Digitalfire Ceramics Technical Articles
|Logged in as Level 2 access: Logout|
Firing: What Happens to Ceramic Ware in a Firing Kiln
Section: Firing, Subsection: General
By understanding what sorts of change art taking place in the ware at each stage of a firing you can tune the curve and atmosphere to produce the best possible ware.
A kiln is not a microwave oven and firing one is not like heating Chinese noodles! It is not a rote timed process where you can just set the oven and go shopping. Firing a kiln is much more like baking an angel food cake. It requires awareness of kiln contents, the process, and the objective. Inflexible schedules are out; flexibility and sensitivity are in.
As a potter or industry, you are basically making rocks in your kiln; metamorphic rocks. You are changing the form of matter just the way a metamorphic rock has been changed from another by the forces of heat and pressure. As I have highlighted elsewhere, there are physical and chemical things that happen in the kiln. The physical changes give us the headaches, but the magic of the chemistry makes it all worthwhile. Let's review the simple stages in a firing.
The Stages of a Firing
The ware has to dry in preparation for bisque or single fire. If you don't have a dedicated drier, then you are using your kiln as a drier. If your drier does not exceed the boiling point of water, then you are using your kiln as a drier. If your ware sits in the studio or plant after drying, then its hydroscopic nature results in the absorption of water from the air and once again, your kiln is the final drier. Whatever the case, all water has to come out and even though a piece looks and feels like it is dry, there can still be plenty of water present. If just 2 or 3 percent needs removal, a typical industrial periodic kiln setting could have hundreds of pounds of water that must escape, all of which expands a thousand times when it turns to steam. Needless to say, this makes for a damp atmosphere during this stage and proper ventilation is a must. Many modern driers (an kilns) have fans that impose a virtual hurricane of draft on the ware in the kiln to remove this water efficiently.
In a fine grained clay (one containing bentonite or ball clay), it requires time to vent the moisture out, especially if ware is thick. If you fire too fast at this early stage, the water within boils, generates steam, and just blows the piece apart. If you heat just a little slower, a few chunks will be blown off at sites of thicker cross section. A little slower yet and maybe just a few cracks. Still slower and only micro cracks that will weaken the ware and encourage failure in later stages or during use. Slower yet and you have it right. Slower yet and you have some margin for getting it right on a continual basis.
Another matter, which must not escape your notice, is the drying of glazed biscuit ware. The absorbent biscuit can pick up considerable water during glazing and this must not be driven out too quickly. A rapid warm up will loosen the glaze from the biscuit, causing it to fall off or crawl during the firing. A moisture-laden atmosphere during early stages can even re-wet fully dried ware, and the subsequent sudden drying and associated steam and pressures associated with rapidly increasing temperature will likewise compromise the fragile glaze-body bond.
How do you tell what firing schedule is right? Experiment. There are a number of variables that make it very difficult to establish rules. The most important are the weight of the ware, the density of the setting, the water content of the clay and its ability to vent this water, and the air flow within the kiln to remove the vapor as it is generated.
Large hand-built sculptural pieces weighing hundreds of pounds can require weeks or even months of protected air drying. These must be fired over two or three days, most of this time at the boiling point stage. Lighter industrial ware like mugs can be humidity force-dried in special chambers in minutes and fired in hours. In general, for dry ware and good airflow in the kiln, most ware, including large porcelain items, can be brought through this stage in several hours. Ware that is not dry may require much more time. In a worst case scenario, namely a densely packed electric kiln having no airflow and large pieces that have not been boil dried, this stage could take 24 hours or more.
Whatever the case, as long as you understand the importance of thoroughly dry ware, air flow in the kiln, clay venting ability, and density of pack, you will be able to adjust matters to encourage success. In this stage, it is not so much a matter of even firing but speed and atmosphere. It is all just common sense.
Before continuing, I would like to mention the matter of clay bodies. Certain ill-conceived clays are much more vulnerable to failure during this stage of firing. Clays, which lack particle size diversity, made only from ball clays, kaolins, and very fine ground feldspar and silica, do not perform well even if grog is included. On the other hand, natural native materials with limited processing will sometimes tolerate very fast firing. I have seen clays that can survive to 2300 °F in less than an hour at thicknesses of 1 cm. To make your bodies tolerant, use large particle size kaolins and ball clays, minimize bentonite, and keep your eyes open for quality special-purpose clays and fireclays, which are known to open up the body.
