Glazes - for the Self-reliant Potter (GTZ, 1993, 179 p.) 11. Glaze problems (introduction...) 11.1. Introduction to glaze problems 11.2. Trouble-shooting checklist 11.3. Specific problem explanations

Glazes - for the Self-reliant Potter (GTZ, 1993, 179 p.)

11. Glaze problems

Anybody with even a little experience with glazes will realize that problems often arise. Knowing what to do about them requires a lot of experience, and even expert glazers often find it difficult to establish the source of a problem.

11.1. Introduction to glaze problems

It is one thing to develop a nice glaze but quite another to keep it working. One potter may want a glaze that crazes, whereas another wants his glaze to be craze-free. A glaze fault may not mean that the glaze is ugly, just that it reacts differently and does not look like the desired effect.

When a glaze suddenly starts to react differently from what we want, we call it a glaze fault. Solving the problem is seldom easy, and usually several factors are involved. The first thing to check is what changes have occurred since the glaze last worked without problems. The following things should be checked:

1. Was the right recipe used and were the materials weighed out correctly? (Always use batch cards for glaze weighing).

2. Were the right raw materials used? Was there any chance of mistaking one material for another? Was the labeling of materials in stock correct?

3. Have new raw materials or a new frit batch been used since the last fault-free glaze batch was produced? If so, check if the new material is different from the original material in stock.

4. Has the body been changed in any way? For example: new preparation method, change of clay material, change of body recipe, higher or lower biscuit firing?

5. Was the preparation of the glaze done as usual? For example: same ball milling time, same screening, same specific gravity of glaze slip?

6. Was there any change in glaze application and were the products clean and dust-free before application?

7. Were there changes in the kiln setting? Was the glazed ware dry before firing started? Were there any changes in fuel, firing schedule, firing atmosphere (reducing/oxidizing) and was the correct top temperature reached (setting of cones, draw trials) ?

Once we know which conditions have changed, we may already be close to establishing what caused the glaze problem. The following trouble-shooting lists may be helpful to find solutions to the problem.

11.2. Trouble-shooting checklist

Problem: The glaze or part of the glaze settles too fast in the bucket.

Causes:

- High amount of frit in the glaze.
- Glaze materials too coarse.
- Water content of glaze slip too high.
- Ball milling time was too short.
- Too little clay in the slip.
- Metal buckets cause fast settling.

Solutions:

- Add 5-10% plastic clay or 0.5-2% bentonite.
- Reduce amount of frit by removing some of the insoluble materials from the frit recipe and adding these to the glaze batch instead.
- Longer ball milling of glaze materials.
- Add a small amount of vinegar (acetic acid) to the glaze.
- Use plastic buckets or wooden containers.

Problem: Glaze slip is too thin, low viscosity.

Causes:

- Water content too high.
- Alkali materials from frit or feldspar have been dissolved in the water, so the slip is deflocculated.

Solutions:

- Let the glaze stand for a day and decant the clear water off the top. Because some materials may be removed with the water, it is better to allow excess water to evaporate.
- Add flocculant (magnesium sulfate, calcium chloride), but only if the ware has too low a porosity. This is often done in production methods which use high-fired biscuit and lower temperature glaze.

Problem: Glaze layer too thin after dipping.

Causes:

- Water content of glaze slip too high.
- Clay body absorbs too little water.

Solutions:

- Increase density of slip (decrease water).
- Biscuit-fire at a lower temperature.
- Glaze only one side of the article at a time and allow it to dry before glazing the other side.
- Add flocculant to the slip so that glaze layer becomes thicker.

Problem: Glaze layer becomes too thick.

Causes:

- Glaze slip density is too high (too little water).
- The glaze does not contain enough clay materials.
- The biscuit body absorbs the water too fast.
- The glaze slip releases its water too fast.
- Dipping or pouring is done too slowly.

Solutions:

- Reduce density by adding water.
- Add plastic clay, bentonite or cellulose (CMC) binder.
- Biscuit-fire to a higher temperature or moisten the pots before glazing.
- Dip.

Problem: Glaze layer cracks during drying.

