The principles behind formulating a porcelain are quite simple. You just need to know the purpose of each material, a starting recipe and a testing regimen.
Incredible strides in porcelain ware production and firing equipment have occurred in recent decades. Robotics and computer controllers have revolutionized the whole ceramics industry. However, clay bodies themselves have tended to resist change. While the understanding and troubleshooting of industrial machines may be common, the understanding of the clays put into them is much more rare. The dynamics of powder, slurry, and wet materials processing, forming, drying, and firing are key factors to the optimization of a production process and the quality of the finished product. Let's try to take at least some of the mystery out of common porcelain formulation.
A traditional functional ware porcelain is actually just a vitrified clay body with low Fe2O3 contamination. A general porcelain recipe is fairly easy to derive. Unlike glazes, we do not really consider the chemistry of porcelains, we are interested in the physical properties. Let's talk about formulating your own porcelain (without getting too technical). While I am talking here about making a plastic porcelain, the principles outlined are certainly applicable to other types, just think outside the box when you need to.
General purpose functional porcelains contain
Special purpose porcelains may also contain
The whitest and most translucent porcelains are made using the most expensive and cleanest (low iron) materials. However some applications do not require a high degree of whiteness, for these lower cost or more readily available clays and feldspars can be used.
The "Universal 25 Porcelain" recipe typically produces an inexpensive cone 10 not-very-white porcelain having moderate plasticity and near zero porosity (depending of course on the specific properties of the ingredients). It is made from 25% each of ball clay, kaolin, feldspar, and silica, or more simply, 50% clay and 25% each of silica and feldspar. Thousands of potters and companies use this "as is", or alter it to accommodate specific materials or circumstances. Let's consider some physical properties to look for in a porcelain (not ordered in any specific way).
Porcelains can be compared in all of the above areas. Logically, you cannot have the best of all of them. There are always trade-offs, compromises, especially if you do not want it to be too expensive. These physical properties can all be measured or at least compared using simple equipment, methods and observations as described in the Test area of this site. Let's look at each of the materials in the recipe to understand their functions and how they can be changed.
A true porcelain would normally derive all its plasticity from a kaolin. Since kaolins are often of limited plasticity, this limits the workability of throwing or modeling porcelains made from them. Still, there are some surprisingly plastic kaolins, although the limited range of particle sizes can mean less than ideal drying performance. Do not accept that a kaolin is plastic just because its name includes the word plastic or a supplier says it is plastic, test yourself. For casting porcelains, an all-kaolin approach is quite feasible (using 50% kaolin rather than 25% kaolin and 25% ball clay) since these bodies benefit greatly from the reduced drying shrinkage and increased water permeability associated with the larger particle size of kaolins. However an all-kaolin casting porcelain may not have the strength to pull away from the plaster mold without cracking, add a little plasticizer if needed.
Kaolins can differ widely in maturity. British kaolins require the use of less feldspar because they already have some natural fluxes as part of the mica mineral they contain. So even though these might be less plastic, less flux is needed, so more kaolin can be used in a recipe. It is an good idea to make up the kaolin complement of a recipe using more than one brand, this provides a better distribution of ultimate particle sizes and minimizes the effect on the body if one kaolin changes.
Since kaolins vary quite widely in their plasticity, maturity, soluble salts, particle size and whiteness, it can be quite a challenge to test and classify them all. Data sheets often are not that helpful because they present information in different ways and seldom does a company explain its materials in terms of other well known alternatives. The burden of picking the best kaolins thus rests on you and your ability to evaluate and compare them using tests that document the appropriate physical properties. We recommend you use the DFAC, SHAB, and LDW tests to evaluate kaolins and ball clays.