The Ceramic Change
Crystal bound water has to escape during bisque or single fire. At earlier stages, mechanically bound pore water, that is water between clay and mineral particles, is expelled. However, H2O is bound right into the clay crystal itself, as well as into other minerals that may be in the clay body. For example, kaolin loses 10%+ of its weight on firing due to this crystal water. This “water smoking” phase occurs over a wide range of temperatures that can extend past the red heat stage. Since large quantities of water can be generated, there is ample reason not to push the kiln too fast up to red heat. For the clay body, there is no going back after the changes that occur during this phase, thus the term “the ceramic change”.
How much of a concern is this? Well, it turns out that it is not nearly as critical as the expulsion of mechanical water which occurs earlier. By the time this stage is in full swing, pores within the body matrix provide a good network of channels through which the steam can be vented. Although shrinkage is not occurring during this period, the ware is very fragile; as it lacks the particle bonding mechanisms it had in the green stage. For this reason, there is one matter of concern. Proper airflow in the kiln for single fire ware should vent all escaping steam to prevent any upset in existing glaze-body bonds.
Quartz Inversions and Conversions
Crystalline solids are rather temperamental and quartz is no different. Quartz is a crystalline form of silica in that it has a three dimensional regular pattern of molecular units. These form naturally in nature because lengthy cooling times allow arrangement. Quartz is made of a network of triangular pyramid (tetrahedron) shaped molecules of silicon combined with four oxygens. Unfortunately, the quartz delights in changing the orientation of the tetrahedron shaped molecules with respect to each other, thus loosening or tightening the whole mass (and thus changing its total size). It exhibits twenty or more personalities called “phases” and these show a remarkable range of physical properties. A change to another phase is called a “silica conversion”. The most significant phases are quartz, tridymite, crystobalite, and glass. The material does not even melt to change phase (except to produce silica glass of course). Only an elevated temperature to increase molecular mobility along with the required time is needed. What is more, each of the above crystal phases has two or more forms (alpha and beta, beta one, etc.). Changes which occur between these are reversible, that is, the change which occurs during heat-up is inverted during cool down. These changes are thus called “quartz inversions”. These inversions, unfortunately, often have associated, rather sudden, volume changes. That means that quartz conversions are something to consider when optimizing the fired properties; quartz inversions are something to consider when firing to prevent cracking losses.
There are two important inversions you need to know about because of their sudden occurrence during temperature increase and decrease. The first is simply called ‘quartz inversion’ and it occurs quite quickly in the 570°C range (1060°F). In this case, the crystal lattice straightens itself out slightly, thus expanding 1% or so. The second is crystobalite inversion at 226°C. This is a little more nasty because it generates a sudden change of 2.5% in volume and it occurs at a temperature within the range of a normal oven. This material has many more forms than quartz, so it is a complex animal to say the least. However, while all bodies will have some quartz, you won’t have a problem with crystobalite inversion unless there is crystobalite in your body. Crystobalite forms naturally and slowly during cooling from above cone 3. It forms much better if pure crystobalite is added to the body to seed the crystals or in the presence of catalysts (e.g. talc in earthenware bodies).
You can ignore these phases. But you will never be able to fully optimize fired properties of your ware and will never fully address “inversion” related firing problems without at least a partial knowledge of silica phases. We could just melt quartz, cool it quickly, and the resulting glass (irregular arrangements of molecules) could be ground into a powder having very stable firing behavior. This would really make things much simpler. Unfortunately, silica melts at a very high temperature, so this is impossible. So we have to live with the stuff and learn to cooperate with it during the firing process.
As noted already, individual particles of quartz in the body change from alpha to beta form of the quartz phase and back during heat up and cool down. It is important to realize that it is not the whole piece of ware, or even the silica within it, that undergoes the associated volume change. It is the small and even microscopic particles of the quartz that do. This behavior is, of course, dampened by the structure in which they exist. During heat up, these particles are in a non-glass bound matrix surrounded by other particles and pore space, so there is much tolerance for the volume change associated with the inversion. However, during cool down or subsequent heat ups, where the clay matrix is a solid mass of glass melted around each particle of quartz, sudden volume changes in the quartz particles are much more likely to cause micro cracks radiating around each. Since the quartz can form the skeleton of the entire structure, waves of change occur through a piece which tend to extend the micro cracks into major cracks.
