Carbon Sequestration Via Steel Slag Accelerated Carbonation: The Influence of Operating Conditions on Process Evolution and Yield

Slags from steel manufacturing have been proven to be suitable for CO₂ sequestration due to their remarkable reactivity related to the typical alkaline nature. In previous work, the research group demonstrated that maximum values of 280-400 g CO₂/kg slag (depending on the slag type) could be attained in slurry phase (L/S = 5 l/kg) at p = 10 bar and T = 100 °C. Stemming from this, since accelerated carbonation may represent a valid option for CO₂ control only if the process has net negative CO₂emissions, further investigations were carried out in order to identify the most appropriate operating conditions in view of maximizing the sequestration yield at the same time keeping the energy demand at low levels.

The present work was aimed at further exploring the effects of the operating conditions on process evolution and yield in slurry-phase carbonation experiments. The effect exerted by temperature, total pressure and CO₂content of the gas on the carbonation yield at L/S=5 l/kg and a residence time of 4h, was estimated through a number of lab-scale accelerated carbonation tests. The experimental campaign was conducted on a size-separated (63-100 μm) fraction of basic oxygen furnace (BOF) slag, and arranged according to a three-level full factorial design in order to evaluate the main and interaction effects of the operating variables on the carbonation performance. The untreated and carbonated slag was characterized for elemental composition, carbonate content, mineralogy (XRD and FT-IR analyses) as well as grain size distribution in order to relate the observed carbonation yield with the changes induced by the process on the main physical and chemical properties of the material under concern. Additional analyses of the chemical composition of the liquid solution recovered at the end of the process were also carried out in order to derive information about the chemical equilibrium conditions between the solid and the liquid phases. To this regard, geochemical modelling of the speciation of constituents in the liquid solutions at the end of the carbonation experiments was performed to identify the most likely carbonate phases being formed during the process and corroborate the findings of mineralogical observations.

The statistical analysis of factorial experiments showed a relevant combined effect of the investigated factors which acted either synergistically or antagonistically depending on the specific ranges of values adopted. The results showed that, among the investigated variables, temperature was the most relevant factor affecting the carbonation process, promoting dissolution of reactive metals from the slag. A significant interaction between total pressure and CO₂ concentration was also observed, which indicated that different combinations of the two factors that corresponded to the same CO₂ partial pressure in the system resulted in different carbonation yields. Likely, this is related to the indirect effect exerted by the combination of total gas pressure and CO₂saturation on the chemical conditions of the slurry system.

The particle size distribution of the untreated slag was compared with that of three samples displaying different conversion yields, and revealing that, irrespective of the CO₂uptake achieved, particle size always decreased upon carbonation. This was assumed as an indication of the fact that the dissolution process taking place to appreciable degrees in all the observed cases, and the limiting stage of the carbonation process was the precipitation phase, which was variously affected by the operating conditions adopted.

Geochemical modelling showed that the most probable candidates governing the thermodynamic equilibrium mostly included different forms of carbonates such as calcium carbonate monohydrate (CaCO₃xH₂O), CO₃-hydrotalcite, vaterite, aragonite, magnesite (MgCO₃), as well as silica (in both the amorphous and crystalline form) and Cr oxides.