Avoiding catastrophic climate change requires dramatically decreasing greenhouse gas emissions and removing already-emitted CO₂ from the atmosphere paired with permanent CO₂ storage. Through carbon mineralization, CO₂ can be stored as carbonates which are environmentally benign and stable, and thus make mineral carbonation a permanent and leakage free CO₂ disposal method. Calcium and magnesium are the most common alkaline earth metals in nature, indicating that these are the most suitable feedstocks for carbonate formation. In natural minerals, the two metals usually appear in the form of silicates, mostly Antigorite (Mg₃Si₂O₅(OH)₄), Lizardite (Mg₆[(OH)₈|Si₄O₁₀), Forsterite (Mg₂SiO₄), Augite (CaMgSi₂O₆+Fe,Al) and Wollastonite (CaSiO₃). In addition, there are many industrial wastes rich in calcium and magnesium, which can be used as feedstocks for mineral carbonation, waste cement (1 Gt/yr), coal fly ash (600 Mt/yr), steelmaking slag (400 Mt/yr), platinum group mineral (PGM) mine tailings (77 Mt/yr), and red gypsum (1.25 Mt/yr). Ex-situ mineral carbonation takes place above ground using mined rocks and wastes. It can proceed through direct and indirect processes. In general, direct mineral carbonation can be performed through direct gas solid mineral carbonation or aqueous mineral carbonation. Indirect mineral carbonation takes place in more than one stage, and can be accomplished via gas-solid mineral carbonation, the pH swing process, the molten salt process, acid extractions, ammonia extraction, caustic extraction, and bioleaching. From a technical point of view, mineral carbonation is more favored through the indirect route because higher purity products can be obtained. Furthermore, the calcium and magnesium conversion rates to carbonates are significantly higher in indirect processes. Indirect gas–solid mineral carbonation requires large amounts of heat input, which inherently increases the energy consumption of the process. Therefore, indirect liquid phase mineral carbonation processes are the much better choice. They consist of a process chain with four essential steps: 1) liquid-solid extraction of metal ions (Mg²⁺,Ca²⁺), 2) solid-liquid separation by filtration, 3) carbonation of metal ions contained in the filtrate by use of CO₂, 4) separation of solid carbonate particles from the post-reaction mixture obtained in the carbonation step.
 

The aim is to develop the individual process steps for CO₂ storage and the production of recyclable carbonates and to link them to form an autonomous process chain. To this end a Self-Learning Robust Autonomous Control(ler) (SLARC) algorithm, consisting of three interconnected parametric statistical estimators, is introduced. The estimators are primarily meant to be implemented via (deep) neural networks (DNN) and thus can be trained using data and process knowledge (model). SLARC is to be understood as a kind of hybrid model which fuses physical laws with the information gathered from data. The former makes it possible to detect unfavorable/unphysical states by constraining to physics of the process while the latter makes it possible to detect changes in particle properties during operation and to adapt to the current state of the process. These two combined with the reinforced online learning feature make SLARC a self-learning algorithm. Additionally, since SLARC is built using a  stochastic optimization formulation, it is also designed to be robust to random disturbances. 
 

• Autonomous control of filtration and selective precipitation:
  One of the core objectives of this project is to design and implement a self-learning and robust controller (SLARC) to enable autonomous functioning of a complex process in general and specifically the filtration and precipitation processes to be developed.
• Development of an observable and controllable vacuum belt filter to apply maximum filtrate flow:
  The temporal monitoring and control of the filtrate flow with simultaneous maximum ion concentration of Mg²⁺ and Ca²⁺ is important for the subsequent selective precipitation. Thus, we integrate measurement technology to observe conductivity, cake height and filtrate flow online for the processed slurry on the lab-scale vacuum belt filter.
• Development of an observable and controllable selective precipitation process for high purity carbonates:
  A dynamic particle population balance (PBE) model of the selective precipitation of MgCO₃ and CaCO₃ particles is to be formulated in WP 2.3 as a basis for the design of SLARC (WP 3). Supported by inline sensors (FTIR probe, pH probe, spectral extinction probe for the mean particle size) and online measurement technology (ion chromatography, flow-through microscopy), the PBE model shall allow the prediction of the temporal evolution of the particle size distribution as well as the chemical composition of the particle population.
   

Project website
 

• CO₂ carbonation from mine waste
• Belt filtration
• Stochastic controller
• Reinforcement learning
• Autonomous control
• SLARC

Jan 2023 - Dec 2025

Prof. Dr.-Ing. Naim Bajcinca (TUK)
Dr. Sandesh Hiremath (TUK) 
Prof. Dr.-Ing. Kai Sundmacher (OvGU)
Dr.-Ing. Marco Gleiß (KIT)
 

Prof. Dr.-Ing. Naim Bajcinca
Gottlieb-Daimler-Str. 42
67663, Kaiserslautern
+49 (0)631/205-3230
naim.bajcinca(at)mv.uni-kl.de