A batch of Assam leaf reaches the drier dark, damp, and technically still alive: its enzymes are still turning, its water content still high enough to rot given the chance. Firing is the step that ends that motion for good. It is also the last of three long dehydrations the leaf goes through, after a full day losing close to a third of its own weight in the withering trough and a shorter spell losing more of it to oxidation on the fermenting floor. Twenty minutes or so of hot air, forced through the leaf at 90 to 130 degrees Celsius (194 to 266 Fahrenheit), drives what is left down to somewhere close to 3 percent of the leaf's own weight, a target precise enough that missing it by a couple of points changes what the batch becomes. Get there, and the tea is finished, stable, and ready for the warehouse. Miss it, and no amount of good work upstream saves what comes out the other end.

The driers that finish the leaf

Two machines do almost all of this work, whichever road the leaf took to get here. CTC and orthodox part ways long before firing, over how hard the leaf gets crushed, and both still leave through the same kind of drier at the end. The endless chain pressure drier, ECP for short, carries the leaf down through several stacked mesh trays on a moving chain, hottest air at the bottom and cooler air higher up, so the leaf meets a falling temperature as it descends; a full pass runs close to 23 minutes in a baseline measured by a 2023 study in the journal AgriEngineering. The fluid bed drier, FBD, does the same job a different way: it forces air up through a perforated floor hard enough to lift the tea into a churning, floating layer, moving heat into the leaf faster and more evenly than a static tray can.

A 2014 study of a Kilburn-made vibro fluid bed drier, the vibrating cousin of the plain FBD, measured the whole job done in 20 minutes flat in a Bangladeshi factory, cutting a batch from 69.1 percent moisture to 2.8 percent, with inlet air at 130 degrees Celsius cooling to 90 by the time it left. The choice between the two machines carries a real cost beyond the cup: a published energy audit of tea processing units in Tripura found that switching an ECP line for an FBD line saves 10 to 15 percent of the drying stage's energy bill on its own, no change to the leaf required.

Three percent, not zero and not six

The final number is not arbitrary, and it is not a nice round target either. Dried too little, left above roughly 3.5 percent, the leaf will not keep: enough water survives for the enzymes firing was supposed to have killed to limp back into action, and mold finds a foothold too, per processing guidance published by Agriculture.Institute. Dried too far, below about 1 percent, the leaf holds its shape but starts giving up the very flavor compounds fermentation just spent an afternoon building, scorched away instead of preserved. Between those two failures sits a narrow band, close to 2.5 to 3 percent, dry enough to sit in a warehouse for two years without turning and wet enough to still taste of something.

Firing fixes the leaf's weight to that number. In the same twenty minutes, it also fixes the leaf's smell. The malt note is not a fermentation product, whatever the trade assumes: it forms mostly in these final drying stages, when heat converts a stockpile of amino acids the withering trough spent a full day assembling into the aldehydes that read as malt, as traced in full elsewhere on this record. Everything upstream of the drier is provisional. The fired batch is the one a factory actually certifies and ships.

The rate that can ruin a batch anyway

Reaching 3 percent is necessary and still not sufficient, because the speed of the drying matters as much as where it lands. Tea-industry drying references, tracing the figure to Tocklai's own trials, put a number on it: lose moisture faster than about 4 percent of the leaf's weight a minute, and the batch turns bitter and harsh regardless of the final reading; hold the loss between roughly 2.8 and 3.6 percent a minute, and the identical final moisture reads as good tea instead.

Overshoot that rate and the failure has a name: case hardening. The outer skin of each granule scorches and seals before the water trapped inside has had time to leave, so the drier's exit reading says the batch is dry while the middle of the leaf is not, a mismatch that shows up later as mold in the warehouse rather than at the factory gate. Undershoot it, running the drier too cool or too slow, and the opposite failure sets in: a dull, flat liquor, described in the trade as stewed, the water gone but the aroma cooked off along with it. A drier's operator is not only watching a clock and a thermometer. The loss rate is the third dial, and it is the one that actually decides the cup.

What actually heats the drier

None of this runs for free. Thermal energy, the heat behind the wither, the drier, and everything else on the floor that is not an electric motor, is the overwhelming share of what a tea factory spends on power: close to 85.65 percent of total energy consumption, a 2023 study of estates across Assam and West Bengal found in the journal Frontiers in Sustainable Food Systems. The same study put a number on firing specifically: for every 1,000 kilograms of made tea, the average factory burns through roughly 630 kilograms of coal (about ten ounces of coal for every pound of tea), some diesel to keep a backup generator running, and just over 100 kilowatt-hours of electricity, and the drying stage alone produces close to 2,762 kilograms of carbon dioxide for that same 1,000 kilograms, nearly three pounds of gas for every pound of leaf. A separate energy audit of tea units in nearby Tripura found a single coal-fired drier burns through about 1.4 kilos of fuel for every kilogram it dries.

That bill has been outrunning what the tea sells for. The Indian Tea Association has calculated that tea prices grew at a compounded 4 percent a year over the past decade while the coal and gas a factory burns climbed 9 to 15 percent a year over the same stretch, Business Standard reported in December 2023. A drier that has always run on coal or firewood is no longer only an environmental question. It is a line item getting worse every year, which is most of why the fuel underneath the leaf is starting to change.

The trees the driers were eating

Coal covers only part of the load; wood makes up much of the rest, and that wood carries a cost the balance sheet does not show directly. A 2022 study of an Indian tea plant's energy use, published in Frontiers in Energy Research, states plainly that the use of forest firewood for thermal energy is causing deforestation in India, since the principal source of that firewood remains the natural forest rather than any managed woodlot. Rachel Cracknell, environment and climate lead at the Ethical Tea Partnership, has called the result hidden deforestation, wood cut and burned out of sight of anyone holding the finished cup, Mongabay reported in 2024.

Government money and industry money are both now pushing the other way. India's Ministry of New and Renewable Energy runs a national programme that funds new biomass briquette and pellet plants with direct central assistance, capital a small factory would rarely spend on its own boiler otherwise. The same swap has already been tried, and studied, in East Africa's tea belt, and the honest result is mixed. A 2022 study in the journal Sustainability, pricing Kenyan firewood at what it actually costs on the market rather than treating it as free, found briquettes save real trees and cut real carbon, but do not yet save money: at those market prices, briquettes are not cost-competitive, and the study found their price would need to fall by roughly half before the swap paid for itself on cost alone. The Tripura audit found a version of the same catch from the fuel side: a drier that burns 150 kilos of coal an hour needs roughly 250 kilos of biomass instead to do the identical job, because biomass simply carries less heat for its weight than coal does. Biomass wins on carbon and on standing trees. It does not win on cost or on volume, and a factory has to haul, store, and feed about two-thirds more tonnage to get there.

The bush burning itself

The most complete version of this swap does not import fuel from anywhere. It burns the tea bush's own trimmings. A 2025 study out of the National Institute of Technology in Arunachal Pradesh, working with pruned tea-plant material from Northeastern India, pressed the woody waste every garden already generates at pruning time into fuel pellets carrying under 5 percent moisture and a heating value of 14 to 15.5 megajoules per kilogram, in the same range as seasoned firewood. Proven at scale or not yet, a garden that fires its next flush on wood cut from its own rows this pruning season closes a loop a coal-fed or firewood-fed drier never could.

Every other figure on this record measures the leaf: a wither percentage, a catechin range, a grade code. This is the one that measures the garden feeding itself, the bush's own waste wood coming back around as the heat that dries the bush's own leaf.