Petrology is a branch of geology.  A petrologist examines rocks and tries to figure out how they were formed.  The refractories engineer has access to very high temperatures and so you can speed up the simulation of metamorphic petrology.  The chemists rule of thumb is that for every 20°C higher-temperature, chemical reactions take place twice as fast.  In metamorphosis not much happens at 100 degrees centigrade.  If you treat a rock at 200°C twice as much will happen every 20°C higher, which means that 32 times as much will happen during the same period.  But 300 is the reaction will take place another 32 times faster, etc.

Now consider a metamorphic reaction that took place over 10 million years at 100 degrees centigrade.  If you put sample in a furnace at 200 degrees centigrade the reaction can take place in 312 000 years.  At 300°C it will take 10 000 years, at 400°C it will take 300 years, at 500°C 10 years, at 600°C 4 months, at 700°C it will take 1 day and at 800°C 45 minutes.  This is not really literally true, but the point is that geological processes can be simulated to some extent in a furnace.  The refractories engineer has access to temperatures unimaginably higher than ever found in the earth’s crust.  In an electric arc furnace temperatures of 2600 degrees are attainable.  

So the refractories engineer can design his own rocks and manufacture them in a practical space of time.  Unfortunately high pressure is not commercially available for the manufacture of refractories.  This means that dissolved water and gases are not available in the synthesis of rocks.  This is a severe limitation in synthesizing metamorphic rocks, but is not relevant in refractories because these elements would be expelled in service anyway.

Phase diagrams are the refractory technologist’s basic tool In design.  Unfortunately phase the diagrams do not predict the actual microstructure of the rock, which is determined in part by wetting angles and other factors.

Diffusion cannot be accelerated to the same extent as chemical reactions.  Unfortunately we cannot freeze synthetic rocks in their high temperature format and they change during cooling.  In some cases this can be used to advantage.  An example of this is Chrome magnesia refractories.  Periclase is magnesium oxide and Periclase refractories generally have the periclase crystals surrounded by low melting point silicates.  If you add chrome oxide to Periclase refractories the chrome oxide goes into solid solution with the periclase at high temperatures.  In rapid cooling the chrome oxide exsolves from the periclase brick in lamelli throughout the crystal.  On slower cooling the chrome oxide exsolves and forms chromia crystals which join the periclase particles together through the low temperature melting silicates.  This makes the structure more solid and no longer prone to high temperature creep.  If calcium oxide is present it does not lower the melting point too much because calcium silicates are all quite refractory.  Silica forms forsterite with magnesia, which melts at 1890°C.  If calcia and silica are both present, Merwinite and Monticellite are formed, with melting points around 1400°C, so they are in glass phase at higher temperatures.

If carbon is present, as in resin bonded carbon magnesia refractories, it gets really interesting.  The graphite reduces the periclase to magnesium metal, which escapes as gas.  As soon as it contacts oxygen, it turns back into magnesium oxide.  This phenomenon allows the brick to seal its pores.  One of the most intriguing phenomena I have observed, was when pitch impregnated Periclase bricks were fired at very high temperature in a tunnel kiln.  They came out bloated and cracked.  From the cracks, hollow tubes emerged, looking like potatoes which have started to germinate.  Weirder still, the tubes curved around the corners of the bricks, following the same direction.  This super weird phenomenon kept a lot of refractories technologists puzzled for a long time.  What had happened was the following;  The graphite burned out on the surface of the bricks.  At higher temperature the graphite reduced the magnesia to metal gas, which sealed the pores when it contacted oxygen in the decarburised zone.  This sealed off the black core of the bricks.  The gas pressure caused the centre to bloat, and wherever the gas emerged through the cracks, it oxidised again, forming a solid MgO tube around the escaping magnesium gas.  These tubes grew outwards, and followed the gas flow in the kiln, bending around the bricks.

I wrote my thesis on Periclase refractories.  My supervisor, Jean Taylor, found literature references claiming that Periclase bricks with smaller pore size distribution performed better, and got me to investigate the effect of particle size distribution on pore size distribution.  One of my bricks fired at 100°C lower than the others, gave superior properties.  This is to refractories technologists highly counter intuitive.  I thought it must be due to an experimental error.  Jean got me examine polished specimens for about a week until I could explain the result.  The research is on Keramicalia’s web site under “Textbooks”.  To press such bricks would create difficulties in the manufacturing plant, and the project never reached fruition.  If it did, it would save a huge expense on fuel by firing to 100K lower temperature.  The cost of firing at high temperature roughly doubles for every 100K higher temperature.

When I studied Geology, the first mention of refractories was in second year mineralogy, regarding andalusite.  We all blurted out “What are refractories?”  We were told that they are bricks for melting glass in furnaces, and there is a whole factory in Olifantsfontein making them.  It later transpired that I and several of my colleagues spent part of our careers working in that factory and other refractories factories