• Glazes

    Iron Saturate Glazes

    During the next few days I’ll be releasing a series showing how to create a glaze using Glazy and volumetric blending.

    The first step is familiarizing oneself with the glaze type. For this demonstration I’m interested in creating a Cone 10 Iron-Saturate Red microcrystalline glaze also known as “Kaki”, “Tomato Red”, and “Persimmon”. Some historically examples are Chinese Song Dynasty Ding Persimmon-glazed wares as well as many of Shoji Hamada’s works.

    For each search, Glazy shows Recipe Cards with photos as well as a Stull Chart. The Stull Chart has five major regions: Unfused, Matte, Semi-Matte, Bright/Gloss, and Under-fired. There is another area, Crazed, that overlaps the other regions. In the next step, I will create a Biaxial Test using the Stull chart as a guide.

    From the analyses of Iron-Saturate glazes in Glazy, it is not clear what the ideal amount of Silica and Alumina (and the Si:Al ratio) should be. So our first step will be to re-create the Si:Al Stull Chart with a prototype Iron-Saturate glaze. In an Si:Al grid we can only adjust the amounts of Silica and Alumina, so we must set in stone the other characteristics of the glaze. Looking at the analyses of recipes in Glazy, it is apparent that we will need a good deal of Iron as well as Phosphorus. For our fluxes, apparently some MgO is required. As an educated guess, or initial prototype Iron Saturate glaze will have variable Silica and Alumina, while the following are set for all glazes: KNaO 0.2, CaO 0.6, MgO 0.2, Fe2O3 0.22, P2O5 0.12.

    We will use volumetric blending to magically create 25 glaze tests from only 4 batches of glaze. The four corner glazes composed using the Glazy Recipe Calculator. The columns originate from the Origin so that each column represents a specific Si:Al ratio. It is decided to put the “Left” column in the Stull Matte Region, while the “Right” column is pushed close to the Under-Fired Region.

    Batches of 500 grams are created for each corner glaze, and the test glazes are mixed using volumetric blending with samples of 20mg. The tests are ready to be fired!

    Note: The “educated guess” for our initial prototype glaze is informed in large part by the amazing work of Carol Marians. Carol has posted 8 years of glaze research on her website at:

    I made a mistake on the Si:Al ratios for each biaxial column. The ratios should be approximately 4.3, 5.4, 7, 9.3, and 12.8

    Here are the results of the Iron-Saturate Biaxial. While the reduction firing with uncontrolled cooling results in brown, metallic surfaces, the oxidation firing with a controlled cool and hold at 1700°F (925°C) gives us more interesting results. The biaxial reveals that an Si:Al ratio of around 9-9.5 (the fourth column) promotes redder glazes. In particular, tile D4 (4th row, 4th column) seems promising.

    The oxidation firing schedule is adopted from Carol Marians, but simplified to:
    150°F/hr to 250°F (65°C/hr to 120°C)
    400°F/hr to 2050°F (200°C/hr to 1120°C)
    120°F/hr to 2250°F (50°C/hr to 1230°C)
    60°F/hr to 2290°F (16°C/hr to 1250°C)
    40°F/hr to 2310°F (4°C/hr to 1265°C)
    Hold of 10 minutes at 2310°F (1265°C)
    400°F to 1700°F (925°C) – Down-fire
    Hold 2 hours at 1700°F (925°C)

    See for many more examples of firing schedules for iron red glazes.

    From the Iron-Saturate Biaxial we choose tile D4 to work with.  We can now “zoom in” and refine the Si:Al ratio for this tile.  At Si:Al 8-8.5 the glaze seems more evenly covered in crystals.  As the Si:Al ratio is increased the coverage becomes more splotchy.

    It seems the Si:Al ratio for biaxial tile D4 was already pretty good.  Now we can move on to testing factors other than Si:Al in the prototype glaze.  In this test, we increase the level of R2O (KNaO, or K2O & Na2O) while decreasing the amount of Calcium.  I was surprised by the result for 0.3 KNaO, perhaps there would have been a better result if both Calcium and Magnesium were decreased?  Or decrease Si & Al? Anyway, based on this test I’ll just stay at R2O:RO 0.2:0.8

    Test of Iron-Saturate Biaxial tile D4 replacing Mahavir Potash Feldspar with Minspar 200 Soda Feldspar.  Not a 1-to-1 percentage replacement, but maintaining the same UMF (except K2O and Na2O).

