Ruthenium

Market Overview: The Smallest Traded PGM Runs on a Roughly 30-Tonne Annual Float

Ruthenium (Ru, atomic number 44) is one of the six platinum-group metals (PGMs) alongside platinum, palladium, rhodium, iridium, and osmium, and by traded volume it is the smallest, alongside iridium, of the metals that see meaningful industrial and investment turnover. The USGS Mineral Commodity Summaries 2026, Platinum-Group Metals chapter groups ruthenium with palladium, platinum, iridium, osmium, and rhodium under a single commodity heading, reflecting the fact that all six are mined together and only separated at the refining stage. According to Johnson Matthey's PGM Market Report, May 2025, global primary ruthenium supply was 954,000 troy ounces (29.7 tonnes) in 2025, down from 1,113,000 oz (34.6 tonnes) in 2024 and roughly in line with the 930,000–1,417,000 oz range seen over 2020–2024. The International Platinum Group Metals Association's May 2026 ruthenium factsheet separately puts global refined ruthenium output at around 31 tonnes in 2025, up about 3 tonnes from 2019 (IPA Ruthenium Factsheet, May 2026).

Why the market is this thin: ruthenium is never mined as a target metal in its own right. It is recovered only as a minor byproduct stream within platinum-palladium-rhodium PGM concentrate, and its output therefore rises and falls with decisions producers make about platinum and palladium — not with ruthenium demand itself. The UK Competition and Markets Authority's review of the Sibanye-Lonmin merger found that ruthenium and iridium combined accounted for only roughly 0–10% of the combined group's PGM revenues, and that “production decisions would not realistically be determined by these two by-products” (UK CMA, Sibanye/Lonmin merger decision). This byproduct structure is the single most important fact about the ruthenium market: supply is structurally inelastic to price.

Reserves and resources: concentrated almost entirely in the Bushveld Complex

World PGM resources are estimated at more than 100 million kilograms, with the largest reserves and resources located in South Africa's Bushveld Complex, whose UG2 Chromitite reef is described as the largest known repository of PGM resources globally, containing more than 75% of known platinum resources (IPA Ruthenium Factsheet, May 2026). Ruthenium itself is one of the least abundant PGMs in the Earth's crust, occurring at only about 0.57 parts per billion overall (0.34 ppb in the upper crust), and its availability is a direct function of Bushveld UG2 ore grades, which typically run 2–6 grams of combined PGMs per tonne (IPA Ruthenium Factsheet, May 2026). USGS MCS 2026 reports U.S. imports for consumption of ruthenium rose to an estimated 14,000 kg in 2025, up from 9,780 kg in 2024, alongside a 53% year-on-year jump in the average ruthenium price — the largest price increase of any PGM USGS tracked for 2025.

Supply-demand balance: a persistent, widening physical deficit

Source: Johnson Matthey, PGM Market Report, May 2025; underlying supply/demand history spreadsheet at Johnson Matthey PGM market data (ruthenium and iridium). Negative stock movements indicate the market drew down producer and merchant inventories to cover the shortfall between mine supply and industrial consumption; Johnson Matthey's May 2026 report confirms the ruthenium market moved into an even deeper deficit in 2025, and that unless further drawdowns occur, 2026 fundamentals remain tight (Mining Weekly, 14 May 2026, reporting Johnson Matthey's 2026 PGM Market Report).

2023→2025 momentum: demand growth outpacing a byproduct-constrained mine base

Total ruthenium demand rose from 1,064,000 oz in 2023 to 1,197,000 oz in 2024 (+13%) and to 1,224,000 oz in 2025, driven by a rebound in hard-disk-drive purchasing tied to AI-era data centre buildout and record consumption in Chinese chemical process catalysts (Johnson Matthey, PGM Market Report, May 2025). Over the same period, primary mine supply was essentially flat to lower, constrained by weather disruptions at major South African producers and by the byproduct economics described above. Afriforesight (15 Aug 2025) reports that Johnson Matthey expected the 2025 ruthenium supply deficit to widen to 240,000 oz, up from 84,000 oz in 2024, attributing the widening gap partly to weather-related production disruptions at Valterra Platinum (formerly Anglo American Platinum) and Impala Platinum in the first quarter of 2025.