What does this all mean? It means there is not too much to worry about with quartz inversion in first fire ware on the way up, or about cool-down for bisque ware. In both cases, the open body is quite tolerant. However, take it easy on second-fire earthenware, very easy on second-fire stoneware, and super easy on second-fire porcelain. Watch for excessive amounts of quartz powder in dense bodies that do not fire to full vitrification. In these, the quartz has not been dissolved by the corrosive action of the fluxes, but remains part of a non-homogenous fired matrix. If possible, use the finest quartz powder available and this will make dunting (cooling cracks) during firing less of a problem.
Almost all bodies contain some organic matter that must decompose and then burn at some point to produce carbon gases (the dark color of ball clays, for example, is due to their coal content). As expected, this burning occurs at red heat and beyond. It is of interest in the firing process because proper oxidation and sufficient time are needed to prevent black coring of the body and associated expansion and strength problems. This means you should provide adequate time for this part of the process, namely, at least a few hours with some draft for thicker ware. In addition, glazes or fine slips should not flux and seal the surface too early as this could result in bloating when the last remaining gases encounter blocked escape routes.
When all the water has been removed from a clay, there is really little left to hold individual particles together other than intimate contact. At some point during heat-up, chemical bonds begin to develop between particles. These processes do not involve melting yet, but a rearrangement of the molecular structure does seem to occur as a result of the increased mobility afforded by the rising temperature. This is the sintering point of a clay. You can demonstrate this by putting a sample of powdered clay into a kiln, and firing to a temperature necessary to bond the pile of powder into a cohesive and solid lump. The sintering point is normally around red heat. When a body has reached this point, it becomes impervious to water, thus resisting slaking (particle disassociation in water). If heated a little higher, ware demonstrates considerable ability to withstand thermal shock, a property that is lost to some degree as the glassy phase develops at higher temperatures.
At some point in the firing, fluxes begin to react strongly in bodies and glazes, and chemical changes begin to occur. “Decomposition” refers to the first stage of oxide rebuilding undertaken by the Kiln God. Although materials like feldspar and kaolin are put into the kiln, the kiln fires gradually “deconstruct” these materials into their basic oxide building blocks. In some cases (e.g. single fire glaze ware), this deconstruction yields gases like sulphur and carbon dioxide which must escape, typically by bubbling up through or out of the molten glaze (e.g. whiting, dolomite lose up to 40% of their weight during firing). Reconstruction of the glaze and body matrix occurs using the pool of oxides available after decomposition. As you might expect, when the temperature begins to drop, either after shut-off or soaking, decomposition stops and recomposition to a glass begins. Thus, the most interesting part of the show really begins when the kiln starts cooling, and up until that point the stage was just being set. The glass that forms during normal cooling is a random arrangement of oxide molecules, unlike a crystalline solid which has a regular repeating structure that requires extended cooling time to form.
Many potters and a few industries use reduction firing to achieve rich iron brown and earthtone colors and special effects like copper red glazes. Reduction firing is somewhat of a ‘black art’ and is difficult to describe on paper. The basic idea is to supply only enough oxygen in the kiln to burn the fuel. But rules tend to break down in actual practice, because many people do what they call a “heavy reduction” by supplying even less oxygen and thus introduce unburned carbon from the gas into the developing body and glaze chemistry. The reduction process denies the iron compounds in the body of the oxygen molecules that they would like to have. This forces them into the reduced form, thus producing the desired colors. Many potters begin a body reduction around 1000°C, holding this for a time. Then they move to a neutral atmosphere to bring the kiln up to glaze melting temperature; when they again apply reduction to shape the final iron chemistry of the glaze. Others begin a light reduction at 1000°C and fire this atmosphere right to final temperature. Some practitioners close the firing with a short soaking period in oxidation to clear any carbon residue, others do not. Reduction is difficult to maintain consistency within, and thus potters, who tend to love the mystery and surprise of each firing, have embraced the technique.