    Now that we’ve established an R2O:RO ratio of 0.2 using Potash Feldspar, we can test the best proportion of Calcia to Magnesia.  With our R2O set at 0.2, 0.8 remains for the RO (including Calcia and Magnesia) portion of our UMF fluxes.  The educated guess of 0.6 CaO and 0.2 MgO in our original Iron Saturate biaxial turns out to be a good choice.  More than 0.2 MgO also gives some interesting glazes with a more metallic surface.

    Same UMF, different sources of Magnesia.

    Glazes are often split into two parts:  A base glaze and additives like colorants and opacifiers.  It’s like ordering a pizza (base) with toppings (additives). We’ve already tested many aspects of the Iron-Saturate base glaze including Silica:Alumina, R2O:RO, and Calcia:Magnesia.  Now we can move on to the additives: Iron and Bone Ash. (The decision between what is part of the base recipe vs. an additive is somewhat arbitrary.  Just as with a pizza that’s made it to your stomach, it all eventually ends up mixed together.) Additives are added in addition to the base glaze recipe.  So in this test, I did not alter the base glaze at all, the only difference is increasing Red Iron Oxide and Bone Ash. I was surprised to see that additional bone ash didn’t alter the glaze a lot more.

    A line blend of biaxial test D4 blended with the same recipe without Bone Ash (but maintaining the same fluxes to account for the missing CaO from the Bone Ash). Without P2O5, our glaze is a nice tenmoku.  At 0.03 P2O5 very faint traces of crystallization appear.  At 0.06 P2O5 crystallization is much more evident, and somewhere between 0.06 and 0.09 P2O5 there is a dramatic transformation.

    Testing different sources of iron using tile D4 from the Iron-Saturate Biaxial.  I’m not sure what’s going on with Yellow Iron Oxide.  It would be interesting to see other sources of iron, especially iron phosphate.

    Adding Titanium Dioxide in 1% increments to tile D4 from the Iron-Saturate Biaxial. Titanium Dioxide is often present in analyses of Song Dynasty Russet/Persimmon glazes as well as Japanese Kaki glazes, but usually in amounts of less than 1%.

  • Traditional

    Traditional celadon

    Recent firing with traditional porcelain stone glaze.  In the past I’ve tried but failed to use modern materials like feldspar and kaolin to capture the beautiful, unctuous surface and depth of porcelain stone celadons.  In this glaze the coloration is completely due to iron occurring naturally in the material.

  • Glazes


    A Jun glaze on stoneware from the last kiln

  • Glazes

    Qing Dynasty Test Tiles

    Spotted in a friend's book. Unfortunately I forgot to record the title.

  • Glazes

    An old Porcelain Stone Mine

    It’s surprising to me how often archaeological discoveries seem to be made in Jingdezhen, but then I remember that wherever I walk in this place there are deep layers of shards beneath my feet.

    A friend of mine was given samples from a recently found porcelain stone mine dating from the Five Dynasties Period.  Apparently the find has not gone unnoticed- professional antique makers have been secretly mining the site.  Luckily we have the chance to acquire some of this porcelain stone.

    I’m often dealing with unfamiliar, traditional materials of which chemical analyses are lacking or unreliable.  In these cases, I usually create a series of line blends to get a basic idea of what I’m working with.  From those first tests, one can further refine glazes using more line blends and triaxials.

    For this porcelain stone I created the following initial tests:

    • Pure porcelain stone, crushed, milled and sieved.
    • Porcelain body using porcelain stone and kaolin at 15-45%.
    • Lime-fluxed celadon glazes:
      • Porcelain stone and 10-20% Er Hui (Glaze Ash)
      • Porcelain stone and 10-20% Wollastonite
      • Porcelain stone and 10-20% Whiting

    Idealized “traditional” recipes are also based on two-component mixtures.  For glazes, porcelain stone was mixed with a flux like glaze ash.  For porcelain bodies, porcelain stone was simply mixed with a proportion of kaolin.

