Lithium process reagents and their last mile to the puna (Catamarca)
Lithium's most expensive input is not energy or labor: it is reagents, and Argentina barely produces them. Rio Tinto published that they were 42% of its production cost in 2025; the national State measured them at USD 1,918 per tonne of carbonate and warned that ~90% of that money is a single product: soda ash. South America's only producer sits in Río Negro, some 1,700 km from the salar in a straight line — and the State itself wrote that this freight «weighs very heavily on the final cost of a tonne delivered to the salar area» and that a complementary plant should therefore be installed near the lithium projects. With four Catamarca operations starting up in 12 months, that leaves a market of USD 55-95 M/year of reagent delivered at the plant gate. The molecule is fought over abroad; what is left on this side is the last mile —storage, silo, big bags and the final 145 km of dirt at 4,000 meters— which today is solved project by project: three operations do not publish who provides it and the fourth's feasibility study declares that it outsources it to a local company. Nobody sells it as a shared service to all four.
This is not a window niche: the reagent is opex and gets consumed every day for the ~40 years of mine life. It starts with four start-ups in 12 months and grows with every ramp-up.
On this page
What the market is made of
The TAM is activity, not what you will invoice — and in reagents the difference is brutal. The molecule travels from Wyoming, Turkey or Río Negro; the purchase is decided by a global desk (Rio Tinto controls two of the three plants that consume, and planned to import half of its soda ash through a Chilean port); and on-site conditioning is already built, with three separate warehouses and two forklifts dedicated to reagents alone. All of that is captive. Your real gap is the last mile: storage, silo, bagging, quality control and the final 145-215 km of dirt — the addressable segment, small in money and large in accessibility.
~USD 62 M/year (~89% of the reagent TAM), in three layers that do not compete with one another.
See the full breakdown
USD 4-12 M/year addressable by a local/domestic entrant = essentially the entire last-mile TAM (puna freight + storage + conditioning + bagging + dosing: USD 7-30 M/year) plus the clearance and bonded-warehouse fee on USD 48-83 M/year of nationalized merchandise, plus the unquantified upside of regional quarry lime (the lime ratio per t of LCE is missing). The basis for the SAM is not inference: the project's primary source states that reagent and product logistics are OUTSOURCED TO a local company in three independent mentions.
USD 1.5-5 M/year for ONE entrant within 2-3 years. Honest arithmetic: you do not win all four operators (Rio Tinto contracts in bulk for two operations and Zijin brings its own chain), so the wedge is 1-2 clients over 30-50% of the SAM. A realistic path, in this order: (1) a handling and last-mile subcontract with the logistics company already approved — 2-4 months to first invoice; (2) your own storage terminal with one operator under contract — 6-12 months; (3) scaling up to supply delivered at the plant gate, with the molecule bought from an importer.
Who really pays?
«The mine» is not a single door, and in reagents the money splits into three flows contracted separately — one of them does not even pass through Argentina:
Rio Tinto controls two of the three plants that consume carbonate in Catamarca and buys in bulk prob; Sal de Vida's feasibility study planned to import 50% of its soda ash through the Port of Antofagasta, Chile, and Livent reinstated the Pocitos-Antofagasta rail link used, verbatim, «to import reagents» verif. South America's only producer, ALPAT, is in Río Negro: it does produce —170,000 to 210,000 t/year declared to the State— but freight to the salar weighs so heavily that the official study itself says a complementary plant near the lithium projects should be considered verif.
Sal de Vida's feasibility study says it without hedging: «lithium carbonate and reagent transport logistics will be outsourced to a local company». It also clarifies twice that the operator provides fuel and servicing to all of its fleet except reagent and product logistics. Reagents arrive in 1 t big bags on 28 t flatbed trucks. It is the link in the niche that the study puts in writing as outsourced and local thesis.
Sal de Vida built a reagent storage warehouse and a reagent and consumables preparation building with three separate storage areas —quicklime, caustic soda and sodium carbonate—, plus its own lime slaking plant; it has 4 forklifts, 2 dedicated exclusively to reagents, and a forklift fleet that goes from 5 to 10 when it scales up. The four operators on the same salar replicate the same infrastructure. You beat it on price at the next scale-up, not by regulation.
