Of all the different types of possible glaze defects, the defect known as crawling probably has the greatest number of potential contributing causes. Why is that and what can you do to prevent it?
Cause and Effect
Crawling is characterized by the appearance after firing of generally irregularly-shaped bare or almost bare patches of the underlying clay body, surrounded by a somewhat thicker outline of the glaze (shown in the images below). While there are a lot of factors that can contribute to the occurrence of crawling, there is one predominant cause, a poor bond between the dried glaze layer and the clay beneath. If the adhesion is poor, when the ware is heated and the glaze melts, instead of adhering to the clay, the glaze tends to adhere to itself (because of the surface tension of the glaze) and pulls away from the area of low adhesion with the clay.
Crawling is not caused by the firing, but only occurs during the firing as a result of pre-existing conditions. However, diagnosing the exact cause of crawling can be difficult because of the fact that anything in the processing of the ware before firing that results in a weakening of the bond, or a failure of a good bond to develop, can contribute to the occurrence of crawling. These factors can include composition of the glaze, application of the glaze, and drying and subsequent handling of the glaze. Furthermore, because the boundary between the dried glaze layer and the clay is hidden from view, the existence of a weakened bond is not visible early in the process.
If a glaze recipe contains too much clay or any other very fine or fluffy material that has the potential to absorb a lot of water, as the wet glaze dries, it can shrink sideways along the surface of the clay, as well as in thickness. This sideways shrinkage movement results in a weakening of the adhesion between the glaze layer and the clay. Alternatively, if a glaze recipe does not contain enough clay or other fine powders, but only relatively coarse or granular ingredients, there may not be enough fine material to create good contact and a good bond between the dried glaze and the clay.
If a glaze slurry is not well mixed and all the ingredients are not well suspended in the liquid before application of the glaze, the portion of the glaze that is used may contain mostly fine ingredients (that haven’t settled out), resulting in greater than normal drying shrinkage and sideways movement.
Any loose debris or dust present on the surface of the clay before glaze application, as well as the presence of grease, oil, or any other water-repellent material, can reduce the bond. In order for a good bond to form, the water in the glaze has to wet the surface of the clay and be easily absorbed by the porous clay, pulling the solid ingredient particles into close contact with the clay surface. Wiping the bisqueware with a damp sponge just before glazing can eliminate dust and promote wetting.
If the surface of the clay has been smoothed by techniques such as burnishing or the application of fine-grained slips, the smooth surface can contribute to the failure of a good bond to form. This same principle is well known in the field of adhesives where it is frequently recommended that surfaces to be glued together be roughened slightly before joining (to increase the contact area and “interlocking” of the surfaces).
If a glaze is applied too thickly, or the porosity of the clay has been reduced by a higher-than-normal bisque firing, and the water cannot be absorbed rapidly by the clay, the slower drying can lead to sideways shrinkage, loosening the original bond.
If the inside and outside surfaces of a pot are glazed separately, the water absorbed by the clay from the first application, which is still present in the clay, can prevent rapid absorption during the second application, resulting in a failure to form a strong bond. Alternatively, if the glaze from the first application has completely dried, the water that is absorbed by the clay during the second part of the application can soak through the wall and loosen the bond of the first layer, especially if the wall is thin. A similar situation can arise when glazes are layered. The water that is absorbed from the second application can soak through the underlying glaze layer and weaken the adhesion of the underlying layer, resulting in crawling of the bottom layer, carrying the upper layer along with it. Yet another related situation occurs when thin-walled ware is glazed by dipping. Water is absorbed from both sides of the wall and can very quickly saturate the wall, reducing further water absorption and preventing tight contact of any adsorbed glaze layer.
Base glazes that are naturally viscous (stiff) when melted, or the presence of excessive amounts of opacifiers or other additives that tend to increase the viscosity of the glaze, can result in poor wetting and adhesion of the melted glaze to the clay, and thus to crawling. Poor adhesion can also be caused by the presence of thick coatings of refractory (non-melting) underglazes and stains such as chromium oxide and cobalt oxide on the surface of the clay which prevent the melted glaze from sticking to the surface.
The important principle is that crawling is actually caused by conditions set up before, during, or right after glaze application. The solution is to examine the glazing process for any of the possible conditions described and make appropriate changes in the recipe or procedures.
As an example of a glaze recipe change, calcined kaolin can be substituted for part of the high kaolin content in a glaze recipe (86% as much calcined kaolin is needed to replace raw kaolin) in order to reduce drying shrinkage. When clay is calcined (heated to approximately bisque firing temperatures), bound water is driven off and the clay loses the ability to absorb large amounts of water and shows reduced drying shrinkage. Generally, only part (roughly 50%) of the raw clay is replaced, in order to retain the contribution of the fine, uncalcined clay to the formation of good adhesion, although this depends upon the actual amount of clay that is in the recipe. It is advisable to retain at least 10% of the total recipe as uncalcined clay.
the author Phil Berneburg is a professor of ceramic arts at Hood College in Frederick, Maryland, where he specializes in teaching the scientific background and technical problem-solving in the Certificate, MA and MFA programs. He is also a ceramic engineer and studio potter.