Techno File: Loss on Ignition Dave Finkelnburg

Appears in the February 2022 issue of Ceramics Monthly.

Loss on ignition, or LOI, gives valuable, but often underused information about ceramic raw materials. Here are some tips about why LOI matters and how ceramic artists can use it to their advantage.

Defining the Terms

Bound Water: Regarding a clay mineral, that water that, when it is driven off by the heat of a kiln firing, irreversibly alters the structure of the mineral. This is unlike free water that escapes from the clay at or below 212°F (100°C).

Carbonate: Any of a variety of minerals used in ceramic art that contains one or more carbonate (CO₃) ions. When heated, the carbonate decomposes into carbon dioxide (CO₂) gas and an oxide compound.

Coefficient of Linear Thermal Expansion: The increase or decrease in length per unit of length of a material when heated or cooled over a specified range of temperature. This property of
a ceramic material indicates how much it will expand or contract as it changes temperature.

Cone: A physical (rather than electrical) device that indicates the peak temperature that has been reached in a kiln firing at the time the cone is viewed. Each cone has a triangular pyramid shape and is formulated from ceramic materials so that it deforms at a specific temperature. The cone is placed in a kiln, upright but leaning at a small, specified angle. Cone manufacturers provide tables showing the temperature equivalent to a fully bent cone as a function of cone size and kiln heating rate. Cones referenced in this article are made by Orton (www.ortonceramic.com). 

Cones are given, rather than temperatures, in this article, because they’re easier to relate to and because chemical decomposition of minerals occurs over a range of temperatures depending in part on particle size and composition. The cones shown are the most accurate upper limits currently available and emphasize that both heating rate and temperature affect the decomposition.

Weight Loss Science

Just what is loss on ignition (LOI)? As the name implies, LOI is the loss (of weight) as a percentage of the total weight of a sample of a raw material when fired to a given temperature. Not all ceramic materials lose weight in a firing. Silica is a good example. Its LOI is zero.

When a material does lose weight in a firing, why does this happen? When heated to peak kiln temperatures, some minerals decompose. All clays, for example, are crystals of alumina and silica that contain some water in their crystal lattices. This bound water is only released at the elevated temperatures of a kiln firing. Kaolinite, the most common ceramic clay mineral, contains almost 14% water by weight. All of it will be driven off in a firing by the time the firing reaches cone 016. 

Very pure kaolins have LOIs that approach 14%. That’s why LOI is a better indicator of the purity of a kaolin than fired color. Kaolins mined in North America all contain some iron and titanium dioxide (TiO_²) that cause the kaolins to fire a bit off-white. The whitest-firing kaolin available in the US and Canada is Grolleg, imported from Great Britain. It has almost no titania, so it fires brilliantly white, yet it still has a LOI of about 12%. The lower LOI indicates it’s not pure kaolinite. Closer examination reveals the presence of a potassium feldspar. Small amounts of feldspars and quartz are often found in kaolins. The feldspar in Grolleg kaolin is neither good nor bad, but it does slightly affect the melting temperature, as well as the coefficient of linear thermal expansion, of glazes made using the material.

Talc also yields water vapor when fired. Pure talc contains about 9% water. It’s driven off by the time a firing reaches cone 05. Several minerals containing boron—ulexite, Colemanite, and Gerstley borate—also contain significant bound water that can evolve during a kiln firing.

Why does the temperature at which bound water is driven off vary between minerals? It’s simply because the strength of the chemical bonds between different minerals and the water bound in their crystal lattices varies from one mineral to another. That’s because the crystal structure and composition vary between the minerals. To break bonds of higher strengths and release bound water requires more energy, in this case higher temperature (see graph below).

Whiting is the finely ground mineral calcium carbonate (CaCO_³). In a firing, it breaks down into calcium oxide (CaO), which stays in the ceramic, and carbon dioxide (CO_²). The CO_² is vented out of the kiln. Dolomite, which has variable amounts of magnesium carbonate and calcium carbonate, is also a source of CO², as are magnesium, barium, and lithium carbonates. Almost all the LOI of importance to artists is from the loss of bound water, carbon dioxide, or sulfur.

Sulfur dioxide (SO_²) and some sulfur trioxide (SO_³) are produced when sulfur in the mineral iron pyrite (FeS_²) burns in an oxidizing kiln atmosphere. The combustion is mostly complete by cone 010. Pyrite is a common but minor contaminant of ball clays.

In the Glaze Lab

Understanding which ceramic materials lose part of their weight as vapor during a firing, and at what temperature, is particularly important in controlling the quality of glazed ceramic surfaces. Frits, feldspar, and silica/flint have no significant LOI. The LOI of pure whiting is about 44%. Pure magnesium carbonate has a LOI of over 52%. Dolomite is in between the two. There’s obviously a lot of carbon dioxide being given off when any carbonate is an ingredient in a glaze or clay body. If any of them is used in a glaze in combination with a low-melting-temperature frit or ingredient like Gerstley borate, the evolving vapor can be trapped in the glaze and cause pinholing, dimpling, or blistering of the glaze. By itself, Gerstley borate has a LOI of about 30%, all of it as vapor from bound water.

Water vapor given off when bound water decomposes from crystals of clay is also a potential source of glaze problems. The issues occur, of course, when a glaze melts at a lower temperature than the temperature at which the evolution of vapor from bound water is complete. 

Manufacturers of high-quality, raw ceramic materials measure the LOI of their products. If a supplier does not provide that information, it may be necessary to ask for it specifically. Where native materials are being used, it may be necessary to weigh a bone-dry sample, fire it, and weigh the sample again. LOI can then be calculated using the following formula:

the author Dave Finkelnburg is a studio potter, practicing engineer, and a regular contributor to Ceramics Monthly. He earned his master’s degree in ceramic engineering from Alfred University.