Larger particle size kaolins tend to be dirtier, less plastic, more expensive and have lower dry strength, so their benefits come at a cost. It is possible to make fairly fast-casting zero-ball-clay bodies with ordinary light burning kaolins like EPK. Some people use large-particle-size kaolin when they also have ball clay in the mix, the latter may be canceling the benefit of the former! Of course large particle ball clays are also available, but remember that they are still considerably finer than standard kaolins. Also, fine-tuning a kaolin mix makes little sense if the body is not deflocculated properly. The best approach is to use a standard white burning kaolin, deflocculate it properly and learn to work with it. Then fine tune it by the substitution of some large particle material to speed casting rate while watching for any deleterious properties introduced.
Following is a sample of a Foresight report on a kaolin (Foresight was the predecessor to insight-live.com). I have tested using DFAC, SHAB, LDW and SIEV tests.
TESTDATA REPORT FOR A RUN ======================================== NUMBER: L2497 DESCRIP: K&T DIAMOND KAOLIN DATE: 07/27/93 LOCATION: BD 725 ======================================== This is a montmorillonitic intermediate particle size material. I received this sample 7/93 for testing to compare to pioneer kaolin. This has 1-2% lower fired shrinkage and 4-5% higher absorption at the higher temperature than pioneer kaolin. It fired color is similar. This kaolin would be valuable to augment pioneer in our bodies to minimize shift if one material changes. DRYING FACTOR (ID-DF, ABBR-DFAC) DRY_FAC - A000 LOI/Water Content (ID-LW, ABBR-LDW ) WET-WT DRY-WT FIRE-WT OIL-WT IMM-WT PERCENT LOI DENSITY +-------+-------+-- ------+-------+------+ 1 | 25.99 | 15.03 | 13.10 | 15.53 | | 42.2% 12.8% 1.00 g/cc +-------+-------+--------+-------+------+ SHRINKAGE/ABSORPTION/H2O (ID-SA, ABBR-SAWL) DRY-LEN FIR-LEN FIRE-WT BOIL-WT CONE FIRE-SHR DRY-SHR ABSORP +-------+-------+-------+-------+-----+ 6 |96.3 |89.4 |31.68 |36.61 |6.4 | 7.17% 3.7% 15.6% 7 |96.17 |88.43 |31.87 |35.98 |7.0 | 8.05% 3.8% 12.9% 8 |96.15 |88.16 |32.65 |36.59 |7.4 | 8.31% 3.8% 12.1% 9 |96.18 |86.29 |31.09 |33.37 |8.9 | 10.28% 3.8% 7.3% 11 |96.38 |86.2 |30.96 |33.16 |10.8 | 10.56% 3.6% 7.1% 12 |96.39 |86.42 |32.74 |35.52 |10.0R| 10.34% 3.6% 8.5% +-------+-------+-------+-------+-----+ SOLUBLES (ID-SL, ABBR-SOLU) FIRED-CLAY GLAZ-CLAY DRY-CLAY +----------+---------+-------+ 1 | NIL| | NIL | +----------+---------+-------+ SEIVE ANALYSIS (ID-SV, ABBR-SIEV) TOTAL PLUS-35 PLUS-48 PLUS-65 PLUS-100 PLUS-150 PLUS-200 PLUS-325 +------+-------+-------+-------+--------+--------+--------+--------+ 1 | 100 | | | | | .01| .04 | .6 | +------+-------+-------+-------+--------+--------+--------+--------+
Ball clay is much finer and thus much more plastic than kaolin. A very wide range of ball clays are available, some are much more plastic. White burning ball clays often do not have near the plasticity of their dirtier counterparts. Bentonite is much finer than ball clay (the ultimate particles are much smaller). It is incredibly plastic, adding only 2% to a recipe can drastically improve working properties. However, these materials have a down side. Ball clays can have ten times the amount of brown-firing iron oxide that kaolin has and many have heavy soluble salts that produce a dark colored scum on the burned surface. Many also contain lignite particles that can produce glaze imperfections. Raw bentonites can be downright dirty, burning brown or red with possible specking and soluble salts sometimes so heavy they form a glaze. White firing bentonites come in a wide range of plasticities. Some have so little plasticity you will wonder why anyone would use them for ceramics, others are incredibly plastic. But beware, even though you might only need 3-5%, the cost of this one material could be more than the other 95% combined! A study of all the ball clays and bentonites available to you can be a real education! Again it is very important that you have a well defined testing program to compare these materials (like what you can administer in an account at https://insight-live.com). There is another up-side of ball clay worth mentioning. Ball clays contain free silica, so if there is significant ball clay in a recipe you can reduce the silica and that leaves room to add more clay for plasticity or feldspar for maturity.
