We have moved our education section

Please Click Here

Super-Refined Terra Sigillata

Vince Pitelka - vpitelka@dtccom.net
Home 615/597-5376 Work 615/597-6801 FAX 615/597-6803
Appalachian Center for Crafts, 1560 Craft Center Drive, Smithville TN 37166

I expect that my approach is very similar to that of others who go for a highly-refined terra sig. Too many of the recipes out there do not separate out the large particles adequately, and the product is not a true terra sig. I started researching this when I began doing my "ancient clay" classes and workshops at U-Mass about twelve years ago. First experiments were from the standard recipes in books and CM, and the results were unsatisfactory. I wanted what I had seen on ancient Greek and Roman pots. Finally, via Parmalee, I discovered the work of a German ceramic chemist named Schumann, who researched terra sigs as a coating for sanitary sewer pipe - not very romantic. Schumann discovered the real secrets of terra sig, which had been lost for about 1600 years. Schumann's info guided me to the current system I use.

A glaze hydrometer is required below. A winemaker's hydrometer will not work, because it measures fluids lighter than water. A proper glaze hydrometer should have a scale reading from 1.00 (the weight of water) to 2.00, in 100ths.

I usually use redart, ball clay, or goldart as a starting point. Start with a small amount of hot water, and into it dissolve the deflocculant - 0.25% (1/4 of 1%) soda ash AND .25% sodium silicate (the two together seem to work better than either by itself), based on the dry weight of the clay to be made into terra sig. Once dissolved, add this to some cold water, then add the clay, and add more water, blending with a jiffy-mixer, until the mixture is very thin, checking with the hydrometer until the reading is 1.2 for the redart or goldart slip or 1.15 for the ball clay slip. For a full five-gallon bucket of initial mix, this will take about 16 lbs of redart or goldart, and about 12 lbs of ball clay. Place the bucket with this mixture up on a table and LEAVE IT UNDISTURBED FOR EXACTLY 20 HOURS. After the 20 hours has passed, siphon off the uppermost, thinnest liquid with a winemakers siphon - this tool is ideal for the task, because it has a length of flexible clear plastic hose connected to a stiff clear plastic length about 20" long, with a small "cap" on the end so that the tip of the siphon sucks from above rather than below. This makes it much easier to tell when you begin to get to thicker material, which is absolutely critical.

The 20-hour settling time may seem arbitrary, but it's not. There are several forces at work in this deflocculated mix. The deflocculant introduces same electrical charges to the clay particles, causing them to repel one another and stay in suspension longer. Also, there is the ever-present atomic vibration which causes particles in liquids to naturally disperse. Working against these forces is gravity, causing particles to settle out. At 20 hours, gravity has caused all the heavier particles to settle out, while the finest particles, generally those less than one micron (1/1000 of a mm.) are still in suspension, due to atomic vibration and deflocculation. The top layer IS THE TERRA SIG. Do not discard ANY MATERIAL AT THE TOP, even if it seems quite clear, because it will contain the very finest particles.

Start the siphon with the tip just barely immersed in the settled mix. DO NOT AGITATE THIS CONTAINER OR THE SETTLED MIX AT ALL. As mentioned above, place it up on a table BEFORE the 20-hour settling period, and do not move it for any reason. If you must move it, remix it and start the 20 hour settling period again. Once the siphoning is started, slowly feed the tip of the siphon down into the mix as the thin liquid is siphoned off. Keep the tip close to the surface, so that it periodically sucks a bit air. If it sucks too much the siphoning action will stop, but having it suck a little bit of air is critical, because it gives a good indication of how thick the liquid is. As soon as you get to thicker liquid the siphon will begin to suck much more air. As soon as this happens, STOP SIPHONING. Resist the temptation to keep siphoning, because the product will be inferior. I have never tried to do anything with what remains in the bucket, which is MOST of what you started out with.

