Serbia has a credible opportunity to establish an electronic-waste processing operation capable of recovering more value from discarded computers, telecommunications equipment and industrial electronics. The market is substantially large, the country has a deeper industrial base and existing recyclers already collect and dismantle significant volumes of electrical equipment. Yet those advantages also make the investment more complicated. A new operator would enter a functioning market rather than an empty one, and another conventional facility processing refrigerators, washing machines and televisions would struggle to distinguish itself.
The more attractive opening lies further along the recycling chain: secure dismantling of corporate electronics, grading of printed circuit boards, mechanical concentration of copper and precious-metal-bearing material, laboratory-controlled sampling and preparation of traceable batches for specialist refiners. Serbia has collection and primary-treatment capacity, but much of the highest-value material still leaves the country before the more profitable stages of metallurgical recovery, brings Clarion.Engineer
That gap is where a Serbian urban-mining operation could develop a defensible position. It would not initially need to produce finished gold bars. Its first commercial objective would be to turn mixed and inconsistently valued electronic scrap into documented, homogeneous concentrates that command higher prices from European copper and precious-metals refiners.
Serbia already has several established electrical and electronic waste processors. The most visible include E-Reciklaža 2010 in Niš, SET Reciklaža in Belgrade, Božić i Sinovi in Pančevo and Eko-Metal in Vrdnik. Together with smaller collectors, transport companies and storage operators, they form a network covering most of the country. Available sector assessments put the combined nominal capacity of the principal plants at close to 75,000 tonnes annually, although actual collection and treatment volumes are considerably lower.
Formal WEEE collection is generally estimated at approximately 15,000–20,000 tonnes a year. Other industry assessments suggest that as much as 30,000 tonnes of obsolete electrical and electronic equipment may still enter municipal waste, remain stored in households and company warehouses or pass through informal channels. This discrepancy is commercially tempting, but it cannot be treated as immediately available feedstock. Waste sitting in basements or reaching informal collectors does not become bankable supply merely because a processing line has been installed.
The central investment challenge is therefore not equipment capacity. It is control over material flows, explains Clarion.Engineer
A new facility would need binding relationships with telecommunications operators, banks, insurance companies, hospitals, public institutions, data centres, retailers, manufacturers and industrial companies replacing control systems and automation equipment. These sources are more valuable than mixed household collections because they produce a higher proportion of computers, servers, routers, mobile devices, switching equipment and printed circuit boards. They also require services that extend beyond recycling, including asset registration, secure data destruction, serial-number tracking and documented confirmation of final treatment.
This service layer could become as important as the recovered metals. Serbian companies operating under international ESG policies increasingly need evidence showing what happened to obsolete IT equipment, where data-bearing devices were destroyed, which waste codes were used and how the resulting fractions were treated. Banks, telecommunications groups and foreign-owned manufacturers cannot rely indefinitely on a handwritten collection certificate. They need an auditable chain of custody capable of supporting internal sustainability reporting, information-security controls and group-level environmental assurance.
Printed circuit boards are the economic centre of the proposed operation, but they are not a uniform commodity. Depending on the equipment category, circuit boards may represent less than 1% of the weight of a refrigerator, around 10–15% of a laptop and more than 35% of some mobile devices. Their metal content varies even more sharply.
Low-grade boards from household appliances may contain only modest amounts of copper and very small concentrations of gold. Server, telecommunications and mobile-phone boards can contain materially higher concentrations of copper, gold, silver and palladium. Copper may account for approximately 7–33% of board weight, while gold concentrations can range from less than 20 grams per tonne in low-grade material to more than 1 kilogram per tonne in selected high-grade boards.
The difference determines whether a batch is worth hundreds, thousands or tens of thousands of euros per tonne. Visual sorting alone cannot reliably establish that value.
A Serbian operation would therefore need to create clearly defined commercial grades. Power boards, television boards, computer motherboards, server boards, telecom boards, CPUs, memory modules, connectors and other components should be separated, weighed and sampled independently. Portable X-ray fluorescence equipment could assist with rapid screening, but final settlements would need to rely on representative batch samples and accredited laboratory analysis.
