Everything You Need to Know About Cobalt Blue Glaze Recipes

Cobalt: The Bluest of Possibilities

Cobalt glaze recipes are in most potters’ repertoires because they produce beautiful blue glazes. And what’s not to love about that? But blue is not the only color cobalt produces. This colorant is actually quite versatile and can make glazes that run the gamut from green to purple, pink to blue violet, blues mottled with red, pink, and even an intense black.

In this post, an excerpt from the “Technofile” department in Ceramics Monthly, Dave Finkelnburg discusses the many possibilities that are possible with this multifaceted little colorant. Plus he shares some sweet cobalt blue glaze recipes. – Jennifer Poellot Harnetty, editor.

There are four major factors that can affect a glaze’s color: the clay and slip beneath it, kiln atmosphere, fired temperature, and the glaze composition, including the colorants.

Defining the Terms

Oxide: A chemical compound containing at least one oxygen atom as well as at least one other element. Oxides result when an element combines with oxygen. Some raw materials are used in oxide form (like black cobalt oxide) while others form oxides during the firing process. In a glaze firing the oxygen usually comes from air, though in a fuel-fired kiln it may also come from carbon monoxide and carbon dioxide.

15+ Tried & True Cone 10 Glaze Recipes

Get more than 15 high-fire glaze recipes when you download this freebie15+ Tried & True Cone 10 Glaze Recipes.

Carbonate: The compound of carbon and oxygen in a 1:3 ratio that gives a negatively charged ion. It is generally insoluble. When fired, the carbonate breaks down and CO2 goes up the chimney, leaving behind an oxide.

Valence Electrons: The unbound, shared electrons that move among atoms rather than moving within a single atom.

Mole: The base unit of measure for the amount of substance, either atoms or molecules. This unit is used in unity molecular formulas for glazes.

Cobalt Blue Glaze Recipes – Like a Slice From a Rainbow

When a fired glaze is exposed to light, the color of the glaze will depend entirely on which wavelengths of the light are absorbed by the valence electrons in the glaze colorants. The energy level of the valence electrons determines which photons it will absorb (absorb them all and you have black) or emit (emit them all and you have white). Emit only one wavelength and you have that color, just like taking a little slice out of a rainbow. Cobalt in a fired glaze usually absorbs all wavelengths of visible light except blue and this is why a cobalt blue glaze is blue.

However, two or more colorants in a glaze interact so that the wavelengths of light absorbed are different for the combination than for either colorant by itself. The interaction between atoms of one colorant, say cobalt, and another, say chrome, alters the energy level of the valence electrons of both elements. That is why we may add both cobalt (blue) and chrome (green) to get a glaze that is turquoise.

Flux elements such as sodium, stabilizers such as alumina, and even some glass formers also influence valence electrons. Because the glass formed in the glaze firing controls the interaction of the glaze elements, what you see in the mixed glaze is almost never the color you get in the fired glaze.

cobalt glaze recipes

More Cobalt Blue Glaze Facts

• Cobalt Oxide: CoO is 1.4 times stronger than cobalt carbonate

• Cobalt Carbonate: Co3O4 (mix of CoO and Co2O3)

• CoO and Co3O4, although mainly thought of as colorants, are also strong fluxes

• Melting point (oxide): 328°F (1805°C), not yet volatile even at 2552°F (1400°C)

• Strongest coloring oxide: 0.25% noticeably blue in a transparent base glaze, used in a wide array of decal inks, underglaze colors, body stains, and colored glazes

• Soluble in glaze melt, thus it has little or no opacifying effect

• Very fine particle size, gives more uniform color in glazes

• Atmosphere and firing temperature do not change the color

• Very dependable color results in both oxidation and reduction conditions, and in fast and slow firing

• Toxic (inhalation and ingestion)

finished cobalt glaze test tiles

Cobalt Blue Glaze in Practice

If we use lots of talc, dolomite or another source of magnesium in a cobalt glaze, a beautiful bubble-gum purple glaze can be the result! Magnesium oxide (MgO) shifts the wavelengths of light emitted from our fired cobalt glaze from blue to purple. Every mole of flux should include more than 0.2 moles of MgO to get purple. Make a line blend varying MgO content to test for the shade of purple desired. Lesser amounts of MgO will produce lavender, larger amounts combined with an opacifier will produce a strong grape purple.