Supply Chain: A Byproduct Riding on Platinum and Palladium Mine Economics

South African Bushveld producers: Valterra Platinum, Impala Platinum, Sibanye-Stillwater, Northam

South Africa's Bushveld Igneous Complex is, per Heraeus, the source of roughly 72% of global primary PGM supply and by far the largest source of ruthenium (Heraeus, PGM Recycling for Hydrogen whitepaper, April 2026). Valterra Platinum (formerly Anglo American Platinum) reported a 36% year-on-year rise in its realised ruthenium price in Q1 2025, reflecting the metal's sharp price appreciation (Mining Weekly, 24 Apr 2025, citing Anglo American Platinum's Q1 2025 production report). Impala Platinum (Implats) mines and refines PGMs from its Rustenburg, Marula, and Impala Bafokeng operations in South Africa, together with its Zimbabwean subsidiary Zimplats and Canadian refining assets, historically using an ion-exchange technology to produce ruthenium sponge (Implats, Impala fact sheet, 2024; Materials journal, Reclaiming Ruthenium review, 2026). Sibanye-Stillwater, which describes itself as a top-tier PGM and gold producer, “also produces and refines iridium and ruthenium, nickel, chrome, copper and cobalt” alongside its South African and U.S. (Montana) operations (Sibanye-Stillwater media release, 13 Feb 2026; Sibanye-Stillwater, Southern Africa operations). Northam Platinum operates the Zondereinde, Booysendal, and Eland Bushveld mines and is a smaller but growing contributor to South African ruthenium byproduct output alongside the larger three.

Nornickel: Russia's byproduct ruthenium from Norilsk-Talnakh nickel-copper ore

Nornickel (MMC Norilsk Nickel) is the world's largest palladium producer and a major nickel and copper producer; ruthenium is one of thirteen metals and elements the company lists as byproducts of its Norilsk Industrial District (Taimyr Peninsula) and Kola Peninsula operations, alongside cobalt, rhodium, silver, gold, iridium, selenium, and tellurium (Nornickel, corporate profile; Nornickel, 2025 production results, 28 Jan 2026). Because ruthenium and iridium ride on nickel-copper-palladium mining economics rather than their own price signals, Nornickel's 2026 guidance for a slight decline in overall PGM output — due to a temporary shift toward lower-grade disseminated ore — implies a corresponding constraint on Russian byproduct ruthenium supply (Nornickel, 2025 production results and 2026 guidance). Russian palladium has separately become the subject of a major U.S. trade action: in February 2026 the U.S. Department of Commerce issued a preliminary antidumping determination of 132.83% against unwrought Russian palladium, followed by a preliminary countervailing duty of 109.10% in March 2026 (a combined 241.93%), after a petition filed by Sibanye-Stillwater and the United Steelworkers Union in July 2025 (Miningmx, 18 Mar 2026). These duties target palladium specifically, but underscore that Russian-origin PGM byproducts, including ruthenium, now trade under a materially different trade-policy shadow in the U.S. market than they did before 2025.

Zimplats: the smallest of the three primary sources

Zimplats, Implats' Zimbabwean subsidiary operating on the Great Dyke, reports its production as “6E” ounces — platinum, palladium, rhodium, ruthenium, iridium, and gold combined. In the quarter to September 2025, Zimplats' 6E metal-in-final-product fell to 142,535 ounces, with ruthenium production down 22% year-on-year to 4,968 ounces and iridium down 18% to 2,353 ounces (Mining Zimbabwe, 1 Nov 2025). Zimplats' own fact sheet shows ruthenium at roughly 3.5% of its 6E ounce basket by weight ratio, a smaller share than platinum (46.6%), palladium (37.3%), or rhodium (3.9%) (Implats, Zimplats fact sheet, 2024). Zimplats has separately invested over US$444 million in smelter and SO₂ abatement upgrades, with phase two targeted for completion by June 2028, aimed at processing more of its own concentrate domestically rather than exporting matte for offshore refining (Mining Zimbabwe, 23 Apr 2025).

Refining leaders and specialist finishers: Johnson Matthey, Heraeus, Umicore, BASF, Furuya Metal, TANAKA