Following is a sample report for a ball clay:
======================================== NUMBER: L2553D DESCRIP: GLEASON BALL CLAY DATE: 04/29/94 LOCATION: BD 763 ======================================== Mixed 50:50 calcine:raw. This is quite white in the raw state and is much more plastic than 49'r. Fired bars are the whitest of the ball clays tested this round and are very clean. There is a little brownish scum at 10r, but this is a very nice looking ball clay, although it is shrinking much more. This is a very refractory ball clay. DRYING FACTOR (ID-DF, ABBR-DFAC) DRY_FAC - A000 | LOI/Water Content (ID-LW, ABBR-LDW ) WET-WT DRY-WT FIRE-WT OIL-WT IMM-WT PERCENT LOI DENSITY +-------+-------+--------+-------+-------+ 1 | 33.58 | 24.22 | 22.82 | 24.57 | 8.94 | 27.9% 5.8% 1.57 g/cc +-------+-------+--------+-------+-------+ SHRINKAGE/ABSORPTION/H2O (ID-SA, ABBR-SAWL) DRY-LEN FIR-LEN FIRE-WT BOIL-WT CONE FIRE-SHR DRY-SHR ABSORP +-------+-------+-------+-------+-----+ 5 | 94.81 | 84.65 | 35.1 | 37.8 | 9.8 | 10.72% 5.2% 7.7% 6 | 94.81 | 86.05 | 38.70 | 43.15 | 6.3 | 9.24% 5.2% 11.5% 7 | 94.8 | 85.1 | 37.07 | 40.65 | 6.9 | 10.23% 5.2% 9.7% 8 | 94.9 | 85.03 | 35.52 | 38.68 | 8.3 | 10.40% 5.1% 8.9% 9 | 94.94 | 84.18 | 33.47 | 35.99 | 8.8 | 11.33% 5.1% 7.5% 11 | 94.79 | 84.13 | 32.31 | 34.79 |10.7 | 11.25% 5.2% 7.7% 12 | 94.85 | 84.51 | 32.24 | 35.01 | 10R | 10.90% 5.2% 8.6% +-------+-------+-------+-------+-----+ SOLUBLES (ID-SL, ABBR-SOLU) FIRED-CLAY GLAZ-CLAY DRY-CLAY +----------+---------+--------+ 1 | MED | | MED | +----------+---------+--------+ SEIVE ANALYSIS (ID-SV, ABBR-SIEV) TOTAL PLUS-35 PLUS-48 PLUS-65 PLUS-100 PLUS-150 PLUS-200 PLUS-325 +------+-------+-------+-------+--------+--------+--------+--------+ 1 | 100 | | | .01 | .01 | .03 | .28 | 1.1 | +------+-------+-------+-------+--------+--------+--------+--------+
So the easiest thing you can do to the standard '25 Porcelain' recipe to increase its plasticity is add 2%-3% raw bentonite. Even though most inexpensive 200 mesh bentonites can be quite dirty, this small amount may not affect the fired color as much as you might expect. Even if the bentonite has 5% iron, adding 5% of it to a recipe only adds 0.25% iron to the body as a whole. Do not overlook one detail: One bag of bentonite may not contain iron specks while the next will. Bentonite processors generally guarantee that a certain percentage is finer than 200 mesh, but they do not identify what the coarser material is. You can buy microfine ceramic grade (e.g. 600 mesh) raw bentonites, they are much more expensive, but you will never get fired specks. If whiteness is not all-important, you can increase the ball clay at the expense of kaolin to produce a plastic whiteware (but you may need to reduce the feldspar also, as ball clay is less refractory than kaolin). If you want a very white porcelain, you have to meet the challenge of reducing or entirely eliminating the ball clay. Where super whiteness is secondary to having a good general purpose body, some ball clay can usually be tolerated. If having the most plastic body possible is the most important, then you will use all ball clay and no kaolin (e.g. B-Mix from Laguna).
