城市环境研究所在循环经济助力大宗材料碳减排研究中取得进展----中国科学院上海分院.pdf

nature climate change Analysis https://doi.org/10.1038/s41558-023-01782-6 China’s bulk material loops can be closed but deep decarbonization requires demand reduction Received: 23 January 2023 Accepted: 25 July 2023 Lulu Song 1,2,3, Stijn van Ewijk 4, Eric Masanet Fanran Meng 9, Jonathan M. Cullen 7, Zhi Cao Wei-Qiang Chen 1,2,3 , Takuma Watari & 5,6 , 7,8 10 Published online: xx xx xxxx Check for updates China, as the largest global producer of bulk materials, confronts formidable challenges in mitigating greenhouse gas emissions arising from their production. Yet the emission savings resulting from circular economy strategies, such as improved scrap recovery, more intensive use and lifetime extension, remain underexplored. Here we show that, by 2060, China could source most of its required bulk materials through recycling, partially attributable to a declining population. Province-level results show that, while economic development initially drives up material demand, it also enables closed loops as demand approaches saturation levels. Between now and 2060, improved scrap recovery cumulatively reduces greenhouse gas emissions by 10%, while more intensive use, resulting in reduced material demand, reduces emissions by 21%. Lifetime extension offers a modest benefit, leading to a 3% reduction in emissions. Alongside the large potential for recycling, our findings highlight the importance of demand reduction in meeting global climate targets. Materials are indisputably the backbone of our modern civilization1. Bulk materials, such as cement, steel, aluminium, copper, glass and various chemicals, are consumed in large volumes and provide essential services, which are critical for fulfilling basic human needs: shelter, workplace, mobility and communication. While bulk materials are indispensable for modern society, their production carries a high environmental price. Recent studies calculate that the production of bulk materials accounts for almost 60% of the energy consumption and ~70% of the direct CO2 emissions from the global industrial sector2. Unless measures are urgently taken to change the way materials are produced or consumed, it is expected that soaring needs for housing and infrastructure development will drive up global demand for bulk materials, placing ambitious climate targets at risk3,4. Here, we analyse the technical potential and greenhouse gas (GHG) emission savings of several critical measures designed to shift China toward where societal demand for bulk materials is drastically reduced without compromising the level of human well-being5. The 2015 Paris Agreement has called for international efforts to limit the increase in the global average temperature to well below 2 °C above pre-industrial levels and pursue further efforts to limit the increase to 1.5 °C (ref. 6). The 1.5 °C vision entails a transition toward industrial and energy systems with net-zero emissions by Key Lab of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen, China. 2Xiamen Key Lab of Urban Metabolism, Xiamen, China. 3University of Chinese Academy of Sciences, Beijing, China. 4Department of Civil, Environmental and Geomatic Engineering (CEGE), University College London (UCL), London, UK. 5Bren School of Environmental Science and Management, University of California, Santa Barbara, CA, USA. 6Department of Mechanical Engineering, University of California, Santa Barbara, CA, USA. 7Department of Engineering, University of Cambridge, Cambridge, UK. 8Material Cycles Division, National Institute for Environmental Studies, Tsukuba, Japan. 9Department of Chemical and Biological Engineering, Faculty of Engineering, University of Sheffield, Sheffield, UK. 10College of Environmental Science and Engineering, Nankai University, Tianjin, China. e-mail: zhicao@nankai.edu.cn; wqchen@iue.ac.cn 1 Nature Climate Change Analysis https://doi.org/10.1038/s41558-023-01782-6 Process emission More intensive use t per capita S-curve GHG SSP2 Population 31 provinces Disposal and incineration In-use stocks Primary material production Energy-related emission End-use sectors 5 sectors Lifetime Lifetime Material demand 103 products End-of-life material Materials Secondary material production 13 materials Inter-provincial trade Lifetime extension Improved scrap recovery Fig. 1 | Overview of the IMAGINE Materials model. IMAGINE Materials: integrated modelling of the material–energy–emission nexus associated with bulk materials. SSP2 refers to shared socioeconomic pathway 2. Table 1 | Summary of scenarios and key assumptions Scenario CE strategies Population GHG emission intensities Recycling rates Per-capita material stocks Lifetimes FP No CE strategies are considered Future demographic characteristics broadly follow their historical patterns GHG emission intensities remain unchanged from 2019 to 2060 Recycling rates remain unchanged from 2019 to 2060 Future trends of per-capita material stocks broadly follow their historical patterns Lifetimes remain unchanged from 2019 to 2060 RA No CE strategies are considered Future demographic characteristics broadly follow their historical patterns Moderate improvements take place in bulk materials production, emulating the IEA ETP 2017 Reference Technology scenario31 Recycling rates remain unchanged from 2019 to 2060 Future trends of per-capita material stocks broadly follow their historical patterns Lifetimes remain unchanged from 2019 to 2060 Future trends of per-capita material stocks broadly follow their historical patterns Lifetimes remain unchanged from 2019 to 2060 Material stocks per person in 2060 will be reduced by 0–25% Lifetimes remain unchanged from 2019 to 2060 Material stocks per person in 2060 will be reduced by 0–25% Product lifetime will be gradually prolonged by 30–55% from 2019 to 2060 Improved scrap recovery More intensive use CE Lifetime extension Future demographic characteristics broadly follow their historical patterns Moderate improvements take place in bulk materials production, emulating the IEA ETP 2017 Reference Technology scenario31 EoL material recycling rate will gradually increase and reach the theoretical maximum by 2060 The recycling rate represents the proportion of recycled EoL materials, measured as a percentage of the total EoL materials available. mid-century7–9. In light of the magnitude of annual CO2 emissions arising from bulk materials production (8.4 Gt per year in 20202), scientific