Which projects move this demand
Expansion (Phase 1B) of the historic Fénix lithium-carbonate project in the Salar del Hombre Muerto, operated by Rio Tinto via Minera del Altiplano…
see the project →Tres Quebradas stage 2: plant and infrastructure for 40,000 t/year of lithium carbonate via direct extraction (DLE), per art…
see the project →When the window opens
In reagents what you watch is where the molecule comes from and how it can be moved. The plant of 40,000 tonnes a year approved at Tres Quebradas on 29 July 2026 has construction from June 2026 and operation scheduled for 1 April 2029. And there are two rules that are never announced because they are already written into the regulation: from 1 May to 30 September a winter procedure applies, which pushes you to have the input stockpiled up there beforehand; and since January 2026 the reagent travels in a dedicated convoy, escorted and only in daylight. These four numbers are the ones that give notice.
See the evidence
See the evidence
If the reagent comes in through Chile (port of Antofagasta) or via the Pocitos-Antofagasta railway under an international trader's contract, the last mile stays inside that contract and the local entrant never sees it.
See the evidence
Today the wind is strongly at its back: Q2-2026 spot USD 22,043/t (+125% y/y in the first half), Rio Tinto's realized price USD 18,960/t and Fénix's C1 below USD 5/kg verif Rio Tinto Q2-2026 filing with the SEC.
See the evidence
Tres Quebradas' Phase 2 now has its act: resolution published on 29-Jul-2026, with USD 594,136,000 of computable investment verif — the USD 709 M was the announced amount.
See the evidence
Fénix went from a 2-3 month to a 1-2 day concentration cycle with Phase 1B. Less evaporation and less softening push the ratio from the 2.081 actually observed toward the stoichiometric 1.435: up to 31% less soda ash volume at the same output.
See the evidence
See the remaining 5 risks
The Sáenz-Jalil agreement (May-2021) splits the taxes and royalties of the disputed Hombre Muerto strip 50/50, and lapses once Congress settles the boundary: the winning province becomes the sole authority.
See the evidence
Three spills in seven months in Fiambalá and a $254.1 million fine on Zijin-Liex from the Ministerio de Minería de Catamarca prob converging provincial press after the brine spill of 25-Mar-2026.
See the evidence
The storm of Jul 21-23, 2026 left RP 43 Antofagasta-Salar del Hombre Muerto impassable, closed the Paso San Francisco with winds above 170 km/h, recorded -27 °C at 3Q's camp and forced the rescue of five miners with a helicopter brought in from Buenos Aires.
See the evidence
If it executes its expansion to 550,000 t/year verif projection declared by ALPAT in the official study of the Secretariat of Mining, the domestic price falls and the trading margin compresses.
See the evidence
This is what sets this niche apart from camps, geomembranes or earthmoving: the reagent is perpetual OPEX, not construction capex.
See the evidence
The opportunity in depth
The opportunity in depth
The shared storage and conditioning terminal. Today the same service is solved four times over, one per operation, and nobody sells it shared: covered shed, one-tonne big bags on 28 t flatbed trucks and dedicated forklifts. A terminal in the area with bulk receiving, silo, bagging, moisture control and shared freight turns four sunk costs into a service with a single asset. Ticket of USD 1-4 M of capex. And since January 2026 there is a rule that makes it dearer and more necessary at the same time: a provincial resolution requires the convoy of mineral and chemical substances —the text names the reagents— not to be mixed with other products, to carry an outsourced escort for every one to three trucks, to travel only in daylight and to respect lower weight caps. Pooling the same input from several operators into one convoy is still possible; what is no longer possible is mixing the reagent with other cargo to fill the truck.
Regional quarry lime, the only reagent with a possible 100% Catamarca chain. Sal de Vida receives granular quicklime and slakes it at its own slaking plant: it buys the product, it does not make it. The NOA has limestone quarries and the province taxes limestone at 0.75%. The lithium market asked for ~250,000 t of lime in 2023 and projects more than 800,000 t by 2027 prob — and nobody has documented who delivers it at Hombre Muerto. Here you compete against a freight rate, not against Wyoming.
