Industries / Mining and materials / Glass and ceramics
Firing and kilns
Firing is the heating of shaped clay, glaze or ceramic powder in a kiln until it hardens into its final form, along with the kilns built for it.
7
Milestones achieved, from c. 1700–1027 BCE to 2010.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
c. 1700–1027 BCE
Glazes fired past 1,200 °C
During China's Shang dynasty, which a 2012 UCL doctoral thesis dates to about 1700 to 1027 BCE, potters in the lower Yangtze region began making proto-porcelain, a glazed ceramic that matures at more than 1,200 °C. Analysis of 61 sherds from kiln sites at Deqing, Zhejiang, found bodies made from local porcelain stone and glazes made by applying wood ash rich in lime, fired at about the temperature later lime glazes need. The thesis concludes these potters could already reach that heat in their kilns.
Credit
Shang and Zhou dynasty potters of Zhejiang; analysis by Yin Min (UCL Institute of Archaeology)
1858
A fire that moves
The German engineer Friedrich Hoffmann began patenting his annular kiln in 1858. Bricks stay in place in a ring of chambers while the fire, fed with coal, moves from chamber to chamber, so the kiln never has to be cooled and relit; heat moving through the ring fires one batch and lets others cool slowly. Hoffmann won a prize for it at the 1867 Paris Universal Exposition, and it was soon adopted for firing brick and lime, according to the Musée des arts et métiers in Paris, which holds a model of it.
Credit
Friedrich Hoffmann
1912
Ware rides through the fire
In a tunnel kiln, cars loaded with ware travel slowly through a long tunnel that is hot in the middle, so firing never stops. Conrad Dressler built his first tunnel kiln in 1912 for the tile maker J. H. Barratt in Stoke-on-Trent, England, to replace bottle-shaped periodic kilns, which were slow to heat and cool and hard to control. Universal Sanitary of New Castle, Pennsylvania, then ordered one for sanitaryware (toilets and basins), and by the mid-1930s more than 400 Dressler tunnel kilns were in service, by Ceramic Industry's account.
Credit
Conrad Dressler; Dressler Tunnel Ovens of Fenton, Stoke-on-Trent
1953
Overheated glass turns ceramic
At Corning Glass Works, S. Donald Stookey found in 1953 that a glass he overheated became harder, stronger and more electrically resistant. The new crystalline material, called glass-ceramic, is shaped as glass and then heat-treated so that crystals grow inside it. CorningWare cookware made from it went on sale in 1958, and glass-ceramics later went into smooth-top cooking surfaces for stoves, according to the National Inventors Hall of Fame, which inducted Stookey in 2010 (US Patent 2,920,971).
Credit
S. Donald Stookey, Corning Glass Works
1974
Tiles fired in thirty minutes
Pietro Marazzi of Sassuolo, Italy, filed a patent on 2 July 1974 for a line that dries tiles and fires body and glaze together in a single pass, carrying them one layer deep on driven rollers through a gas-fired tunnel kiln. The US version of the patent gives a whole firing of "the order of 30 minutes" at generally 1,050 to 1,100 °C. A heritage record of the Emilia-Romagna region says this fast single firing cut the tile production cycle from about 24 hours to about one hour, and dates Marazzi's patent to 1975.
Credit
Pietro Marazzi; Marazzi Ceramiche
2001
A furnace change breaks hips
Zirconia ceramic balls for artificial hips began fracturing inside patients, and in 2001 Health Canada told surgeons to stop implanting zirconia heads made since 1998 by the French firm Saint-Gobain Advanced Ceramics Desmarquest. A retrieval study by Masonis and colleagues (2004) found the failures were confined to heads sintered in a tunnel furnace introduced in 1998, counted 343 documented fractures since 2000, and found fracture marks unique to heads from that furnace.
LessonChanging how a proven part is fired can change its strength, so a new kiln or furnace calls for the part to be qualified again.
