Back in April I received a collection of wood-fired test tiles from Ann-Charlotte Ohlsson and Anne Mette Hjortshøj (Bornholm, Denmark).  There are three sets, one from a small test kiln and one from each of the chambers of their larger two-chamber kiln (see Featured Image for this post).  In each set the tiles were ordered from “top” to “bottom” according to where they were stacked in the kiln.  As described by Ann-Charlotte, they use a variety of clay bodies formulated using both local and commercial clay components.  After laying out the full collection of tiles I decided to start by looking at one that was made from a Hasle-blend clay body and fired near the top of the first chamber in the large kiln.

This tile seems to have been stacked flat in the kiln and there appears to be a round wad-mark visible on what I am considering its bottom (the side that faced a shelf).  Some ash speckle is clearly visible on the top (the side was more open to the kiln atmosphere).  The thumbnail below links to a high-resolution image (approx 7MB) of the bottom side of a small piece that I broke off from the full tile to study:

The faint trace of a wad (I believe) is visible; the electron microscope image below was taken near the lower-right corner of the piece, outside the wad mark.  Looking at the full-size image (click on the thumbnail), I would say that the matte regions of the surface take on a predominantly brown-olive cast while towards the edges there are glossy areas tending towards purple-black.  The wad mark includes some tan-yellow patches that look as if they might constitute a sort of “crust” on the surface of the clay—I’m not sure what those are but I tend to associate that sort of feature with wood ash (was there any sawdust in the wadding?).

I’m only just starting to look at and think about these tiles but I wanted to share one particularly nice electron microscope image (thumbnail below, click on it for the full-size image approx 7MB):

This image was acquired using the FEI Magellan scanning electron microscope (SEM) in the Stanford Nano Shared Facilities of Stanford University.  (For those in the know, this is a secondary electron image at 5000x taken at 10kV in field-free mode with an ETD detector.)  You may notice a darkened square region near the center of the frame—this is an artifact caused by a kind of “burn in” caused by the sequence of steps I took while focusing this image.  For those new to looking at these sort of images, let me characterize it as something like a black-and-white aerial view of the fired clay surface at VERY high magnification.  The scale bar in the lower right-hand corner shows 10 microns, so the area shown in this electron microscope image is similar to the cross-section of a human hair.  I would encourage you to try to see it as a collection of crystalline blocks (of various shapes and sizes) embedded in a glassy surface.  As a very loose analogy this could be like looking down on a collection of icebergs, or as a somewhat more rigorous analogy you could think of mixing a bunch of Lego blocks in an epoxy resin and then letting it set.  My interpretation of this kind of image is based on the assumption that exposed clay surfaces are fluxed by sodium/potassium gases in the kiln atmosphere (produced by burning wood) and develop a thin “molten glass” layer at top temperatures of the firing.  When the firing is over and the pieces cool, this thin glass layer sets (hardens) but it is possible for various elements in the glass to crystallize within the glassy matrix.  Again as a loose analogy you can think about starting with boiling-hot water in which you dissolve a lot of sugar, then if you cool this down you would expect some sugar crystals to reform, and if you even imagine flash-freezing this mix you could imagine having small sugar crystals trapped in glassy ice.

Of course the molten clay surface is chemically rather complex—among other things we expect it to contain silicon, aluminum, magnesium, calcium, iron, sodium, potassium, and titanium (not to mention oxygen).  I think that a major part of understanding color formation on wood-fired clay surface has to do with understanding how various kinds of micro-crystals contribute to what we see by eye, and in understanding how details of our clay compositions and firing/cooling schedules influence the kinds of crystals that form.  In particular, it seems to matter a lot where the iron in the clay ends up—in what kinds of crystals, what the sizes of those crystals are, and how those crystals are clustered.  To give you just a quick glimpse of how we can start to study such things, the image below (shown at full scale, it is not super-high resolution) shows a color-coded composite of approximately the same frame as in the SEM image above:

As indicated by the legend, the coloring of this image indicates the relative amounts of several key elements in various parts of the field of view.  This type of “map” is created by an Energy Dispersive Spectrometer (EDS) tool incorporated within the SEM.  It is not super accurate but gives us at least some rough clues.  For starters, we see here that many of the large crystal “plaques” in the field of view are iron-bearing, while the background soup that they float in is (crudely speaking) some sort of silicate glass.  One key point I take from this imaging is that the iron-bearing crystals are relatively large and sparse on this tile surface.  Looking at reduction-cool surfaces from wood firings that show stronger iron red colors, one tends to see massive clusters of much tinier iron oxide crystals… more on that in another post!

Part of this work was performed at the Stanford Nano Shared Facilities (SNSF), supported by the National Science Foundation under award ECCS-1542152.