    Usually a single line blend of either Whiting or Wollastonite could tell you a lot about a porcelain stone.  However, porcelain stone mixed with Glaze Ash or Whiting often results in fuming/carbon trapping, so I wanted to test each flux separately.  I usually also create Dolomite or Talc tests.

    I also prepared two sets of test tiles for cone 10 and 12 firings.

    Stones of all types can be used in glazes.  Joseph Grebanier’s Chinese Stoneware Glazes lists many recipes that use locally sourced granite.  And Brian Sutherland’s Glazes from Natural Sources contains a wealth of information on the subject.

    A sledge hammer is used to break off pieces of the hard porcelain stone

    A very hard mortar and pestle is used to further break down the porcelain stone.

    The crushed porcelain stone is ball milled for four hours.

    Even after milling, the mixture needs to be sieved.

    After sitting and decanting excess water, the mixture is dried on a plaster slab.

    Because I’m in a hurry, the mixture is further dried on the stove.

    Porcelain body tests with increasing proportions of kaolin.

    Fired porcelain body tests.

    Fired porcelain glaze tests with increasing proportion of flux.

  • Glazes

    Tea Dust Glaze

    Tea Dust Recipes on Glazy

    The following glazes and more can be found on my new website, Glazy:

    The Complete Guide to High Fire Glazes Tea Dust Recipes

    Some tea dust glazes from The Complete Guide to High Fire Glazes.  I fired these glazes according to my usual firing schedule, probably too hot and not ideal for the development of crystals necessary for good tea dust glazes.  I crash cool to 1000C and then completely shut up the kiln.

    Currie 10 Tea Dust

    Stoneware, Reduction Cone 10

    Porcelain, Reduction Cone 10

    Porcelain, Reduction Cone 11

    Currie 11 Tea Dust

    Stoneware, Reduction Cone 10

    Porcelain, Reduction Cone 10

    Porcelain, Reduction Cone 11

    Stoneware, Reduction Cone 10

    Porcelain, Reduction Cone 10

    Porcelain, Reduction Cone 11

    Stoneware, Reduction Cone 10

    Porcelain, Reduction Cone 10

    Porcelain, Reduction Cone 11

    Stoneware, Reduction Cone 10

    Porcelain, Reduction Cone 10

    Porcelain, Reduction Cone 11

    Chinese Traditional Tea Dust Glaze

    Although the glazes above call all be considered tea dust, in China teadust glaze usually refers to Qing dynasty wares like the ones below.  Although tea dust can result in a range of colors and with varying crystal sizes, usually the crystals are very evenly dispersed.  (The example below is a Qianlong vase sold through Sotheby’s.)

    Qianlong Teadust Vase

    Various Teadust Colors

    Chinese Traditional Tea Dust Glaze Recipes

    The recipe above comes from 颜色釉 (Colored Glazes), published in 1978 by the Jingdezhen Ceramics Company.  The components of the recipe can be difficult to find today, even in Jingdezhen.  Those materials that are still produced probably have different compositions than in 1978.  For instance, I’ve noticed differences in Glaze Stone (釉果) and Zijin Stone (紫金石) from year to year.  What this recipe has in common with the Western tea dust glazes above is the addition of magnesium oxide in the form of talc, which helps form the tea dust crystals.

    The glaze preparation is quite interesting.  The ingredients (excluding the white clay (白土)) is ball milled for 30 hours, then white clay is added and the full glaze is milled for another four hours.  After adding water, the glaze is sprayed approximately 1mm thick.  Firing is to 1250°C in a weak reduction atmosphere.

    Tea Dust Triaxial - First Attempt

    In part because I’ve had a difficult time finding a reliable chemical analysis for the glaze stone I use, but mostly because I often don’t have much clue what I’m doing, I often use triaxial blends to find interesting glazes.  I’m also interested in making glazes for my normal firing schedule, rather than changing my firing for a single type of glaze (like the recipe above that fires to 1240°C).

