search  current discussion  categories  glazes - faults 

bubbles (long)

updated wed 4 mar 98

 

Gavin Stairs on tue 3 mar 98

I'd like to take a somewhat more careful look at the bubble question.

First, bubbles are little pockets of gas in a liquid or solid. If the gas
pockets are sufficiently numerous, they may coalesce to produce a foam,
which is pockets of gas contiguously distributed through a liquid or solid.
Isolated bubbles are usually round because the surface tension of the
liquid pulls any shape to round. Foam bubbles have curved or flat facets
where bubbles contact each other. How bubbles pack into a foam, and the
rules governing their contact faces is an interesting but slightly
irrelevant topic.

In glazes, bubbles may form in one of four ways:
1. as trapped gas from voids in the unmelted glaze, which get incorporated
into the glaze melt as it develops.
2. as gas escaping from the underlying body as the glaze melts and forms.
3. as volatile fractions present in the immature glaze, such as bound
water, lignite, carbonate, sulfate, etc.
4. as dissolved gas coming out of solution.

Gas expands when heated. The expansion is linear with respect to an
absolute temperature scale. The usual scale is Kelvin, which is like
Celsius or centigrade, only shifted by 273deg. So room temperature is
about 300K. That means the low cones are about 3 times hotter, and cone 10
is approximately 4 times hotter. So ordinary trapped gas in the pores of
bisque and green glaze will expand to about 3-4 times original volume
during heating in the kiln.

If a glaze is made of, say, 10% volatiles, in calcium carbonate, ball clay
lignite, bound water in kaolin or clay or bentonite, and as various
fractions of colorants, many of which are carbonates or hydrates, what
becomes of this stuff? Gas occupies about 1000 times the volume of an
equivalent solid mass, and quite a lot more than that at kiln temperatures.
So the 10% loi becomes perhaps 50,000% by volume after dissociation.
About 5000 times greater volume, or 500 times the volume of the glaze.
Suppose you have glazed a 300 liter (10cu.ft.) kiln load of ware, using 0.5
liter (2 cups) of glaze with an loi of 10%. The volume of ware is 5 litres
(1.3 gal) of bisque with a porosity of 5%. That means that in firing to
cone 04 or so, you will liberate about 0.75 liters of trapped air from the
bisque pores, plus some water, say a total of 1.5 liters, plus over 250
liters (9 cu.ft) of volatiles from the glaze. So, you will liberate about
the kiln volume of volatiles in a glaze firing.

For most of the heating curve, the glaze coat is immature and does not
seal. Sealing begins when the first melting begins, at the contact points
between the most active fluxes and glass formers. In borated glasses, this
can be quite low.

Liquids, like glazes, have some capacity to dissolve gases. As the liquids
rise to their boiling temperatures, their gas solubility decreases. So
some gasses which may be dissolved at low temperatures may come out of
solution at higher temperatures, like soda water.

Bubbles formed in the first way are like pinholes and crawls. They usually
disperse during the first, foaming phase of the glaze formation. This is
aided by the wettability of the bisque in liquid glaze. If there is
something in the bisque which does not wet in liquid glaze, then the glaze
is likely to crawl or pinhole. If a pinhole doesn't break the glaze
surface, it will appear as a bubble.

When does the gas form? First, the free water boils out at a very low
temperature, around 100C (212F). Next, the bound water begins to be
liberated at up to about 500C (930F). The carbonates begin to dissociate
quite early as well, but some may persist for a while. Free carbon begins
to burn out at red heat, perhaps around 750C (1400F). That takes care of
most of the volatiles.

Glazes with lots of boron and/or alkalis begin to form at around red heat.
That means that some very low melting point glazes will foam like mad while
they are in the glaze formation stage. Above maybe cone 06 or so, most of
the volatiles are gone, or at least well on the way to going. Are there
any volatile sources above this? I don't know.

Once a bubble has formed, what keeps it a bubble? What breaks it? In
physics, we say there is a tendency to move from a low entropy state to a
high entropy state. Crudely, that means there has to be some energy
liberated in order to drive a process. The surface tension in a bubble is
such an energy source. The higher the surface tension, the more a bubble
will tend to break at a surface. Countering that is viscosity. A bubble
once formed usually has to travel to a surface in order to break. Then,
the fluid must flow away from the thin wall separating the bubble from the
surface. Viscosity impedes both of these motions. So a high viscosity
glaze will tend to retain bubbles.

In a liquid with many bubbles, small bubbles tend to dissolve and large
bubbles tend to grow. This is the mechanism by which gasses eventually
escape from the body of liquid, as small bubbles cannot move very much,
while large bubbles can span the distance to a surface.

So highly fluid, borated alkali glazes probably are prone to bubbling at
low temperatures. Depending how viscous they are, they may tolerate this
well, or you may end up with a mass of foam.

In the mid range, alkali earth glazes are less prone to foaming, but may be
prone to crawling, as they are less polar.

In the high range, dissolved gas bubbles may appear. These will appear as
small bubbles, which may become suspended in the glaze and act very much
like an opacifier.

In any range, if the bisque is outgassing from pore gas or volatiles,
bubbles may appear and become trapped. In any viscous glaze, a bubble
formed from any mechanism may be unable to migrate to a surface or dissolve.

So, if I wanted a recipe for a bubbling glaze, it would be high viscosity,
low temperature maturation, high loi, on a porous bisque or greenware,
applied thickly and fired fast. Fast up and fast down again, never making
the glaze runny.

If I wanted a high temperature bubble glaze, I think I'd have to depend on
dissolved gas, unless someone knows of an ultra stable volatile glaze
constituent. That would probably mean an alkali glaze, holding at a
temperature just above liquid phase on the way up, and then up to a peak
with the glaze just about to break down and volatilize, hold long enough to
grow bubbles, then a fast cool. The glaze would probably only show bubbles
in pools, like in the bottoms of bowls. It may be possible to shoot up to
cone 10 before all the volatiles are gone, but it would have to be a very
fast fire.

Evan suggested that this might be a good question for research. I think it
might be. First, some paper study in the glass archives. Bubbles are the
bane of glass makers, so they will know a lot about this subject. If you
get a bubble in the middle of a pot of glass, you are not likely to get it
out, by any means.

Gavin

Gavin Stairs
Toronto, Canada