Optimizing Greenhouse Gas Emission Quantification and Assessing Mitigation Effectiveness in Semiconductor Manufacturing

Fluorinated greenhouse gases (F-GHGs) from the semiconductor industry represent a rapidly growing climate driver, yet localized accounting and abatement analyses remain scarce. To improve estimation accuracy and assess mitigation potential, we developed an improved accounting model by incorporating previously neglected sources, experimentally measuring destruction and removal efficiencies (DREs) for major process gases, and establishing gas usage coefficients by wafer size and process type. Applied to Shanghai's semiconductor sector in 2022, the model estimated 3.35 Mt CO2e emissions, dominated by process gases (47.93%) and electricity consumption (43.88%). Emissions based on measured DREs were 7.4-23.0 times higher than those from the Intergovernmental Panel on Climate Change (IPCC) 2019 default DREs, revealing overestimation in defaults. The share of indirect emissions increased markedly with wafer size, from 20.68% to 55.46%, driven by higher power intensity in advanced nodes and greater abatement efficiency. Plasma-based abatement achieved the highest DREs, yet their NOx byproducts warrant further attention. Without additional controls, emissions could increase to 4.20 Mt CO2e by 2035, whereas defined mitigation scenarios could achieve 37.49-77.69% reductions, primarily through enhanced gas abatement and low-carbon electricity adoption. The study emphasizes localized methodology and scenario analysis as essential for developing effective climate strategies in emerging industries.