THE ART OF ENAMELING: RAKU

Raku enameling is an exciting process. It is smelly, and dirty, and one is much more kinesthetically involved than when sitting at the cloisonné table. The outcome of the process is, in the first attempts, whimsical—to say the least. After the enamelist has done a few pieces, however, the delight in the unpredictable results can easily turn into frustration at the lack of control one has over the materials. The question becomes how to push the raku process from being a “let's throw this in the raku pot and see what happens" event into an intentional use of materials and what we know in order to produce cohesive work with some life and integrity—works of art which will convey visually a message or feeling.

Simply stated, the enameling raku process parallels the clay raku process with some adjustments made for the differences in the materials. Materials used are the usual ones: a metal base (usually copper or pre-coated steel) and vitreous enamel. Work is developed in the same way that enameled pieces are. Once a piece has its final firing, it is raku-fired in one of two ways. It can be placed immediately on removal from the furnace into a covered metal container that has been filled with combustible material. Once the material is flaming well, the bucket is covered tightly. The fire burns the oxygen out of the air in the container and then draws the metallic oxides to the surface of the work resulting in iridescent metallic colors and/or a fumed surface, depending on the enamels used. Or the work can be put between layers of wet newspaper, with the paper pressed down tightly around the piece. It is then left to “steam" for a few minutes and then removed from the paper. One of the results sometimes achieved using the wet paper is a swirling effect in the iridescent copper surface.


OBJECTIVES:

The objectives here are to achieve a variety of tactile (or physical) and visual surface effects, to discover some methods of controlling design, and then to bring some materials other than glass, copper and/or steel into the work.

  1. Tactile surface control: Depending on the combustible material in the container, various physical surfaces will result. Dry leaves, pine needles, sawdust or shavings, shredded paper, and wet layered paper are the most commonly used material. Many potters are using shredded paper to avoid the lines and indentations which result from the other materials. I have been using layers of wet paper for flat pieces for the same reason. However, those other materials each leave their peculiar signature on the work. Dry leaves will often leave the imprint of their skeletal structure in the still fluid enamel, pine needles give a very pleasant series of long lines on the surface, and the sawdust provides a fine, more even texture. This latter material is good for pieces which have had some well-defined line and color work developed in the enameling, since it does not add a further design element which could prove visually confusing. Working with placement of pine needles can be used as a design control element, also. Lay the pine needles out in a pattern on some flat wet newspaper. Place your piece face down on the needles and then cover with more wet newspaper. The needles will imprint your design in the molten glass surface.
  2. Visual surface control: The visual surface effects depend on use of color, line and oxides in the enamels. Through the testing of colors it can be determined which colors will remain “constant", i.e. will not produce the metallic glints which can be strong enough to obscure the original color. Generally the opaques are constant. They do pick up some fuming from the carbonaceous material produced in the fire, most of which can be scrubbed off with a glass brush or a 3-M green scrubber (which will sometimes leave a scratched surface—the glass brush is better). But there is often a pleasant soft gray cloud pattern which will be thoroughly embedded in the enamel.
  3. Design control: Carefully considered placement of the constants and of the enamels which will give metallic surfaces also help in design control. Stenciling a pattern, which will leave fairly large areas of the constant, provides some definition of design. Sifting a base coat of one color, with a sifting of a second color only over part of the work plate helps control design also.
  4. Elements other than enamel and copper or steel: Gold leaf may be placed on the piece for the last firing in the furnace. This gives a very strong design element which can help with the “intentional" aspect of the work. I have used 22k gold leaf successfully (available from Daniel Smith Art Supplies in Seattle). For a large defined square, leave the gold leaf on its loose backing paper and place it face down on a surface painted with an adhering solution such as Klyr-Fire, then gently pull the backing paper off. For freer gold areas try picking up the gold leaf with a damp brush—and good luck to you! It helps to watch a person who works with the material. A wide soft dry brush can be used which, after being brushed across your hair several times, has enough static electricity to cause the gold leaf to cling to the brush making it easier to control. In the raku process the gold leaf may darken some, and do use a gentle hand when scrubbing the carbon off it. It also tends to separate in the kiln firing, making a more organic pattern.

Silver nitrate solution has some interesting effects on the enamel in the raku firing. By careful placement of the silver nitrate it can become an element in design control also. When fed with a brush into unfired enamel, the silver nitrate results in heavy cratering and bubbling, giving a very rough silvery surface. If painted on previously fired enamel the surface is much smoother. The optimum effect occurs with this smoother surface when, as the enamel cracks when placed into the furnace for the last firing, the silver nitrate liquefies and runs down these hairline cracks, leaving a fine tracery of silver. This is especially effective when done on a dark blue or black base. Fed into some Kly-Fire, the silver nitrate will give a light, feathered look, while painting it on to a dry surface gives a well-defined edge to the lines. Silver nitrate on some of the whites will result in beige and pink colors. On one of the test pieces two crystals were put dry on the enameled plate. On that piece the result was that the crystal exploded, leaving a starburst of silver in small areas. We did not achieve that effect again, however. Usually a few crystals put into a teaspoonful of water will provide enough solution for a good day's work—experiment to get the right strength. Do not mix up a lot—its potency diminishes as it sits, and it is a bit pricey, being silver. The dry crystals should be kept in a dark, dry place. The silver nitrate can be bought in crystal form at a pottery supply house—or occasionally through a pharmacist, though they have been known to be rather sticky about it.