These are the fluxes, or more correctly, contain the fluxes. Fluxes are the oxides that help develop fired maturity by liquefying and slowly dissolving some of the clay and silica. The total flux amount necessary is easily determined by simply firing to a range of temperatures above and below the one you intend to work to; studying the absorption, strength, and firing shrinkage curves; and adjusting the amount of feldspar to give the desired maturity. The amount of feldspar for a cone 10 body vary from 15%-30%, depending on the flux content of other materials in the recipe. For a typical American kaolin, it takes about 25% for cone 10 and up to 50% for cone 6.
Feldspars are not without potential problems. While some brands can be relatively iron free, others fire surprisingly darker. Some can present flocculation problems due to slight solubility (i.e. nepheline syenite). Sodium feldspars are generally cleaner and more potent. Use two or three together if possible (to dampen changes that could occur in one).
Silica tends to be a very consistent and inexpensive material. Quartz grains act primarily as a micro-aggregate or framework structure for the fired matrix. In addition, some of the silica is dissolved by the fluxes to produce aluminum-silicate glasses. Too much silica in a recipe could mean lower plasticity (since less room is left for clay). However, there is also much discussion about the detrimental effects of crystobalite (i.e. dunting), whose development during high temperature firing is related to available free quartz. Thus there is some merit to lower silica amounts, especially if you have the ability to adjust your glazes to lower their expansion. The use of less silica means more clay can be added resulting in higher plasticity. A finer silica (300 mesh) reacts better with the fluxes and thus less is needed. Too little silica in a body can mean crazing glazes since the quartz mineral contributes to the low expansion that assists glaze fit. Of course, lack of silica will also mean more of a tendency to warp during firing. For cone 10, many technicians aim at 20-25% for expansion reasons and to provide firing stability over a range of temperatures.
Most people have noted that the '25 Porcelain' recipe has flaws that can be corrected for individual situations and materials. To improve this recipe for cone 10:
To formulate a white translucent cone 6 feldspar porcelain:
If you need to formulate a casting porcelain, remember that much lower plasticity is needed. Common consensus in industry is that it is much more important to use kaolins of large particle size, so water can easily be drawn out by the plaster mold. However for a smaller operation that does not need to really optimize mold release time, normal kaolins are fine. The cleanest kaolins and ball clays are also the least plastic, thus casting porcelains (that can tolerate this) can achieve whiter and more translucent effects than their plastic counterparts. It is imperative that you understand the principles of 'deflocculation' so that the amount of water in the slurry can be minimized and proper mold release and casting time can be achieved. If you have not witnessed the magic of adding a few drops of a dispersant to a hopelessly thick clay-water mix that defies agitation, you have not lived! If you need to cast very thin ware you may have to add some bentonite (start with 1%) to the slurry to give it the strength to pull itself away from the mold during drying.
If you are making a plastic porcelain for use in modeling, throwing or machine forming; pay careful attention to its drying properties. Since porcelains are fine-grained, they don't usually dry well, thus plastic porcelains are even worse. Add enough bentonite to give the plasticity needed but no so much that ware cracks on drying. It is important to have a good test to rate and compare drying performance (see the DFAC test in the Tests area of this site). Some people add molochite grog to porcelains for better drying, but do not assume this will work until you try it yourself.
