Geochemical modeling of CO2 sequestration in ultramafic mine wastes from Australia, Canada, and South Africa: Implications for carbon accounting and monitoring

Abstract

Passive carbonation in ultramafic mine wastes results from the spontaneous reaction of gangue minerals with carbon dioxide (CO2) to form stable carbonate minerals that store CO2. Mines can benefit from unintentional CO2 sequestration to offset greenhouse gas (GHG) emissions. Quantitative X-ray diffraction (QXRD) analysis of numerous samples of mine wastes has typically been used to quantify passive carbonation rates. This analysis yields valuable information about carbon sinks; however, extensive mineralogical assessments are technically demanding and cost-prohibitive for routine carbon accounting. Alternatively, we explored using inverse geochemical modeling to estimate passive carbonation rates and monitor CO2 sequestration in active mines. Water chemistry data, tailings mineralogy, and operational information routinely collected by mines were used as inputs for models. The predictive capabilities of the models were tested for Mount Keith nickel mine (Australia), Diavik diamond mine (Canada) – for which rates were previously determined using QXRD. A new site — Venetia diamond mine (South Africa) — was used to illustrate the potential of geochemical modeling for carbon accounting and as a long-term monitoring tool for CO2 sequestration. Mount Keith models predicted passive carbonation rates (3900 g CO2/m2/yr) consistent with previous QXRD assessments (2400 g CO2/m2/yr; 172 samples) using only two water samples. CO2 removal rates for Diavik were found to be impacted by seasonality and to range between ∼375 and 510 g CO2/m2/yr, showing similarities with previous QXRD rates estimates (313–350 g CO2/m2/yr). With the long-term water chemistry records (2009–2018) available for Venetia mine, we predicted calcite as the main CO2 sink and that ∼14,875 t CO2 were stored in the kimberlite residues impoundments (3.5 km2) over 9 years at a rate of ∼470 g CO2/m2/yr. Moreover, long-term monitoring shows a decline in CO2 removal rates at Venetia, possibly due to changes in kimberlite reactivity and current mine waste disposal practices. Overall, inverse modeling proved that it can be a viable alternative or complement to QXRD techniques for carbon accounting in active mines, substantially reducing the need for mine waste sampling and permitting long-term monitoring. Moreover, the model shows how passive carbonation rates change dynamically with the mine production and waste management practices. Simultaneously, the model provides insights into potential mineral reactions operating in tailings and that regulate carbon sequestration mechanisms. For that reason, this accounting technique has an enormous potential to assist in decision-making for implementing enhanced CO2 sequestration strategies (e.g., enhanced rock weathering) and to support carbon sequestration monitoring, validation, and reporting in active mines.