Beyond the mine-site refineries in South Africa, Russia, and Zimbabwe, a small group of global precious-metals refiners and fabricators handle the bulk of ruthenium's downstream refining, chemical conversion, and market-making. Johnson Matthey is both a refiner and the PGM industry's principal published data source, publishing the annual PGM Market Report used throughout this deep-dive (Johnson Matthey, PGM market reports). Heraeus publishes daily PGM reference prices and, together with Sibanye-Stillwater, co-developed a ruthenium-based PEM electrolyser catalyst discussed in Section 3 (Heraeus, current precious metal prices). Umicore and BASF both operate PGM refining and catalyst businesses that process ruthenium-bearing scrap and virgin metal into industrial catalyst and electronic-grade forms. In Japan, Furuya Metal is a specialist ruthenium and iridium refiner that states its raw material is “typically mined in South Africa,” often containing significant impurities that require proprietary purification before use, and that it has developed direct relationships with “three major mining companies” in South Africa given that the major source of iridium and ruthenium is concentrated there (Furuya Metal, ruthenium processing technology; The Worldfolio, Furuya Metal interview, 2021). Furuya Metal invested to expand ruthenium refining capacity at its Tsuchiura plant as early as 2018, positioning it as a key upstream supplier into the Japanese electronics and chemical industries (ACN Newswire, 21 Jun 2018). TANAKA Holdings, another major Japanese precious-metals group, similarly supplies ruthenium sputtering targets and chemical compounds into the electronics supply chain, though primary-source production data for TANAKA's ruthenium volumes are not separately published.

Vertical integration example: Furuya Metal's business model illustrates the whole chain end to end — South African mine concentrate is shipped to Japan, refined into high-purity ruthenium sponge and powder, converted into sputtering targets and chemical precursors for the semiconductor and chemical industries, and ultimately recycled back through the same refining circuit once end-of-life electronic scrap and spent catalysts are collected (Furuya Metal, Ruthenium Wiring).

End Uses: Chemical Catalysts and Data Storage Dominate, Semiconductors Are the Wildcard

Source: Johnson Matthey, PGM Market Report, May 2025. Ruthenium use is “overwhelmingly in industrial applications, with only a small amount of investment and jewellery consumption” captured in the “other” line.

1. Hard disk drives (HDD): the historic anchor demand, now bifurcating between MAMR and HAMR

Ruthenium's highest-profile end use has long been as an ultra-thin interlayer in perpendicular magnetic recording (PMR) and, more recently, microwave-assisted magnetic recording (MAMR) hard-disk media, where a nanometre-scale ruthenium film sits between the magnetic recording layer and the soft underlayer to control magnetic grain orientation (Precious Metals Commodity Management, HDD technology transitions overview). Heraeus's 2026 forecast states plainly that “hard disk drive manufacturers have seen a recovery and acceleration in demand as AI datacenter buildout increases,” and that while heat-assisted magnetic recording (HAMR) — which uses much lower amounts of ruthenium — is now being sold commercially with rising volumes, MAMR “still represents by far the majority of sales” (Heraeus Precious Forecast 2026). The IPA's May 2026 factsheet frames this as a long-run substitution risk: “ruthenium-based technologies are expected to reach a capacity limit,” and HAMR’s advance to deliver higher storage densities implies ruthenium content in new media will trend lower than in current technologies even as overall drive shipment volumes rise with AI data-centre demand (IPA Ruthenium Factsheet, May 2026). Johnson Matthey's May 2025 report explicitly credits data-centre investment for a 2024 rebound in HDD-driven ruthenium purchasing, reversing several years of decline as SSDs took storage share from HDDs (Johnson Matthey PGM Market Report, May 2025).

2. Semiconductor barrier and liner layers: the emerging structural demand driver

As copper interconnect pitches scale below roughly 20–25 nanometres, the industry-standard tantalum nitride/tantalum (TaN/Ta) barrier-liner stack becomes disproportionately resistive, because TaN and Ta occupy a fixed minimum thickness (historically >4 nm) regardless of how narrow the copper line becomes, eating into the conductive copper cross-section (PatSnap, Ruthenium Interconnect Barrier Layer Patents 2026). Ruthenium’s low electron mean free path (roughly 6–8 nm versus copper's ~40 nm) makes it far less sensitive to the resistance increase that plagues narrow copper wires at advanced nodes, while its strong copper wettability enables direct electroplating without a separate physical-vapor- deposition copper seed layer (Vik's Newsletter, Ruthenium interconnects, Jan 2025). Applied Materials' ruthenium-cobalt (RuCo) liner technology reduces liner thickness by 33% (down to 20 angstroms) while cutting interconnect resistance by up to 25%; in a 2 nm test chip, Applied's process delivered a 2.5% performance boost over a standard process flow, and Applied states that “RuCo has been adopted by all leading logic chipmakers for the 2nm node” (USGS Mineral Commodity Summaries 2025 — PGMs; Allan Chemical Corporation, barrier layer advances, Nov 2025).