No matter what type of porcelain you make, give careful thought to how mature it needs to be. If zero absorption is not necessary, consider reducing the feldspar to get more plasticity from the clays (and be able to reduce the bentonite). Such a body can still be considered functional and vitreous and it will resist warping in the kiln. In addition, material changes, which result in more maturity will be less likely to cause trouble in a body that has some 'room to move'. When you evaluate fired absorption, measure the property at a variety of temperatures. A 'measure-only-at-the-working-temperature' philosophy is a tunnel vision approach that will almost certainly get you into trouble, especially in situations where the working temperature is far above the point at which zero porosity is attained. For example, if your body achieves zero porosity at cone 8, firing it to cone 10 will give you alot of headaches with warping.
True, we have only scratched the surface but this is certainly enough for a good start. In summary, I have prepared a chart which follows. It lists some of the trade-offs you must consider when formulating or adjusting a pottery porcelain.
|Silica||20%-25%||Use at least 20% (unless you have really low expansion glazes). Use more for high expansion glazes. For maximum translucency use less (less silica permits more feldspar)|
Use the finest particle size available to help reduce the amount needed. Try calcined alumina as a substitute to produce a stronger (but more expensive) product. Use calculation to lower the expansion of your glazes so silica can be reduced.
Cone 10: 25% feldspar
Cone 6: 35% feldspar
Cone 6: 20%
|Feldspar (and Nepheline Syenite) are by far the most common fluxes. The percentage needed depends on what you want. Use the amount required for the degree vitrification needed at the temperature you want to fire at. If translucency is paramount, mostly frit and a little feldspar (frit is much more potent, less is needed leaving more room for clay and silica). Ferro Frit 3110 is more most common body frit.
Measure maturity using tests defined at insight-live.com (absorption, fired shrinkage and strength over a range of temperatures) and adjust to required amount. Try different feldspars (including Nepheline Syenite), mixes of feldspars and frit, or just frit to reach the compromise you want between maturity, whiteness and cost.
|Kaolin||Cone 10: 25%-50%
Cone 6: 20%-40%.
|Ideally, use as much kaolin and as little ball clay as possible (for firing whiteness). Test as many types as you can to find the best combination of whiteness and plasticity. English kaolins are less plastic (and more expensive), but also fire to greater maturity. American kaolins are not as white but more mineralogically pure and require more flux. Highly processed delaminated kaolins are whiter but more more expensive. Halloysites (like those from New Zealand or Applied Minerals) are the whitest available, but also very expensive (and require more feldspar). Use a mix of two or three kaolins to minimize impact of a change in one.|
|Ball Clay||Up to 25%||Ball clays are higher in iron and detrimentally effect fired whiteness. Use them as less expensive sources of plasticity. Higher ball clay percentages slow drying and can increase drying cracks. If whiteness or translucency are important use only kaolins (with added plasticizers). Doing detailed fired and workability comparisons of brand names can really pay off will ball clays. If necessary, add 0.3% barium carbonate (for the ball clay proportion) to precipitate solubles that cause surface scumming. Combine two or more ball clays to minimize the impact of changes in one.|
|Plasticizers||Plastic bodies: 2-4%
|A wide range are available. Bentonites are cheapest. Highly processed smectites and hectorites are most expensive. The best quality products are twenty or thirty times more expensive than the cheapest (but you get what you pay for). Do tests to compare the options and find a compromise between impact on fired color and the plasticity gained. Check drying performance, bloating, specking, solubles and consistency of the brand you will be using. The highest quality porcelains employ the very white non-plastic clays, that means this component is critical to get the needed working properties. If money is no object, use New Zealand Halloysite with VeeGum as a plasticizer and Frit 3110 as the flux (this will give you alot of flexibility to maximize the silica for glaze fit).|
These are fired around 1250C bisque and glazed between 1050 and 1100C. One composition we know of is 45% bone, 30% clay and 25% mix of potash feldspar and quartz. But lower bone versions are used also, it being as low as 10% (the clay is the same but the feldspar obviously has to be much higher). It is not completely clear why companies do not simply add frit to their porcelains to get more translucency. If you know this, please email me.