and policy communities have sought opportunities to decarbonize the production of bulk materials10. Yet, decoupling emissions from bulk materials production is challenging for three reasons. First, bulk materials production requires high-temperature heat, which is economically challenging to provide without combusting fossil fuels. Second, a substantial fraction of CO2 emissions from bulk materials production results from process chemical reactions. Avoiding these process emissions entails deployment of carbon capture, use and storage or Nature Climate Change switching to alternative processes, with both options currently not ready to be deployed at scale4,11,12. Third, facilities for producing bulk materials are designed to operate over long periods, in many cases for decades, posing infrastructural lock-ins that delay or prevent the transition to low-carbon alternatives13. Production-centric emissions reduction strategies may fall short of addressing emissions from the production of these ‘hard-to-decarbonize’ materials, highlighting the need for broadening the portfolio of decarbonization levers to include measures that reduce societal demand for, and promote recycling of, bulk materials. Analysis 2050–2060 2030–2040 After 2060 2040–2050 No data 0 2050 2060 Heilongjiang –8 Tibet -4 Mt yr –1 20 15 Sichuan –11 10 5 0 2020 2030 2040 2050 2060 Ningxia 10 –1 Mt yr –1 100 2050 2060 Guangxi 2020 2030 2040 Year 2050 2060 2020 2030 2040 Year Difference in inflow between FP and CE Taiwan 2050 2060 Sichuan Outflow (CE) 200 0 2020 2030 2040 Year 2050 2060 200 0 2020 2030 2040 Year 2050 2060 2030 2040 2050 2060 50 0 0 2020 2030 2040 2050 2060 –1 Mt yr –1 2030 2040 2050 2060 Year Anhui 400 2020 2030 2040 Year 0 2050 2060 2020 2030 2040 Year 600 Shanghai 2050 2060 Jiangsu 2020 2030 2040 400 200 0 2050 2060 Year Hainan 2020 2030 2040 Year 600 Hubei 2050 2060 Hunan 400 –1 300 200 2020 2030 2040 0 2050 2060 Year 400 Guizhou 80 Guangxi 300 2030 2040 Year 2050 2060 Hainan 400 200 0 2030 2040 Year 2050 2060 40 0 2030 2040 Year Guangdong 0 2050 2060 2020 2030 2040 Year 300 Jiangxi 2050 2060 Fujian 300 800 20 2020 2020 400 1,000 –1 Mt yr 200 100 2020 1,200 60 –1 Mt yr 100 0 2020 600 200 100 100 200 –1 2020 –1 Mt yr –1 Mt yr 100 Tianjin 100 0 200 200 2050 2060 150 300 100 300 Yunnan Year 100 0 Shandong 200 Zhejiang 200 Year 300 Mt yr –1 Mt yr 200 0 100 0 Year 2050 2060 400 200 Mt yr –1 Mt yr Outflow (FP) 300 2030 2040 500 Chongqing 2030 2040 50 2020 Year 300 400 Inflow (CE) 400 400 Fujian –10 Year 500 Mt yr 2030 2040 0 –1 2020 Guangdong –6 50 50 0 Yunnan –6 150 100 an eji Zh -1 2020 600 Mt yr 20 150 Shanghai –5 150 150 800 Henan 600 g Jiangxi –4 Hunan –5 Guizhou Shanxi 200 200 30 Mt yr –1 Mt yr Shaanxi 250 40 0 250 Year 800 ng -8 su Beijing 100 0 2050 2060 –1 300 Jia Anhui –4 Hubei –14 Year 50 Henan –7 Shaanxi –11 2030 2040 Liaoning 100 0 –1 Tibet 2020 2050 2060 Year 200 50 100 0 Year 2050 2060 –1 25 150 Mt yr Year 200 300 200 2030 2040 –1 Qinghai –9 2050 2060 400 2030 2040 50 2020 –1 2030 2040 Hebei Mt yr 2020 Ch on g –8 qin g 0 Tianjin Hebei –7 –9 Shandong –11 Shanxi –11 Ningxia 100 250 Mt yr 10 –1 –1 Mt yr 20 500 Mt yr Ga n –9 su Liaoning –7 2020 250 200 150 0 –1 Qinghai 30 Beijing –5 Inner Mongolia –6 0 2050 2060 50 Mt yr Year 40 Jilin –6 Xinjiang 2050 2060 Mt yr 2030 2040 Year Inner Mongolia 200 50 2020 2030 2040 Mt yr Mt yr 100 0 2020 250 –1 Year Mt yr 2030 2040 –1 2020 Gansu Mt yr 0 2050 2060 50 600 400 –1 Year 100 50 –1 2030 2040 Mt yr –1 Before 2030 Mt yr 2020 150 Jilin 100 Mt yr 5,000 150 Heilongjiang 150 2,500 50 –1 200 –1 Mt yr –1 Mt yr 150 100 0 When outflow meets inflow in the FP scenario China 7,500 –1 10,000 Xinjiang 200 Mt yr 250 https://doi.org/10.1038/s41558-023-01782-6 200 100 100 200 2020 2030 2040 Year 2050 2060 0 2020 2030 2040 Year 2050 2060 0 2020 2030 2040 Year 2050 2060 0 2020 2030 2040 Year 2050 2060 Fig. 2 | Material demand (inflow) and EoL material availability (outflow) between 2019 and 2060 across China. The number under each province name represents the time difference between the time when outflow meets inflow in the FP scenario and the time when this becomes possible in the CE scenario. Several examples from the literature point to the importance of circular economy strategies (sometimes referred to as material efficiency strategies14–17). In a recent International Energy Agency (IEA) report, circular economy strategies for buildings and vehicles contribute ~30% of the combined CO2 reduction for three bulk materials: steel, cement and aluminium18. Other studies have revealed the potential of circular economy strategies in decarbonizing concrete11,19, steel20, residential buildings14,21, commercial buildings21 and passenger vehicles14. Despite the welcome inclusion of circular economy strategies in climate mitigation roadmaps and policy formulation, our understanding of the efficacy of circular economy strategies is still limited to a few specific sectors or materials22. The extent to which circular economy strategies will contribute to bulk materials decarbonization remains an open question, calling for examining the opportunities carried in circular economy strategies for a panoply of bulk materials. China is an ideal testbed for exploring how a circular materials system might help achieve deep emission cuts for bulk materials. The recent decades have witnessed a rapid growth in China’s appetite for bulk materials, with China now producing ~60% of the global cement23, primary aluminium24 and crude steel25, as well as ~30% of global plastics26. In 2020, China’s bulk materials production accounted for >60% of the energy consumption and ~75% of the direct CO2 emissions from China’s industrial sector27. Previous studies show that regional differences in bulk materials use exist between China’s western and eastern areas28,29. For example, the use of steel by society, over time, results in the buildup of steel stocks, where steel is embedded in products like vehicles and buildings for long periods. In the less-developed western provinces of China, steel stocks have grown to around 3–4 t per capita, comparable to steel stocks in Argentina and Bulgaria. In the more-developed eastern Nature Climate Change provinces of China, steel stocks have reached around 8–9 t per capita, comparable to steel stocks in many developed economies, such as Norway and Ireland. A provincial-level analysis of bulk materials production, use and stocks in China can provide insight