Intel demonstrated a “subtractive ruthenium” interconnect process with air-gap integration at IEDM 2024, and its June 2026 process roadmap update confirms Intel demonstrated subtractive ruthenium with airgap integration achieving up to ~35% capacitance reduction and measurable frequency gains versus copper for its Intel 18A-P process family (Intel Newsroom, VLSI Symposium update, 16 Jun 2026; Tom's Hardware, IEDM 2024 coverage). TSMC patent filings describe ruthenium-based liner layers for its 2 nm (N2) and beyond nodes, including a 2024 patent on a ruthenium-based liner for copper interconnects that reduces via “pinch points” and a 2025 patent on selective TaN barrier deposition combined with ruthenium liners engineered with a bottom-to-sidewall thickness ratio of 2:1 to 8:1 (PatSnap, Ruthenium interconnects at sub-10nm BEOL nodes, Apr 2026; US12094770B2, Ruthenium-based liner for a copper interconnect). Samsung's 2025 patent filings describe a hybrid metallization scheme where narrow upper metal lines below 10 nm use ruthenium while wider lines retain copper, citing scaling data showing ruthenium, rhodium, and iridium outperform copper at sub-10 nm pitches (PatSnap, Ruthenium Interconnect Barrier Layer Patents 2026). Deposition equipment for these processes is supplied principally by Applied Materials and Lam Research, both of which have introduced ruthenium- and cobalt-capable deposition tools aimed specifically at sub-3nm logic metallization (Applied Materials, semiconductor products).

Why this matters for the ruthenium market: semiconductor barrier/liner demand is still a small fraction of total ruthenium consumption relative to chemicals and HDDs, but it is additive, structurally growing (tied to the number of metal layers per leading-edge chip and the number of leading-edge fabs), and largely insulated from the same substitution pressure (HAMR) that threatens HDD demand. Ruthenium demand data reported by Johnson Matthey do not yet break out semiconductor interconnect volumes separately from the broader “electrical & electronics” category, so the precise tonnage attributable to leading-edge logic chips is not publicly disclosed.

3. Chemical catalysts: China's caprolactam boom is now the single largest end use

Chemical-sector demand for ruthenium set a new all-time high in 2024 and grew further in 2025, and Johnson Matthey attributes the growth overwhelmingly to China: “the use of ruthenium process catalysts is dominated by China,” where ruthenium is used in bulk chemical applications, “in particular by the caprolactam sector, which produces feedstocks for the nylon industry” and “has seen substantial growth in recent years, driven by capacity additions for nylon 6 resins” (Johnson Matthey, PGM Market Report, May 2025). Outside China, ruthenium sees comparatively limited chemicals-industry use, mainly in speciality chemical catalysts and, alongside iridium, in the Cativa acetic-acid process. The IPA factsheet adds further specific applications: catalytic wet air oxidation for industrial wastewater treatment, asymmetric and transfer hydrogenation, homogeneous catalysis, and ammonia cracking/decomposition/ synthesis (IPA Ruthenium Factsheet, May 2026).

4. Electrochemistry: chlor-alkali DSA anodes and the PEM electrolyser opportunity

Ruthenium oxide, mixed with titanium oxide, is the electrocatalytic coating at the heart of the dimensionally stable anode (DSA), the standard electrode used across the global chlor-alkali industry since the 1960s. The benchmark commercial coating composition is approximately 30% RuO₂/70% TiO₂, chosen because it combines high catalytic activity for chlorine evolution with a lower cost than a pure-PGM-oxide coating and strong long-term stability (DiVA/Chalmers, Theoretical and Experimental Studies of DSA Electrodes). One academic review estimates the chlor-alkali industry consumes roughly 3 tonnes of ruthenium per year for anodic electrocatalytic coatings — on the order of 10% of annual global ruthenium production — (Academia.edu, Anodic Electrocatalytic Coatings for Electrolytic Chlorine Production: A Review). More recently, ruthenium oxide has emerged as a candidate to reduce iridium loading in PEM (proton-exchange-membrane) water electrolysers for green hydrogen: RuO₂ is the most catalytically active pure-metal-oxide electrocatalyst for the oxygen evolution reaction, and roughly one-tenth the cost of IrO₂, but suffers durability problems (dissolving to soluble, volatile RuO₄) under the harsh acidic conditions of a PEM cell, which is why commercial catalysts blend Ru with Ir rather than using Ru alone (RSC Advances, Ruthenium-based electrocatalyst for acidic water oxidation, Sep 2025). Heraeus, working with Sibanye-Stillwater, launched a ruthenium-iridium oxide catalyst in November 2023 that the companies say enables up to an 85% reduction in iridium loading versus a pure IrO₂ catalyst (below 0.1 tonnes of iridium per gigawatt of installed electrolyser capacity), while delivering up to 50 times higher mass activity than iridium oxide alone and matching IrO₂'s stability after 30,000 accelerated degradation cycles (Heraeus, Ru-Ir catalyst press release, 14 Nov 2023; Heraeus, ruthenium catalyst for PEM electrolysis technical sheet). Academic groups continue to pursue similar chemistry, including a Rice University-affiliated team that demonstrated a Ru₆IrOₓ catalyst with an iridium-to-ruthenium atomic ratio of only 1:6, stable for over 1,500 hours of continuous electrolysis and validated at 50 amps in a scaled-up cell by industrial electrolyser maker De Nora (Rice University-affiliated research, low-iridium RuO₂ PEM catalyst, 2025).

