The photovoltaic glass industry is expanding rapidly, and furnace life directly affects the profitability of the entire production line. Among all furnace refractories, sidewall blocks are particularly critical. They remain in continuous contact with molten glass at extreme temperatures while enduring both chemical corrosion and mechanical erosion.
An unsuitable material can accelerate wear, shorten furnace life, and even force an early cold repair. Yet no single configuration has become standard for photovoltaic rolled-glass furnaces: some use fused α-β alumina throughout, while others combine fused AZS and α-β alumina blocks.
The right choice is not simply the most expensive material. It depends on glass temperature, composition, flow pattern, furnace structure, and operating conditions. This article compares the two principal materials and explains how they can be configured more effectively.
Fused α-β corundum bricks are made from high-purity alumina, typically containing more than 90% Al₂O₃, and are melted at approximately 2,000–2,200°C.
Their crystalline structure consists of interlocking α-Al₂O₃ and β-Al₂O₃ phases. The high proportion of β-alumina refines the microstructure and improves resistance to thermal deformation.
Their main advantage is chemical cleanliness. The nepheline-rich reaction layer formed at the glass-contact surface has relatively low viscosity and can dissolve more readily into the glass. As a result, small amounts of dissolved material are less likely to cause persistent glass defects.
This makes fused α-β alumina an effective choice for areas operating below approximately 1,350°C. At higher temperatures, however, its reaction layer may be removed too easily by glass flow, reducing its ability to protect the refractory surface.
Fused AZS belongs to the Al₂O₃–ZrO₂–SiO₂ system. A common example is oxidation-cast, void-free AZS-33.
Its main crystalline phases are densely interlocked corundum and baddeleyite. Because zirconia has an extremely high melting point, the material offers excellent resistance to corrosion by molten glass.
During service, ZrO₂ becomes incorporated into the surface reaction layer, giving it much higher viscosity than the reaction layer formed on α-β alumina. This viscous layer adheres to the block surface and acts as a protective barrier against further corrosion.
The trade-off is that if material from this layer enters the glass, its higher viscosity can make it more difficult to dissolve and homogenize. Therefore, fused AZS offers stronger protection in high-temperature areas but must also be managed carefully to minimize the risk of glass contamination.
Photovoltaic rolled-glass furnaces differ from conventional float-glass furnaces in several important ways.
The throats of photovoltaic glass furnaces are becoming shorter and narrower. In furnaces of a similar capacity, throat dimensions may decrease from approximately 7,500 to 5,000 mm in length and from 4,000 to 2,200 mm in width.
A shorter, narrower throat reduces the heat-dissipation area and increases glass-flow velocity. Consequently, the temperature drop through the throat may be only about 90°C, compared with approximately 110°C in a typical float-glass furnace.
The glass near the throat outlet may therefore remain hotter than expected.
Float-glass furnaces commonly use horizontal or vertical stirrers, dilution air and water-cooled equipment to control glass temperature.
Many photovoltaic rolled-glass furnaces eliminate stirrers and dilution air while increasing forehearth insulation. These changes reduce downstream heat loss and extend the high-temperature zone.
Photovoltaic glass has very low iron content and high thermal transmission. This can intensify horizontal convection within the molten glass, increasing the flow-related erosion of sidewall refractories.
Together, these factors make the area around the throat outlet one of the most demanding parts of the furnace lining.
The most important reference temperature in sidewall-block selection is approximately 1,350°C.
Operating data from two structurally identical 700 t/day furnaces illustrate this point. With the same glass specification and pull rate, a difference in refining temperature resulted in a roughly 20°C difference at the throat outlet. One furnace operated near 1,370°C, while the other was close to 1,350°C.
The furnace using fused α-β alumina near 1,370°C experienced noticeably faster corrosion. At this temperature, the nepheline-rich reaction layer did not provide sufficient protection, whereas the more viscous reaction layer formed on fused AZS remained effective.
A practical selection guideline is therefore:
The 1,350°C value should be treated as an engineering reference rather than an absolute universal limit. Actual selection should also consider glass chemistry, flow velocity, block quality, cooling conditions and furnace design.
Based on temperature distribution and operating experience, a practical “9,000 mm rule” has been developed for configuring sidewall blocks.
Use the two corner blocks at the throat outlet as the centres of two circles, each with a radius of 9,000 mm:
In a 700 t/day furnace, the average temperature drop from the furnace hot spot to the throat may be approximately 7.5°C per metre. For conservative design near the throat outlet, a value of 6°C per metre can be used.
Assuming a relatively high outlet temperature of 1,400°C, approximately 9 metres are required for the glass temperature to fall below 1,350°C. This forms the basis of the 9,000 mm radius.
For a four-line furnace, the two main channels within the 9,000 mm radius use fused AZS. By the time the glass reaches the nearest branch channels, its temperature should have fallen below 1,350°C, allowing fused α-β alumina to be used for the branch-channel sidewalls.
The same principle applies to a five-line furnace, although the AZS-covered area will normally be larger because of the additional channels.
In a typical four-line arrangement, fused AZS may account for approximately 30–40% of the total sidewall-block quantity.
Compared with a single-material design, a properly optimized mixed configuration may reduce overall material cost while improving durability in the most severely exposed areas. Exact savings and service-life improvements, however, must be verified against the specific furnace design and operating conditions.
Temperature is the main selection criterion, but it is not the only one.
Larger furnaces generally cool more slowly. A 700 t/day furnace may show an average glass-temperature drop of about 7.5°C per metre, while a 1,000–1,200 t/day furnace may have a more gradual temperature decline.
In larger furnaces, the high-temperature zone extends farther downstream. The 9,000 mm radius may therefore need to be increased.
Operating observations suggest that a 20°C increase in refining temperature may produce a similar rise at the throat outlet, depending on furnace design and operating stability.
If the refining temperature exceeds approximately 1,455°C and the throat-outlet temperature approaches 1,370°C, the standard material configuration may require additional protection, such as controlled air cooling of the sidewalls.
Improved channel insulation slows the cooling of molten glass and extends the high-temperature zone. This may support flow stability and bubble removal, but it also increases the area requiring higher-corrosion-resistance refractories.
Furnace design must therefore balance energy efficiency, glass conditioning and refractory life.
If the throat-outlet temperature remains consistently high, replacing the refractory is not always the first or most effective response. Three operating measures should be considered:
Effective temperature control and material configuration are usually more economical than dealing with an early cold repair.
Sidewall blocks are only one part of the photovoltaic glass furnace lining. Reliable operation requires coordinated material selection throughout the furnace:
A weakness in any critical zone can shorten the operating life of the entire furnace.
Material selection and configuration can only deliver the expected results when the blocks themselves have stable and consistent quality.
Zhengzhou Firebird New Material Co., Ltd. has more than a decade of experience supplying high-temperature refractory and insulation materials. Its quality-control process covers raw-material inspection, production, forming and finished-product verification.
For photovoltaic glass furnaces, working with an experienced manufacturer provides better material traceability, more reliable delivery and stronger technical support throughout product selection and supply.
Firebird can support projects involving furnace sidewall blocks, fused AZS, fused α-β alumina and other high-temperature refractory products, with material recommendations based on the furnace structure and operating conditions.
Selecting sidewall blocks for a photovoltaic glass furnace is not a simple choice between two materials. It requires balancing temperature, glass chemistry, flow conditions, furnace structure, investment and expected service life.
Three principles provide a practical starting point:
Choosing the right sidewall-block configuration does more than reduce refractory costs. It protects furnace availability, production capacity and the long-term profitability of the photovoltaic glass line.