into the associated GHG emissions and mitigation options for countries across the globe. Against this background, we develop an integrated modelling framework IMAGINE Materials (short for integrated modelling of the material–energy–emission nexus associated with bulk materials), which is populated by the provincial material stocks and flows database (PMSFD) for China30. The amassed database keeps track of the production, use, stocks and disposal of 13 bulk materials (cement, steel, aluminium, copper, rubber, plastic, glass, lime, asphalt, sand, gravel, brick and wood) and 103 product types, grouped into five end-use segments (building, infrastructure, transport equipment, machinery and household appliances) for domestic consumption during 1978–2018 (Fig. 1). The data collectively account for 80% of all bulk materials produced in China (Methods). With these comprehensive datasets, we investigate patterns of bulk materials production, use, stocks and disposal across China’s provinces. We then explore the viability of creating a closed-loop system for bulk materials and its potential contribution toward achieving net-zero emissions for bulk materials in China from 2019 to 2060. To model the GHG savings by circular economy strategies, we pair our database with life cycle assessment (LCA) results and assess the GHG emissions associated with bulk materials production in three distinct future scenarios. Scenarios and narratives Using the IMAGINE Materials modelling framework, we design three scenarios to compare the GHG emissions arising from bulk materials Analysis https://doi.org/10.1038/s41558-023-01782-6 a Production 5.8 Gt Primary materials b Use 6.5 Gt Material demand 7.8 Gt EoL materials Production 0.6 Gt Primary materials 7.0 Gt Disposal and incineration Use 3.2 Gt Material demand 4.2 Gt EoL materials 1.7 Gt Disposal and incineration 100 1.8 Gt Lifetime extension 2.4 Gravel Sand 1.2 Cement Brick 0 Wood Others Aluminium Copper Steel 2.3 Mt Traded EoL materials 1.6 Gt More intensive use 93.7 Mt Traded EoL materials 50 0 2.4 Gt Recovery 0.8 Gt Recovery c d 300 40,000 20,000 150 AH BJ CQ FJ GS GD GX GZ HI HE HL HA HB HN JS JX JL LN NM NX QH SD SH SX SN SC TJ XJ XZ YN ZJ AH BJ CQ FJ GS GD GX GZ HI HE HL HA HB HN JS JX JL LN NM NX QH SD SH SX SN SC TJ XJ XZ YN ZJ Exporting provinces AH BJ CQ FJ GS GD GX GZ HI HE HL HA HB HN JS JX JL LN NM NX QH SD SH SX SN SC TJ XJ XZ YN ZJ 0 5,000 10,000 500 AH BJ CQ FJ GS GD GX GZ HI HE HL HA HB HN JS JX JL LN NM NX QH SD SH SX SN SC TJ XJ XZ YN ZJ Exporting provinces 250 Importing provinces Importing provinces 25 50 75 100 125 Unit: 1,000 t 0 1,000 2,000 3,000 4,000 5,000 Unit: 1,000 t Fig. 3 | Material demand, EoL material availability, material savings and interprovincial EoL material trade in 2060. a, Material flows in the FP and RA scenarios. b, Material flows and savings in the CE scenario. c, Interprovincial trade of EoL materials in the FP and RA scenarios. d, Interprovincial trade of EoL materials in the CE scenario. Definitions of two-letter codes are provided in Supplementary Table 3. production: (1) a frozen progress (FP) scenario in which all model parameters remain constant from 2019 to 2060; (2) a recent ambitions (RA) scenario, which is consistent with the IEA ETP 2017 Reference Technology scenario31; and (3) a circular economy (CE) scenario in which circular economy strategies are expected to play a crucial role in decarbonizing bulk materials production from 2019 to 2060. The FP scenario reflects a future where no technological improvements in the bulk materials system take place and the historical trends for material stocks continue through 2060. The RA scenario depicts the expected joint efforts (for example, improving energy efficiency and switching from fossil fuels to renewable energy) taken by governments and industry, reflecting the recent ambitions of stakeholders involved in decarbonizing bulk materials production. The CE scenario considers three strategies32: (1) improved scrap recovery, (2) more intensive use and (3) lifetime extension. We choose to model the preceding three CE strategies because evaluating the potential of other CE strategies (for example, remanufacturing, material substitution and lightweighting) requires more fine-grained data and models, which are currently unavailable. As opposed to the FP and RA scenarios, the CE scenario envisions a less material-demanding future, where discarded materials are circulated back into the economy while the total societal throughput of materials is minimized. Whenever possible, deployment levels of CE strategies are derived from roadmaps and scenario analyses in the literature, which estimate achievable deployment levels of each strategy (Table 1). Nature Climate Change Surging EoL materials make closing material cycles possible Our simulations show that the gradual saturation of material stocks leads to peaks and subsequent declines in material demand (Fig. 2). Notably, in the FP scenario where no interventions are taken, the national availability of secondary materials matches, and then Analysis MtCO2e yr–1 2,000 1,500 1,000 500 0 2020 2030 2040 2050 2060 Year c 100 Share of GHG emissions by source (%) b 80 2,500 Cumulative GHG saving (Gt CO2e) a https://doi.org/10.1038/s41558-023-01782-6 70 60 50 40 30 20 10 0 Lifetime Remaining Frozen Recent Improved More progress ambitions scrap intensive extension use recovery GHG emissions Frozen progress Circular economy GHG savings Recent ambitions Improved scrap recovery Remaining GHG emissions Primary material production Secondary material processing 80 60 40 20 0 2020 2030 2040 2050 2060 Year More intensive use Lifetime extension Fig. 4 | GHG savings by three CE strategies and remaining GHG emissions. a, Annual GHG savings by three CE strategies and remaining GHG emissions from 2019 to 2060. b, Cumulative GHG savings by three strategies and remaining GHG emissions from 2019 and 2060. c, Breakdown of remaining GHG emissions by source between 2019 and 2060. Solid lines represent the GHG emissions in the FP scenario. The dashed line represents the GHG emissions in the CE scenario, where three CE strategies are synergistically considered. Areas represent the annual GHG savings by recent ambitions or three CE strategies. Stacked bars represent the cumulative GHG savings by recent ambitions or three CE strategies from 2019 to 2060. overtakes, the total demand for materials around 2050. As material stocks in China start