    Because triaxials are limited to three variables, the following tests make some assumptions about tea dust glazes.  The top axis of the triaxial will be glaze stone (the “unknown” ingredient), the left axis whiting (the flux), and the right axis silica.  I’m certain I need magnesium in the form of talc, dolomite, or magnesium carbonate- I choose 8% talc.  I also need some iron, based on past experience and other glaze recipes I choose red iron oxide at 8%.

    From the resulting triaxial it seems that crystal growth improves as silica is increased while glaze stone and whiting are decreased.

    Tea Dust Triaxial - Second Attempt

    For each row of the first triaxial, I like what’s happening in the third column.  For the next test, I extend this third column down a few rows to see what happens.

    Extended Triaxial. Left- Porcelain, Right- Stoneware

    It’s nice to see my educated guesses are working out.  Beginning from the fourth row of this second test, it seems like I’m coming close to a glaze that I like.

    You’ll notice that the glazes look completely different on the two types of test tiles (porcelain on the left and stoneware on the right).

    Just to check that my initial amounts for Iron and Talc are in the right ballpark, I choose the fourth row of this test and make some adjustments.  It seems that Red Iron Oxide at 8% additional and Talc at 10% are ideal.

    After a few more tests further adjusting the silica, iron, and talc I ended up with the glaze at the left.  While it’s not an ideal tea dust glaze by Qianlong standards, what’s important is that I like it and I can fire it using my usual temperature and schedule.

    Photomerged microphoto of the same test tile.

    The teadust glaze on a small bowl

    August 2015 Update

    Just to see what would happen if I continued adding silica to the recipe, I extended the bottom-right corner of the original triaxials.  As you can see below, incrementally replacing glaze stone with silica leads to an increasingly uniform crystal distribution more closely approximating Chinese teadust glazes.  (A feldspar/whiting/silica triaxial for some of the High-Fire Glazes recipes above may have similar results.)  Note that in the following tests I increased Red Iron Oxide to 9% (additional).

    Second triaxial extension for teadust test glaze

    Complete tested triaxial for teadust glaze.

    Microphoto of the extended triaxial's bottom-right test

    Adjusting Coleman Tea Dust Black

    As I did with the traditional tea dust glaze above, I wanted to adjust one of the tea dust recipes in John Britt’s High Fire Glazes to increase coverage of the crystals.  The most promising candidate I had found was Coleman’s Tea Dust Black.  Starting with that recipe, I designed a triaxial that incrementally increased silica while reducing feldspar (also lowering the alumina).

    The original recipe for Coleman Tea Dust Black:

    Custer Feldspar 39.81
    Silica 25
    Whiting 15.74
    Kentucky OM #4  12.04
    Talc, Magnesium Silicate 7.41
    Red iron oxide, RIO 9.3
    Total: 109.3

    Since I can only change three variables in the triaxial, I choose to modify Custer Feldspar, Silica, and Whiting.  I keep the Ball Clay at 12%, raise Talc slightly to 8% (which was a good amount in my traditional tea dust glaze) and set Red Iron Oxide at 9.5%.

    Triaxial adjusting Coleman Tea Dust Black. Porcelain, Reduction Cone 10

    The most instructive column is the last, with Whiting set at 14%.  As silica is increased and feldspar decreased, crystal coverage becomes more even.



    The above triaxial was slow-cooled.  Below is a comparison of the same glaze with a crash-cool to 1000°C.

    John Sankey has an excellent article about iron glazes and cooling:


    Teadust glaze with slow cooling. After reaching cone 10, immediately close up kiln entirely.

    Teadust glaze in fast cool to 1000°C. After reaching 1000°C, completely close up kiln.

    I added the recipe for this glaze on Glazy at

    Potash Feldspar: 28%
    Silica: 40%
    Whiting: 12%
    Ball Clay: 12%
    Talc: 8%
    Red Iron Oxide: 9.5%

    Actually, this glaze is probably too high in silica, and the resulting stony surface not at all like the satiny Chinese teadust glazes.  I will try again, increasing the talc and adjusting iron levels rather than simply adding silica.  Ideally I would also attempt different firing/cooling schedules, but I won’t change my firings just for a single glaze.

