Making and Using Pottery Washes
A wash is the cheapest, oldest and most forgiving way to make a carved surface read. Metal oxide, a little flux, water, brushed on and wiped back off. Here is what each ingredient does, how the same wash behaves in oxidation and in reduction, and a calculator that weighs out any batch you want.
Mix colorant and flux half and half by dry weight, add water until it looks like watercolor, brush it onto bisque and wipe the high spots back off. The recesses keep the color. Test one tile before you touch a pot.
Jump to a section
Color that hides in the low spots
A wash is metal oxide and a little flux, thinned with water until it brushes like watercolor. You flood it over a carved or textured bisque surface and then wipe the surface back with a damp sponge.
The sponge only reaches the high spots. Color stays in every line, dimple and undercut, and the raised areas come back clean. That is the whole effect, and it is why a wash makes carving read from across a room when a flat glaze would swallow it.
The flux is the part people skip. Without it the oxide sits on the surface as dry powder and rubs off on your hands after firing. With it, the oxide fuses into the clay.
The route that works first time
Seven steps, iron oxide on bisque, which is the easiest wash there is.
- Weigh dry, with a mask on. Equal parts colorant and flux is the standard starting split. The calculator further down does the arithmetic for any batch size.
- Add water until it looks like watercolor. Roughly two and a half times the dry weight in milliliters. Thin enough to flow into a line, not so thin it runs off.
- Stir, then sieve if you have one. Lumps of dry oxide fire as dark specks.
- Brush it on bisque, generously. Flood the carved area rather than painting it neatly. You want it in every recess.
- Wait a minute, then wipe back with a damp sponge, rinsing the sponge often. The high spots come clean and the lines keep the color.
- Let it dry, then glaze over it if you want, or fire it bare for a matte stained surface.
- Fire as normal. Ours go to cone 5½ with everything else.
Test one tile first. A wash looks almost nothing in the bucket like it does out of the kiln.
Three things ruin nearly every wash
You left out the flux
Oxide and water alone fires as dry, chalky powder that rubs off on hands and on everything in the cupboard. The flux is what fuses it to the pot. Half and half by dry weight is the standard fix.
You used too much cobalt
Cobalt is the strongest colorant in the room by a wide margin. A wash mixed at the same strength you would use for iron comes out a flat black-blue that kills every line you carved.
You wiped it back too late
Once the wash dries hard on the surface, wiping stops lifting it evenly and starts scrubbing it out of the lines too. Wipe while it is still damp.
What a wash actually is
Six facts, and the rest of this page comes out of them.
- It is a colorant suspended in water, not a glaze. There is no glass former in it worth speaking of. It stains a surface rather than coating it, which is why a fired wash is matte and slightly rough where a glaze is smooth.
- The flux is a small percentage, not the point. Enough to fuse the colorant to the clay, not enough to melt into a glassy layer. Push the flux too high and you have accidentally made a thin runny glaze.
- It works by being wiped off. The technique is subtractive. You apply everywhere and remove selectively, so the surface itself decides where color lands.
- Bisque is the usual home for it. Porous bisque pulls the water in and grips the colorant, which makes the wipe-back controllable. On leather-hard clay it bleeds, and on a fired glaze it sits on top and mostly wipes away.
- It is not food safe on its own. A bare fired wash is a thin layer of metal oxide on a porous surface. Anything that meets food or a mouth needs a glaze over it.
- It goes under or over a glaze, and they look nothing alike. Under a clear, the color deepens and gains gloss. Over a fired glaze, it stays on the surface, catches texture and reads much drier.
Wash, underglaze, slip: the difference in one line each
A wash is colorant plus water, made to be wiped back. An underglaze is colorant in a prepared body, made to stay exactly where you brushed it. A slip is liquid clay with colorant in it, thick enough to build a layer you can carve back through.
The ingredients, and what each one does
Three jobs: something that makes color, something that makes it stick, and water. Everything on this page is a variation on those three.