saturating, national material demand falls to a low point of ~7.3 Gt per year in 2036, remains steady from 2037 to 2046 and is expected to decline further due to the combined effect of a shrinking population and saturated per-capita material stocks. By 2060, national material demand is expected to be as low as 6.5 Gt per year. At the same time, the supply of secondary materials rises over time, since increasing amounts of materials become available at the product end-of-life (EoL). As a result, the gap between material demand and secondary supply quickly shrinks between 2019 and 2060. Despite the potential for closing the material cycles through secondary supply, it is still thermodynamically challenging to reach high recycling rates for several materials, such as brick, glass, rubber and plastics (Supplementary Table 7). The time when a closed-loop bulk materials system becomes viable varies by region and material. In the FP scenario, the more-developed eastern provinces will attain a closed-loop material system a few decades earlier, whereas for the less-developed western provinces, matching material demand with regional secondary supply is possible only after 2040. This difference can be explained by regional inequality in asset accumulation and infrastructure development. The supply of secondary materials is unevenly distributed across provinces due to different stock patterns. From 2019 to 2040, several higher-income provinces, including Beijing, Tianjin, Jiangsu, Shanghai, Zhejiang and Fujian, are projected to double their availability of EoL materials, creating more opportunities for the recycling and remanufacturing of secondary materials. These provinces will be in the position to fully close their steel, copper and aluminium cycles from 2040 to 2060 (Supplementary Fig. 14), providing that adequate collection and new alloy separation technologies and infrastructure are in place. By contrast, the supply of secondary materials in less-developed provinces, most of which are located in Northwest China or Southwest China, is insufficient to match material demand before 2060. The provinces in inland China will face severe shortages of secondary materials required for closing the material cycles. This is caused by both the rapid rise in material demand, driven by population growth in lower-income provinces and the reduced availability of secondary materials due to smaller in-use material stocks. As fertility rates— key parameters governing population growth—often fall alongside economic development and urbanization, population declines are expected to arrive later in lower-income provinces. However, the gap between secondary materials and material demand in lower-income provinces may be bridged by transporting the surplus EoL materials from wealthier provinces. In the CE scenario, we envision a less material-demanding future, where three CE strategies bend the curves of material demand and secondary supply (Fig. 2). Our simulations show that the decline in the national demand for materials parallels the decline in the national supply of secondary materials, yet the gap between them closes by 2040, ~9 years sooner than in the FP scenario. The exact timing of when material demand is matched with secondary supply varies by region, with several provinces—including Shandong, Shanxi, Shaanxi, Hubei, Fujian and Sichuan—seeing an even earlier breakeven point. In the FP and RA scenarios, non-metallic materials, including gravel, sand, cement and brick, account for most of the societal throughput of materials but only a small fraction of the demand for these materials is sourced from secondary supply and interprovincial trade (Fig. 3a). In the CE scenario, more intensive use and lifetime extension combined reduce material demand by 3.4 Gt in 2060, bringing down the national demand to 3.2 Gt (Fig. 3b). In the same year, 4.2 Gt of EoL materials is available, of which >80% (2.4 Gt) is reprocessed and circulated back into the economy; the remainder of EoL materials (1.7 Gt) is sent to landfills or incinerators. Given the high recycling rates in the CE scenario, only 0.6 Gt of material demand is sourced from primary production in 2060 and 93.7 Mt from interprovincial trade. In the FP and RA scenarios, interprovincial trade of EoL materials is projected to reach 2.3 Mt by 2060, with Beijing, Heilongjiang, Jilin, Liaoning, Shanghai and Zhejiang emerging as the primary exporters (Fig. 3c). In the CE scenario, while these provinces maintain their important role as exporters of EoL materials, Guangdong, Jiangsu, Shandong and Sichuan assume a dominant position in trading EoL materials (Fig. 3d). Nature Climate Change Analysis https://doi.org/10.1038/s41558-023-01782-6 a b Aggregates 500 12 400 Cement 1,000 14 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 Cumulative 800 6 600 Gt CO2e 200 8 MtCO2e yr–1 300 Gt CO2e MtCO2e yr–1 10 400 4 100 2030 2040 c 2050 500 Gt CO2e MtCO2e yr–1 14 200 12 175 300 200 8 6 100 2030 2040 e 2050 2060 2050 2060 Non-ferrous metals 3.5 3.0 2.5 100 75 2.0 1.5 1.0 50 2 25 0 Cumulative 0 0.5 2020 2030 2040 f Plastic 10 2040 125 4 2020 2030 150 10 400 0 2020 d Steel 600 0 0 Cumulative 2060 Gt CO2e 2020 MtCO2e yr–1 0 200 2 0 Cumulative 2060 Glass 7 0.4 2050 0.25 6 0.1 2 4 Gt CO2e 4 0.2 5 MtCO2e yr–1 6 0 0.20 0.3 Gt CO2e MtCO2e yr–1 8 3 0.10 2 0.05 1 2020 2030 2040 2050 2060 0 Cumulative GHG savings Recent ambitions Improved scrap recovery 0.15 0 2020 2030 2040 2050 2060 0 Cumulative GHG emissions Frozen progress Circular economy Remaining GHG emissions More intensive use Lifetime extension Primary material production Secondary material processing Fig. 5 | GHG savings by three CE strategies and remaining GHG emissions across materials. Solid lines represent the GHG emissions from 2019 to 2060 in the FP scenario. Dashed lines represent the GHG emissions from 2019 to 2060 in the CE scenario. Areas represent the annual GHG savings by recent ambitions or three CE strategies from 2019 to 2060. Stacked bars represent the cumulative GHG savings by recent ambitions or three CE strategies from 2019 to 2060. More elaborate discussions and detailed results pertaining to other materials are provided in Supplementary Section 3.4. CE strategies can deliver substantial GHG savings 25.4 GtCO2e over the period 2019 to 2060 (Fig. 4a,b), equivalent to 34% of the cumulative GHG emissions in the FP scenario. In the FP scenario, GHG emissions peak at ~2.3 Gt per year around 2022 