When porcelain is fired it is not just a simple gradual melting that is occurring. The feldspar creates a liquid phase in which some of the kaolin (and quartz) dissolve, but also in which the kaolin crystal transforms into another form of different shape, mineralogy and amazingly, of a higher melting temperature. This crystal-matrix bonded by feldspar-glass on a silica-skeleton creates the incredible hardness and toughness that porcelain can have. To make this all happen you need a good testing regimen and a good place to maintain the audit trail of your work. Where? An account at Insight-live.com.
These are two cone 6 transparent glazed porcelain mugs with a light bulb inside. On the left is the porcelainous Plainsman M370 (Laguna B-Mix 6 would have similar opacity). Right is a zero-porosity New Zealand kaolin based porcelain called Polar Ice (from Plainsmanclays.com also)! The secret to making a plastic porcelain this white and translucent is not just the NZ kaolin, but the use of a very expensive plasticizer, VeeGum T, to enable maximizing the feldspar to get the fired maturity.
These bowls were made by Tony Hansen using a mixture of white and stained New-Zealand-kaolin-based porcelain (Plainsman Polar Ice) fired at cone 6. The body is not only white, but very translucent.
This cast bowl (just out of the mold and dried) is 130mm in diameter and 85mm deep and yet the walls are only 1mm thick and it only weighs 89 gm! The slip was in the mold for only 1 minute. What slip? A New Zealand Halloysite based cone 6 translucent porcelain. This NZ material is fabulous for casting slips (it needs a little extra plasticizer also to give the body the strength to pull away from the mold surface as it shrinks).
This is 70% kaolin and 30% feldspar. The fired body looks very porcelaneous but right out of the kiln the crazing is like this. The thermal expansion of the kaolin:feldspar mix is too low, no ordinary glaze will fit it. The other issue is with the flat particle shape of kaolin. The throwing process has tended to line up the particles concentric to the centre. During drying and especially firing, more shrinkage occurs along the radius lines than along the concentrics. Ten of these were made at the same time and all ten cracked like this during the bisque firing (the drying process would have initiated the stresses). Added silica (preferably 20% or more), although having particles that are very small, acts like aggregate in concrete, strengthening the matrix and forcing the clay particles to orient more randomly.
Out Bound Links
Some starting recipes for stoneware and porcelain with information on how to adjust and adapt them
The term "sintered" refers to the particle-to-partice bonding and packing that occurs within a ceramic matrix as temperature increases. Sintering is a process as well as a state. Sintered bodies are not vitrified; the process occurs without any glass development (melting) to glue particles together....
Veegum T, Veegum CER, Veegum Pro, VGT
Montmorillonite, Bentonite USA
Crazing refers to small hairline cracks in glazed surfaces that usually appear after firing but can appear years later. It is caused by a mismatch in the thermal expansions of glaze and body. Most ceramics expand slightly on heating and contract on cooling. Even though the amount of change is very s...
F3110, Frit KFG 4110, Ferro Crystal Frit, Ferro Frit 3110-2
In Bound Links
How can two potters have completely different opinions about the plasticity and workability characteristics of the same clay body
Learn to test your clay bodies and recording the results in an organized way and understanding the purpose of each test and how to relate its results to changes that need to be made in process and recipe.
There are many methods to flux and densify clay bodies and these are based on a number of different mechanisms.
Traditional utilitarian porcelains are comparatively white burning and vitreous clay bodies that are made from feldspar, clay and quartz. When fired, the feldspar flows and dissolves many of the other particles into a viscous glassy melt that bonds the quartz particles and, and if temperatures are s...
2018-06-24 - The starting point for a cone 10 porcelain. Such porcelains require about 25% silica for glaze fit and about 25% feldspar to achieve near zero porosit...
By Tony Hansen