In recent years, oxygen monitoring devices (i.e. Australian Oxytrol Systems at www.cof.com.au/useoxy.htm) have become available which enable users to exercise tight control over the kiln atmosphere. Manufacturers provide detailed instructions on what oxygen level to maintain for each glaze type. This and the general appeal of high temperature reduction have made those using electric kilns feel they are ‘second class citizens’. However, this is not necessarily true for you.
Earthenware and low-temperature whitewares are not fired to maturity, thus they never vitrify completely. This is not to say they lack strength. A body of considerable porosity and pore space can be remarkably strong by virtue of the glass weld between its particles. Think of vitrification as a process that develops in a clay body during firing. We take it far enough to produce the desired strength and color, but not so far that ware begins to warp excessively. Thus each person arbitrarily decides what ‘vitrified’ is for himself and his own circumstances. Some bodies vitrify over a wide range of temperatures, others do so over a very narrow range and thus require close firing control.
Knowledge of this process helps us to see the importance of testing a body at temperatures below and above the actual working temperature, and testing at slower and faster rates of rise. This helps you to see it in the context of the vitrification process and alerts you if your firing is miss-targeted. Soaking the firing takes on much more meaning when you understand that vitrification is a process. The body is composed of quartz mineral and clay crystal particles (and possibly grog or alumina) which form a physical skeleton around which the flux-containing materials flow. As firing proceeds, the silica hungry fluxes become more active and begin to dissolve the quartz particles and remove silica from the metakaolin (originally the hydrated clay crystal). As this happens, the melt forms silicates and thus stiffens, needing yet higher temperatures to continue the process. Given these higher temperatures (above 1000°C), the formation of long mullite crystals from the decomposing metakaolin occurs. This rearrangement happens without melting of the crystal, and the higher Al2O3 form melts much later, further stabilizing the vitrified mass. On a chemical level, the alumina oxide present acts as somewhat of a chemical skeleton as silica comes into solution, further stabilizing the clay mass. This helps us understand part of the magic of why the piece does not end up lying on the kiln shelf in a heap.
Remember then, the firing is not just the melting of a glass cement to glue together a bunch of microscopic rocks, it is a matter of silica conversions and inversions, mullite development, chemistry development of the melt, silica dissolution, and a multitude of other things. These make it possible to produce fired products having a great variety of physical properties from the same piece of clay by adjusting only the firing schedule. One excellent reference on this complex subject is the Potter’s Dictionary under ‘silica’, ‘mullite’, and ‘crystobalite’.
As the final stages of firing arrive, the glaze reaches its full viscosity and mobility. Unlike the body, it melts fully and often all oxides move about freely in anticipation of arranging themselves in some semblance of order at freezing. During this period, interaction between body and glaze accelerates and an interfacial layer is formed. The chemistry of the glaze and body, and the time available determine how transitionary this layer becomes. Likewise, the fluidity and surface tension of the glaze determine its ability to wet the surface to heal minor bare spots, and its ability to pass gaseous bubbles percolating up through the melt from the ever shrinking and vitrifying body. During a soak, the glaze has further opportunity to even itself out and develop an optimum interface with the body.
Glaze Cool and Freeze
Cooling is an integral part of the firing process, since this is where the actual glass-building occurs. For the body, on the other hand, the building occurs during heat-up and the beginning of cooling cements this new form of matter. In the simplest possible case, the ware cools, the glaze solidifies as a glass and it is done. However, an element of crystal formation often accompanies cooling. This is especially the case for slower cooling or where the chemistry of the melt encourages the formation of nuclei for crystal growth. The growth actually occurs right around the freezing temperature, which can be much lower than you might think. Some stoneware glazes may take many hundreds of degrees to set and crystallization continues to occur until all molecule mobility is stopped. This means you should be aware of this process and if undesired crystalization (devitrification) happens, adjust the chemistry of the glaze (i.e. raise alumina) or speed the rate of drop during the critical range.
So the firing process is slightly more complicated than most people think. Admittedly, you can fire a kiln without knowing any of this and you can bake a cake by just setting the oven timer and going shopping. But realistically, how likely is it that you are going to achieve optimal results? Or even passable results? •
In Bound Links
Copyright 2003, 2008 http://digitalfire.com, All Rights Reserved
INSIGHT is ceramic chemistry