The first development phase should combine reception, depollution, manual dismantling, battery and capacitor removal, cable separation, data destruction and board grading. The facility would produce conventional ferrous and non-ferrous fractions while retaining higher-value electronic components for further processing.
A second phase would introduce shredding, controlled size reduction, magnetic separation, eddy-current separation, air classification and electrostatic separation. This would produce copper-rich and mixed precious-metal-bearing concentrates while removing a portion of the plastics and fibreglass. Dust extraction, explosion prevention and fire protection would be critical because fragmented boards contain fine combustible material, brominated resins and potentially hazardous metal-bearing dust.
A combined first- and second-stage facility capable of treating approximately 3,000–5,000 tonnes of mixed WEEE annually, including 600–1,200 tonnes of separately sourced printed circuit boards, would require estimated investment of around €2.5–4.5 million. This range assumes the use of an existing industrial building, installation of a medium-capacity mechanical line, enclosed storage, environmental controls, laboratory and sampling equipment, digital inventory systems and commissioning support, explains Clarion.Engineer
A greenfield development requiring land acquisition, major utility connections and purpose-built structures would move above that range. A suitable project would need approximately 5,000–8,000 square metres of land and 2,500–4,000 square metres of enclosed processing, storage and administration space.
The most practical location would be within the Belgrade–Novi Sad industrial corridor, particularly around Inđija, Stara Pazova or Pećinci. These locations provide access to Serbia’s largest concentration of corporate waste generators, proximity to both Belgrade and Novi Sad, motorway connections towards Hungary and Croatia and efficient routes to specialist refineries in Central and Western Europe.
The location would also avoid the need to place an early-stage mechanical operation inside an established heavy-industrial or chemically contaminated zone. Pančevo offers stronger industrial infrastructure but already hosts recycling activity and carries cumulative environmental sensitivities. Niš has good access to southern Serbia and North Macedonia but is also the home market of the country’s largest established WEEE operator. Šabac could become relevant for a later chemical-processing phase because of its industrial infrastructure, though it is less attractive for the initial collection network.
At stabilised utilisation, a combined dismantling and PCB-concentration facility could generate annual EBITDA of approximately €850,000–€2 million. The wide range reflects the importance of feedstock composition. A plant receiving mostly low-grade appliances would generate very different margins from one supplied with servers, telecom equipment and carefully separated boards.
The base case should assume a project-level return of approximately 16–25%, with a simple payback period of four to six years. A financing structure combining 60% equity and 40% debt could lift equity returns towards 18–28%, provided the company secures multi-year supply agreements and downstream refinery contracts before financial close.
Lenders would probably require a minimum debt-service coverage ratio of 1.35–1.50 times without relying on Serbian recycling subsidies. Inventory controls would also receive close attention. A relatively small volume of high-grade telecom boards can carry substantial value, making theft, substitution, sampling error and settlement disputes material credit risks, brings Clarion.Engineer
Working capital could become more demanding than the physical plant. Suppliers may expect payment shortly after delivery, while European refiners can require sampling, assay, treatment and settlement periods extending over several weeks or months. At an annual PCB throughput of 1,000 tonnes, the business could require several million euros of revolving working-capital capacity, depending on board grades and supplier-payment terms.
A transparent assay-and-settlement system would therefore be central to the company’s competitive position. Existing Serbian recyclers would have little reason to divert valuable boards to a new intermediary unless it offered a stronger netback, faster payment, reliable analysis and lower logistics costs. The project should be structured as a processing and commercial partner to licensed operators, rather than as an attempt to displace them from conventional collection.
Serbia’s revised support system for recyclers adds potential upside. Under the 2026–2028 auction-based incentive framework, support for electrical and electronic waste can reach approximately RSD 24.8 per kilogram, equivalent to about €212 per tonne. At 5,000 tonnes of eligible annual treatment, the theoretical maximum would exceed €1 million.