    I hope this article shows how to refine a glaze using triaxial testing.  With some general knowledge about the type of glaze, educated guesses, experience and a bit of luck, one can design triaxials that reveal interesting new glazes.

    Note:  I did not cover glaze limits in this article.  There are various glaze limits that describe “good glazes”.  For instance, on the Glazy page for the triaxial-derived Tea Dust glaze above (, you will notice that the glaze sits just outside the Hesselberth & Roy Δ9-10 Glaze Limits.  It is up to you to ensure your glazes are functional, especially if they contain oxides that are toxic when leached.  However, I find that a lot of interesting glazes are outside of established glaze limits.

  • Glazes

    Porcelain Stone Glazes

    Jingdezhen Porcelain Stone

    There are a number of types of porcelain stone mined throughout Jingdezhen and the surrounding countryside.  Some are more suitable for making porcelain clay, while others are traditionally used for glazes.  It is difficult to know how similar modern-day porcelain stone is to traditional materials.  During the few years I have lived in Jingdezhen, some mines have been closed off to private mining, while others have simply run out of material.  Those still operating often mix poor-quality material with good material in order to increase production.  And plaster is added in ever increasing amounts in order to make the porcelain stone bricks less likely to break in transit.

    Below are three types of porcelain stone fired to 1310 Celsius in a reduction atmosphere.  From left to right:  San Bao porcelain stone, Yaoli glaze stone #1, Yaoli glaze stone #2.

    Most porcelain stone is made from a combination of rocks.  The stones below are used to make Yaoli glaze stone.  On the left is the more common, dirtier stone, in the middle is the higher quality stone, while the right is the mixed, washed and purified stone before adding plaster.  All examples were fired to 1310 Celsius in reduction atmosphere.

    San Bao Porcelain Stone (三宝瓷石)

    Below are some simple tests of porcelain stone from San Bao Village (三宝瓷石).  This stone is often used for making traditional porcelain clay, but it can also be used for glazes.  (However, usually “glaze stone”, or 釉果, is used for glazes.)

    The only chemical analysis I have found for San Bao Porcelain Stone comes from 陶瓷艺术釉工艺学 published by 江西高校出版社:

    SiO2: 7.013, Al2O3: 17.64, Fe2O3: 0.69, CaO: 0.54, MgO: 0.09, K20: 4.02, Na2O: 4.68.  LOI 2.01

    (The SiO2 amount is most likely a typo, they probably intended to write 70.13.)

    However, this analysis can only be used as a basic guideline.  There are noticeable differences between the porcelain stone produced by various families in San Bao, and quality seems to change every year.

    Traditional glaze recipes usually call for whiting, but unless fired very carefully, porcelain stone has a tendency to carbon-trap when whiting is the flux.  Replacing whiting with wollastonite eliminates this problem.

    Below are a few glazes containing San Bao porcelain stone, silica, wollastonite and kaolin.  It is quite easy to make a very nice celadon with porcelain stone.

    Yaoli Glaze Stone (瑶里釉果)

    Yaoli Glaze Stone (瑶里釉果) is traditionally used for creating glazes.  As with San Bao porcelain stone, there are a number of families mining, cleaning, and creating the glaze stone bricks sold in Jingdezhen.

    陶瓷艺术釉工艺学 has the following analysis:

    SiO2: 73.99, Al2O3: 15.55, Fe2O3: 0.37, CaO: 1.76, MgO: 0.33, K20: 2.88, Na2O: 2.63.  LOI 2.88

    But again, the material varies from seller to seller and from year to year.

    At the left is a simple melt test of 75% glaze stone with 25% Wollastonite and 80% glaze stone with 20% Wollastonite.  The 20% Wollastonite version is already a perfectly useable celadon.

    Below are a few simple celadons made with glaze stone, silica, wollastonite, and kaolin.

    Porcelain Bodies

    A test of 80% porcelain stone and 20% kaolin (New Zealand halloysite).  This porcelain is particularly white and fairly translucent.  It is quite nice to throw but fragile when bone dry.  Mixtures of various types of porcelain stone with between 10-40% kaolin produce porcelain bodies suitable for a range of temperatures.