Causes:

- Glaze has too high a drying shrinkage due to high content of clay or zinc oxide or due to over-grinding in the ball mill.
- The glaze is applied too thickly.
- Single-fire glaze was applied to biscuit ware.
- In double glazing, the second glaze may tend to crack.
- Glaze is poured over a sprayed glaze.

Solutions:

- Less ball mill grinding of glaze materials.
- Calcine zinc oxide or add it to the frit.
- Replace part of the clay with calcined clay or introduce alumina (Al2O3) as feldspar.
- Reduce viscosity of glaze by adding water.
- Apply single-fire glaze to leather-hard pots.
- When double glazing, reduce clay content of the second glaze.
- When double glazing, apply the second glaze before the first one dries completely.

Problem: Glaze layer does not stick to the body.

Causes:

- Glaze adhesion to body is too low due to:

greasy or dusty body surface
too fast and too thick application
too high body porosity
too fine grinding of glaze
dusty surface from underglaze colors.

- In single-fire glazing, the glaze shrinks less than the moist body.

Solutions:

- Clean the body surface by brushing or sponging.
- Reduce speed of dipping and, if spraying, apply less glaze at a time.
- Biscuit-fire at a higher temperature.
- Add cellulose binder (CMC) when single-fire glazing.
- Add clay, borax, soda ash or glaze to the underglaze colors.
- Reduce grinding time of glaze.

Problem: Glaze dusts off easily after drying.

Causes:

- Too little binding power of the glaze and low adhesion to body.

Solutions:

- Add 2-5% plastic clay or 0.5-1% bentonite.
- Add a binder like cellulose (CMC) or 12% borax.

Problem: Glaze runs.

Causes:

- Firing temperature too high.
- Viscosity of glaze too low.
- Glaze layer too thick.

Solutions:

- Adjust firing temperature.
- Add alumina (clay, feldspar) or silica (quartz, zircon) to the glaze.
- Glaze thinner.

Problem: Glaze does not melt properly.

Causes:

- Firing temperature too low.
- Too much silica or alumina.
- Not enough glass formers (SiO₂ or B₂O₃) and to much CaO, MgO, BaO.
- Glaze materials too coarse.
- Evaporation of fluxes in the firing (extended firing time).

Solutions:

- Fire at higher temperature.
- Slower firing and longer soaking at the end.
- Increase amount of fluxes, reduce content of alumina.
- Longer milling of glaze materials.
- Add the evaporating fluxes to the frit.

Problem: Pinholes or eggshell surface.

This is one of the most common glaze problems and often the most difficult to cure. It is usually caused by gas escaping from the body or glaze, leaving small holes that do not have time to smooth over.

Causes:

- Firing temperature too low, glaze does not have time to melt completely.
- Firing temperature too high, glaze reacts with body, forming gas bubbles.
- Glaze has high surface tension and viscosity, which do not allow gas bubbles to escape.
- Glaze is too thick, not allowing gases to escape.
- Release of gases from body or engobe.
- Early reduction firing forms carbon and sulfates in the body, which cause pinholes when they are later released.
- Body contains organic particles which burn out, leaving small pits.
- Body contains air bubbles from incorrect slip casting.
- Glaze contains zirconium opacifier, which causes large amounts of gas.
- Contamination from ball milling, usually by linings set in lime cement.
- Contaminated water used to mix glazes.
- Dirty biscuit or dust on the glaze.

Solutions:

- Fire to the correct temperature.
- After final glaze temperature is reached, "soak" the kiln by holding at the same temperature for about half an hour.
- Reduce viscosity and surface tension by changing the glaze recipe.
- Less reduction, especially in early stages of firing.
- Thinner glaze application. Higher biscuit firing.
- Body must be prepared to eliminate organic materials (sometimes long aging to decompose the material and repugging will solve the problem).
- Slip must be mixed and poured carefully to eliminate air bubbles.
- Increasing viscosity of zirconium-opacified glazes by addition of clay or talc may reduce the problem.
- Ball mill linings should be fastened with high-alumina cement mortar.
- Use only clean water to mix glazes.
- Biscuit should not be stored too long and should always be cleaned before glazing. Glazed ware should be put in the kiln as soon as possible.

Problem: Glaze crawls.