Alumina and titania in a cobalt glaze will shift the fired glaze color from blue to green. Because a significant amount of alumina can be dissolved from the clay body by the glaze during firing, glaze thickness can cause the same glaze to turn blue (where thick) and green (where thin) on the same piece. Glaze layering can have a similar effect with layering of the cobalt glaze over a white glaze firing blue but the cobalt glaze alone firing green. Cobalt greens are invariably satin to matte rather than glossy glazes. These glazes are typically flux saturated and the matteness comes from precipitating crystals of the flux in combination with aluminum and silicon. The amount of titanium oxide used, either as rutile or titanium dioxide, influences the green color. While cobalt greens have been reported using as much as 7.5% rutile in a cone 9 glaze, 2% rutile is far more typical. Less rutile also helps avoid pinholes in the glaze. The two glaze examples given here, Reitz Green Glaze and Emily’s Purple, are both cone 9–10, but cobalt’s spectrum of effects can be seen at all temperatures.

cobalt glaze recipe

Black glazes are typically achieved using cobalt oxide or cobalt carbonate plus a blend of iron and other metallic oxides. Typical cobalt levels are between 1 and 3% and iron levels up to 9%. The total of all the colorant oxides need not be more than 10 or 11%. Cobalt should be used with care—it is expensive, and in thick applications, too much cobalt can make a glaze fluid enough to flow off the ware. Iron is not required to make a black glaze, but as an alternative, it is inexpensive, readily available, and non-toxic. One or more of the oxides of copper, manganese, and chrome are added in many black glazes. Glazes high in iron black tend to fade brown and glazes with high amounts of cobalt tend to fade blue over a white glazes. A simple black glaze can be made with 9% red iron oxide plus 2% cobalt. If other oxides are used, a good starting point is 4% iron, 2% cobalt, 2% manganese dioxide, and 2% copper oxide.

**First published in 2014
  • Neal L.

    I have been trying to achieve red, blue and yellow glazes i my electric kiln with not even a hint of success. the red glaze formulas turn brown; the blue formulas turn green ad the yellow glaze formulas turn a light beige. Can you suggest a solution?

  • Teresa A.

    This post illustrates one of my pet peeves on this site. When there are errors in a post, such as a misprinted glaze recipe, and they are corrected in the Comments section – the errors should be corrected in the article BEFORE it is re-posted.

    • Daniel B.

      I am a french amateur ceramist and I wanted to find a cobalt rose with tests feldspar + talc and a low dose of cobalt without success real. Can anyone give information about the cobalt rose.
      thank you
      sorry for my English

  • Martina F.

    I’m absolute beginner in glazing.. why easy, when it can be difficult.. so, I want to explore high transparent porcelain glossy glaze and mix in some cobalt oxide.
    How does that sound to you experts?
    It should be fired at 1240 C, cause I’ve used dark black clay schamott body.
    It’s supposed to be a cup so I just want a rim of that coup in cobalt blue glaze.
    safe or not?
    may thanks

  • Glenda T.

    Please address the disposal of cobalt glaze residue to meet Federal/state Department of Health and Environment hazardous material regulations.
    All educational institutions must follow these regulations.

  • David Z.

    Hi Charlotte, as you probably know, toxicity is a relative matter that depends: first, on human body reaction to material, the amount and /or the exposure time to the agent, and the way you are in contact with. For example, you can surely eat or touch Kaolin without a problem, but you cannot breath it at all. Toxicity of heavy metal oxydes and salts depend upon their degree of solubility in different media. Sweat – for example – contains little amounts of ureic acid that can help heavy metal ions to disperse and be absorbed through skin, and definitely they shouldn’t be eaten.

    When I wrote a book on ceramic materials, trying to know mineral toxicity I found in a Medical encyclopedia the following reference: when beer started to be sold, little amounts of cobalt salts were added to improve foam formation when poured. Later, medical research noticed that beer drinking people got a too big sized hart that caused heart insufficience and, lastly, death. Cobalt then, was totally eliminated from food and drinks. Unfortunately, I don’t have the source reference with me, but what is for sure is that heavy metal derivatives have a specific degree of toxicity and must be handled carefully. Hope this will help.

  • David Z.

    Hi Kathi, on january 4th, I offered a way to convert thr Reitz green glaze to a cone -6 formula. I am sorry I haven’t had the chance to try it but, certainly you could. Serch for information above and good luck.

  • David Z.

    Hi Natalie, some glaze formulations are quite sensitive when applied. As Donna stated on Jan. 9th. high amounts of magnesia introduced here as dolomite and talc) are needed to make cobalt components turn into a violet – almost pink – shade. The first problem you have is we don’t know the substrate (porcelain?) composition and its thermal expansion coeficient. Mole formula of this glaze is
    0.32 CaO
    0.46 MgO – 0.26 Al2O3 – 3.04 SiO2
    0.14 K2O
    0.08 Na2O

    with a thermal expansion coeficient of 6.5.