THE RAKU FIRING PROCESS:

The studio should be set up so that the work can be taken out of the furnace after the last firing and immediately dropped into a garbage can or other container with a lid. (In some studios it is possible to have one kiln outside on a porch or patio.) I usually flip the work over so it lands face down in the material. It may also be dropped face up, with a little combustible material then dropped on top. If the piece is small I often will throw rack, trivet and work into the can, on the theory that more hot metal will ignite the leaves or wood more easily.  Wait a few seconds until the material bursts into flame and then put the lid on tightly. Occasionally here in the Northwest the leaves will be damp. If they don't flame up, quickly throw a little dry material on top of the work. If they still haven't flamed but are producing a lot of smoke, then close up the can anyway—it will often work well. A note: if you are doing small jewelry pieces, use a small saucepan with a lid for your canister. It is easy to lose the small pieces in the large can. Also, instead of using a closed can, small pieces may be slipped in between layers of the wet newspaper previously mentioned, with the newspaper immediately pressed tightly down around the piece. Or, to obtain metallic surfaces on small pieces, such as pendants, you may try the reduction firing by torch instead of the raku container. Remove the piece from the furnace to a soldering tripod and immediately play a soft, smothering flame from the torch over the whole surface. Watch for the metallic glint to appear and immediately remove the flame and quickly cool the work. Do this by picking it up with the pliers and waving it in the air. You may spray with a little water lightly and in short blasts, using the PreVal power pack. (PreVal packs are available in the paint section of most hardware stores. They are both a great boon and a great source of aggravation.)

As mentioned in Experimental Enameling by Fred Ball (the only published work I have seen on raku-fired enamels), when using sawdust it is not necessary to put the lid on the can. The sawdust is usually densely packed and will easily smother out the oxygen quickly.

There are some noticeable differences in the pottery and the enamel process at this point. Enamels should be left in the container for only a few minutes. I wait until the smoke (which inevitably seeps out under the lid) has slowed down. You will notice a difference between “active" smoke which is being driven out by the heat, and the slower smoke when the heat is diminishing. Quickly take the work out with tongs. Do not quench it. I hold it in the tongs and wave it back and forth in the air to cool it. If it is left hot, or left in the can, the metallic surface will sometimes retreat. (N.B. Do not display these raku enamels in sunlight for any length of time. The copper and glass absorb the heat and can become hot enough so the metallic surface disappears again.)

Once cooled, the work is ready for cleaning and, as mentioned before, a glass brush is good for this. At this point you may want to make some aesthetic decisions. Some of the carbon patterns are interesting. If the work is going to receive any handling, then I want to be certain that what is on the surface is going to stay on the surface. If it is a wallpiece, then I am more comfortable leaving some of the carbonaceous material in place. The glass brush does not seem to damage the metallic surface, but do use a softer hand on gold or silver foil or leaf.

Three safety notes: Work in a well-ventilated space. Keep a bucket of water handy for safety purposes. And for those of you who can't tolerate the smoke, contact Rio Grande in Albuquerque. Their half-mask respirator (catalog #201-645) with pink/yell.chem..cart.filter/pr"—their description—(catalog #201-650) works very well. I smell no smoke when wearing it. The mask was about $10 when I got it, and the replaceable filter pack was $17.95 (January, 2001 prices).


ENAMELS TO USE FOR THE BEST EFFECT:

Some years ago I tested some of the lead-free enamels at Thompson Enamel with the help of Woodrow Carpenter and his knowledge of the make-up of these enamels. All of the following enamels mentioned are Thompson enamels. Gold and cadmium-selenium reds and yellows were test with disappointing results. Though occasionally there are gold glints in the metallics, I think they are more apt to be coming from some copper content, not from the gold. Cobalt-based colors also give no metallic surfaces. Some iron can be drawn out of 2110, but that color is more useful as a constant. The two best oxides are manganese and copper, with copper being by far the most successful. Nile green (2305) was tested for its nickel content, but was more useful for its copper oxide.

For manganese, try chestnut brown (2190), russet brown (2140), rose purple (2715) and oil gray (2910). These enamels will give a very slightly dulled surface and the enamel appears to have crystallized slightly. (The best manganese metallic surface, however, comes from lead-bearing 834 smoke gray.)

