Overview
Palladium is the critical metal for which no Western mine exists and none can be built. It is not mined for itself: it arises as a by-product of nickel-sulphide mining in Russia and of platinum-group mining in South Africa. No ore body capable of supporting primary production exists in the United States, the European Union, Japan or China. Approximately 40% of primary output originates in Russia and most of the remainder in South Africa. Roughly 85% of demand is automotive catalysis, where palladium oxidises carbon monoxide and unburned hydrocarbons in three-way catalysts and where no substitute performs the function at equivalent cost outside partial platinum thrifting.
The 2026 price record shows the consequence. Palladium traded at $1,802/oz on 3 February 2026 (approximately $58,000/kg), up 79.75% year on year, having taken a three-year high of $2,200/oz in January. Heraeus had forecast a $950–1,500 corridor for 2026; the market passed through the upper bound within the first month of the year — inside a market that is structurally oversupplied, because automotive palladium demand is in decline as battery-electric vehicles displace petrol powertrains. A metal in surplus that doubles in a year is not responding to consumption; it is responding to who holds it.
For palladium, the entire recoverable Western resource is anthropogenic and already inside the border. Three aqueous streams carry it today: the leachate produced when recyclers dissolve end-of-life autocatalyst washcoat in acid; the leachate of electronic scrap (PCBs, connectors, multilayer ceramic capacitors); and the plating and activation baths of PCB fabrication together with their rinse waters, where palladium leaves the process continuously and is classified as effluent. In all three, dissolution has already been performed and paid for by another process. What is missing is a separation route whose cost does not scale with dilution — classical recovery (solvent extraction, ion exchange, cementation, selective precipitation) is configured for concentrate, and below roughly single-digit mg/L its reagent and consumable cost exceeds the value of the recovered metal even at $58,000/kg.
Applications
Autocatalyst recyclers. After the washcoat is dissolved — routine practice — the resulting liquor holds palladium, platinum and rhodium together with base metals and dissolved support. ARBOK recovers the metal value directly from that liquor rather than requiring the recycler to sell mixed concentrate to a small number of integrated refiners.
Electronic scrap recyclers. Palladium appears in multilayer ceramic capacitors as internal electrode material (historically pure palladium, now often palladium-silver or base-metal alternatives) and in connectors as a palladium-nickel plating layer under a gold flash. Leaching of electronic scrap is established practice; the recovered liquor is copper-dominated with gold, silver and palladium in trace.
PCB fabrication and connector plating. Palladium colloid or ionic palladium activates non-conductive through-hole walls before electroless copper deposition; connector plating uses palladium electrolytes directly. Metal leaves the process with drag-out into rinse tanks and with spent-bath disposal — a continuous stream, generated every operating day, inside a facility with full process knowledge of its own bath chemistry.
PGM refiners. The compact second-stage refining unit converts the ARBOK concentrate into palladium alongside platinum and rhodium.
Operating Principle
ARBOK-ZWD inverts the target of separation: rather than extracting a solute from water, it removes water from the solution. The stream is processed whole under deep vacuum at ambient temperature, where water separates from the dissolved load as a phase-change process rather than as thermal evaporation — the origin of the low specific energy. Clean water is returned at up to 100% of intake by volume; the entire dissolved load leaves as dry separated fractions.
Two consequences matter for palladium. First, process cost is set by volume, not by concentration: a dilute stream and a more concentrated stream are handled identically and at identical cost, which removes precisely the constraint that renders classical recovery uneconomic on rinse waters. Second, the process is non-selective by design, so platinum and rhodium in an autocatalyst liquor, or copper, gold and silver in an electronic-scrap liquor, are recovered in the same pass — no single fraction has to carry the economics alone.
The module delivers a concentrate, not a finished metal. Conversion to individual metals requires a compact second-stage refining unit developed on the same platform. World palladium demand is measured in low hundreds of tonnes, so this refining stage is a modest industrial facility rather than a national programme, and can be sited in any jurisdiction that wants to hold its own separation capacity.
Key Parameters
Working pressure: deep vacuum. Working temperature: ambient; no thermal input beyond the vacuum duty. Specific energy: under 1 kWh/m³ of stream. Water return: up to 100% of intake by volume; zero discharge; no brine. Solids output: dry separated fractions. Consumables: none — no membranes, no filters, no reagents, no ion-exchange media. Form factor: containerised module at a nominal 200 m³/day, deployable singly or in cascade at the recycling yard or inside the fabrication plant.
Market context for the metal:
| Parameter | Value |
|—|—|
| Price, 3 February 2026 | $1,802/oz, approximately $58,000/kg |
| Year-on-year change | +79.75% |
| 2026 high (January) | $2,200/oz, a three-year high |
| Heraeus 2026 forecast corridor | $950–1,500/oz — exceeded within the first month |
| Primary supply, Russia | approximately 40% |
| Primary supply, remainder | mostly South Africa |
| Demand, automotive catalysis | approximately 85% |
| Autocatalyst PGM loading | on the order of 1–2 kg per tonne of monolith, palladium dominant in petrol applications |
| Autocatalyst value at $46,000/tonne scrap | 1 kg palladium per tonne of scrap, valued at 20% below prevailing 2026 prices |
Architecture and Components
Containerised module at a nominal 200 m³/day (73,000 m³/year), installed singly or in cascade at the recycling yard, inside the electronic-scrap processing line, or at the PCB fabrication plant on the plating/activation rinse circuit. Deep-vacuum separation stage, condensate return, density-separated dry fractions, palladium-bearing concentrate take-off. The concentrate is routed to a compact second-stage refining unit, developed on the same platform, that yields palladium alongside platinum and rhodium.