Normally this is caused by cracking or lifting of the glaze layer before firing (see section 11.2.2. for additional causes and solutions).

Causes:

- The viscosity and surface tension of the melted glaze are too high.
- Too fast firing.
- During firing, the body released gases (steam, carbon, sulfur), which lifted the glaze layer off.
- Drying and sintering shrinkage high.
- Glazed ware was still wet when fired.

Solutions:

- Reduce surface tension and viscosity by firing higher.
- Reduce alumina, magnesia and zircon.
- Increase biscuit temperature.
- Correct what causes cracking and lifting of dry glaze (see above).
- Reduce grinding time of glaze.
- Reduce drying and sintering shrinkage by calcining part of clay and zinc oxide content.
- Dry the glazed ware before firing.
- Add 1-2% borax to the glaze.
- Add clay, borax or frit to underglaze colors.

Problem: Glossy glaze turns matt.

Causes:

- Glaze is absorbed by the body (glaze layer is too thin).
- Flux materials evaporate in firing.
- Sulfates from fuel are deposited on the glaze surface.
- Too much steam in the kiln. Glaze is underfired
- Crystals form due to very slow cooling.
- Glaze slip was not properly mixed.

Solutions:

- Glaze thicker.
- Glazed ware set next to biscuit ware or on new shelves, which may attract volatile fluxes from the glaze. Do not mix glaze and biscuit firing.
- Introduce volatile fluxes in the frit.
- Fire with more draft and less reduction and cool quickly.
- Always stir glaze immediately before application and screen it through at least 60 mesh.

Problem: Matt glaze turns glossy.

Causes:

- Glaze is overfired.
- Too fast cooling.
- Oxides from the body combine with the glaze.

Solutions:

- Slower firing, or longer soaking period at the end of the firing.
- Cool slowly, closing all dampers and fire-boxes.
- Use another matting agent.

Problem: Glaze color changes.

Causes:

- Wrong firing temperature, often over-firing.
- Change in reduction/oxidation atmosphere.
- Impurities in glaze or body.
- Coloring oxides not ground fine enough.
- Color pigment from engobe or underglaze melts into the glaze.

Solutions:

- Better control of firing temperature, oxidation/reduction.
- Check materials for impurities.
- Control the grinding time of coloring oxides.
- Add clay or quartz to underglaze pigments and engobes.

These are problems that only appear right after firing or after the article has been used for some time.

Problem: Crazing of the glaze

Causes:

- Thermal expansion of the glaze is higher than the body, which causes the glaze to contract more in cooling. This may be caused by:
- Too much alkali (soda and potash).
- Too little silica, alumina or zinc oxide.
- Too little cristobalite formation in the body.
- Lack of strong clay/glaze interface.
- Glaze is too thick.
- Fast cooling.

Solutions:

- Apply glaze more thinly.
- Higher glaze firing temperature (to increase cristobalite).
- Higher biscuit firing temperature.
- Reduce amount of soda and potash in glaze by replacing with boron (to decrease thermal expansion of glaze).
- Add additional boron, silica, zinc oxide or calcium carbonate to the glaze (to decrease thermal expansion of glaze).
- Add quartz or talc to the body. Quartz forms cristobalite in the body, which has a high thermal expansion.
- Longer soaking at top temperature, slower cooling.

Causes:

- Moisture expansion (sometimes called delayed crazing). Porous bodies expand when they absorb moisture from the air and force the glaze to craze. This is a common problem with earthenware.

Solutions:

- Fire at higher temperature or add flux to the body, making the body more vitreous.
- Add calcium carbonate, talc or dolomite to the body.
- Reduce the thermal expansion of the glaze.

Problem: Shivering of glaze

This is usually seen as particles of glaze falling off the pot after firing (sometimes after a few weeks). It happens most often on sharp edges, but the entire glaze may shiver or the pot may crack.

Causes:

- Thermal expansion of the glaze is less than the body, which leaves the glaze under strong compression. This may be caused by:
- Too high content of silica, boron or zinc oxide in the glaze.
- Too high content of silica in the body.
- Too low content of soda or potash in the glaze.