You can siphon into any container, and the siphoned liquid will of course be far thinner than the original specific gravity, and will be unusable, so the question then is how to concentrate it. I am the proud owner of a 36"-diameter restaurant wok, which I bought at a flea market for one buck. Using this as a slump-mold, I made several 24"-diameter terracotta evaporating dishes with a raised 2" edge. Another excellent mold for such dishes is one of those round dished plastic snow-sleds. I usually just siphon into a five-gallon bucket, and pour that into one of the evaporating dishes. The water soaks into the terracotta (any bisque-fired claybody would work) and evaporates from the back and from the rim. It takes about a week for the terra sig to get back to a useable specific gravity (dependent on temperature, humidity, and air movement). Don't cover the evaporating dish (unless you are welding or grinding or woodworking in the vicinity). Anything that settles into it out of the air won't do it any harm. Don't worry if it seems to be solidifying around the edges. When it has thickened considerably, scrape the solidified stuff loose with a clean rubber scraper, and agitate the mix with a whisk. If necessary, work the lumps against the bottom of the dish with a very clean sponge to bring all the terra sig back into suspension. Decant some into a tall container and check the specific gravity. If it is still thin let it evaporate some more. If it is thicker put it in an appropriate container and add water.

If you wish, you can let the sig dry completely, and keep it around until you need it. When you anticipate needing it, slake it in water for several days, mix well with a jiffy mixer, and adjust to the desired specific gravity.

I originally used sig at a specific gravity of 1.2 or even higher, but ran into trouble with it peeling and chipping. Now I thin it to 1.13 to 1.17, depending on the clay used and the desired results. I apply the sig to bone dry clay, and get the best results when the clay is sanded. I use a wide soft brush, and I simply brush on repeated flowing strokes until I get an opaque buildup which begins to conceal the sanded texture (still very thin). As soon as I get as much sig buildup as I want, and the surface wetness has soaked in, I polish with a soft piece of flannel or T-shirt material. Remove all buttons and seams before using the cloth to polish. I usually get a glassy shine in one polish. It's magical. The amount of terra sig I brush on depends on how opaque I want the coat to be, and how much I want it to smooth out the texture of the clay. It is possible to get a very high shine with an almost transparent coat, because the shine results from the clay platelets laying flat on the surface, and in this refined terra sig the particles are so fine that a distribution of them over the surface will give a good shine and yet still allow the clay beneath to show through. A good terra sig may be the world's most perfect substance.

Also, a properly prepared terra sig makes the very best burnishing slip. For burnishing larger forms, I apply a very thin smear coat of lard, which retards the drying of the sig and allows you to completely burnish the pot. But once you start burnishing, you must finish it in one sitting. If you leave it incomplete and allow it to dry, you must sand the surface, re-coat it, and start over.

Expect to use a LOT OF CLAY to get a good terra sig, but the results will be worth it. To get a gallon of redart terra sig takes about 50 pounds of redart clay. Goldart gives about the same yield, while ball clay gives a higher yield, since it is finer to begin with. As I mentioned above, I have never tried to do anything with the deflocculated residue left from the settling process. It would be good for making thick slip. If you add it to a claybody you would be deflocculating the clay, which will reduce plasticity.

I have fired all my terra sigs to a maximum of ^02. I have applied very thin coats to bisqueware and fired them with adequate results, but never as good a shine or as durable a surface as when applied to bone dry. Terra sig applied to leather hard tends to loose it's shine when it dries. At ^04 the redart sig gives a bright brick-red-orange color, the goldart gives an off-white, and the ball clay gives a PURE white. The redart sig, when properly made, is denser, and in a blackware bonfire gives beautiful brown-to-black colors. The ball clay sig in the blackware firing or in raku post-firing smoking gives intense jet-black.

Recently, some of my students who are using very gritty clays in high fire have tried coating the feet of their wares, and occasionally the contact surface between jar and lid with terra sig, to give a smoother surface than the base clay. Personally, I like the base clay showing in these areas, but it is a matter of personal taste. As Louis Katz indicated to me, goldart sigs do retain a bit of shine in high-fire, but nothing like low-temp polished terra sig. Ball clay sigs in high fire simply give a white satin finish - not really a shine at all. Of course, at high-fire temperatures a redart sig would turn to a glaze, but more refractory clays might give great results. Kaolins and fire clays give extremely low yield in terra sig, because of the coarseness of the particle size. Stoneware clays like goldart give better results, depending on the fraction of fine particles. Experiment away.

I always like to work with pure clay terra sigs, because they give the best shine. It is of course natural that others will want more color, but unless you can ball-mill the mixture the shine will be reduced. At U-Mass we experimented with both oxides and mason stains and got good results by ball-milling the thickened evaporated terra sig and colorants for a day or so. I have used both oxides and mason stains without ball-milling, and the shine is reduced slightly, but the results are still satisfactory.