Red iron oxide, the colorant to start with
Warm browns, ambers and rusts, and it behaves at every temperature. Forgiving about quantity in a way nothing else here is. If you have never mixed a wash, mix this one.
Cobalt, the strongest thing in the room
Blue, and far more of it than you expect. In a glaze, a quarter of one percent already reads blue. In a wash that means a small fraction of what you would use of iron. Cobalt carbonate disperses more evenly than the oxide and is the easier of the two.
Copper, the one that depends on the kiln
Greens and turquoise in oxidation. In reduction it is the basis of copper reds. Above cone 8 it turns volatile and can travel to the pot standing next to it, which is somebody else's pot.
Manganese dioxide, for purple to near black
Browns, purples and blacks, and largely unbothered by whether the kiln is in oxidation or reduction. It also gasses as it decomposes, so a thick coat can pinhole the glaze above it. The one to be strictest with your mask about.
Rutile, for movement
Titanium with iron in it, giving tan and amber with a visible variegation nothing else produces. It is a mined mineral rather than a refined chemical, so one bag differs from the next. Test each new bag.
Mason stains, the predictable option
Pre-made colorant blends. What you mix is roughly what you fire, in a far wider color range than the raw oxides reach. Less surprising and less alive, which is the trade. Check the maker's temperature range before you commit a pot.
The flux, which is not optional
A small addition that melts and glues the colorant to the clay. Skip it and the fired wash dusts off on your hands. Which flux depends on temperature: frit for low fire, Gerstley Borate through the middle, feldspar up high.
Water, and how much
Watercolor consistency is the target everybody describes and nobody defines. In practice, around two and a half times the dry weight in milliliters gets you there. Thinner flows further into fine lines. Thicker sits heavier and wipes back harder.
Oxide or carbonate?
Many colorants come both ways. The carbonate is lighter and finer, disperses more evenly and suspends better, but it carries less metal per gram, so you need roughly twice as much for the same color. Carbonates also give off gas as they fire, which is worth knowing if you are putting a glaze over the top. For washes, most people reach for the carbonate.
The batch calculator
Pick what you are making and how much of it. This splits the dry weight between colorant and flux and tells you how much water to add.
Red iron oxide with Gerstley Borate.
What that gives you
Our range. Gerstley Borate gives a soft, diffused edge and glues the wash to the pot so it does not dust off.
The forgiving one. Warm browns and ambers, and it behaves at every temperature. Start here.
These are starting points, not laws. Every wash gets judged on a fired test tile, and a batch you like is worth writing down.
Mixing a batch
Ten minutes, a scale and a jar. The only part that needs real care is the dry part.
- Mask on before the lid comes off. These are fine metal powders. A respirator with particulate filters, not a cloth mask. The mask stays on until the powder is wet and the bench is wiped.
- Weigh dry into a dry jar. Colorant first, flux second, both on the scale. The calculator above gives you both numbers.
- Stir the dry powders together before any water goes in. Wet lumps never fully break up afterward.
- Add water in stages and stir as you go. It is far easier to thin a thick wash than to thicken a thin one.
- Sieve it if you can, through 80 mesh. Not essential, and it is the difference between an even wash and one with dark specks in it.
- Label the jar. Colorant, flux, ratio, date. Two brown jars of unlabeled wash are indistinguishable and one of them is cobalt.
- Wipe the bench wet, never dry. A dry cloth puts the powder back into the air. Damp sponge, then rinse it out at the sink.
It will settle, and that is normal
Oxides are heavy and they drop to the bottom within hours. Stir every single time before you use the jar, right down to the bottom. A wash used off the top without stirring is mostly water, and the pot will show it.
Oxidation and reduction
The same jar of wash gives different colors depending on whether the kiln has plenty of oxygen or is being starved of it. Our electric kilns run in oxidation, so the left column is what you will get here.