Solutions:

- Reduce the amount of quartz in the glaze.
- Reduce the amount of quartz in the body.
- Add more soda or potash to the glaze.

Problem. Lime popping

This is seen as small pieces of glaze popping off the pot, often several weeks or even months after firing. Under each flake of glaze a small white particle can be seen imbedded in the body.

Causes:

- Small pieces of limestone or plaster in the clay body. These slowly absorb moisture from the air and expand, forcing the glaze off the pot.

Solutions:

- Find the source of the lime pieces.
- Replace contaminated material or screen it through 40 mesh.
- Plaster that has gotten into the clay. All contaminated clay must be thrown out, and better care taken in clay mixing. The problem often comes from recycled clay, which has picked up plaster in the forming section.

11.3. Specific problem explanations

As already mentioned, crazing and shivering are caused by differences in thermal expansion/contraction between the glaze and body.

Crazing

Crazing appears as cracks in the glaze. This occurs during cooling, if the glaze contracts more than the clay. Cures for crazing are mentioned above.

Shivering

Shivering is the opposite of crazing and occurs during cooling if the clay contracts more than the glaze. Cures for crazing are mentioned above.

Thermal expansion

All materials expand when heated. This is called thermal expansion. Some materials expand more than others, and the degree of expansion can be measured. Numbers are used as a scale of thermal expansion, and this is called the coefficient of expansion (CE). The glaze layer on a pot has one coefficient of expansion and the body has another.

Glaze-body tensions

After a pot is fired and taken out of the kiln, it will be exposed to a sudden decrease in temperature. The glaze layer and the body will contract, but most often at different rates. Below is shown what happens when 1) glaze contracts more than body and 2) body contracts more than glaze.

This figure shows a body (white) with a glaze on top (black). The glaze and the body have contracted at the same rate. We say: they have the same coefficient of expansion (CE).

Figure

This figure shows a glaze that has a higher coefficient of expansion (CE) than the body.

Figure

The glaze contracted more, so it is shorter and therefore the glaze is under a tensile stress (it is pulled apart). If the body is very thin it will bend as shown. The arrows show the direction of the stress the glaze is under.

Figure

More often the tensile stress is relieved by cracks in the glaze as shown in this figure. This is called crazing. The stress caused by high CE of the glaze may be relieved by crazing as soon as the pot is taken out of the kiln or it may take days, months or years. The longer it takes, the closer is the CE of body and glaze.

Figure

This figure shows a body with higher CE than the glaze. The body contracted more than the glaze. The glaze is under compression, and if the clay is thin it may bend as shown to relieve the pressure. If body contraction is only slightly greater than glaze contraction, nothing will happen.

Figure

If a glaze contracts much less than the body, the compression on the glaze becomes too much and the glaze will start to flake off like this (shivering). This may not happen by itself, but only if something hits the pot. Typically, the rim of a cup will easily chip off.

Figure

High compression of the glaze may also be relieved by cracking of the body.

Moisture swelling

When the body has been exposed to humidity for a long period, water enters the body, which expands slightly (moisture swelling). This expansion causes the glaze to go into tension and it will craze. This kind of crazing is called delayed crazing.

Solutions

As we saw above, crazing and shivering are caused by different rates of contraction and expansion (different CE's). The problems are cured by adjusting the CE of body and glaze, so that the two contract and expand more closely. It is best if the glaze is left under slight compression.

Coefficients of expansion for various materials

Ceramic materials have different coefficients of expansion (CE). The following list shows the relative values for the most common:

Na2O	High CE	contracts more in cooling.
K2O	­
CaO	|
BaO	|
TiO2	|
Fe2O3	|
Al2O3	|
PbO	|
CuO	|
MnO	|
ZrO2	|
SnO2	|
P2O5	|
ZnO	|
MgO	|
SiO2	¯	contracts less in cooling.
B2O3	Low CE

Adjusting CE of glaze

From this list we can see that if we replace soda (Na₂O) with boron (B₂O₃) in a glaze we will lower the CE of the whole glaze. This can be done without changing the melting point of the glaze. Addition of silica will lower the glaze's CE but will also raise its melting point.

If shivering occurs, it means the CE of the glaze is too low. Adjusting it means adding soda (Na₂O) and reducing boron (B₂O₃).