Credit
Health Canada (warning); J. L. Masonis, R. B. Bourne and colleagues (retrieval study)
2010
Firing in five seconds
Marco Cologna, Boriana Rashkova and Rishi Raj, of the University of Colorado, reported that zirconia could be sintered to full density in a few seconds at about 850 °C by applying an electric field across it, where normal firing takes several hours at 1,450 °C. They called it flash sintering. In 2013 the UK lab Ceram built an 80-foot kiln to try the method on a commercial scale, with tile making as a first target and results expected that year. A 2021 review in MRS Bulletin said flash sintering had still been applied only at laboratory scale, because heat concentrates in hot spots and gives uneven results in larger parts.
LessonA firing method proven on samples a few millimeters across does not yet work on a full-size tile.
Credit
M. Cologna, B. Rashkova and R. Raj (Journal of the American Ceramic Society)
Watch for
Flash sintering with light
Lucideon and the AMRICC Centre in Stone, England, are testing xenon flash sintering, which densifies ceramic coatings and thin layers with short, intense pulses of light from Excelitas Noblelight lamps and no surrounding furnace. Lucideon says its pilot equipment handles areas from 25 mm to more than 300 mm, has densified zirconia pellets in as little as 15 minutes, and has densified the surface of silicon carbide.
Achieved whenan outside lab measures the density and strength of light-sintered coatings against furnace-fired ones, or a manufacturer uses the process in production.
Claimed by
Lucideon, reported by CompositesWorld, November 2026
Open
01
Open kiln-to-quality data
A public dataset that links roller-kiln logs (temperature in each zone, cycle time, pressure) to the quality of the tiles that came out (share sold as first grade, size class, flatness, shade), with a fixed test split and a baseline model that predicts quality from the kiln data.
The briefCredit · unclaimed02
Map firing across the kiln
An open method, and the first public results, for measuring how evenly a roller kiln fires across its width and along its length, by sending a row of standard test tiles through side by side and measuring each one's shrinkage and water absorption afterward.
The briefCredit · unclaimed
Open 01 · the brief
Open kiln-to-quality data
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public dataset that links roller-kiln logs (temperature in each zone, cycle time, pressure) to the quality of the tiles that came out (share sold as first grade, size class, flatness, shade), with a fixed test split and a baseline model that predicts quality from the kiln data.
Why it is still open
The closest work, by Heydari, Kamyabi and Ghayoumi Zadeh (2026), predicted first-grade share from the data of one industrial roller kiln operation, noted that data acquisition in kiln operation is "costly and limited", and trained on perturbed copies of the real records. No public, multi-plant dataset turned up in searches of Hugging Face, Crossref and the web.
Built with what exists
Kiln control-system exports (OPC UA or CSV), the size and shade readings that tile sorting lines already record, Python with pandas and CatBoost or LightGBM, and Zenodo for an open, citable release.
First step
Write the data format and an anonymizing script, then ask one tile plant or a university pilot line for one week of kiln logs and matching sorting-line records.
Done looks like
At least six months of minute-by-minute kiln logs matched to at least 100,000 sorted tiles from at least two plants, published under an open license, with a baseline whose error on held-out months is stated, and one outside group that reproduces it.
Open 02 · the brief
Map firing across the kiln
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open method, and the first public results, for measuring how evenly a roller kiln fires across its width and along its length, by sending a row of standard test tiles through side by side and measuring each one's shrinkage and water absorption afterward.
Why it is still open
Kiln makers describe uniformity in words: SACMI says its new FMN roller kiln, shown in June 2026, "guarantees high firing uniformity" and gives no figures. Published studies of temperature across roller kilns found in Crossref are mostly computer simulations, and no measured comparison across several kilns turned up.
Built with what exists
Test tiles pressed from the plant's own body, calipers, the boiling water-absorption test of ISO 10545-3, temperature-indicating ceramic rings sold for kiln checks, and Python for plotting maps.
First step
Fire one row of ten test tiles across the width of a university or pilot-line kiln and measure them.
Done looks like
Published maps from at least five roller kilns at three or more plants, each from at least three runs, giving the spread of shrinkage across the width in percent and the range of water absorption, with the method written so anyone can repeat it.