The global imperative to decarbonize economies and transition towards low-carbon energy systems is fueling an unprecedented and accelerating demand for essential metals. Projections, such as those reviewed by Watari et al. (2021), forecast significant demand growth by 2050 relative to 2010 levels for key metals like aluminium (215%), copper (140%), nickel (140%), and iron (86%). In this context, the Circular Economy (CE) – broadly understood as a transformative approach to slow, narrow, and close socioeconomic material cycles, retaining value and minimizing primary resource use, waste, and emissions has emerged as a critical paradigm. Rooted in concepts like industrial ecology and industrial symbiosis that gained traction in the early 2000s, and adopted by policymakers in China (2008) and Europe (2015) later that decade, the CE offers a potential pathway to reconcile growing material needs with planetary boundaries.
Within CE scholarship and policy, metals have been a prominent focus, largely because, unlike many materials, they retain their intrinsic properties through recycling and can be reused multiple times without quality degradation (Hagelüken et al., 2016). Consequently, End-of-Life (EoL) recycling – recovering metals from products discarded at the end of their useful service – has become the predominant strategy emphasized in efforts to meet metal demand more sustainably. However, rigorous analysis using dynamic Material Flow Analysis (MFA) combined with probabilistic forecasting, specifically examining the global copper cycle, reveals profound limitations to relying solely on EoL recycling, demanding a more comprehensive and realistic approach to achieving circularity for metals.
Copper's history illustrates the scale of material accumulation. Global primary copper production has exploded from approximately 0.85 million metric tonnes (Mt) in 1910 to over 20 Mt in 2020. Due to its durability, a vast amount of this cumulatively produced copper remains within the technosphere. The global in-use stock of copper grew from around 60 Mt in 1960 to exceed 460 Mt by 2020 – a figure validated to within 4% of comparable studies. Worryingly, however, over 100 Mt of copper that reached its EoL since 1910 has been lost, landfilled, or otherwise dissipated. While an estimated 29% of the current in-use stock has been recycled at least once, the sheer size of this stock, projected to grow further to potentially 650 Mt by 2040 (a 41% increase from 2020 under the IEA's STEPS scenario), represents the theoretical potential for future secondary supply.
Outflows from this stock – copper contained in products reaching EoL – are also increasing, projected to rise from about 16 Mt per annum in 2020 to between 30 Mt (STEPS scenario) and 33 Mt (NZE scenario) by 2050. However, a critical factor limits the immediate availability of this material: copper's long average in-use residence time, estimated at 23 years. This means copper reaching EoL today was, on average, fabricated around the turn of the millennium, when primary production was significantly lower (~13 Mt). Similarly, copper reaching EoL in 2040 was likely produced around 2017. This inherent time lag disconnects the timing of scrap availability from the timing of contemporary demand surges. Consequently, future production growth driven by the energy transition has a relatively small impact on scrap outflows until the mid-to-late 2040s.
This time lag directly translates into limitations on the growth of secondary copper supply relative to total demand. While the absolute annual production of secondary copper is projected to more than double from around 6 Mt in 2020 to over 15 Mt by 2040 (with average annual growth rates around 3%), its ability to satisfy the total demand remains constrained. The peak growth rate of secondary supply occurred in the early 2020s, reflecting the unprecedented primary production boom of the late 1990s (over 5% p.a.).
Crucially, the share of total copper demand met by EoL secondary supply is projected to increase only modestly from its current 23%. Across the next three decades (2021-2050), it is expected to average only 33.4% under the lowest demand scenario (IEA's STEPS) and potentially less, around 31%, under the high-demand Net Zero Emissions (NZE) scenario, primarily because higher overall demand dilutes the share contribution from secondary sources. Even under the most optimistic NZE scenario assumptions, including very high collection and recycling rate growth (95th percentile outcome), the annual share of EoL secondary supply only reaches 49.6% by 2050. This starkly contrasts with the hope that recycling alone could close the loop in the near term. Existing studies forecasting secondary supply often failed to explicitly investigate this potential shortfall or adequately account for uncertainties in future EoL collection and recycling rates, thus providing limited guidance.
The unavoidable conclusion from this analysis is that despite rapid growth in recycled copper volumes, the sheer scale of projected total demand increases means that the need for primary copper production will not slow down soon. Under all three IEA scenarios (STEPS, APS, NZE), primary production is projected to increase at least into the late 2020s. Under the medium (APS) and high (NZE) demand pathways, primary production continues to grow until at least 2040.
The likelihood of reducing primary production below current levels by mid-century solely through increased EoL recycling is extremely low. Across all modelled outcomes, 87% show primary copper production in 2050 still being above 2020 levels. Only under the STEPS scenario is there a modest possibility (28% of outcomes) of dipping below 2020 levels by 2050, and even less likely under APS (2.2%) and NZE (9%). Overall, the analysis indicates less than a 7.5% probability that EoL recycling efforts alone will be sufficient to reduce primary copper production below 2020 levels by 2050. This implies that dozens of new copper mines will need to be opened, a challenge compounded by the depletion of existing mines and declining average ore grade.
The findings for copper, a metal with comparatively high recycling rates and a vast in-use stock, cast serious doubt on the sufficiency of EoL recycling as the primary CE strategy for other critical energy transition minerals. Materials like lithium, with much smaller existing in-use stocks and less developed recycling pathways, face even greater constraints in generating significant secondary supply in the short term, despite potentially shorter product lifespans.
This reality necessitates a fundamental broadening of the CE approach for metals beyond the predominant EoL focus. While maximizing collection and recycling efficiency remains crucial, it must be recognized as only one piece of the puzzle. Evidence suggests demand-side strategies, such as extending product lifetimes (keeping copper in use longer, though this further delays its return as scrap), implementing material efficiency measures, and fostering changed consumption practices (like sharing electric vehicles), may offer equivalent or even greater potential for reducing primary resource demand.
Furthermore, the unavoidable continuation and likely increase in primary metal extraction demands that CE principles be applied rigorously within the mining and processing sectors themselves. This involves moving beyond the siloed approach where CE scholarship often overlooks the upstream realities addressed by resource governance and responsible sourcing debates. Applying CE thinking could involve optimizing extraction processes, reusing mining waste products, promoting regional sharing of mining infrastructure, and developing innovative business models within the primary sector to minimize its environmental and social footprint. Future research must actively explore these avenues, assessing existing CE practices in mining, evaluating policy readiness, and adapting successful strategies from other sectors.
The detailed analysis of the global copper cycle provides compelling evidence that End-of-Life recycling, while indispensable, faces inherent limitations due to material lifespans and time lags that prevent it from single-handedly meeting the exponential growth in metal demand driven by decarbonization or significantly displacing primary production by 2050. Relying primarily on this single strategy risks falling short of sustainability goals. Achieving a truly circular and sustainable economy for metals requires a paradigm shift towards a more comprehensive, systemic approach. This must integrate aggressive EoL recycling efforts with equally ambitious demand-reduction strategies and the innovative application of circular principles throughout the entire value chain, crucially including the primary extraction and production phases. Only through such a multi-pronged effort can we realistically hope to navigate the resource challenges of the energy transition.