    The second problem is you have a 0.46 MgO glaze: this ammount is too high and provokes raw glaze to retract too much showing kind of a crackle and, in some cases to flake out when heated. This should be compensated by gerstley borate and bentonite presence, but it might not be enough, so addition of a gum will become necessary. When making your tests, put only two or three pieces on only one shelf to avoid a kiln catastrophe. Also, make some test pieces to find out how much your clay body (substract) shrinks when totally fired and ask someone to helping you calculate the thermal expansion coeficient of your clay body. That will give you more answers. Good luck.

  • Charlotte A.

    I have recently been hearing that cobalt is toxic, but have yet to hear any substantiating facts. How does cobalt effect the body? Does it remain in the body and accumulate like lead, cadmium or mercury? Or is it like barium which is present in food and water,absorbed and stored in the bones/teeth and excreted over time and the problems arise when a massive quantity of barium absorbed overwhelms the bodies natural processes, most typically in industrial accidents? At what percentages does a problem develop? Can you direct me to source material and studies on its’ toxicology?

  • Donna K.

    I really do not see how you can ‘convert’ a ^10 glaze to ^6 or the reverse. To change the melt you have to change the chemical compostion of the glaze and by doing so you now have a different glaze. You would for example have to increase the Gerstley a great deal and this would throw off the balance of all your other fluxes. I can see how you would keep a purple glaze purple by keeping the MgO at the proper level but I don’t understand how you keep the other characteristics the same (and then there is the entire other kettle of fish of the electric kiln rather than a reduction firing).

  • Kathi H.

    I like the Reitz green also, but is there a cone 6 oxidation receipe that will produce that same color?

  • Hi from Perth Australia (it is dry on this side of the continent…)
    I have tried the Emily’s Purple and fired in a L.P.G. kiln but the glaze actually spat off the pots and all over my shelves……… – it is the most beautiful ‘Pansy Purple’coloured glaze I have seen, I love it and treid it on some of my teapots with a turquoise glaze , the 2 look stunning together in funky teapots – but dont want to try it again because of the mess – do you have any clues as to why this might have happened

  • David Z.

    Sorry, I clicked something wrong.
    Nepheline syenite 61.1 %
    Whiting 5.4 %
    Gerstley borate 5.2 %
    Lithiun Carbonate 1.6 %
    Silica 23.7 %
    Bentonite 3.0 %

    Please try this. While alunina – silica ratio is close to 1 : 10 mat glaze may turn into a gloss one. Decrease silica amount a bit and try again. Good luck with this. Hope it helps.

  • David Z.

    Lowering glaze temperature is never an easy task. There are two things that must be very clearly understood: First: the mole ratio of alkaline oxydes determines the greatest part of the glaze appearance, so it shouldn’t be modified; Second: glaze “hardness is determined by the amount of alumina – silica ratio. When it is low, the glaze might run; when it is too high, the glaze won’t melt properly; when alumina – silica ratio approximates 1 : 10 glaze will turn gloss, if silica ratio is lowered enough it will become matt. So let’s take the REITZ GREEN GLAZE as an example. The mole composition for the given batch formula is as follows:

    0.28 CaO
    0.09 Li2O
    0.02 MgO – 0.86 Al2O3 – 4.03 SiO2
    0.15 K2O – 0.03 B2O3
    0.47 Na2O

    Of course, 0.86 mole of alumina is too high for a cone – 6 glaze, so I made it decrease to get the new formula,where I tried to respect alkaline oxide moles as much as I could

    0.34 CaO
    0.09 Li2O
    0.03 MgO – 0.59 Al2O3 – 4.28 SiO2
    0.12 K2O – 0.08 B2O3
    0.41 Na2O

    That gives me the following batch formula:

    Nepheline Syenite
    0.41 Na2O

  • David Z.

    Some useful answers. Material’s leaching depends always on the “hardness” of the conformed glass, Sonya. That renders your question difficult to answer; but as you have certainly seen, cone 10 recipes proposed here contain Gerstley borate and boron present within tenderizes glass structure very much so small amounts of acid present in food could attack the web and disolve the oxyde. Maybe a hard porcelain glaze could avoid this. Although I recommend you not to taking risks.

    The most commercial version of cobalt oxyde is Co2O3 with aproximately 71% metallic cobalt. It comes generally in a dark Oxford gray color, but it can even be brownish according to impurities and provenance, or black. So Janine, don’t worry about the oxyde color; in general you are getting the same stuff and it will work the same way.

  • Sonya P.

    I work in a community studio where we are urged not to use any cobalt glazes inside pots used for food. At cone 10, can cobalt in a glaze be leached and toxic?