5. Fuel cell anodes, alloys, and jewellery: smaller, well-established niches

Platinum-ruthenium (PtRu) alloy catalysts, typically at a roughly 1:1 atomic ratio, have been the standard anode electrocatalyst for direct methanol fuel cells (DMFCs) since the mid-1960s discovery that ruthenium alloying dramatically improves platinum's tolerance to carbon-monoxide poisoning during methanol electro-oxidation (Journal of Electroanalytical Chemistry, PtRu DMFC anode review; Fuel Cell Store, platinum-ruthenium catalysts). Titanium-ruthenium (Ti-Ru) alloys are used where corrosion resistance in aggressive chemical environments is required, drawing on ruthenium's role (also seen in DSA coatings) as a stabilising alloy addition to titanium. In jewellery, ruthenium is used both as a plating layer on white gold to achieve a bright, neutral white finish, and as a hardening alloy addition to platinum (typically “950Pt/Ru,” 95% platinum/5% ruthenium) and palladium (“950 palladium,” 95% palladium/5% ruthenium) jewellery alloys, improving hardness, castability, and wear resistance (London Assay Office, Palladium: An Introduction; CBIjoux, white gold plating explainer). Ruthenium polypyridyl dyes (notably the compound known as N719) have also been studied extensively as photosensitisers in dye-sensitized solar cells (DSSCs) since the 1990s, though DSSCs remain a research- and niche-commercial-scale technology rather than a material consumer of ruthenium at PGM market scale (University of Milan, Ruthenium Complexes for DSSCs review).

Prices: No Exchange Listing, But 2025 Delivered the Steepest Rally in Nearly Two Decades

Why there is no formal exchange benchmark

Not applicable in the LME/COMEX/LBMA sense — the market is too small and too illiquid to support exchange-traded futures or a fixing panel. Unlike platinum and palladium, which trade on NYMEX/COMEX futures and have LBMA/LPPM-affiliated benchmark auctions, and unlike gold and silver, which have deep LBMA fixings, ruthenium trades exclusively over the counter through refiners and bullion banks. Johnson Matthey and Heraeus both publish reference prices derived from their own dealing books and market intelligence, and these figures are the closest thing the market has to a benchmark (Johnson Matthey, PGM prices and trading; Heraeus, current precious metal prices). Commercial data platforms such as metal.com (Shanghai Metals Market) and Investing.com separately publish daily "Ruthenium 99.99% CIF North West Europe" spot assessments compiled from dealer surveys, but these are market-intelligence products, not exchange-cleared prices (Investing.com, Ruthenium 99.99% CIF NW Europe historical data; Shanghai Metals Market, ruthenium historical price chart).

Price history, 2024–2026

Sources: USGS MCS 2026, platinum-group metals chapter; Johnson Matthey PGM Market Report, May 2025; Heraeus Precious Forecast 2026; Afriforesight, 15 Aug 2025. Independent price trackers corroborate the scale of the move: Strategic Metals Invest shows ruthenium closing 2025 at $48.68/gram (roughly $1,515/oz at the December 2025 close cited), up 162.57% for the year, before easing to around $62.59/gram (about $1,946/oz) by early July 2026 on that particular series — illustrating how far reference prices can diverge across data providers given the absence of a single official benchmark (Strategic Metals Invest, ruthenium price history).

What drove the 2025 rally

USGS attributes the 2025 PGM price increases broadly to “decreased production and increased demand, particularly for rhodium in the hard disk and chemical catalyst industries” — language that in USGS's underlying data table applies most directly to ruthenium, given ruthenium posted the largest percentage price gain (53%) of any PGM that year (USGS MCS 2026). Johnson Matthey's contemporaneous market commentary cites “strong consumer demand, and talk of strategic buying in China” behind the March 2025 push to a three-and-a-half-year high of $630/oz (Johnson Matthey, PGM Market Report, May 2025). Afriforesight separately links the continuation of the rally through mid-2025 to weather-related production disruptions at Valterra Platinum and Impala Platinum, reinforcing that on a byproduct metal, even modest mine-level disruptions can have an outsized price impact given the market's small absolute size (Afriforesight, 15 Aug 2025).

