Steve Harrison investigates native porcelains developed from Australian Bai-tunze of the Southern Highlands of Australia.

This article details my recent research in the use of local materials – specifically, woodfired porcelain and proto porcelains made from the native porcelain stone (bai-tunze) of the old Mittagong Shire. During the development of this work I intended to make a 100 per cent local product. In this regard I have to admit that I have failed, as one of the clay bodies contains one per cent of bentonite sourced from outside the Mittagong shire. In a time when international cultural boundaries are being smudged or collapsed by free trade agreements and globalisation, one result is that simple consumer items such as ceramics are conceived in a design studio in Europe or the US and produced in a factory in China, then shipped to every first world country in the developed world. The result is less choice, less individuality. This work is my attempt to produce a fully local product – with all its limitations and faults, and all its local character. The French have a word, ‘terroir’ that expresses some of this quality. Better porcelain clays can be bought from the local art supply shop. However, they are made from mostly imported ingredients, from all over the world, New Zealand, USA, Europe, Canada and China, at least the water used is Australian. The net result of this international roast is that everyone’s work has a sameness about it. What was once exceptional has now become commonplace. Globalisation has resulted in a situation where we are spoilt by the availability of excellent materials – the best of each class of ingredient is available here now from every market on the globe.

For all its faults, and I am sure that there are many, this work has a character imposed on it by my choices of technique and my ability to extract suitable materials from my local geology. At this point this is the best I can do with what I have discovered. I have used and developed some of the limitations such as the impurities in the raw material to give an exquisite surface flash that is enhanced by the woodfiring process. Usually rich red flashing and translucency are mutually exclusive qualities. However, in this case I have managed to coax both from my materials.

From one point of view this work can be seen as an intellectual exercise, a strike against globalisation, and that might be true and all that there is to it, until you see and handle the pieces. This work is special to me, it has the most wonderful quality, a juicy beauty that elevates the soul. The play of light off the glaze and clay surfaces and through the translucent clay bodies is inspiring. There is a pleasant interaction between the orange flash on the clay surface caused by the woodfiring technique and the subtle blue/grey of the protoceladon/ guan-like glaze surface. Some of these pots are so translucent that you cannot only see through them, but you can even see the colour of the flashing through the bowl.

I have consciously avoided firing these bowls on the rims to keep them flat and symmetrical, rather, I have allowed and even encouraged some warping, which I believe expresses some of the simple natural beauty of this ancient technique. I also work on a slowly revolving treadle-style kick wheel. This method of throwing also has its own natural character. There is something about the slow rhythmic pulse of the kick wheel that can cause the thin wet rim of a finely thrown porcelain bowl to result in a slight eccentricity, which I find engaging. Not bent or slumped, just slightly out of whack. Likewise, I have avoided using ribs or profiles in throwing. I decided long ago that I preferred the hint of soft undulations left by my fingers on the surface.

I also lift all of the pots off the wheel head with my fingers, I don’t use batts, this also adds its extra touch of character and in firing I avoid the use of porcelain setters to encourage a gentle movement in the foot and consequentially the rim. Some pieces do not survive this combination of trials looking their best. However, others survive looking ‘natural’. Of course they are not natural, far from it. These pots are contrived but they look like they have that something that we have lost in our postmodern, post-industrial society. They show a direct link to the place and methods of their making. I believe that all this gives the work a greater authenticity. It certainly is rewarding for me as the maker and after all, I am making them for me.

Fifty years ago the well known Australian potter, Ivan McMeekin, discovered a weathered igneous dyke associated with a coal mine in Mittagong, from which he was able to coax a similar porcelain body. Interestingly although these bai-tunze materials are not associated with that geological event, which is no longer accessible since a freeway was built through the site, there are, however, other clays and rocks in the area that are now shown to have the same or even greater potential. The Southern Highlands turns out to be an interesting geological area in that it has produced similarly useful porcelain, and proto-porcelain materials.

