Why Minsur’s Pisco Refinery Upgrade Is About Slag, Not Scale

Minsur has just asked Peru’s environmental regulator to approve 21 modifications to its Pisco tin refinery, backed by US$9.7 million in capital. The timing is not coincidental.

The facility set an all-time processing record in 2025, and it is now pressing against the physical limits of its own infrastructure.

That request arrives at a moment when the global tin market is running a sustained deficit and Minsur holds the position of the world’s second-largest refined tin producer. The strategic question for the company is not whether to expand, but how to keep a record-breaking plant running without it becoming a victim of its own throughput.

What follows here separates signal from headline. This piece unpacks what the 21 modifications actually involve, why slag management and safety infrastructure are the real constraints behind the headline figure, and what Minsur’s incremental brownfield strategy signals about the economics of sustaining high-output tin refining in a supply-constrained market.

A record year that pushed infrastructure to its limits

In 2025, Minsur’s Pisco smelting and refinery facility in Ica, Peru, processed 87,527 tonnes of tin concentrate. That was an all-time record for the plant, and it converted into 32,981 tonnes of refined tin, a year-on-year increase of roughly 6.6%.

A record of that size is also a stress test. It reveals exactly where a facility runs out of room.

The 2025 result was not a single lever pulled. It came from three converging operational improvements, aided by the plant’s artificial intelligence optimisation platform, Optimus TIN:

• Greater workforce productivity across the smelting and refining lines
• Increased furnace processing throughput
• Enhanced metallurgical recovery rates, extracting more refined metal from each tonne of concentrate

The record anchor: 87,527 tonnes of tin concentrate processed in a single year, the highest volume in the facility’s history.

The International Tin Association (ITA) ranked Minsur as the world’s second-largest refined tin producer in 2024, at 36,300 tonnes, behind China’s Yunnan Tin. That scale is what makes Pisco’s throughput ceiling a matter of global relevance, not just corporate housekeeping.

The cost of running flat out

The first half of 2026 showed what sustaining that ceiling actually demands. Pisco produced 8,314 tonnes of refined tin in Q1 2026 and 7,034 tonnes in Q2 2026, a modest step down from the record pace.

According to SMM coverage, the dip traced back to a longer annual maintenance window at the smelter: 11.5 days in 2026, compared with 6.5 days in 2024.

The read here is straightforward. When a plant runs near its infrastructure ceiling, structured downtime is not optional, it is the price of reliability. The record tells you Pisco is operating close to the top of its current envelope, and the H1 2026 pause confirms that holding that level requires active, ongoing management rather than a single capital event.

Why slag is the bottleneck no headline mentions

The US$9.7 million figure grabs attention. The chemistry behind it explains the money.

High-throughput tin smelting produces large volumes of slag, the molten waste layer left after metal is separated from concentrate. In Minsur’s Ausmelt furnace, a submerged-lance smelting system, the feed mixes tin oxide concentrate, coal, fuel, iron ore, limestone, and recirculating process materials. Slag is the unavoidable by-product, and it carries meaningful metal content: typically 3-25% tin oxide and 10-40% iron oxide.

That residual metal is the commercial catch. Some research estimates that roughly 5% of all smelted tin globally is discarded as slag because of processing inefficiencies, which is why slag handling is a revenue question, not only a waste question.

Metallurgical slag management sits at the intersection of regulatory compliance, metal recovery economics, and environmental permitting; across base metals smelting, how operators account for residual metal content in slag determines whether the material is classified as waste or a recoverable resource.

Here is how the current loop works at Pisco:

1. Concentrate, coal, iron ore, and recirculating materials are fed into the Ausmelt furnace and smelted
2. Slag with iron and residual tin content is produced as an output
3. The slag is temporarily stockpiled on-site
4. It is then transported to the San Rafael mine for use as paste-fill, a material that backfills mined-out voids

As throughput climbs, that loop tightens. More refined tin means more slag, and the on-site deposit is now approaching its safe storage capacity limit. Minsur’s submission includes a proposed 20% enlargement of the slag deposit for exactly this reason.

This is the single modification that most directly ties Minsur’s ability to repeat its 2025 record in future years. Without added storage headroom, the constraint stops being a smelting limit and becomes a storage limit. The plant could keep running hot and still be forced to slow down simply because it has nowhere to put the waste.

The upgrade programme addresses several of these physical and safety pressure points at once. The table below sets out five of the most significant changes among the 21 proposed.

The 21 modifications: debottlenecking, not expansion

Read as a whole, the 21 modifications are not a random list. They form a structured response to three distinct gaps that emerge when a plant runs at full capacity: recovering value from waste streams, expanding physical infrastructure where constraints have appeared, and retiring safety risks before they become liabilities.

Minsur’s submission to Senace, Peru’s environmental regulator, is explicit on one point. None of the 21 modifications are expected to generate significant new environmental impacts.

Senace’s ITS evaluation framework designates the Informe Técnico Sustentatorio as the appropriate instrument for modifications that generate non-significant environmental impacts or involve technological improvements, which is precisely why Minsur’s 21-modification submission fits this regulatory pathway rather than requiring a full environmental impact assessment.

None of the 21 proposed modifications are expected to generate significant new environmental impacts, positioning the programme as compliance-preserving rather than a greenfield-scale expansion.

That framing matters. It signals the programme is designed to sustain the 2025 record, not to push output beyond it.