  • Erin P.

    Apologies everyone for the missing info on the Reitz Green Glaze recipe! It has been corrected.

    The two specific missing items were:
    Ball Clay…..8

    editorial assistant

  • Amarilis A.

    Excelente artículo.

    ..Sólo comentar que al vidriado Reitz Green Glaze, pareciera faltarle datos. Si no es así por favor aclararmelo. Estoy interesada en probarlos.

    Un saludo y felicitaciones.

  • Jharnetty@ceramics.org H.

    Hello folks! For those interested in converting the cone 10 glazes to cone 6, check out this free download: Making the Switch from Cone 10 to Cone 6 Ceramic Glaze Recipes: A Little Knowledge of Ceramic Glaze Chemistry and Raw Materials Goes a Long Way. Here’s the link:


    You can download it even if you are a current subscriber!

  • Yael Z.

    cobalt carbonate chemical formula is CoCO2, not as you have stated. upon heating it releases CO2 to become CoO, cobalt oxide.

  • Genevieve N.

    you wrote : “Melting point (oxide): 328°F (1805°C), not yet volatile even at 2552°F (1400°C)”. This does not make sense at all! How can 328 F be 1805 C ?! And then 2552F be much less ?!
    Somehow this article would be easier to follow if there were not such mistakes, and also if all the ingredients were clearly listed as in the Reitz green recipe which has two blanks for 15 and 8.
    Can someone please proofread the articles before they are posted ?

  • William W.

    Re my previous post – as iron is an element it should read Iron has an atomic weight of 55.845 sorry if I caused any confusion.

  • William W.

    This might help Teresa
    In order to reduce the complexity of comparing and measuring quantities of elements and compounds a common unit is used called the Mole which is the number of fundamental units (atoms) contained in 12 grams of Carbon 12

    that is 6.022×10 to the power of 23 (602,200, 000,000,000,000,000,000u), and is called the Avogadro Number after the scientist Amadeo Avogadro

    A mole of any substance contains the same number of atomic mass units as 12 gms of carbon 12.

    One mole of lead = 6.022×10 to the power of 23 atoms of lead

    One mole of potash feldspar = 6.022×10 to the power of 23 molecules of potash feldspar.

    In everyday practical terms the Mole is that quantity of a substance whose mass in grams is the same as it’s atomic/formula weight.

    For example, iron has a formula weight of 55.845 u, so a mole of iron has a mass of 55.845 grams.

  • Kate B.

    Fantastic article. Thanks for consolidating all that information in one spot and tying it together so well. Both the general basics and the Co specifics are really helpful. You brought up things I hadn’t thought about in the glaze context.
    This one is definitely a keeper.

  • Carol M.

    As to the several comments about converting recipes to cone 6, there have been a number of good articles that have been made available through Ceramic Arts Daily. Click on their “midrange glaze recipes” button to get links to several articles.

  • I would also like to convert to cone 6 oxidation. “Back in the day” everyone fired cone 10-12 reduction, but today many of us use lower temp electrics for practicality and energy savings. I loved the technical explination but would like to see some cone 6 alternatives. Thanks

  • Anne M.

    I would love to know how you could convert the cone 9-10 recipes to cone 6. Thanks….. it’s great to have the technical stretch.

  • Barbara M W.

    Also like help on reformulating C 9-10 glazez to C 5-6. because I love the green/blue glaze given above.
    Niec to know the logic behind the results.

  • Anne M.

    I would love to know how you could convert the cone 9-10 recipes to cone 6. Thanks….. it’s great to have the technical stretch.

  • Rose H.

    Can someone clarify what the 15 and 8 values are for following the Nepheline Syenite the end of the Reitz Green Glaze recipe? Thanks.

  • Glenn M.

    >• Cobalt Carbonate: Co3O4 (mix of CoO and Co2O3)

    Co3O4 (Cobalt (II,II) oxide) may be what Cobalt Carbonate converts into during a firing, but CoCO3 is Cobalt Carbonate, not what was listed.

    With regards to previous comments, the melting points for cobalts oxides are: 895°C, 1900°C, and 1933°C, for Co3O4, Co2O3, and CoO respectively.

  • Edward S.

    very useful info; in the More Cobalt Facts section I’m assuming a “zero” has been left off the Fahrenheit temperature for melting point of CoO; should be 3280F. (?)

  • Eric S.

    thanks for this review. helped to clarify some questions I had. pretty technical, but nice to have some more education for those of us who are “community-trained”.

Leave a Reply

You must be logged in to post a comment.

Enter Your Log In Credentials
This setting should only be used on your home or work computer.

Larger version of the image
Send this to a friend