Advantages
Cost independent of concentration. The mechanism that makes classical recovery fail on dilute streams does not operate here; a lean stream and a richer stream cost the same per cubic metre to process.
Non-selective co-recovery. Platinum and rhodium in autocatalyst liquor, or copper, gold and silver in electronic-scrap liquor, are captured in the same pass — no single fraction carries the economics alone.
No consumables. No membranes, filters, reagents or ion-exchange media; nothing to replenish against the volume processed.
Zero discharge. Water is returned at up to 100% of intake by volume, removing the disposal/permit-compliance obligation for the rinse stream.
Dissolution is already paid for. In all three streams the operator has already dissolved the palladium at their own cost; ARBOK adds only the separation step.
Integrations
Installs on autocatalyst-recycler leach liquor, electronic-scrap leachate, or PCB plating/activation rinse circuits, ahead of or in place of conventional dilute-effluent treatment. Concentrate routed to a PGM refining partner or to the ARBOK second-stage refining unit. Related PGM-recovery cases on the same platform: ARBOK-Rhodium, ARBOK-Iridium, ARBOK-Ruthenium. Shared feedstock context with plating and metal-finishing wastewater: ARBOK-GALVANIX. Platform-level brine and tailings recovery: ARBOK Critical-Materials Recovery.
Deployment & Operation
Deployment starts with a site assay of the target liquor or rinse stream to establish palladium (and co-metal) concentration and flow, from which module count is set. The module installs on the existing leachate or rinse circuit without changing the recycler's or fabricator's own dissolution or plating chemistry, and the resulting concentrate is offtaken to a refining partner or to the ARBOK second-stage unit for conversion to individual metals.
The process is additive to the recycler's or fabricator's existing operation: it changes what happens to the metal-bearing liquid afterward, not the chemistry that produced it.
TRL
[требует уточнения из базы] — neither the preprint nor the post states a TRL figure for the palladium case specifically; both describe the ARBOK-ZWD platform and its second-stage refining unit in operating-principle and economic terms without a stage classification.
Market Potential
Streams described in the source:
- End-of-life autocatalyst leachate — the richest palladium-bearing material in circulation, with mature collection under end-of-life vehicle directives in the EU and Japan and commercial scrap economics in the US.
- Electronic scrap leachate — PCBs, connectors, multilayer ceramic capacitors; grade varies with equipment vintage.
- PCB plating and activation baths and rinse waters — the only continuous (not batch) stream of the three, generated every operating day inside a facility with full chemistry knowledge.
Siting logic from the source: the European Union and Japan mandate end-of-life vehicle collection but lack separation capacity at the scale of the arisings; the United States has catalyst manufacturing and a large vehicle fleet but effectively no primary ore; China, Korea and Taiwan hold the largest board and capacitor manufacturing base in the world, and therefore the largest plating and activation bath volume. Any jurisdiction with emissions legislation and no platinum-group resource fits the pattern: the standard is written domestically, the metal is imported, and the metal has already arrived.
Buyers: autocatalyst recyclers, electronic-scrap recyclers, PCB fabricators and connector platers, PGM refiners.
Typical Project Economics
Basis: recovered volume valued at 20% below prevailing 2026 prices.
| Line | Value |
|—|—|
| Scrap autocatalyst at 1 kg palladium per tonne | approximately $46,000 in palladium alone, per tonne of scrap |
| Module at 73,000 m³/year on plating rinse water, 1 mg/L palladium | approximately 73 kg/year, approximately $3.4 million |
| Same module at 5 mg/L palladium | approximately 365 kg/year, approximately $17 million |
| Platinum and rhodium from the same solution | additional; in autocatalyst service rhodium can exceed the palladium line on value despite far lower mass |
| Removed discharge obligation and returned water | a separate line, material in board fabrication where water is often permit-constrained |
Two qualifications carried directly from the source: palladium concentrations in rinse waters are facility-specific and not published, so the figures above are illustrative and site assay is a precondition for any commercial commitment; and autocatalyst loadings vary by vehicle vintage, engine type and market, so 1 kg per tonne is a working figure rather than a specification.
Risk Factors
Structural surplus vs. security. Automotive palladium demand is in structural decline from electrification and the market is described as oversupplied; this is real but does not address who holds a given surplus — the 2026 price record (a doubling inside a declining-demand market) is presented in the source as the empirical form of that distinction.
Grade dependence in the rinse-water case. Concentrations are low in absolute terms and are facility-specific and unpublished; because ARBOK process cost is independent of concentration, low grade determines required throughput rather than viability, but site assay is still a precondition for commercial commitment.
Separation, not collection, is the bottleneck. Recycling of autocatalysts and electronic scrap already exists at scale; what does not exist outside a small number of integrated refiners is the separation step that determines who captures the value.
Related Technologies
ARBOK-Rhodium · ARBOK-Iridium · ARBOK-Ruthenium · ARBOK-GALVANIX · ARBOK Critical-Materials Recovery