Waste recovery and recirculating material upgrades

Three changes carry direct metal-recovery or revenue implications:

• Big bag co-processing: Large bags containing residual tin will be incinerated and co-processed in Ausmelt Furnaces 1 and 2 to recover the remaining metal, replacing external disposal
• Used refractory bricks: These will now be sold commercially rather than reprocessed at San Rafael, a route previously ruled out due to insufficient recovery rates
• Vacuum furnace installation: The most technically significant single change, a new vacuum furnace will handle recirculating materials currently processed through electrolytic cells

Each converts a cost or a waste stream into recoverable value, which is the quiet economic logic running through this category.

Infrastructure additions and safety modernisation

The infrastructure changes expand physical capacity where the plant has run short of room: the 20% slag deposit expansion, a new billet cooling zone, a cayana (refractory) heating area, the repurposing of existing coal storage to also hold tin concentrate, and a new external storage area for incoming materials.

The safety upgrades retire identified risks. Mechanised ladle skimming and metal cutting tools replace manual equipment to cut ergonomic hazards for workers. A fire detection, alarm, and suppression system will be added to the main conveyor belt, which currently has no such protection.

Alongside these, Minsur plans to replace aging components: a post-combustion blower, four obsolete compressors, and a regenerative dryer. These are infrastructure longevity measures, the kind that prevent unplanned failure rather than add output.

This submission follows the same pattern as the eighth Supporting Technical Instrument (ITS), which Senace approved on 18 September 2025 at an investment of about US$5.5 million. Both sit within a broader phased upgrade programme reportedly totalling around US$55 million.

The distinction between debottlenecking and expansion is the analytical takeaway. This programme is not built to produce more tin than Pisco’s 2025 record. It is built so the plant can reliably repeat that record without being halted by storage limits, equipment failure, or a compliance gap.

Brownfield logic in a structurally tight market

Step back from the machinery and the US$9.7 million figure starts to look larger than it reads. In tin refining, the alternative to incremental upgrades is prohibitively expensive.

Industry analyses contrast brownfield retrofits against greenfield builds. A brownfield retrofit can often be completed in roughly 18 months at a cost of US$50-100 million per site. A comparable greenfield refinery has historically cost US$400 million to US$1 billion and taken three to five years to commission (a figure that should be treated as an unverified industry estimate).

That cost gap is why sustaining existing high-output plants matters so much right now. The market has no cheap or fast way to add fresh capacity.

The return profile of brownfield mining investment returns differs structurally from greenfield builds: existing permitted infrastructure, known geology, and an operational workforce compress both capital intensity and time-to-production, which is why incremental upgrades to proven facilities tend to deliver more predictable outcomes than new builds at comparable headline costs.

The deficit picture underlines the point, tightening across three consecutive readings:

• 2024: ITA reported a global refined tin deficit of approximately 2,200 tonnes
• 2025: Guosen Securities forecast a deficit of approximately 13,000 tonnes, with demand at 388,000 tonnes against supply of 375,000 tonnes
• H1 2026: Rzzro estimated a deficit of approximately 8,000 tonnes

Price context: As of early September 2026, the cash-settlement tin price traded near US$55,000 per tonne, confirmed by Westmetall, Alumeco, and Business Insider. At that level, every tonne Pisco fails to produce carries a meaningful revenue cost.

The supply backdrop sharpens the risk. Exports from Myanmar’s Wa State have been estimated at just 60% of pre-2023 levels, and Indonesian refined tin exports reportedly fell about 33% in 2024 (both figures flagged as unverified estimates). With two major producing regions running below historical output, the reliability of remaining high-output refiners becomes a market-level variable.

The deficit trajectory compounds the exposure, because tin supply chain disruptions at individual refinery nodes propagate quickly through downstream electronics and semiconductor manufacturing, where substitution options are limited and inventory buffers are thin.

That is the strategic weight behind a modest headline number. Minsur’s ability to hold Pisco at its 2025 record is not only a company objective. In a widening deficit, it is a meaningful input into global supply adequacy.

What the upgrade programme resolves, and what remains open

The US$9.7 million programme does specific, bounded work. It is worth separating what it settles from what it leaves untouched.

What the 21 modifications resolve:

• The slag storage bottleneck, via the 20% deposit enlargement
• Identified safety risks, including the unprotected conveyor and manual ladle handling
• Recovery of metal value from waste streams previously disposed of externally
• Regulatory compliance, maintained without seeking greenfield-scale environmental approvals

What the programme does not address:

• Concentrate supply from San Rafael, a separate operational variable. Q1 2025 feed from San Rafael rose 23% year-on-year, with the B2 plant up 44%, so the feed side has been building independently of this submission
• Global tin price movements, which sit entirely outside Minsur’s control
• The unresolved gap between national and facility-level production data

That last point deserves a caveat rather than a conclusion. EY’s Peru mining and metals investment guide reports national tin production at 28,517 tonnes in 2025, yet Minsur reported 32,981 tonnes at Pisco alone. Whether that reflects a data reporting gap or a definitional difference in what counts as national production is unclear, and anyone tracking Peru’s role in global tin supply should hold both figures with that uncertainty in mind.

One more variable sits above all of this: the Senace approval itself. The submission is a request, not a decision, and the regulator’s timeline will determine when the modifications can proceed.

Peru’s mining regulatory environment has been shifting on multiple fronts in 2026, with concession tenure reforms and environmental permitting timelines creating a backdrop against which Senace’s review of the Pisco submission will unfold.

A programme that clears infrastructure bottlenecks while leaving concentrate supply and price dynamics as open questions is a measured, bounded investment. Understanding those boundaries gives you a more accurate basis for judging what Minsur’s production trajectory through 2026 and 2027 can realistically look like.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. Several supply and cost estimates referenced here are flagged as unverified and are subject to change based on market developments.