Iron
Oxidation: warm browns, rust, amber. Predictable and very usable. Reduction: the same iron turns toward green and grey, and it is the whole basis of celadon. The single biggest color shift on this list.
Copper
Oxidation: green to turquoise, depending on the glaze beneath. Reduction: copper red, the ox blood the Chinese perfected, which is difficult and prized and needs the right base glaze underneath to appear at all.
Cobalt
Blue in both, which is why it turns up everywhere. In a high-magnesium glaze it can shift toward purple in oxidation and pink in reduction, but the plain answer is that cobalt is blue whatever your kiln is doing.
Manganese
Largely indifferent to atmosphere. Browns, purples and blacks either way. If you want a result that does not care which kiln it goes in, this is it.
Rutile
Oxidation: tan and amber with movement. Reduction: the same rutile goes bluer and more broken, the effect people call rutile blue. Its iron content is doing what iron does.
Chrome
Green in both, and worth mentioning for what it does to its neighbors. Chrome and tin together make pink, and chrome starts to fume just above the temperature we fire to, which can pink the tin-white glaze on the next shelf. Keep them apart.
What this means at The Clay Hole
Every kiln here is electric and fires in oxidation to cone 5½. So a recipe written for cone 10 reduction will not give you the color it promises, however carefully you weigh it. That is not the recipe being wrong. It is a different kiln. Read any wash recipe you find online for what atmosphere it assumes before you buy ingredients for it.
Putting it on and wiping it back
Applying a wash takes a minute. Wiping it back is the skill, and it is where the surface stops looking muddy and starts looking carved.
- Stir it, every time. Right to the bottom of the jar. This is the step people skip and then wonder why the color is weak.
- Flood, do not paint. Load the brush and push the wash into the texture. Neat brushwork is the wrong instinct here, because you are going to remove most of it anyway.
- Give it a minute. Bisque needs a moment to pull the water in. Wipe immediately and you lift it straight back out of the lines.
- Wipe with a damp sponge, rinsed often. A dirty sponge just moves color around. Rinse, wring, wipe, repeat. Each pass takes a little more off the high spots.
- Stop earlier than feels right. The fired result is lighter than the wet one. Wipe until it looks slightly too strong and it will land about right.
- Decide about glaze before you fire. A clear glaze over the top deepens the color and makes it glossy and food safe. Bare, it stays matte and stained, and it is not food safe.
- Over a fired glaze is a different technique. The wash sits on the surface instead of soaking in, catches only the texture, and comes out drier and more fragile. Worth doing deliberately, frustrating by accident.
What goes wrong, and what caused it
It rubs off after firing
No flux, or nowhere near enough. The colorant never fused to the clay. Remix with a proper flux addition and refire the piece; a wash is one of the few things that takes a second pass well.
The whole pot went dark and the carving vanished
Too strong a colorant, not enough wiping, or both. Cobalt and manganese do this fastest. Mix a weaker batch rather than trying to wipe harder next time.
It looks patchy and streaky
The jar was not stirred, or the sponge was not rinsed. Both put uneven amounts of colorant on different parts of the same pot.
Dark specks all over it
Dry lumps that never dispersed. Sieve the batch through 80 mesh and they disappear.
The glaze over it pinholed
Too heavy a wash underneath, especially with manganese or a carbonate. They give off gas as they decompose, and if the glaze has already sealed over the top, the gas has to come through it. Thinner wash, and wipe back further.
Nothing like the recipe promised
Check what atmosphere the recipe assumed. An iron wash written for cone 10 reduction is a celadon green there and a plain brown in our electric kilns. Neither is wrong; they are different firings.
The bench checklist
Left before you mix, right before you wipe.
