Reactive crystallisation studies of CaCO3 processing via a CO2 capture process : real-time crystallisation monitoring, fault detection, and hydrodynamic modelling

Abstract

Currently, a broad spectrum of decarbonisation efforts are ongoing in fundamental areas of human lifestyle to mitigate the global concerns materialising from climate-related transformations. Within the framework of carbon capture, utilisation, and storage (CCUS), the present study investigated the integration of two separation processes, namely, carbon dioxide (CO2) capture and the reactive crystallisation of calcium carbonate (CaCO3). In this work, aqueous sodium hydroxide solution was used for CO2 absorption. The scalable and economically feasible production route of micron-sized CaCO3 occurs by the direct addition of the CO2-loaded liquid solution to a crystalliser containing aqueous calcium chloride. The coupled processes eliminate the demand for high-temperature regeneration and promote high product purity. From the CO2 capture perspective, small-scale demonstration experiments were scaled-up by utilising a hollow fibre membrane contactor–based CO2 capture system developed in-house. The overall mass transfer enhancement due to reaction was investigated.

The fast kinetic reactive crystallisation process was studied at two different scales and a range of operating conditions by utilising multiple process analytical tools. The employed image analysis–based in-line digital microscope camera provided real-time crystal size data and images from the crystal suspension. The in-line probe was utilised to develop an agitation–based feedback control for the CaCO3 precipitation process. The real-time control scheme influences the crystal formation process by manipulating the energy dissipation and spatial supersaturation distribution of the stirred tank reactor. The investigation provided preliminary insights into the challenges of real-time micron-sized reactive crystallisation measurement and control.

A first of its kind framework was proposed to develop a real-time process fault detection and diagnostics utilising 1D electrical resistance tomography (ERT) measurements. The most sensitive measurement point was experimentally justified as a data transmitter position for process diagnostics. In parallel to the ERT, the application of a novel ultrasound tomography (UST) system was investigated in the reactive crystallisation process. The bulk crystal concentration and the reagent feeding region were successfully visualised using ultrasound tomographic reconstructions.

Computational fluid dynamics (CFD) was used to perform 3D simulations of the mixing hydrodynamic and to visualise the spatio-temporal distribution of the chemical reaction in the stirred tank reactor. The experimentally validated CFD simulations deliver a framework that can be employed as a virtual tomography tool in parallel with ERT and UST measurements. The main idea was to directly utilise the crystal size distribution data of the small-scale experiments for model development. The empirical modelling approach inherently contains the effects of the operating conditions and the kinetics of the precipitation and uses less computational resources by reducing the design parameters.