Founding members of the Potters Society, Ivan McMeekin, Ivan England and Peter Rushforth were keen users of local materials, not always because of their special qualities, but rather because there was no reliable supply of Australian materials such as felspar at that time. McMeekin’s students Col Levy, Gwyn John (Hanssen-Pigott) and Les Blakeborough have also developed bodies and glazes from local sources. However, to the best of my knowledge pots like these have not been produced in this area for 50 years, not since McMeekin’s time at Sturt Workshops.

These materials have been here a long time, and I have lived and worked here for close to 30 years. It has taken me almost that long to track down and discover all of these materials. I’ve come to realise that it’s not just the place, it’s the head space. It is only now that it has all started to come together for me, the slow combination of me homogenising with country, like making a frit, these things take time to fully integrate. I have systematically searched the entire shire, collecting hundreds of materials and making thousands of tests from them. So far I have filled many notebooks and made many individual batches of porcelains, from four bai-tunze-like materials, two primary kaolins and several igneous rocks such as basalt, syenite and granite.

Bai-tunze No1 is a weathered felspathic material. This weathering has resulted in a small amount of clay mineral being present in the rock, such that after milling it develops poor plasticity. However the large amount of felspar that remains unweathered develops a remarkable translucency. The remainder of the material is quartz which helps develop the glaze fit.

When I first collected this rock I thought that it might be a useful candidate as a glaze ingredient. However it looked so unpromising that it sat on my bench for a year until I finally began testing it. Being a hard rock and finely crystalline, I thought that it might be a sandstone or a quartzite. It required primary crushing in a jaw crusher and then several hours of ball-milling to achieve the desired degree of fineness for testing. The 100 per cent position on the glaze test tile revealed a vitreous grey surface with a slight sheen. Initial testing as a glaze ingredient involved line blending with calcium carbonate and silica, both individually and together in various percentages. Line blending with calcium carbonate, between five and 25 per cent, turned the rock into a range of useful glazes, 15 per cent resulted in a guan-like glaze. I was quite surprised at this result, but pleased, as I thought that it might just turn out to be just another sandstone sample. The only negative result was the crawling of this viscous glaze, a result that I am not unfamiliar with in this type of glaze, However, crawling is most usually associated with plastic materials, and this was obviously hard rock and not at all plastic – or so I thought. Reflecting on the crawling I decided that it might be possible that the rock could have some plastic quality. I took a small handful of the fine milled powder and mixed it with a small amount of water in the palm of my hand, and unbelievably it became marginally plastic, sufficient to allow a poor coil and push-pull test to be made. I have to admit that I was so surprised that it took me a little time to gather my thoughts, but not too long. I soon had a set of shrinkage bars and translucency wedges made and drying. The fired results of this indicated a grey, vitreous shiny clay body with translucency up to 5.4 mm thick in reduction and perfect glaze fit with a standard limestone glaze. I thought about this Australian bai-tunze-like material, trying to understand its nature. Examination with a hand lens showed a fine crystal structure with opaque crystals that I guessed to be felspar, some obvious shiny quartz fragments and only a few fine black dots that I assumed to be biotite mica.

I also compared samples of the broken rock with clean geological samples of felspar in the lump form and compared cleavage planes and fracture angles. Some of these angles fitted together like a jigsaw puzzle, this comparison led me to believe that my bai-tunze was primarily a highly felspathic rock. The fact that there was little or no flashing on the exposed clay body sample in the wood firings led me to believe that the felspar was one of the plagioclase series, with a small calcium content, probably at the albite end of the series. The calcium content must be small, as there is only a slight tendency for the pots to squat during the stoneware firings in the wood kiln to between cone 8 and 9. However, firing to cone 10 and above caused squatting.

