Sequelograph

A real event, read carefully. Then a fiction that does not pretend to be the news.

The record, 18 Sep 2026

Rio Tinto's Argentine lithium plants post early output in commissioning

On September 18, the Direct Lithium Extraction Argentina 2026 news site reported that Rio Tinto's Fenix 1B and Sal de Vida projects produced first material in the second quarter of 2026 ahead of schedule but remained in commissioning, with undisclosed volumes, while the Rincon starter plant delivered 385 tonnes as engineers worked to stabilize its processing reactors. The report says Rincon uses direct lithium extraction, which Rio Tinto says conserves water and yields battery-grade carbonate above 99.5% purity, and notes the project's $2.5 billion investment, 60,000-tonne annual target, and larger expansion expected in 2028. Early output during commissioning is not steady production, and the company gave no cause or timeline for the reactor work.

Read the source at Direct Lithium Extraction Argentina 2026

The record ends here. Everything below this line is invented.

What if a plant could make 385 tonnes but not the same tonne twice?

Part one

The Starter Plant

Lot 41 failed at 99.4, and Rosa stared at the number until it stopped looking like a number and started looking like a locked door. Battery grade meant 99.5 or better – half a percent of everything else, at most – and the starter plant's whole week had been balanced on that line. She pulled her gloves off one finger at a time, the way she did when she needed her hands to stay honest, and ran the sample again. The ship sailed on her numbers.

It was her nineteenth day on rotation at the Vientos starter plant, eighteen hundred kilometers from her daughter's school parade. The plant sat on the edge of the Salar de los Vientos, a white flat so wide the horizon looked poured, a cluster of low buildings and pipe racks that looked, from the ridge road, like a ship that had mistaken the desert for the sea. It was small on purpose: three thousand tonnes a year, a trial-sized machine built to prove the chemistry before the big one rose beside it – fifty-seven thousand tonnes, billions of dollars, first production in 2028. This quarter the little plant had made 385 tonnes. Across the ocean, a buyer was waiting to load two hundred of them onto a ship.

On the video call that morning, Alma had held up her parade costume – a cardboard vicuña head, ears slightly uneven – and made Rosa promise. "You said the ship can't leave without your numbers."

"The ship can't leave without my numbers," Rosa had agreed, because it was true, and because eight-year-olds deserve at least one true thing a day. "I'll be home for the parade. I promise on the vicuña."

Now Lot 41 sat at 99.4, and Lots 38 through 40 – passed, sealed, stacked by the loading bay – suddenly looked less like a shipment and more like a rumor.

The plant's story about itself was simple, and Rosa had believed it the way you believe a map. Brine came up from the wells. The direct extraction units pulled the lithium out in hours – no evaporation ponds, no waiting on the sun – and the finishing line dried it into white carbonate powder. Battery grade, above 99.5, steady as a metronome. The engineers' word for the spring had been stabilizing: the reactors were new, the crews were new, and every week ran smoother than the last.

So when the assays started swinging – 99.7 Monday, 99.5 Tuesday, 99.4 today – Rosa blamed the reactors like everyone else. New machines hunted. That was commissioning. The map said the chemistry was proven; the territory just needed time.

She spent the afternoon the way quality technicians spend afternoons when the numbers wobble: re-running samples, checking the calibration, cleaning the instrument's lens twice. The re-runs agreed with the runs. The calibration was true. The lens was clean. Small questions opened and closed like doors down a hallway – wrong vial? No, barcodes matched. Wrong shift's sample? No, timestamps matched. Contamination in the drying oven? She tested the blanks. Clean.

At four she did what the procedure didn't say to do: she walked the retained samples back to their wells. Every lot on the shelf carried a tag with its brine source, and she'd never had a reason to read those tags as a pattern. Lots 38, 39, 40 – the good ones – all came from the east wells, the old ones, drilled first. Lot 41, and the two wobbling lots from last week, came from the north wells, brought online in March.

She stood in the sample room with the tags in her hand and felt the map redraw itself. The reactors weren't hunting. They were being fed two different brines and asked to sing one note. The extraction units had been tuned on east-well chemistry – one lithium grade, one set of companion salts – and the north wells ran leaner and saltier. Same desert, different water. The plant could make good powder. It just couldn't make the same powder twice until somebody taught the schedule which well was which.

The fix took two days and no new equipment. Rosa wrote it as a blending rotation – east and north brine sequenced so the units never saw a swing bigger than they could absorb – and the operations chief, a tired man with brine stains on his boots, read it twice and said, "Run it." By Friday the assays lined up like fence posts: 99.6, 99.7, 99.6. Lot 41 was reprocessed through the adjusted rotation and came out at 99.7, and Rosa sealed it herself, pressing the stencil onto the white bag with the flat of her hand.

Two hundred tonnes, qualified on her numbers. The trucks began rolling for the port and the ship, and Rosa booked the flight home that would land her with one day to spare before the parade. She stood at the loading bay watching the white bags go – each one a small rectangular promise – and did the arithmetic the whole valley was doing that spring: A starter plant can ship lithium while its big expansion is still years away. Three thousand tonnes learning to walk so that sixty thousand could run. Her daughter's vicuña ears, she thought, were also slightly uneven, and also exactly right.

The convoy's dust had barely settled when her phone lit up with a message from the buyer's lab – twelve hours ahead, already Monday morning there. A photo of her Lot 39 bag, open on their bench. An attachment: their assay. And one line, polite as a handshake with a stone in it:

Our panel reads Lot 39 at 99.4. Please advise.

What’s real

  • Rincon starter delivered 385 tonnes in a quarter.
  • Engineers worked to stabilize processing reactors.
  • Rio Tinto says DLE conserves water, tops 99.5% purity.

What’s invented

  • Rosa, Alma, and the starter lab are fictional.
  • The well-grade swing and blending fix are invented.
  • The cross-ocean assay dispute is speculative.

Part two: Fourteen Columns

The ship sails on Rosa's numbers, but the buyer's lab reads one lot below battery-grade, and both labs are right.

Part two isn’t on sale yet. Check back soon.

Filed under earth, lithium, brine, mining.

More records

How it works

Each record starts with one real science event, with a link to where it was reported.

A short story then asks what happens next. Part one is free to read.

Part two costs a few dollars, paid once. You keep it.