and, thereafter, On top of the progress envisaged in the RA scenario, CE strategies can deliver substantial GHG savings, amounting to a cumulative total of Nature Climate Change Analysis steadily decrease to 1.4 GtCO2e per year by 2060 due to declines in material demand. In the RA scenario, improvements in materials production cumulatively save 9.8 GtCO2e. Among the three CE strategies, improved scrap recovery saves 322.1 MtCO2e per year by 2060 and results in cumulative savings of 7.6 GtCO2e emissions from 2019 to 2060, by replacing primary supply with secondary supply. Improvements in scrap collection and secondary material processing can realize the potential of available EoL materials, yet the contribution of this strategy has limits: ~23% of the annual emissions in the FP scenario in 2060 and ~10% of the cumulative emissions in the FP scenario from 2019 to 2060. This is because recycling rates are already relatively high for materials like copper, steel and aluminium. To achieve net-zero emissions for bulk materials in China, it is apparent that recycling alone will not suffice. While material recycling eliminates the GHG emissions from bulk materials production, these are partially offset by GHG emissions created in secondary material processing (Fig. 4c). For example, collection, sorting and separation of EoL materials consume appreciable amounts of energy, as these waste-handling activities require energy-consuming vehicles and machinery. More intensive use results in an additional 13% saving in GHG emissions in 2060 or a 21% saving in cumulative GHG emissions from 2019 to 2060 compared to the FP scenario. These emission savings result from activities such as designing reasonably sized buildings, designing lightweight cars, space-sharing and ride-sharing. Lifetime extension emerges as an important option for saving GHG after 2050, resulting in a 15% reduction in annual GHG emissions in 2060 or a 3% reduction in cumulative GHG emissions from 2019 to 2060 compared to the FP scenario. Interestingly, unlike improved scrap recovery, more intensive use and lifetime extension reduce GHG emissions by reducing overall demand and slowing down the turnover of material stocks. Recycling does not always save GHG emissions Our results reveal that material recycling has the greatest GHG mitigation potential for metals, while more intensive use and lifetime extension may be more promising strategies for most non-metallic materials, including cement, plastics and glass (Fig. 5). This results from differing ratios of emission intensities for primary versus secondary production across the various materials. The emission intensity of recycling processes for metals is typically much lower than for primary production. By contrast, for non-metal materials, the emission intensities for recycling processes are much closer to primary production levels, sometimes even exceeding the primary production level, for example, in the case of cement. Discussion Our analysis shows that patterns of material stocks set fundamental boundary conditions for future material demand and EoL material availability, which in turn determine GHG emissions. Moving forward, mitigation analyses must consider the timing of ebbs and flows in material demand, in-use material stocks and secondary supply each region is expected to experience, so as to prepare adequate policy, infrastructure and technology responses33. In line with previous studies34,35, our analysis reveals that if material stocks in each province conform to an S-shaped pattern, GHG emissions associated with bulk materials are expected to peak around 2025 and decline thereafter, coinciding with the projections by IEA27 and Boston Consulting Group36. As the exact timing for promoting CE strategies may differ across regions in China, promoting CE strategies should consider timing and regional differences. For example, East China could be a first mover and act as a role model for other regions by shifting primary production to secondary supply, as East China will see a rising supply of EoL materials starting after 2030 (Fig. 2). Nevertheless, the recycling sector in China is still dominated by small- and medium-sized companies which lack technological progress and environmental awareness, Nature Climate Change https://doi.org/10.1038/s41558-023-01782-6 resulting in low-quality, less-competitive recycled products27. Given this precious window of opportunity, local governments in East China must address urgent issues that currently hinder effective material recycling, such as infrastructural lock-ins, material dilution and quality losses. Less-developed regions have an opportunity to learn from the early adopters in more-developed regions, as the former prepares to adopt the latter’s CE practices. While our results show that material recycling brings substantial GHG savings, pursuing this strategy alone will not deliver net-zero emission targets. The climate benefit of material recycling may be limited due to constraints related to the quantity and quality of recovered materials and thermodynamic factors37,38. Material recycling requires additional energy or material input and emissions involved in the collection, sorting, separation and reprocessing of EoL materials to close material loops, which undermines the emission savings resulting from avoiding primary production. The presence of material linkages and scrap contamination poses major challenges to the efficient recovery of materials and limits the potential for emission reduction. These factors hinder the recycling process by introducing complexities and uncertainties that affect the quality and quantity of recovered materials. For some materials, recycling delivers only limited benefits to GHG emission reduction15,17,39. For example, glass recycling can be impractical or expensive when waste glass is broken, contaminated or blended with different colours17,40. For this reason, developing high-quality streams of EoL materials through better sorting, separation or cleaning is insufficient to eliminate GHG emissions for all bulk materials. Our analysis highlights that demand reduction is essential to decarbonizing bulk materials. Compared with material cycling, CE strategies that minimize the societal throughput of materials by reducing material demand have received less attention to date but have great potential for reducing GHG emissions, particularly for materials without a viable recycling loop32. For