Cement Kiln Transition Zone: Silica-Mullite Brick or Anti-Spalling High-Alumina?
The transition zone of a rotary kiln is an underrated position. It lacks the stable coating that protects the burning zone, and it is far harsher than the preheating zone: 1000–1300°C with frequent swings, alkali and sulphur circulating and concentrating in the brick, coating that forms and drops repeatedly, plus mechanical squeezing from shell ovality. Choose wrongly here and the result is not a slightly shorter campaign but a loose ring within three months. Two families dominate: silica-mullite (SiC-bearing) brick and anti-spalling high-alumina brick. They are not substitutes for each other; they divide the work.
What the transition zone actually does to brick
Four loads act at once. Thermal shock comes from stops, feed interruptions and coating loss, where the hot face can drop several hundred degrees in tens of minutes. Alkali and sulphur attack comes from K₂O, Na₂O and SO₃ carried in raw meal and fuel, condensing in the pore structure between 900 and 1200°C and reacting with alumina and silica to form leucite and nepheline with volumetric expansion above 20%, which swells and spalls the brick. Mechanical stress comes from shell ovality and from deformation near the tyres. And coating is thin and unstable here, so the hot face sees flame radiation and material abrasion directly. Of the four, alkali attack and thermal shock decide the brick type; mechanical stress is handled by joint and keying design; coating behaviour follows from the hot-face character.
Where silica-mullite brick fits, and its numbers
Silica-mullite brick sits in the Al₂O₃-SiO₂-SiC system. SiC carries the alkali and abrasion resistance, the mullite phase carries thermal shock resistance. Workable ranges: Al₂O₃ 60–68%, SiC 8–14%, bulk density 2.65–2.80 g/cm³, apparent porosity 17% or lower, cold crushing strength 80 MPa or higher, refractoriness under load (0.2 MPa) starting at 1550°C or above, and 20 cycles or more at 1100°C water quenching. Its strength is alkali resistance: under reducing or neutral conditions SiC is comparatively inert to alkali vapour and clearly slows leucite formation, while abrasion resistance suits kilns with aggressive raw meal. Its weakness is higher thermal conductivity and poorer coating adhesion than magnesia or spinel grades, so it does not belong in the burning zone. A practical test: if the brick you pull out shows a dense, discoloured hot face over a swollen, porous interior, alkali attack is the driver and silica-mullite is the right direction.
Where anti-spalling high-alumina brick belongs
Anti-spalling high-alumina brick adds mullite, kyanite or a little zirconia to a high-alumina base, using microcrack toughening and controlled residual expansion to survive thermal cycling. Typical ranges: Al₂O₃ 65–75%, apparent porosity 18% or lower, cold crushing strength 70 MPa or above, 15–20 water-quench cycles at 1100°C, refractoriness under load starting at 1500°C or above. It wins on thermal shock and high-temperature structural stability, and loses to SiC-bearing brick on alkali resistance — it will also swell under sustained high-alkali conditions. So place it on the burning-zone side of the transition zone, in the first one or two rings behind the nose ring, and in kilns that stop and start often enough that thermal cycling outweighs alkali load. If your kiln carries a heavy alkali load, with high alkali in the raw meal and limited bypass, and the coating never really holds, this brick in the hottest part of the transition zone will not last a full campaign.
Installation values: joints, keys and retainer rings
Control is similar for both types, but the numbers are not negotiable. Keep joints at 2–3 mm and lay wet with a matching mortar. For silica-mullite brick the mortar must carry a similar SiC content; ordinary high-alumina mortar becomes the first thing the alkali eats, and the failure starts in the joint. Use two to three key plates per ring, 1–2 mm thick, driven at the 45° and 135° positions rather than concentrated at the crown. Retainer rings stay where the drawing puts them; fill up to the ring with cut brick, and never with mortar thicker than 5 mm. Hold shell ovality within 0.5%; above that, alternating squeeze in service loosens joints first, and no brick grade survives it. After installation, verify joint staggering, record keys ring by ring, and keep cut brick below about 8% of the total.
Thresholds in service and the replacement rhythm
Do not wait for brick to disappear. Set three thresholds. Infrared shell scanning: 250–320°C is normal for the transition zone, and any local reading above 380°C gets marked and scanned more often. Residual thickness below one third of original, roughly 80 mm, goes into the next shutdown plan. Three or more loose brick in one ring, or brick dropping out in sequence, means stop and repair rather than run damaged. In practice the transition zone is replaced more often than the burning zone, and a full-section change is not always necessary; segment the replacement along the shell temperature profile, but end each segment at a retainer ring, never mid-ring. After a new lining, heat up at 30°C per hour or less to 500°C, then 50°C per hour, and give the first coating time to build.
A selection table you can use directly
Compressed into decisions: high alkali load, strong abrasion, coating that never holds — silica-mullite brick. Frequent starts and stops where thermal shock dominates and alkali is moderate — anti-spalling high-alumina brick. The one or two rings nearest the burning zone — anti-spalling high-alumina or a spinel transition grade. Large diameter, high rpm, difficult ovality control — silica-mullite first, for abrasion and squeeze resistance. Tight budget with mild conditions — anti-spalling high-alumina offers the better cost balance. One closing point: transition zone failures are rarely single-cause. When you pull the brick, do the failure analysis — swelling from alkali, cracking from thermal shock, or loosening from mechanical squeeze — and keep photographs and samples. Replacing brick without that analysis simply repeats the same mistake in the next campaign.
WeChat · 17513376789