The refractory materials in a glass furnace have to resist not only extremely high temperatures, but also the corrosion coming from the glass melt and the furnace atmosphere. The post-mortem analysis of corroded refractory material after furnace service provides essential information about, and insights into, the correlation between furnace performance and furnace operation on the one hand, and refractory wear on the other. The interpretation of the microscopic observations makes it possible to explain the different corrosion mechanisms and, thus, provides valuable details and clues for a possible optimisation of future furnace performance and operation.

In order to simulate accelerated industrial working conditions and to predict the degradation of refractories, elaborate testing procedures are performed.

Corrosion resistance
The resistance to thermal and chemical stresses is tested with the dynamic finger test and the static plate corrosion test according to the International Commission on Glass, Technical Committee 11 (the so-called TC11).
In the dynamic finger test, a cylindrical sample is rotated in a platinum crucible together with molten glass. The test is carried out with predefined parameters (temperature, time) and the reduction in diameter at fluxline (three phase contact) is measured. The corrosion rate can also be determined by measuring the decrease in length (forced convection at the rotating cylinder face) as a function of time and at various rotation rates.
In the static plate corrosion test, sample platelets are immersed into the glass melt. After the chosen time the fluxline depth of the corroded samples is measured. This test is carried out with four samples under the same test conditions and is, therefore, an appropriate method to compare and evaluate certain refractory grades together. An international round-robin investigation within the Technical Committee 11 of the International Commission on Glass confirmed the good reproducibility of the static plate corrosion test, and the TC11 recommended it as the standard corrosion test.

Stone formation
Due to the interaction of the glass melt with fused cast AZS, solid stones can be generated.
The defect potential is evaluated by counting this kind of inclusion inside the remaining glass after each corrosion test.

Blistering
The tendency of fused cast AZS to form bubbles in contact with molten glass is evaluated with an especially developed test procedure in which standard glass is molten in a fused cast AZS crucible.
The number of bubbles contained in the glass samples after test melting is an indication of blistering potential.
The tendency to release bubbles is expressed in bubbles/gram glass, following the terminology of glass defects of the International Commission on Glass.

Exudation
In order to evaluate the expulsion of glassy phase, a simulation of the load-free preheating stage, which the refractory blocks in a melting furnace undergo, is performed by heating the test material up to temperatures comparable to real furnace operation conditions.
The test procedure follows the recommendations of the Technical Committee 11 of the International Commission on Glass. The experimental data of REFEL refractories show that for temperatures higher than 1500°C there is only limited exudation which decreases slowly with service time.

Vapour attack
This test method according to ASTM is performed to evaluate the resistance of refractories in glass melting furnace superstructures to vapour attack. The vapour test method may also be useful for evaluating refractories in other applications where vapour attack occurs.

The Scanning Electron Microscope (SEM) is used for the high resolution analysis of very fine crystalline structures as well as the analysis of glass chemistry and glass defects such as impurities and inclusions. Fragments of the material to be examined or ground sections – gold-vaporised surfaces for structure, carbon-vaporised surfaces for chemical composition – are scanned by an electron beam and displayed on a monitor for analysis. Due to the high degree of magnification, the special arrangement of the structure components and pores in the material under investigation become visible. It is, therefore, possible to qualitatively determine the existing elements during the investigation of impurities, and to make semi-quantitative statements on the chemical composition of the structure components of ground areas.

Glass defects can arise for various reasons, such as unmolten raw materials or the extensive use of recycled glass cullet, and result in the occurrence of unsightly and partly dangerous inclusions.
Furthermore, also melting conditions and their influence on the refractory materials in the furnace can stimulate glass defects.
Through the identification of glass defects REFEL’s experts can give comments and recommendations based on their long-term experience regarding possible causes as well as possible preventive or corrective measures to avoid glass defects in the future.
In doubtful cases, the results of optical and electron microscopes are verified by X-ray diffraction analysis in REFEL’s state-of-the-art laboratory.