Basis differentials and grade considerations

Public price series generally quote 99.9–99.99% purity ruthenium metal (sponge, powder, or ingot) delivered in Europe (CIF Rotterdam/North West Europe) or ex-works China; there is no widely published geographic or grade-differential structure comparable to, say, cobalt's Chinese versus European premiums, both because trading volumes are too thin to support a liquid basis market and because the overwhelming majority of physical flow moves through a small number of refiner relationships rather than an open spot market. Sputtering-target-grade (semiconductor) and catalyst-grade ruthenium compounds (chlorides, oxides, organometallics) trade at a further, largely undisclosed premium to metal-sponge reference prices, reflecting additional refining and fabrication value-add.

Trade Policy: PGMs Sit Outside Tariff Regimes, But Russia-Specific Actions Are Reshaping Flows

U.S. tariff exemption for PGMs

Heraeus's April 2025 market appraisal confirms that “platinum and other PGMs will not be subject to new US tariffs,” noting that the platinum-group metals were included in a long list of products excluded from the reciprocal tariff regime announced that month, specifically because of their designation as “critical minerals” under U.S. legislation. This exemption covers platinum, palladium, rhodium, and the “small PGMs” — ruthenium and iridium (Heraeus, PGM Appraisal, 7 Apr 2025).

The Russian palladium antidumping/countervailing duty action and its ruthenium spillover

In July 2025, Sibanye-Stillwater and the United Steelworkers Union filed antidumping and countervailing duty petitions with the U.S. Department of Commerce and International Trade Commission against unwrought palladium imports from Russia, arguing Russian producers — principally Nornickel — had been dumping palladium below market value since sanctions following the 2022 Ukraine invasion, with U.S. imports of Russian palladium rising 35% between 2022 and 2024 even as palladium prices fell 50% (Shanghai Metals Market, 31 Jul 2025). Commerce issued a preliminary antidumping determination of 132.83% in February 2026 and a preliminary countervailing duty of 109.10% in March 2026, for a combined preliminary rate of 241.93%; a final ITC determination was scheduled for May 2026 (Miningmx, 18 Mar 2026; Sibanye-Stillwater media release, 13 Feb 2026).

Why it matters for ruthenium specifically: the petitions and duties target palladium, not ruthenium directly — but because Nornickel produces ruthenium as a joint byproduct of the same Norilsk-Talnakh ore that yields its palladium, any structural shift in how Russian PGM output reaches Western markets (via different intermediaries, different countries of first import, or reduced overall throughput if Russian producers respond to duties by curtailing output) has second-order implications for how much Russian-origin ruthenium reaches U.S. and allied buyers, and through which trade channels. No ruthenium-specific antidumping or countervailing action has been filed as of July 2026.

EU Critical Raw Materials Act status

Ruthenium's status under the EU's 2023 Critical Raw Materials Act framework tracks the broader PGM basket: the European Commission's 2023 Critical Raw Materials assessment groups platinum-group metals together, reflecting their shared byproduct production structure and near-total EU import dependence, since there is essentially no primary PGM mining within the EU27 (IPA Ruthenium Factsheet, May 2026). The EU's principal exposure is to South African and Russian supply concentration rather than to Chinese export licensing, in contrast to metals such as antimony, gallium, germanium, tungsten, and bismuth, which face direct MOFCOM export controls; ruthenium is not currently subject to any Chinese export licensing regime, because China is a demand-side (not supply-side) participant in the ruthenium market.

No MOFCOM controls; China is the demand side, not the supply side

Not applicable — China has no meaningful domestic ruthenium mine production and is not a控 net exporter subject to licensing controls. Unlike tungsten, antimony, gallium, germanium, bismuth, and the rare earths, where China dominates mining and/or refining and has used MOFCOM export licensing as trade leverage, ruthenium's supply chain runs almost entirely through South Africa, Russia, and Zimbabwe. China is instead the single largest source of demand growth, through its caprolactam/nylon-6 chemical industry, and Johnson Matthey specifically flags “talk of strategic buying in China” as a 2025 price driver (Johnson Matthey, PGM Market Report, May 2025) — the opposite dynamic from the export-control metals, where China restricts outbound supply rather than competing for inbound supply.