Adjusting body

Adjustment of body CE is not done according to the CE of the materials listed above. The contraction rate of the body depends to a much higher degree on the sudden reversible contraction of silica crystals when these change their crystal structure (cristobalite).

Quartz change

Quartz is a crystal form of silica. Quartz is created in the body during firing when the clay crystal changes form and releases some of its silica. When quartz is heated it changes its crystal structure at 573°C. This happens very suddenly and is accompanied by a 1% expansion. On cooling to below 573°C it contracts again.

Figure Figure 11.3.1.B The volume change of quartz is caused by a rearrangement of the bond between the atoms. At 573°C the angle suddenly shifts as shown.

Cristobalite

Cristobalite is another crystal form of silica. It changes its size around 220°C and the volume change is nearly 3%. Cristobalite is created at temperatures above 900°C from silica released from the clay (Al₂O₃ 2SiO₂) or talc (3MgO · 4SiO₂) or from quartz.

Figure 11.3.1.C. The graph shows volume changes of three forms of silica. The two crystal forms change dramatically but silica in glass hardly changes.

Body-glaze contraction

The two graphs below show how the body and its glaze contract during cooling. The graph in Figure 11.3.1.D shows a body that does not contain any cristobalite. At 573°C the body contracts suddenly due to the contraction of quartz, but at this temperature the glaze is still fluid enough to follow the contraction of the body.

Figure 11.3.1.D. Contraction of body and glaze during cooling. Glaze contracts more so it will craze.

Around 500°C the earthenware glaze hardens and from then onwards contracts according to its own CE. In this example the glaze has a higher CE than the body; it contracts more. This leaves the glaze under tensile stress; the glaze is smaller than the body. This will cause the glaze to craze.

Figure 11.3.1.E. The body in this graph contains crisotobalite and shows a sudden contraction at 220°C. This causes an overall higher contraction of body compared to glaze.

The graph in Figure 11.3.1.E shows contraction of a body containing cristobalite. As above, the glaze first follows the quartz contraction, then hardens and starts to contract more than the body. However, at 220°C the cristobalite change causes the body to contract, and at this temperature the glaze is hard so it is left under compression. This compression will prevent the glaze from crazing.

Moisture crazing

After firing, the porous earthenware body will absorb moisture and this causes the body to expand. If the glaze is not under sufficient compression it will craze. Such delayed crazing may occur a long time after firing. The moisture expansion of the body is reduced by making the body more vitreous. Additions of talc or limestone to the body reduce moisture crazing.

Crazing cure

For both types of crazing the cure is:

- Add quartz (or silica), talc or limestone to the body.
- Biscuit-fire to a higher temperature.
- Glaze-fire to a higher temperature.
- Add silica to the glaze.
- In the glaze, replace fluxes with high thermal expansion, like soda (Na₂O) and potash (K₂O), with boron oxide (B₂O₃).

It may seem strange that the cure for crazing is to add silica to both body and glaze. The reason is that adding silica to glaze makes it contract less, but silica added to the body causes the body to contract more.

Crazing test
There are several ways to test how the expansion of glaze and body fit each other. The most simple ones are:

- Rings of clay with a diameter of 5 to 10 cm are made with a small gap and biscuit-The gap is measured, the ring is glazed on its outer surface and refired. After firing the gap is measured to see if the ring has contracted or expanded. If the gap has become greater the glaze will craze.
- Glazed samples are exposed to thermal shocks by repeated heating and cooling.

The thermal shocks can be from boiling water into ice water. The number of cycles the sample can withstand before crazing indicates its craze resistance.

Another method is to heat the sample at first to 100°C then cool it in 20°C water. This is repeated while raising the temperature in steps of 10 or 20 degrees. The higher the heating temperature the sample withstands without crazing the longer it will be able to stay craze-free under normal conditions.