Foreign trade and the bonded warehouse of western Catamarca. The provincial mining suppliers' chamber listed, in its 2023 category map, a single member in foreign trade for a portfolio that brings in a good part of its main reagent from abroad, with a 9% tariff (Sal de Vida plans to import half of its sodium carbonate through Chile). And the clearance link cannot be provided from within the province: out of the 175 bonded warehouses authorised nationwide, Catamarca has zero. What can change that is the Free Zone in the Tinogasta department, with the national agreement signed in September 2026 and awaiting the provincial legislature's vote.
When you get paid, and what blocks it
already in
split
Imports are the reference supplier even though ALPAT produces: 600,000 t imported for lithium between 2016 and 2023 = 41% of Argentina's total imports of the compound prob; 34.18% of the country's entire import of it goes to LCE verif. Probable origin: the United States (92.12% of world reserves, USGS 2022) and Turkey (3.52%) verif. A non-refundable 9% tariff. NO trader identified by name: that is the stated gap in this map.
Punta Delgado, San Antonio Oeste, Río Negro — ~1,700 km from the salar in a straight line (more by road). Ammonia-soda process with assistance from Krebs Paris.
See the evidence
Sal de Vida planned to import 50% of its requirement through the Port of Antofagasta, CHILE (that half never touches Argentine customs); Livent reinstated the Pocitos-Antofagasta rail link 'used to ship product and import reagents'; Rio Tinto's '30 in 30' plan states 'program based contracting strategies'; Zijin runs 3Q with its own Chinese supply chain. The reagent can reach the puna without passing through an Argentine purchasing decision.
This is the specific captive to break. Sal de Vida has a Reagent Storage (Area 3500) + Reagent and Consumable Preparation Building (Area 3500) with three separate storage areas —quicklime, caustic and sodium carbonate—, 4 forklifts with 2 dedicated exclusively to reagents and a fleet of 5 forklifts in Stage 1 that goes to 10 in Stage 2.
See the evidence
See the remaining 3 players
The lithium market demanded ~250,000 t of lime in 2023 and projects >800,000 t by 2027 prob. We could NOT document who delivers quicklime at the Salar del Hombre Muerto, or from which quarry: it is the most concrete intra-national substitution gap.
Fénix moves ~40 trucks/day of inputs and product. The mining suppliers' chamber's roll listed in 2023 only 4 members in transport and 1 in foreign trade across 19 categories and 123 memberships. Transport was the most contested category of 2026: CAPROMITI threatened road blockades in Fiambalá because Zijin-Liex hired a carrier from Córdoba, and the Catamarca Transport Chamber joined in.
A Salta bonded warehouse where Eramine consolidated 100 big bags into MSC containers. Catamarca has no equivalent; hence the push for a Free Trade Zone in Tinogasta (agreement with the national government of September 2026; its law went back to the provincial Senate on 14-Oct-2026). The foreign-trade link had ONE (1) member in the entire 2023 roll of the mining suppliers' chamber.
The jobs it createsThis is the niche that creates the least employment per dollar billed (70-80% of the TAM is imported molecule, not labor) and the one that creates the MOST STABLE employment in the whole chain: it is opex of an operation with ~40 years of remaining mine life, not an 18-month construction job. La lectura completa para el que busca trabajo, en la hoja de este nicho para la gente →
calculate it
The number is our own estimate. The calculation, step by step, is below.
The full calculation, step by step
Concentration Extreme in the product and fragmented only in freight. The UIA-CAEM-BID report of May 2026 sums it up: domestic supply in 73 of 91 mining categories and a single domestic soda ash producer. That producer —ALPAT, in Río Negro— does produce: version 2 of the Mining Secretariat's official study describes it operating with 225,000 t/year installed and ~400 employees, and with a series declared by the company itself of 170,000 t in 2021 and 2022 and 210,000 t in 2023. The problem is not that it does not exist: it is that it sits at the other end of the country, and the same official study says that transport cost «weighs very heavily on the final cost of a tonne of sodium carbonate delivered to the salar area». So this is a market of one distant domestic producer plus imports, with the largest buyer (Rio Tinto, owner of Fénix and Sal de Vida) integrating its own imports through a foreign port and with on-site conditioning self-supplied redundantly by the four operators. The only layer with real competition and possible entry is the last mile: storage, bagging, quality control and the final 145-215 km of dirt road up to the salt flat.