ESG & Recycling: A Structurally Under-Developed Secondary Supply Chain

Why open-loop recycling barely exists for ruthenium

Johnson Matthey's circular-economy whitepaper states that while roughly 20% of platinum and around 30% of both palladium and rhodium supplied annually for purchase comes from open-loop recycling (chiefly spent automotive catalytic converters), “ruthenium and iridium lack consistent sources of secondary supply as they are typically recycled in closed loop” (Johnson Matthey, Reclaiming the Future: PGM Insights for a Circular Economy). The IPA's May 2026 factsheet elaborates: open-loop recycling — where the original purchaser does not retain ownership of the metal — is “not yet well developed” for ruthenium “because ruthenium content is often very low in final products and recovery processes have not been developed given the economics of dealing with small amounts of metal despite its value.” The IPA notes it publishes recycled-content figures for platinum, palladium, and rhodium but “could not publish data on recycled ruthenium due to limited availability of data,” since confidentiality rules require at least three reporting companies before a figure can be disclosed — itself a signal of how few firms recycle ruthenium at commercial scale (IPA Ruthenium Factsheet, May 2026).

Closed-loop recycling: HDD scrap and spent chemical catalysts

Most ruthenium recycling that does occur is closed-loop, meaning the metal is recovered and returned directly to the original industrial user — for example, a chemical company recovering ruthenium from a spent caprolactam catalyst to manufacture fresh catalyst, or an HDD platter manufacturer recovering sputtering-target and wafer scrap internally. Because this metal never re-enters the open market, it is not counted as part of reported primary or secondary supply (IPA Ruthenium Factsheet, May 2026). A 2026 review in the journal Materials, “Reclaiming Ruthenium: A Comprehensive Review of Recovery,” catalogues the main secondary feedstocks — spent catalysts, metallurgical byproducts, wastewaters, and even spent nuclear fuel — and describes recovery chemistry typically proceeding via hydrometallurgical leaching in acidic or alkaline solution with strong oxidants, followed by multi-stage purification using precipitation, solvent extraction, distillation, ion exchange, electrolysis, pyrolysis, or reduction, often requiring repeated cycles to reach required purity (Materials, Reclaiming Ruthenium review, 2026).

Market-research estimates of ruthenium's specific share of PGM recycling vary widely and should be treated cautiously given the primary-source data gap the IPA itself flags: one market report estimates ruthenium contributed roughly 6.8% of PGM recycling market value, driven by HDD-coating and chemical-catalyst scrap growing at an estimated 8.4% CAGR, against secondary supply's roughly 32% share of total combined PGM supply (Dataintelo, Recycling of Platinum Group Metals Market Report) — but because this is a secondary market-research estimate rather than IPA- or Johnson Matthey-reported primary data, it should be read as indicative rather than authoritative.

Standards and responsible sourcing

There is no ruthenium-specific responsible-sourcing standard comparable to the LBMA Responsible Sourcing guidance for gold and silver. PGM producers and refiners generally operate under the broader OECD Due Diligence Guidance for Responsible Supply Chains of Minerals from Conflict-Affected and High-Risk Areas, and South African, Russian, and Zimbabwean PGM operations are covered by their respective national mining, environmental, and labour regulatory regimes rather than by any PGM-specific international certification scheme. Sibanye-Stillwater, Valterra Platinum, Implats, and Nornickel all publish annual sustainability reports covering worker safety, water use, and emissions at their PGM operations, but no independent third-party ruthenium chain-of-custody certification currently exists.

Environmental and labour hotspots

South African deep-level Bushveld mining carries well-documented safety and labour risks common to the country's broader mining sector, including seismicity, heat stress at depths approaching 2.2 km, and periodic industrial action; these risks apply to PGM mining broadly rather than to ruthenium specifically, since ruthenium cannot be separated from the platinum-palladium-rhodium ore stream at the mining stage (IPA Ruthenium Factsheet, May 2026). South African electricity supply constraints (load-shedding) have also been repeatedly cited as a production-disruption risk factor for PGM smelting and refining, indirectly affecting ruthenium output alongside the other PGMs.