A rough guide is:

120°C	craze-free for	8 days
150°C	craze-free for	100 days
180°C	craze-free for	2 years
200°C	craze-free for	life

Even if a sample survives the thermal shock test it may still craze due to moisture swelling. This can be tested in an autoclave which is simply a pressure cooker that can withstand higher pressures. A pressure cooker can be used instead. The glaze sample is placed in the pressure cooker with some water. It is kept under pressure for a period and then checked for crazing. The time it can withstand pressure without crazing indicates the time it may stay craze-free under normal circumstances. The following rough guide is for testing in an autoclave under a pressure of 3 atmospheres (about 3 bars). If using a pressure with, say, a pressure of 1.5 atmospheres the testing time in the table should be doubled:

Hours in autoclave	Expected craze-free life
1	1 -2 years
2	2 -3 years
3	4 -6 years
4	9 -10 years
5	13-15 years

All the tests provide only a rough indication of craze resistance. When you do the test you will develop your own procedure' which then should always be followed faithfully. In this way you will be able to compare your crazing test with your previous results.

11.3.2. CRAWLING

Crawling appears as areas of clay that are not covered by the glaze. It may be small areas or, in extreme cases, the glaze may pull up into a pattern of small balls or islands, leaving bare clay in between.

Crawling is caused by:

Poor adhesion of glaze

Dusty or oily biscuit prevents the glaze from sticking to the body. Refractory oxides (chrome, rutile) or underglazes that act as a dust layer prevent the formation of an interface. Adding clay, borax or frit to the underglaze colorants helps.

High surface tension

High surface tension of the glaze in melting pulls it into islands before the clay/glaze interface forms. This is caused by certain oxides, especially magnesia, clay and zinc oxide. The solution is to replace magnesia by other materials, to calcine part of the clay or to use calcined zinc oxide instead of raw.

Cracking of glaze layer

Extensive shrinkage of glaze in drying and early stages of firing, usually caused by too much clay content or by overgrinding the glaze, causes the glaze to crack and separate from the body. A thick glaze layer is more likely to crack.

11.3.3. PINHOLING AND BLISTERING

Pinholes appear as tiny holes in the glaze surface. Blisters look like frozen bubbles or craters. They are a problem in utilitarian ware, as they collect dirt. They may be only on the surface of the glaze or may penetrate to the clay layer.

Figure 11.3.3.A. Pinholing.

Pinholes

During firing gas bubbles are formed in the melted glaze. The bubbles will move to the surface of the fluid glaze and be released.

If you watch any glaze metling, you can actually see this process. Some glazes (especially those containing raw borax) foam and boil until they finally smooth out. When the firing is stopped before the glaze has had time to heal over, a pinhole or crater is left (see Fig. 11.3.3.A). Since overfiring also causes pinholes it is better to keep the maximum temperature for some time (soaking period).

The main sources of the gas are:

- After glazing, a large volume of air exists in the space between the solid glaze materials. The air gathers into bubbles during sintering and melting.
- Release of sulfates and carbon in the body and from some of the glaze materials.
- Air bubbles in the body introduced by improper handling of the casting slip.
- Sulfates and carbon from the fuel may deposit in the body during the initial stages of firing. Above 900°C the gas will be released.

It is important to find out if the problem is in the glaze or in the body. Relatively large pinholes that go all the way to the body are usually caused by small holes in the body that do not accept the glaze-this is most common with slip-cast ware, or with common red clay that contains particles of organic matter, sand or mica.

Problems arise if the glaze starts to cool and solidify while bubbles or craters are still forming.

Detailed causes and solutions are given in section 11.2.3.

11.3.4. COLOR CHANGES

Potters are often plagued by changes in glaze color, either within the same kiln-load or from separate firings.

Often this problem can be traced to glaze preparation. The colors may not be ground finely enough, weighing may be incorrect, raw materials may have changed.

Otherwise, the problem usually is due to more or less reduction than usual. This is one of the most difficult conditions to control in firing and depends completely on the skill of the firemaster.

The problem is worst in glazes that contain color oxides that are sensitive to reduction. The most sensitive is copper, which is green in oxidation and red in reduction; and iron oxide, which is yellow, red to brown in oxidation and mottled red-brown, grey to blue or green in reduction. Other oxides change less.

Heavy reduction will darken the iron in the body, which will affect the glaze, also darkening it. Sometimes a pot will be dark on the reduced side and light on the oxidized side.

Other causes and solutions are given above.