Industries / Food, farming and forestry / Food and beverage processing / Brewing and distilling
Malting and mashing
Malting sprouts barley and then dries it in a kiln so that its starch can later be turned into sugar; mashing soaks the crushed malt in hot water to make that sugar, and the sweet liquid, called wort, is then boiled with hops.
4
Milestones achieved, from 1784 to 1979.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1784
Thermometer and saccharometer
Michael Combrune, a London brewer, argued in his Essay on Brewing (1758) and The Theory and Practice of Brewing (1762) that the thermometer was the one instrument "which the brewer can with certainty govern himself by". In 1784 John Richardson published a treatise on the saccharometer, a floating gauge that measures how dense a wort is and so how much sugar the mash has given. His book set out estimates of what malts from English, Scottish and foreign barley yield and how strong different beers are. Mash heat and sugar yield became numbers a brewer could record and compare.
Credit
Michael Combrune; John Richardson
1817
Roasted malt for color
After Britain banned coloring additives in beer in 1816, porter brewers working mostly with pale malt had no lawful way to darken their beer. On 28 March 1817 Daniel Wheeler was granted British Patent No. 4112 for "A New or Improved Method of Drying and Preparation of Malt": roasting malt like coffee beans in metal cylinders turning over a furnace, at 360 to 400°F (about 180 to 205°C) by an 1846 account. A small amount of this "patent" or black malt could darken a large brew without a burnt taste. One account quoted by the Brookston Beer Bulletin says Guinness had replaced all its brown malt with it by 1828.
Credit
Daniel Wheeler
1873
Air blown through the grain
In a floor maltings, sprouting barley lies on a floor and workers turn it by hand. Galland's first pneumatic maltings, at Maxéville near Nancy in France in 1873, forced air through the germinating grain, though the grain was still turned by hand. Two systems followed that also turned the grain by machine: Galland's drum, which rotates, and Saladin's box, in which screws run back and forth through the bed.
LessonThe first Saladin box maltings in England, at Ditchingham in Norfolk (1891), cost "£15,000 and upwards" against an expected £8,000, and its owners failed in 1906; the historian Amber Patrick writes that pneumatic malting was not a success in England until the second half of the 20th century.
Credit
Galland; Saladin, both of Nancy
1979
A carcinogen from the kiln
B. Spiegelhalder, G. Eisenbrand and R. Preussmann reported traces of N-nitrosodimethylamine (NDMA), a compound that causes cancer in animals, in beer. Their survey of West German food found beer gave 64% of men's NDMA intake and traced it to the kilning (drying) of malt, where oxides of nitrogen from the burner react with amines in the grain. Maltsters changed their kilns: a 1989–90 survey put West German men's NDMA intake at 0.28 micrograms a day, against 1.1 micrograms before, and a 2007 survey of 138 beers from 42 countries found none detectable in 79% of them.
LessonThe heat source that dries malt is part of what ends up in the beer.
Credit
Spiegelhalder, Eisenbrand and Preussmann; the malting industry
Watch for
Kilns heated by spare wind
Simpsons Malt says the Energy Centre beside its Tweed Valley Maltings in Berwick-upon-Tweed, England, running since late 2024 and operated by AMP Clean Energy, heats its kilns mainly with a 12 MW electric boiler run on curtailed wind power, meaning power from wind farms that would otherwise be paid to switch off, backed by three 6 MW wood-chip boilers. It says this cut the site's carbon footprint for malt (Scope 1 and 2, the emissions from its own fuel and bought power) by 72% and saves up to 25,000 tonnes of CO2 a year; the same report also gives a 90% figure for the malt made there.
Achieved whenan outside party, such as an auditor or the UK Emissions Trading Scheme's verified emissions for the site, shows the site's emissions down by about that much from before 2024, when the EU's register put Tweed Valley Maltings at 31,387 tonnes in 2019 and 28,176 tonnes in 2020.
Claimed by
Simpsons Malt, 2025 Sustainability Report
Open
01
Maltings' carbon against the register
A public table of every malting plant in the European Union's emissions register, and the UK's where it is published, giving each plant's verified CO2 by year, its owner and stated capacity, and each public carbon claim the owner has made, marked as matching the register, not matching, or not checkable.
The briefCredit · unclaimed02
Fermentability from the malt sheet
An open dataset and model that predict how far a wort will ferment, its attenuation, from the numbers on a maltster's certificate of analysis and the brewery's mash temperatures, so a brewer can see before brewing how a new lot of malt will behave.
The briefCredit · unclaimed
Open 01 · the brief
Maltings' carbon against the register
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of every malting plant in the European Union's emissions register, and the UK's where it is published, giving each plant's verified CO2 by year, its owner and stated capacity, and each public carbon claim the owner has made, marked as matching the register, not matching, or not checkable.
Why it is still open
The EU register lists verified emissions for maltings by name, but nobody has published them next to the companies' claims. In July 2023 Heineken's Mouterij Albert in Belgium said a new heat-pump kiln would cut its footprint from 33,000 to 5,000 tonnes of CO2 a year, while also raising capacity from 275,000 to 375,000 tonnes of malt. The register's 2025 file gives its two kiln installations a combined 28,039 tonnes in 2022, 29,567 tonnes in 2024 and 31,580 tonnes in 2025, so the cut does not show there, though output per tonne and the scope of the claim would need checking. The nearest prior work, a 2021 PBL and TNO study, covered only two Dutch maltings with 2019 data.
Built with what exists
The EU register's yearly verified-emissions spreadsheets (2008 to 2025), company sustainability reports and press releases, a language model to pull capacities and claims out of those documents with a person checking each row (see the Note "Seven more flagged"), Python with openpyxl, and Zenodo to publish with a DOI.
First step
Download the 2025 verified-emissions file and list every installation whose name contains a word for malt in the local language (malt, mouterij, malterie, Mälzerei, sladovny, słodownia).
Done looks like
Every malting installation in the 2025 register matched to its owner, with tonnes of CO2 per tonne of stated capacity for each year it reported, and each public claim from 2020 on marked match, mismatch or not checkable, with the company's reply where it gave one.
Open 02 · the brief
Fermentability from the malt sheet
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open dataset and model that predict how far a wort will ferment, its attenuation, from the numbers on a maltster's certificate of analysis and the brewery's mash temperatures, so a brewer can see before brewing how a new lot of malt will behave.
Why it is still open
Published models need lab measurements that are not on standard certificates. Robert Muller's 2000 model of fermentability after high-temperature mashing uses enzyme and starch data, and its author says it "is not appropriate" for ordinary mashes. Evans and colleagues measured starch- and wall-breaking enzymes in 94 commercial malts and concluded that measures "more predictive of a malt's brewing performance" than the current ones are needed. A search found no public dataset or tool that works from certificate numbers and brewhouse mash temperatures.
Built with what exists
Certificates of analysis and brew logs (gravity at the start and end of fermentation) from licensed breweries, small lab mashes at three temperatures, Python with a regression or gradient-boosting model, and Zenodo to publish the data.
First step
Collect one licensed brewery's malt certificates and matching brew logs for the past year, and measure how much of the variation in attenuation the certificate numbers alone explain.
Done looks like
At least 100 malt lots from at least three maltsters, each with its certificate and measured attenuation at three mash temperatures, published; and a model that predicts attenuation within 2 percentage points for 80% of lots from a maltster it was not trained on.