That amount could transform project cash flow, but it should not underpin the debt case. Access depends on auctions, quotas, documentation, qualifying recovery operations, state-aid conditions and annual budget implementation. The financial model should treat awarded support as upside capable of accelerating debt repayment or financing collection infrastructure, not as guaranteed operating revenue.
Permitting will influence the development schedule almost as much as equipment procurement. The facility would require waste-management authorisations covering the specific waste codes and treatment operations, including hazardous fractions where applicable. Environmental-impact screening, construction approvals, fire-protection consent, occupational-safety systems, water conditions and air-emission controls would need to be aligned with the final technology.
The operator would also need a digital mass-balance system connecting every incoming batch with weighing records, dismantling results, separated materials, residues, exports and final recovery certificates. This documentation is essential both for regulators and for corporate clients purchasing secure recycling services.
A 12-month delay in permitting and commissioning could reduce the project IRR by approximately 3–4 percentage points. An 18-month delay, combined with additional interest during construction and deferred revenue, could remove 5–7 percentage points from returns. The development programme should consequently place environmental permitting, waste-code confirmation and site suitability ahead of equipment ordering.
A full hydrometallurgical plant recovering copper, gold, silver and palladium represents a possible third phase, but its economics require a substantially larger and more stable PCB supply. A credible industrial threshold would be approximately 1,500–3,000 tonnes of circuit boards annually, supported by domestic contracts and regional imports.
Such a plant could require additional CAPEX of €8–15 million, including leaching reactors, reagent storage, solvent extraction or selective precipitation, electrowinning, process-water recycling, wastewater treatment, ventilation, emissions control, hazardous-residue management and laboratory infrastructure. The technology would need to manage copper as the dominant metal while concentrating smaller volumes of gold, silver and palladium.
The metallurgical flow sheet cannot be selected before the feedstock has been characterised. Boards containing high concentrations of copper, tin, lead, nickel, antimony and brominated resin behave differently from clean copper laminates or selected telecom components. A process designed around assumed gold content could underperform once exposed to the actual mixed Serbian feed.
Serbia Zijin Copper in Bor provides a potentially important industrial connection. The complex already produces copper cathode, gold, silver and other precious-metal by-products. A future cooperation could involve testing whether mechanically concentrated PCB material, copper-rich fractions or intermediate residues could be integrated into an approved metallurgical route. That option requires dedicated technical and environmental due diligence; the existence of a copper smelter does not automatically mean it can accept electronic scrap.
Until a domestic solution is technically proven, the most credible downstream counterparties remain European groups such as Umicore, Aurubis and Boliden, which already process electronic scrap through integrated copper and precious-metals systems. Long-term contracts with one or more of these companies would give the Serbian facility clearer settlement terms, verified final recovery and an internationally credible audit trail.
Regional expansion could later draw selected material from Montenegro, North Macedonia, Bosnia and Herzegovina and Albania. Serbia’s central position and road connections support this model, while the Bar port could provide a complementary maritime route for material collected in Montenegro. Cross-border waste movements, however, require prior classification, notification and consent where hazardous characteristics apply. Regional supply should therefore be treated as an expansion case rather than the volume supporting the initial investment, brings Clarion.Engineer
The strongest commercial proposition is a Serbian urban-mining and secure electronics recovery hub, not an immediate attempt to build a miniature gold refinery. An initial €2.5–4.5 million investment could capture value currently lost between basic dismantling and final European refining, while creating a platform for corporate data-destruction services, ESG documentation and regional material aggregation.
The chemical recovery phase becomes bankable after the operation has accumulated several years of verified assays, established stable board grades and secured at least 1,500 tonnes of annual PCB feedstock. Serbia already has the recyclers, industrial clients and logistics needed to support the first stage. The missing component is a specialised operator capable of treating electronic scrap as a controlled metals resource rather than another undifferentiated waste stream.
Elevated by Clarion.Engineer