Two-dimensional assessment of cobalt transport and separation through ionic polymer inclusion membrane: experimental optimization and artificial neural network modeling

In today's world, increasing technological demand and decreasing tenor of cobalt on the earth make the selective extraction and the recycling of cobalt processes indispensable. The study is to highlight both experimental optimization and artificial neural network (ANN) modeling of cobalt separation and transport performances through PVC-based ionic polymer inclusion membrane (IPIM) considering different molecular structure room temperature ionic liquids (RTILs) derivatives as an ion carrier. Some sophisticated investigations on PVC-based IPIMs have been done by changing membrane compositions. Also, the membranes were characterized in different aspects and techniques like SEM-EDX, AFM, contact angle measurement, etc. In optimum conditions, the initial mass transfer coefficient (J(i)) of cobalt was found about 2,44x10(-6) mols(-1)m(-2) by the longest alkyl chain substituted ionic liquid (RTIL4) and also found that the selectivity of the process was found very well for cobalt in the presence of the other metal ions, especially toward nickel and cadmium. ANN modeling has been performed on the effective parameter analysis that was performed on the experimental results. As a result of ANN analysis, the most effective parameters were determined as RTIL type and operation time have higher modeling sensitivity. The promising results show experimentally optimized and ANN modeled process has a cost-effective, environmentally-friendly potential for cobalt separation and transport.