The rule that moves it
Two rules of different rank move this niche: the local-content commitments written into the RIGI resolutions (95.34% of what is paid to suppliers at Hombre Muerto Oeste goes to local suppliers, 60% at Fénix) — which improve the climate but do not cover the operating spend where the reagent lives, and which moreover count goods and works, not services —, and the provincial Re.P.E.M., a secretariat resolution that demands two years of local roots and turns a joint venture with a local partner into the key to entry.
See the underlying reading
The underlying policy that opens up this demand. It is the same inference declared by the rules and the program pillars pushing in this direction.
the RIGI promise is kept + opening and deregulationSee the full legal grounds
Where the number comes from
The band is USD 55-95 M/year on real 2026 consumption, with a midpoint of ~USD 70 M/year. At 2027-28 nominal capacity it rises to USD 85-145 M/year, and the ceiling reaches ~USD 220 M/year if Tres Quebradas' Phase 2 gets built
See the calculation, the variables and how it was validated
The TAM is built from three variables you can watch one at a time: how much carbonate the province produces, how much soda ash each tonne demands and how much the reagent costs delivered up there. Change one and the number is recalculated.
Robustness check: by the other route —tonnes of soda ash times their price— you get USD 50-90 M/year from that input alone, plus USD 25-60 M/year of lime, caustic soda and hydrochloric acid: the two methods cross inside the band. And the calculation runs on real 2026 consumption, not on nominal capacity: with three plants commissioning simultaneously, effective production stays below nominal, which is why the 2027-28 nameplate scenario is published separately, at USD 85-145 M/year.
The number rests on a few variables. The formula shows how it moves when each one changes; and each variable carries its freshness seal — how often it is worth revisiting. estim
The figure rests on two real prices, not on an assumption. The first is official: the Mining Secretariat measured that reagents are 40.9% of the spend of producing brine lithium in Argentina —USD 1,918 per tonne of carbonate— and that actual soda ash consumption is 2.081 tonnes per tonne of product. The second we opened ourselves in Sal de Vida's feasibility study: USD 1,314 per tonne of reagents, and there the document clarifies something decisive —the prices are quoted delivered to site, at 4,000 meters—. On those two anchors we multiply by the carbonate capacity the province has today (Fénix, Sal de Vida and Tres Quebradas), and we leave out Hombre Muerto Oeste because it produces chloride, not carbonate, and therefore does not consume the reagent. The annual total is our own estimate and we say so. The storage and final-leg rates are stated assumptions, anchored in the only published puna freight price that exists: USD 133 per tonne for the outbound leg to port, from the same document.
Coverage: ARCA's official list of authorized bonded warehouses —175 across the country, counted one by one— and CAPPROMIN's trade map, the provincial chamber of mining suppliers (123 memberships with a 2023 cut-off: 1 member in foreign trade, 1 in insurance and 12 in inputs) · Sep 14, 2026 · not reviewed: the Re.P.E.M., which is the mandatory provincial register, publishes the total number of registrants but not the list of names; and no operation publishes who supplies it today with soda ash and lime
How to cite this figure: Despegue (2026). Lithium process reagents and their last mile to the puna (Catamarca) · Catamarca. despegueargentina.com/en/catamarca/reactivos-soda-ash-litio · terms of use
Neighbouring markets2 markets in the same group, from USD 0,8 to USD 11 M a year
We look at what your company does and tell you whether we see a sign that it fits this market, with the evidence behind it. If we do not see one, we say so too and name the condition that would change it. One page as a dated PDF, within 48 working hours.
Open the report ↗A real analysis · a real company, open in full →
- 20+ years in technology, 15 of them in data and AI, for clients across Europe and the Americas
- Certified in AI governance (ISO/IEC 42001)
- Machine Learning (Google Cloud)
- Registered expert with the European Commission