example, while cement is often the most expensive ingredient found in concrete, restoring the properties of EoL hydrated cement would require energy inputs comparable to manufacturing new cement, making it extremely challenging to recycle17. For materials that are difficult to recycle, reducing the societal throughput of materials through more intensive use and lifetime extension appears to be promising emission reduction strategies. While more intensive use and lifetime extension could reduce the need for material stocks, the transition toward a less material-demanding world will require fundamental societal and behavioural changes, improved design, cultural transition and better planning15,41. China’s policy-makers have high hopes for increasing recycling rates, such as increased use of construction waste and electronic waste42. Moving forward, we urge China’s policy-makers to consider not merely increasing recycling rates but also putting in place far-sighted efforts in more intensive use and lifetime extension. Online content Any methods, additional references, Nature Portfolio reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41558-023-01782-6. References 1. 2. 3. Graedel, T. E., Harper, E. M., Nassar, N. T. & Reck, B. K. On the materials basis of modern society. Proc. Natl Acad. Sci. USA 112, 6295–6300 (2015). Net Zero by 2050: A Roadmap for the Global Energy Sector (IEA, 2021); https://www.iea.org/reports/net-zero-by-2050 Global Resources Outlook 2019: Natural Resources for the Future we Want (UNEP, 2019); https://wedocs.unep.org/handle/ 20.500.11822/27517 Analysis 4. Cao, Z. et al. The sponge effect and carbon emission mitigation potentials of the global cement cycle. Nat. Commun. 11, 3777 (2020). 5. Creutzig, F. et al. Demand-side solutions to climate change mitigation consistent with high levels of well-being. Nat. Clim. Change 12, 36–46 (2022). 6. Adoption of the Paris Agreement by the President: Paris Climate Change Conference (UNFCCC, 2019); http://unfccc.int/resource/ docs/2015/cop21/eng/l09r01.pdf 7. Rogelj, J. et al. Energy system transformations for limiting end-of-century warming to below 1.5 °C. Nat. Clim. Change 5, 519–527 (2015). 8. Grubler, A. et al. A low energy demand scenario for meeting the 1.5 °C target and sustainable development goals without negative emission technologies. Nat. Energy 3, 515–527 (2018). 9. IPCC. Special Report on Renewable Energy Sources and Climate Change Mitigation (eds Edenhofer, O. et al.) (Cambridge Univ. Press, 2011). 10. Habert, G. et al. Environmental impacts and decarbonization strategies in the cement and concrete industries. Nat. Rev. Earth Environ. 1, 559–573 (2020). 11. Watari, T., Cao, Z., Hata, S. & Nansai, K. Efficient use of cement and concrete to reduce reliance on supply-side technologies for net-zero emissions. Nat. Commun. 13, 4158 (2022). 12. Habert, G., Billard, C., Rossi, P., Chen, C. & Roussel, N. Cement production technology improvement compared to factor 4 objectives. Cem. Concr. Res. 40, 820–826 (2010). 13. Tong, D. et al. Committed emissions from existing energy infrastructure jeopardize 1.5 °C climate target. Nature 572, 373–377 (2019). 14. Pauliuk, S. et al. Global scenarios of resource and emission savings from material efficiency in residential buildings and cars. Nat. Commun. 12, 5097 (2021). 15. Allwood, J. M. Unrealistic techno-optimisim is holding back progress on resource efficiency. Nat. Mater. 17, 1050–1053 (2018). 16. Watari, T., Hata, S., Nakajima, K. & Nansai, K. Limited quantity and quality of steel supply in a zero-emission future. Nat. Sustain. 6, 336–343 (2023). 17. Allwood, J. M. in Handbook of Recycling (eds Worrell, E. & Reuter, M. A.) 445–477 (Elsevier, 2014). 18. Material Efficiency in Clean Energy Transitions (IEA, 2019); https://www.iea.org/reports/ material-efficiency-in-clean-energy-transitions 19. Cao, Z., Masanet, E., Tiwari, A. & Akolawala, S. Decarbonizing Concrete: Deep Decarbonization Pathways for the Cement and Concrete Cycle in the United States, India and China (Industrial Sustainability Analysis Laboratory, 2021). 20. Wang, P. et al. Efficiency stagnation in global steel production urges joint supply-and-demand-side mitigation efforts. Nat. Commun. 12, 2066 (2021). 21. Zhong, X. et al. Global greenhouse gas emissions from residential and commercial building materials and mitigation strategies to 2060. Nat. Commun. 12, 6126 (2021). 22. Creutzig, F. et al. in Climate Change 2022: Mitigation of Climate Change (eds Shukla, P. R. et al.) 752–943 (Cambridge Univ. Press, 2022). 23. Cement Statistics and Information (USGS, 2021); https://www. usgs.gov/centers/national-minerals-information-center/ cement-statistics-and-information 24. Primary Aluminium Production (International Aluminium Institute, 2022); https://international-aluminium.org/statistics/ primary-aluminium-production Nature Climate Change https://doi.org/10.1038/s41558-023-01782-6 25. World Steel in Figures 2022 (World Steel Association, 2022); https:// worldsteel.org/steel-topics/statistics/world-steel-in-figures-2022 26. Plastics—the Facts 2021. An Analysis of European Plastics Production, Demand and Waste Data (PlasticsEurope, 2022); https://plasticseurope.org/knowledge-hub/plastics-the-facts-2021 27. An Energy Sector Roadmap to Carbon Neutrality in China (IEA, 2021); https://www.iea.org/events/ an-energy-sector-roadmap-to-carbon-neutrality-in-china 28. Pauliuk, S., Wang, T. & Müller, D. B. Steel all over the world: estimating in-use stocks of iron for 200 countries. Resour. Conserv. Recycl. 71, 22–30 (2013). 29. Song, L. et al. Mapping provincial steel stocks and flows in China: 1978–2050. J. Clean. Prod. 262, 121393 (2020). 30. Song, L. et al. China material stocks and flows account for 1978–2018. Sci. Data 8, 303 (2021). 31. Energy Technology Perspectives 2017: Catalysing Energy Technology Transformations (IEA, 2017); https://www.iea.org/ reports/energy-technology-perspectives-2017 32. Morseletto, P. Targets for a circular economy. Resour. Conserv. Recycl. 153, 104553 (2020). 33. van Ewijk, S. et al. Ten Insights From Industrial Ecology for the Circular Economy (ISIE, 2023). 34. Global Material Resources Outlook to 2060: Economic Drivers and Environmental Consequences (OECD, 2019); https://doi.org/10.178 7/9789264307452-en 35. Bleischwitz, R., Nechifor, V., Winning, M., Huang, B. & Geng, Y. Extrapolation or saturation—revisiting growth patterns, development stages and decoupling. Glob. Environ. Change 48, 86–96 (2018). 