For copper, the best enamel seems to be 2435 turquoise, 2430 beryl, 2335 peacock and 2305 nile green. In experiments with 2435 turquoise (containing the highest copper content) there was copper all over most of the surface of this piece. Using it in defined areas helps with design control. In one piece there were some interesting areas in which it looked as though the turquoise color bubbled through the copper on the surface, giving a pleasing break-through surface with the interaction of the warm and cool colors. The copper itself tends to be dark. 2430 beryl, containing a little less copper and including some iron, produced a tile with the majority of the surface covered with copper. This copper was livelier in sheen and color than the 2435 and had a brown cast to it. I did not see the breakthrough effect with this one, which could have been the firing process and not the material. I was happiest with 2335 peacock green, which has even less copper and has some potassium dichromate in it. This also gave a very lively metal surface, and the copper was much redder than the previous two tests. The purest copper color, with a pink cast to it, achieved in these early tests, came from the 2305 nile green.

2110 was mentioned previously. When used in combinations with one of the greens (i.e. 2430 beryl) it gave a very nice result, with the warmth of the tan 2110 playing with the warmth of the copper, against the cool green.

In one of my classes some of the students worked with strong light/dark contrasts, using 1020 white (and experimenting with other whites) and then greens or copper-based blues. Break-through patterns figure heavily in these. Another satisfactory combination was 2110 used with 2715. Work was done with colored opaques as well as the white. One student did a series of very consistent tests, using the same design pattern laid on various light colored opaques, resulting in a set of tiles which could easily be compared with each other. For all of us, some pieces were great successes, some were real dogs, but all are worth keeping for the study of the process and what happens to the materials.


FURTHER LEARNING ALONG THE WAY:

If thin pre-coated steel is used (such as steel blackboards—or more properly now—whiteboards) fire an extra coat of counter enamel on the back to help control cracking.

It is a little harder to pull the metallic oxides from the lead-free enamels to the surface than it was with the lead-bearing. (I do miss the leaded 834 smoke gray and 387 midnight blue).

Even though there is a higher percentage of metal oxides in the opaques, they do not give up their oxides. The transparents listed will give you the colors you want. 

After setting up your areas of color, try to resist covering the whole with an enamel which will give you metallic luster. Place it judiciously. Copper luster especially can be so strong that it obscures all the rest of your design.

Since an absolutely smooth enameled surface is not a concern with this technique, leaded and lead-free enamels may be combined with little or no problem. I've had no more cracking with those combinations that with lead-free on lead-free, etc.

The lead-free enamels don't crack as much when they go into the furnace, so it is difficult to achieve as much of the silver tracery with the silver nitrate. To offset this, have the piece as cool as possible when it goes into the hot kiln. The kiln should be hot when the piece is put in, but can be cooled slightly so that the cracks that naturally appear at this time heal a bit more slowly, giving the silver nitrate a chance to run.

I have been puzzled several times when copper appears on the surface after the raku firing, and I know I did not put any copper oxide enamels on the piece. My feeling has been that copper is somehow being dragged from the copper plate to the surface. One student's plate that was fired with pink transparent and 2110 had a lot of copper around the edges. It looked as if it had been electroformed with copper, the copper extending in over the enamel a couple of millimeters. (The fact that the piece had been cut with a torch helped this "electroformed" effect.) After observing and assessing work done in one class, we decided that it does indeed have to be coming from the base copper.

If you don't like the result, scrub the workpiece and do the raku process again. However, if you have a piece which is 90% pleasing to you, consider leaving it as is, rather than losing in a second firing the nice effects you have achieved. Knowing when to stop is very important with raku enameling. Raku pieces are strong works and it is easy to push them too far so that design elements are lost and their inherent and wonderful “crappy elegance" loses the elegance part of that description. 


FURTHER NOTES:

While working at The Enamelist Society 2005 conference, we discovered that some of the Ninomiya (Japanese, leaded) enamels worked very well at producing the iridescence given by manganese oxide colorants. One of the students used an undercoat #610—and I am sorry that I don't know who makes that enamel—placing Ninomiya L96, L15, L82 and N71 on top in separate areas. Ninomiya enamels are available from Enamelwork Supply Co. (Coral Shaffer), 1022 NE 68th Street, Seattle,WA 98115. Phone is 206-525-9271. Email is ewsco@comcast.net.

I have now done more experimenting using wet paper instead of leaves, etc. Sometimes the result is dramatic swirls of copper, which I feel may be produced by the steam working across the surface, trying to find a place to vent out of the paper layers. I also have had this NOT work when the wet paper was on the cement floor of the garage and the weather was very cold. I believe the cold floor may have cooled the metal and enamel too quickly.

Successful raku-firing is a give and take situation. In preparation, give it all you've got in your store of knowledge about design control, including use of line, color, and texture. In results, take what the material and process gives back, which is very exciting indeed.

Copyright 2006 Jean E. Tudor. All rights reserved.