ceramic barrier on ceramics? If you are using dense alumina crucibles and they still get corroded too soon, your only option is to go for platinum crucibles.

For large quantities of glass an option would be the use of crucibles of zirconium oxide or chromium oxide. These do not corrode as quickly as corundum.

Please find attached a PhD on refractory materials

http://www.pdhonline.org/courses/m158/m158content.pdf

Thanks for your answers.

platinum crucible is very expensive. What do you say about modifying ceramic crucible with a layer of TiN?

Unfortunately, I do not know your glass compositions and requirements to contaminations..., but I believe that graphite or "white graphite" (graphite-like hexagonal boron nitride) is a good solution for your problems...

TiN oxidises in air from about 500°C, as well as graphite or BN.

If it were possible to carry out the process under an inert gas, so one could work with graphite or molybdenum.

An air graphite must not be hotter than 600°C, but the short term, it can tolerate much higher temperatures with little burn.

Thank a lot for your great points. However it seems that I have no choice other than platinum.

So, what do you say about using a thin layer of platinum on a thick layer of another cheaper metal such as vanadium which are deposited on a ceramic crucible?

A layer of platinum (only platinum, no vanadium or something else) can indeed prevent corrosion of the ceramic crucible, but not protected it from thermal shock. By thermal shock explodes the ceramic crucible after a pair of inserts and platinum is lost. Therefore, a thin platinum crucible is a better solution.

To Dr. V. Verlotski: According to your writing, you have no experience in carbon or alpha-BN materials applications. So-called Tammann furnace was invented by Prof. Gustav Heinrich Johann Apollon Tammann in the 19th century. However, all around the world this device with graphite heater is still used in the metallurgical research practice for various purposes in the temperature ranges up to 2500 oC. More likely, you do not know that the graphitized carbon materials are being produced at the temperatures around 3000-3300 oC without any protective gaseous atmosphere (in air!). I have got my personal experience in studies of carbon-carbon composites behaviour under the surface heating at the temperatures up to 10,000 K (!). Currently, no other materials, except carbon or carbon containing composites, could be recommended for such extreme environments. Concerning the crucibles made from graphite: it is very easy and cheap to cover all the graphite surfaces there by painting them with refractory oxide powdered and bakelite based compositions. The only problem is connected with the required level of contamination for the charge, which is being melted in the graphite (coated or non-coated) crucible, so the direct interaction between carbon and charge matter is the most important factor in this application (sometimes BN is less active chemically to melts than carbon, and there is some kind of preferability to its application). Even to apply the noble metals, including Os and Ir - with the highest melting points (please see Volume 1 of my book on UHTM available on my ResearchGate account), is better as a coating on a graphite crucible prepared by electroplating from molten salts (chlorides, fluorides etc) bath.