36. Building a Greener Future: How China can Reach its Dual Climate Goals (Boston Consulting Group, 2021); https:// web-assets.bcg.com/ff/a6/c514e7314190b5cb27b1383fae1b/ bcg-x-cdrf-how-china-can-reach-its-dual-climate-goals-mar2021-en.pdf 37. Corvellec, H., Stowell, A. F. & Johansson, N. Critiques of the circular economy. J. Ind. Ecol. 26, 421–432 (2021). 38. Reuter, M. A., van Schaik, A., Gutzmer, J., Bartie, N. & Abadías-Llamas, A. Challenges of the circular economy: a material, metallurgical and product design perspective. Annu. Rev. Mater. Res. 49, 253–274 (2019). 39. van Ewijk, S., Stegemann, J. A. & Ekins, P. Limited climate benefits of global recycling of pulp and paper. Nat. Sustain. 4, 180–187 (2021). 40. Westbroek, C. D., Bitting, J., Craglia, M., Azevedo, J. M. & Cullen, J. M. Global material flow analysis of glass: from raw materials to end of life. J. Ind. Ecol. 25, 333–343 (2021). 41. Allwood, J. M. et al. Sustainable Materials: With Both Eyes Open (UIT Cambridge, 2012). 42. The 14th Five-Year Plan for Circular Economy Development (National Development and Reform Commission, 2021); https://www.ndrc.gov.cn/xxgk/zcfb/ghwb/202107/ t20210707_1285527.html?code=&state=123 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. © The Author(s), under exclusive licence to Springer Nature Limited 2023 Analysis Methods Modelling framework We develop an integrated modelling framework IMAGINE Materials, which consistently quantifies GHG emissions associated with each bulk material43. The IMAGINE Materials model is populated by the PMSFD for China, covering 13 materials (including cement, steel, aluminium, copper, rubber, plastic, glass, lime, asphalt, sand, gravel, brick and wood) and 31 provinces30. The PMSFD includes 103 products, which are grouped into five end-use sectors (building, infrastructure, transport equipment, machinery and household appliances). We pair the material layer with the GHG emission layer to simulate material demand, EoL material availability and associated GHG emissions from 2019 to 2060. Complementary to previous studies that adopt a sector- or material-specific perspective4,11,12,14,16,20,21, the IMAGINE Materials modelling framework offers a comprehensive assessment of emission reductions stemming from three distinct circular economy strategies at the province level. While the IMAGINE modelling framework aggregates sector-specific nuances, it can shed important light on the efficacy of circular economy strategies across materials and the optimal timing for promoting CE strategies in different regions. This modelling framework serves as a template that allows analysts to explore the combined effect of CE strategies in decarbonizing bulk materials across diverse contexts, should relevant data become available. Material stocks evolutionary mode identification The historical material stocks are derived from the PMSFD database, which includes material stocks estimated by the bottom-up accounting approach for 13 materials in 31 provinces in mainland China from 1978 to 2018. We use the level, speed and acceleration of material stocks, which were recommended by refs. 44,45, to project the evolution patterns of material stocks. The level represents the per-capita material stocks at year t; the speed represents the change or differential in per-capita material stocks between two consecutive years; the acceleration represents the change in speed between two consecutive years. The autoregressive integrated moving average approach is used to analyse the growth patterns of speed and acceleration. On the basis of the per-capita stocks and the order of difference at which the time series is stationary, we identify four evolutionary modes, each of which represents a progression stage of an S-shaped curve. Each of the 31 provinces is classified as one of four evolutionary modes (Supplementary Section 3.2). Future stocks and flows projection We simulate future material demand and EoL material availability using a stock-driven approach where future material stocks are determined by future population and per-capita material stocks. Population projections for each province in China from 2019 to 2060 are derived from a previous study46. Per-capita material stocks are projected as a simplification to follow an S-shaped curve but with differentiated patterns across provinces. The level of per-capita material stocks is deemed an explicit physical representation of service provision to society. As observed in several previous studies28,45,47,48, the historical patterns of per-capita material stocks show similarities across countries: the growth of per-capita material stocks increases rapidly at first, then slows down and eventually levels off. As such, we assume that per-capita material stocks (all materials combined) will eventually saturate at defined levels (200 t per capita in the FP and RA scenarios and 150 t per capita or the present-day level in the CE scenario). A modified Gompertz function is used to simulate the development of per-capita material stocks49. Considering the observed historical patterns of material stocks, we assume that per-capita material stocks follow an S-shaped curve that moves all provinces toward a national convergence of per-capita material stocks (Supplementary Section 3.3). A certain fraction of EoL materials is recycled to replace virgin materials. A normal lifetime distribution with mean and standard Nature Climate Change https://doi.org/10.1038/s41558-023-01782-6 deviation establishes a relationship between material demand and EoL material availability50. CE strategies and related GHG emission mitigation potential We create three scenarios to reflect plausible futures of China’s bulk materials system: (1) an FP scenario in which no technological improvements in the bulk materials system will take place and future trends of material stocks broadly follow their historical patterns; (2) an RA scenario in which the recent ambitions of stakeholders involved in decarbonizing bulk materials production are considered; and (3) a CE scenario in which we consider three circular economy strategies (improved scrap recovery, more intensive use and lifetime extension). More intensive use aims to reduce the total societal need for material-intensive products, resulting in reduced material demand. A recent study exploring a ‘low energy demand scenario’ found that a decent living standard can be provided with 30 m2 