Technofile: Spodumene

Spodumene is a naturally occurring mineral source of the flux element lithium and is commonly referred to as lithium feldspar. It’s lithium that lowers glaze melting temperature and dramatically brightens such colors as cobalt blues. Using spodumene is not without its challenges, however.

Sourcing Lithium

Potash feldspars may contain, in total, more than 14% sodium and potassium oxides by weight. At 7.7% lithium oxide by weight, a familiar commercial spodumene appears to contain much less flux. Such thinking, of course, is entirely wrong.

The most common error made in the formulation of glazes containing lithium in any form is to overlook the critical fact that lithium weighs less per atom than any other flux element on earth. Since fluxing power is due to the number of flux atoms present in a glaze, rather than the weight of those atoms, the challenge of lithium is to avoid adding too much of it.

To visualize this in action, consider simply replacing a potash feldspar in a glaze recipe with an equal weight of spodumene. What will the result be? First, the new recipe will have more alkali flux than before. Recall that lithium, sodium, and potassium are all alkali flux elements.

The recipe will also have more alumina and less silica than before. To bring the glaze recipe back close to what it was originally requires adding silica (sometimes called flint) and reducing alumina by using less clay.

However, of possibly greater importance, the coefficient of thermal expansion (CTE) of the glaze will be sharply lower. As a result, the glaze may shiver off rims and other sharp edges of ware after firing.

The point, of course, is that like many other glaze ingredients, spodumene cannot simply be substituted for another feldspar in a glaze recipe on a weight-for-weight basis without changing the fired result.

In the case of spodumene, careful use of glaze calculation software is the most efficient way to bring lithium into the glaze recipe without producing undesirable changes to the fired glaze.

Don’t Settle For Less or Pay Too Much

Some spodumene sources produce a relatively coarse product. Ideally all glaze ingredients would pass through a 200-mesh sieve. It is wise to at least dry sieve a sample of a new batch of spodumene through an 80-mesh sieve. Any spodumene that will not pass through that sieve would settle out quickly in the glaze bucket, so it should be either discarded or set aside to be ground fine enough to use.

A number of other sources of lithium have been used in glazes and clay bodies including frits, petalite, lepidolite, and lithium carbonate. However, spodumene is easily the most affordable lithium source available.

In a studio situation where a mixed glaze may remain in a tightly closed bucket for a long period of time, lithium carbonate may present serious problems if one is twice-firing. The first firing, of course, burns organics out of the body, drives off water, and strengthens the ware while commonly leaving it sufficiently porous to absorb water from the applied glaze.

The problem with lithium carbonate is it is somewhat soluble in cold water. Left to sit long enough in the glaze slurry, some lithium carbonate will dissolve. The lithium in solution then migrates, in the evaporating water from the glaze slurry, to the surface of the ware as it dries (see diagram at left).

That creates a local concentration of lithium at the glaze surface, which can be high enough to cause shivering at that point. In extreme cases of lithium carbonate dissolving in a glaze, both crazing and shivering have been reported on the same fired piece!

The fact that lithium lowers the CTE of a fired glaze can be an advantage, however, as long as the lithium remains undissolved. Small amounts of spodumene, which contains lithium in a less soluble form than lithium carbonate, are frequently added to glaze recipes to lower glaze expansion and reduce or eliminate glaze crazing.

Recipe Comparison

Glazes that have been designed to fit stoneware clay bodies will typically craze on porcelain clay bodies fired to the same temperature. Likewise, glazes that do not craze on porcelain clay bodies will typically shiver on edges of ware made from stoneware clay bodies fired to the same temperature. Small replacements of the potash feldspar in a glaze recipe with spodumene on a one for one basis have a significant effect on glaze fit. The following calculations illustrate this.

The 4321 glaze has a calculated SiO2:Al2O3 ratio of ~9.38 and a CTE of 7.29. A direct substitution of spodumene for the potash feldspar in the recipe (40% spodumene) produces a calculated SiO2:Al2O3 ratio of ~7.06 and a CTE of 5.53. The silica to alumina ratios are in a range where both glazes will be glossy and any fired difference will not be visible. The spodumene version, however, with its very dramatically smaller CTE, may shiver seriously off any sharp edge.

I’ve used 4321 quite a bit at cone 10. When using it with porcelain,  I have substituted talc for some of the whiting as well as adding a small amount (5%) of spodumene to reduce the CTE. The CTE of porcelain bodies has a wide range, affected both by the amount of silica in the recipe and the firing temperature (degree of vitrification). The point of the tests is not to show that spodumene causes shivering, because it doesn’t unless one uses an excess of it, but rather that from this extreme example it is clear spodumene has a significant effect on lowering glaze expansion and in small, calculated additions can improve the glaze fit of an otherwise crazing glaze.

Subjecting the test tiles to several consecutive cycles of cooling in a freezer for a half hour or so followed by being gently placed directly into boiling water should prove that point. This will be more apparent in glazes with dark colorants fired onto a light stoneware body.

Have a technical topic you want explored further in Techno File? Send us your ideas at

Subscriber Extras: Archive Article

Click here
to read the archive article, “Spodumene for Low Thermal Expansion,” by M.J. Murray, which originally appeared in the December 1995 issue.


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