Before you mix
- Respirator on before any lid comes off
- Colorant and flux both weighed, not guessed
- Flux matches the firing range
- Dry powders stirred together first
- Water added in stages
- Sieved, if a sieve is on hand
- Jar labelled: colorant, flux, ratio, date
- Bench wiped wet, never dry
Before you wipe back
- Jar stirred from the bottom
- Test tile going in the same load
- Wash flooded into the texture, not painted
- Waited a minute before the first wipe
- Sponge rinsed between passes
- Stopped while it still looked a bit strong
- Decided whether a glaze is going over it
The ones that come up at the tables
Do I really need the flux?
For anything that gets handled, yes. Oxide and water alone fires as a dry powder that rubs off on hands and marks whatever it sits next to. The flux melts just enough to fuse the colorant to the clay. If a glaze is going over the top it matters less, because the glaze does the gluing, but it still helps.
Can I put a wash on greenware instead of bisque?
You can, and it behaves differently. Leather-hard clay is wetter, so the wash bleeds and spreads and the wipe-back is much harder to control. Bone dry greenware drinks it instantly and gives you almost no working time. Bisque is where the technique is easiest, which is why nearly everyone uses it.
Is a washed pot food safe?
Not on its own. A bare fired wash is metal oxide on a porous surface. Put a food safe glaze over any surface that meets food or a mouth. On the outside of a mug, under a glaze, it is fine.
Why does my recipe not give the color it says?
Most published wash recipes assume cone 10 reduction, because that is where a lot of studio pottery writing comes from. Our kilns are electric and fire to cone 5½ in oxidation. Iron that goes celadon green in reduction goes warm brown here. Check the atmosphere a recipe assumes before you judge it.
How long does a jar keep?
Effectively forever, as long as the lid is on. There is nothing in it to spoil. It will separate hard, and some fluxes go a bit claggy at the bottom over months, but a good stir brings it back. Label the jar and it will outlast your interest in that color.
Can I mix two colorants together?
Yes, and it is where the good colors live. Cobalt with manganese and iron makes a deep black. Cobalt with rutile gets streaky and mottled. Just move in small steps and write down what you did, because a blend you cannot reproduce is a blend you will mourn.
What is the difference between a wash and an underglaze?
An underglaze is a finished product designed to stay exactly where you put it, in a color that fires close to what you see in the jar. A wash is raw colorant designed to be wiped back off, and it fires nothing like it looks wet. Different tools for different jobs.
Can I wash over a glaze that is already fired?
Yes, and you get a drier, more fragile surface that catches texture rather than sinking in. It needs a refire to fix it. Worth doing on purpose. Just do not expect it to behave like a wash on bisque.
Where to go next
Oxide and carbonate washes
The companion page, with more recipes and more on what each oxide does in a glaze rather than a wash.
Read it →Underglaze techniques, rules and firing
The other way to put color on before the glaze, and the one that stays exactly where you brushed it.
Read it →Carving and sgraffito
The surfaces a wash exists to show off. Carve first, wash second.
Read it →Making and logging glaze test tiles
The cheapest way to find out what a wash does before you put it on a pot you care about.
Read it →Food safe, what you can claim
Why a bare washed surface is not a food surface, and what a glaze over the top changes.
Read it →Clay Hole University
Every lesson and guide in one place, from first day at the wheel through firing and glaze chemistry.
Start there →The cheapest way to make a pot look made
A jar of wash costs a few dollars and lasts years. There is no expensive mistake available here and no firing to book. The only real cost is the ten minutes it takes to weigh it out properly and the one tile you fire before you commit a pot.
So mix a small batch of iron, put it on something carved, and wipe it back. If the surface suddenly reads from across the room, you have understood the technique, and everything else on this page is a variation on it.
Half colorant, half flux, water to watercolor, flood it on bisque, wipe the high spots back. Mask on for the dry part, stir the jar every time, test one tile first.
If you want every ingredient explained, what changes between oxidation and reduction, and the six ways it goes wrong, switch to the Comprehensive Version at the top of the page.
Which version would you like to print?
Videos come out as a labelled box with the title and channel to search for, since paper does not play them.