per capita of floor space, which is far below the current per-capita housing floor area in several high-income provinces in China8. Lifetime extension, which aims to extend the service life of products, requires not only technological measures (for example, more adaptable and durable designs) but also policy actions (for example, better zoning policies and better access to quality repair) because the physical durability of products does not always determine their real lifetime. In choosing values for the deployment level of each strategy, we only consider technical feasibility, with no consideration given to investment or deployment costs. Notably, we assume that advanced collection technologies and infrastructure and alloy separation technologies will be deployed to overcome compositional and quality barriers, ensuring that materials sourced from secondary supply (referred to as secondary materials) can replace virgin materials without a loss of quality. Many previous scenario analyses provide target values by 2050. Therefore, when no deployment values are available for 2060, we extrapolate 2050 values to 2060 on the basis of the previous 5-year growth rate. More details about the methods, data and assumptions are provided in Supplementary Section 3.4. Life cycle assessment results are used to calculate GHG emissions of the primary production (cradle-to-gate) and secondary production (including EoL collection and processing) of each material type. We compile a life cycle inventory database by leveraging life cycle inventories available from existing literature and the Gabi database. Details are provided in Supplementary Section 2.6. Limitations and uncertainty While the potential of CE strategies is analysed with comprehensive datasets, there are opportunities to enhance our analysis by integrating sector- and material-specific insights from previous studies. Additionally, the process-based life cycle inventory database used by our analysis may underestimate the emission factors of some recycling or production processes due to the difficulty of including small quantifiable processes in the model. Addressing these limitations and incorporating these factors into the current study would be a crucial step for future research. Another area for improvement lies in considering the decarbonization efforts in secondary material processing to provide an holistic view of decarbonizing bulk materials. Furthermore, it is important to emphasize that our results do not represent future predictions but rather present potential scenarios or pathways for the implementation of CE strategies aimed at reducing GHG emissions associated with bulk materials production. To assess the uncertainties arising from material linkages and scrap contamination, we have conducted additional analyses, which are detailed in Supplementary Figs. 16–18. These analyses contribute to a more comprehensive understanding of the potential uncertainties associated with our findings. Data availability Data used for populating the model are available from https://doi. org/10.6084/m9.figshare.21837195 (ref. 43). Source data are provided with this paper. Analysis Code availability Codes used for simulating material flows and stocks and GHG emissions are available via https://doi.org/10.6084/m9.figshare.21837195 (ref. 43). References 43. Song, L., Cao, Z. & Chen, W.-Q. Dataset and code for bulk materials flows and GHG emission reduction potential in China. figshare https://doi.org/10.6084/m9.figshare.21837195 (2023). 44. Fishman, T., Schandl, H. & Tanikawa, H. Stochastic analysis and forecasts of the patterns of speed, acceleration and levels of material stock accumulation in society. Environ. Sci. Technol. 50, 3729–3737 (2016). 45. Cao, Z., Shen, L., Lovik, A. N., Muller, D. B. & Liu, G. Elaborating the history of our cementing societies: an in-use stock perspective. Environ. Sci. Technol. 51, 11468–11475 (2017). 46. Chen, Y. et al. Provincial and gridded population projection for China under shared socioeconomic pathways from 2010 to 2100. Sci. Data 7, 83 (2020). 47. Wiedenhofer, D. et al. Prospects for a saturation of humanity’s resource use? An analysis of material stocks and flows in nine world regions from 1900 to 2035. Glob. Environ. Change 71, 102410 (2021). 48. Müller, D. B., Wang, T. & Duval, B. Patterns of iron use in societal evolution. Environ. Sci. Technol. 45, 182–188 (2011). 49. Pauliuk, S., Milford, R. L., Müller, D. B. & Allwood, J. M. The steel scrap age. Environ. Sci. Technol. 47, 3448–3454 (2013). 50. Wolfram, P., Tu, Q., Heeren, N., Pauliuk, S. & Hertwich, E. G. Material efficiency and climate change mitigation of passenger vehicles. J. Ind. Ecol. 25, 494–510 (2021). Acknowledgements This work was supported by the Natural Science Foundation of China (grant nos 71961147003, 52170183 and 52070178 to W.Q.C. and L.L.S.), the National Key Research and Development Program of the Ministry of Science and Technology (grant no. 2017YFC0505703 to W.Q.C. Nature Climate Change https://doi.org/10.1038/s41558-023-01782-6 and L.L.S.), the International Partnership Program of the Chinese Academy of Sciences (grant no. 132C35KYSB20200004 to W.Q.C. and L.L.S.), the National Social Science Fund of China (grant no. 21&ZD104 to W.Q.C. and L.L.S.), Special Research Fund (BOF) of the University of Antwerp (grant no. 41-FA100200-FFB200410 to Z.C.) and the Fundamental Research Funds for the Central Universities (grant no. 040-63233060 to Z.C.). E.M., F.M. and J.M.C. acknowledge support from C-THRU: Carbon clarity in the global petrochemical supply chain (www.c-thru.org). Author contributions W.Q.C., L.L.S. and Z.C. conceived and designed the research. W.Q.C. and Z.C. supervised the project. L.L.S. performed the simulations. L.L.S. and Z.C. produced the figures. S.v.E. and E.M. contributed to the scenario design. T.W., F.R.M. and J.M.C. contributed to the result interpretation. L.L.S. and Z.C. prepared the first draft. All authors reviewed and edited the manuscript. Competing interests The authors declare no competing interests. Additional information Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41558-023-01782-6. Correspondence and requests for materials should be addressed to Zhi Cao or Wei-Qiang Chen. Peer review information Nature Climate Change thanks Raimund Bleischwitz, Qingshi Tu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Reprints and permissions information is available at www.nature.com/reprints.