This theory however turned out to be wrong. Analysis proved the felspar to be sanidine, a disordered potash felspar. There seems to be enough mullite and secondary mullite to hold it up at high temperatures, with the calcium content low enough to increase the translucency without weakening the secondary mullite structure created by the felspar. This Australian bai-tunze although quite translucent fires distinctly grey/white in reduction so I assumed that there was some iron content, although it must be small.

Finding a material like this has convinced me that porcelain isn’t as remarkable a material as I formerly thought. It seems that there are several places around the world where this kind of geology and weathering has occurred and that all that is required is inquisitiveness and ingenuity and translucent porcelain is the inevitable result. When I started this research, I assumed that I would end up making brown clay bodies and tenmoku glazes. So I was pleasantly surprised to find such a range of eminently usable pale/low iron materials.

I have since paid to have this material analysed with X ray fluorescence XRF providing chemical analysis as per cent oxide. This revealed high silica content 76.49 per cent and moderate alumina content 12.93 per cent, nothing exceptional there. The iron content is negligible 0.56 per cent and titanium virtually non existent at 0.13 per cent with sodium and potassium as expected for a weathered felspar 6.53 per cent total. However, the calcium and magnesium percentage was a surprise and lower than I expected at 0.27 per cent in total. This was not enough of the alkali earth fluxes to explain the grey white quality of the fired surface and the lack of flashing. Subsequent firings have shown that this clay/rock does flash in the woodfiring, albeit faintly. It shows a soft pale orange/pink flash around the foot but it is subtle and only reveals itself under close examination in good light. I am assuming that the pale nature of this flash is due to the low iron content. It’s interesting just what can be interpreted from simple physical testing and observation, but it is nice to have chemical analysis as confirmation.

Bai-tunze No 2 is a weathered igneous rock that has intruded through the local sandstone. This weathering has resulted in a large amount of clay mineral being present. The original rock has completely disintegrated such that it has become a grainy, slightly sticky, fine sand-like material. It doesn’t need to be crushed, as it slakes easily, but needs milling to develop its plasticity. It appeared that it had lost its flux content during weathering and had become primarily a kaolinitic material, it didn’t fully vitrify at the 100 per cent sample, it crazed heavily with a standard limestone glaze indicating high alumina and lower silica. Because it isn’t white, it obviously has an iron content. The net result was a material that flashed well in woodfiring, high in alumina, low in silica and with some alkali.

When I first collected this rotten rock, I thought that it might be a useful candidate as a glaze ingredient. It took limestone and silica to make it into a celadon glaze. It became obvious that it was a useful candidate as a bai-tunze – one that could be blended together with my other collected and processed minerals to produce a range of porcelain clay bodies. The bodies that I developed with a high content of No 2 bai-tunze have a tendency to craze but develop a splendid red/orange woodfired surface flash with marginal translucency.

Analysis results showed kaolinite, illite, montmorillonite and quartz, an iron oxide content of 1.8 per cent, titanium low at 0.13 per cent, alkalis at 5 per cent and calcium oxide also low at 0.03 per cent. Hence the good flame response.

The examples above follow the Chinese archetype of porcelain development. I have also produced two series of porcelains based on mixtures of primary kaolin, felspar and silica. These on the other hand follow the European path of porcelain body development.

Kaolin No1 unusually turned out to be extremely pure. It was short with the only obvious impurity being a slight vein of yellow ochre staining passing through it. When this clay was blended with felspar and silica extracted by flotation from a pegmatite dyke material, it proved to be able to be developed into an extremely white and translucent porcelain body showing translucency up to 6.6 mm thick in reduction and 7.5 mm thick in oxidation. Plasticity was virtually non-existent, but this was overcome by the addition of bentonite. To this end I tested 13 different bentonites, natural and synthetic, Australian and imported. The Queensland material ‘Turbobond’, proved to be as useful as any and economical with it, as was the Western Australian product, ‘Bentonite A’.

