the section by saying "But none of these hypotheses are
really capable of explaining glass formation to a more satisfactory
extent than already described and will not be discussed further in this
book." I think it would be fair to describe what he has 'already
described' as a sophisticated oxide-based model--although I would
happily yield to another interpretation by someone more knowledgeable
in the technology than I.
So it may well be that there is extensive information already in the
literature regarding an ionic model. Seems like a thorough literature
search would be a profitable next step.
Hank Murrow on fri 17 oct 03
On Friday, October 17, 2003, at 01:46 PM, John Hesselberth wrote:
> Earl, another thought on the subject. I just picked up "The Chemistry
> of Glasses" by A. Paul, second edition, 1990. While it may be 13 years
> out of date, I found his presentation of 4 theories of glass formation
> interesting. So it may well be that there is extensive information
> already in the
> literature regarding an ionic model. Seems like a thorough literature
> search would be a profitable next step.
My friend David Stannard recommended Weyl's book to me:
"W.A.Weyl wrote a book, collected 1930s series of articles from Ceram.
called "Colored Glasses" (Dawsons of Pall Mall, 1959) which I found
useful for the static "network former/modifier theory" treatment. Good
imagining competing color tendencies."
David and I are big proponents of Ionic Potential as predictor of
behaviour in the melt.
Cheers, Hank in Eugene