Forward Look 2026–2030: Two Demand Curves Crossing, One Supply Curve Barely Moving

Supply pipeline: essentially fixed by Bushveld, Norilsk, and Great Dyke mine plans

Because ruthenium has no primary mines, there is no capacity-expansion pipeline analogous to a new lithium or nickel project. Any increase in ruthenium output requires either (a) an increase in overall Bushveld, Norilsk-Talnakh, or Great Dyke PGM ore throughput, or (b) a shift in ore mix or refining recovery rates that yields more ruthenium per tonne of platinum/palladium produced. Phoenix Refining's industry commentary specifically flags the commissioning of major new PGM mining complexes such as Anglo American's Platreef project in South Africa's Limpopo province as a potential future source of incremental byproduct ruthenium supply, alongside “a growing pipeline of possible new developments” (Phoenix Refining, Ruthenium for the Hydrogen Industry). Zimplats' smelter and SO₂-abatement investment (targeted for phase-two completion by June 2028) will modestly increase Zimbabwean processing capacity but is aimed primarily at environmental compliance and domestic value addition rather than expanding mined tonnage (Mining Zimbabwe, 23 Apr 2025). Nornickel's own 2026 guidance points the other direction — a slight decline in overall PGM output due to lower-grade ore — reinforcing that Russian byproduct ruthenium supply is more likely to contract than expand over the near term (Nornickel, 2025 production results and 2026 guidance).

Demand scenario 1: semiconductor adoption becomes a durable structural driver

If ruthenium-cobalt liners and ruthenium interconnects are adopted broadly across 2nm-class (TSMC N2, Samsung SF2, Intel 18A/18A-P) and future 1.4nm-class (Intel 14A) logic nodes, and if ruthenium use extends further into contact and via-fill applications as several patent filings suggest, then semiconductor demand could become a durable, multi-year growth vector layered on top of (rather than replacing) the existing chemical-catalyst and HDD demand base. TSMC's 2 nm process entered volume production in Q4 2025 and Intel's 18A entered high-volume manufacturing around the same period, with Samsung's SF2 node also ramping — meaning 2026 is the first full year in which multiple leading foundries are simultaneously running ruthenium-containing metallization schemes at meaningful wafer volumes (Wikipedia, 2 nm process, citing TSMC/Samsung/Intel disclosures; Intel 18A high-volume manufacturing update, Jan 2026).

Demand scenario 2: HAMR erodes ruthenium-per-drive faster than data-centre volume growth compensates

The competing scenario is that heat-assisted magnetic recording, which the IPA and Heraeus both confirm uses substantially less ruthenium per drive than MAMR/PMR, scales quickly enough across Seagate and Western Digital's roadmaps that total HDD-sector ruthenium consumption declines even as AI-driven data-centre exabyte demand keeps rising (Heraeus Precious Forecast 2026; IPA Ruthenium Factsheet, May 2026). As of mid-2026, both sources describe MAMR as still dominant by sales volume, meaning this transition — while directionally clear — has not yet materially reduced aggregate HDD-sector ruthenium demand.

Key risks: byproduct inelasticity, South African operating risk, and Russian trade-policy spillover

The dominant structural risk across every scenario is that ruthenium supply cannot respond to price or demand signals the way a primary-mined commodity can. If either the semiconductor-adoption or the AI-data-centre-HDD demand scenario plays out more strongly than currently modelled, the market has essentially one adjustment mechanism left after producer-stock drawdowns are exhausted: price. Johnson Matthey's 2026 report already flags this dynamic, describing a market where “ruthenium and iridium are also expected to be in deficit amid rising demand from the data storage and energy transition sectors” (Mining Weekly, 14 May 2026). Compounding this, South African operational risk (load-shedding, deep-level mining safety, and periodic weather disruption) and Russian geopolitical/trade-policy risk (the palladium antidumping/countervailing case, and the broader overhang of Western sanctions architecture around Nornickel) both sit directly upstream of ruthenium supply, given that South Africa and Russia together account for the large majority of global output.

What would change the picture

A durable de-escalation of Russia-related sanctions and trade actions could ease some Western buyers' sourcing constraints but would not increase physical ruthenium supply, since Nornickel's output is capped by nickel/copper/palladium mine economics rather than by trade-policy access. A faster-than-expected HAMR ramp would reduce HDD-sector ruthenium intensity but is unlikely to be fully offset by exabyte growth in the near term given how dominant MAMR remains as of mid-2026. The single most significant potential structural swing factor is the pace and breadth of ruthenium adoption in leading-edge semiconductor interconnects: if Ru/RuCo liners become standard not just at the tightest 2nm-class metal layers but across a growing share of the full interconnect stack in future nodes, ruthenium could shift from a metal whose demand story is dominated by a declining legacy application (HDDs) to one dominated by a structurally growing one (advanced logic), a transition that would mark a fundamental re-rating of the metal's demand base.