I think that I have been able to determine the provenance of this sandstone lens kaolin. It appears to be a weathered igneous dyke. primarily because of the signs of metamorphism in the surrounding country rock and must have undergone subsequent leaching and weathering. Analysis revealed 47 per cent silica, 37 per cent alumina, 0.5 per cent iron, and all other traces as less than 0.0X per cent or below detectable levels2. This is the whitest and most translucent of the bodies that I have been able to develop, albeit, the most difficult to throw. Another porcelain paste that I have been able to create is based on a weathered igneous dyke. It was developed along the lines of the previous example, that experience informed my judgement and sped up the development process. I found that this clay was a primary kaolin weathered from an igneous intrusion, this was inferred from the remnant effects of upward thrusting on sandstone on either side of the intrusion. This site was contaminated with irony shale bands and lignite and other organics.

A series of bodies were prepared along the lines of those previously discussed.

The best of these yielded a porcelain showing translucency up to 4.9 mm thick in oxidation and 4.2 mm thick in reduction. Although this series of bodies was not as pure and translucent as the others, I persisted with it because it was quite plastic and enjoyable to throw. Analysis revealed that it consisted of kaolinite, dickite (a closely related clay mineral to kaolinite), nontronite, (a smectite, similar to bentonite, where iron replaces alumina in the crystal lattice), the remainder being quartz. The material showed 51 per cent silica, 32.3 per cent alumina, iron at 1.37 per cent with titanium at 1.19 per cent. Soda was below detectable levels and the potash was 0.83 per cent with calcium at 0.03 and magnesium at 0.33 per cent. Although the titanium and iron were both above 1 per cent, it was still possible to develop workable translucency, although less translucent than the other bodies. Possibly putting the lie to the ‘titanium eliminates translucency’ theory.

The above combination of clay minerals resulted in a quite plastic clay body which was capable of bonding large quantities of non-plastic felspar and silica and still remain throwable. Its strength at high temperatures due to its high kaolinite content allowed me to introduce several per cent of calcium carbonate as a secondary flux without sustaining noticeable slumping. The effect of the addition of calcium carbonate bleaches any iron oxide present, increasing whiteness and increases the quantity of body glass developed, increasing translucency. As long as a critical level is not passed, the porcelain will still hold its form at stoneware temperatures.

The processing of these rocks and clays involves a series of steps depending on the nature of the material as it is found in the field. Bai-tun-ze 1 is a hard rock which is minutely crystalline and does not respond to calcining. Calcining is not an option for batches that are destined for use as a body ingredient. This rock requires a jaw crusher to break it down to a workable size for further milling.

I have found that it is beneficial to mill these materials in several small batches over different time periods to gain a material of several different particle sizes. The resulting bodies are more workable if they contain particles of several different size gradings. The greater percentage of the bodies are milled finely with smaller additions of coarser material that has been milled for a shorter time. This results in a clay that is better structured and stands up better on the potter’s wheel.

All the firing was done in an extended throat Bourry box kiln. All the wood was sourced form my property and that of a neighbour. The kiln was built from homemade fire bricks produced from a local deposit of white to off/white bauxite. The bricks were handmade in a slop mould and pre-fired before construction. The basic clay is quite iron stained and few of the resulting bricks are white, more consistently orange and brown. I made them as insulating refractories by incorporating various amounts of used coffee grounds obtained from the local café. These were collected and sun dried before mixing with the hammer-milled bauxite. As this was intended to be a fully local project I felt the need to incorporate the flourishing contemporary café culture/coffee culture that is so all pervasive in modern first world life and culture. My thesis and the work devolved from it is so quaint and old fashioned that this is one of my few concessions to contemporary life.

The results that I have been able to achieve are far better than I could have let myself believe were possible. When I first started out on this project, I didn’t even know of the existence of most of these raw materials. Nor did anyone else. It was common belief in the district among potters that there was no local source of white felspar or granite. But it was there just waiting to be discovered, and what materials they are to work with.

Steve Harrison is a ceramic artist from the Southern Highlands, NSW, Australia.