Browsing by Author "Mgabhi, Senzo"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Open Access
    Investigating the influence of solute type and solute concentration on ice scale formation on Polypropylene Graphite (PP-GR) during Eutectic Freeze Crystallization
    (2025) Matau, Madimetsa; Lewis, Alison; Mgabhi, Senzo; Chivavava, Jemitias
    Eutectic freeze crystallization (EFC) is a promising sustainable method for the treatment of hypersaline streams. However, the formation of ice scale layers on heat exchanger surfaces has hindered its industrial implementation. Previous studies have shown that smooth heat exchanger (HX) materials with low surface energy like Polypropylene Graphite (PP-GR), can help reduce ice scaling, but their anti-scaling performance in the EFC of brines with varying solute types and concentration remains unexplored. Therefore, this research aimed to study how solute concentration and type affect the ice scaling on the HX surface of PP-GR during EFC. The study involved a continuous EFC process with a 3L column crystallizer equipped with a PP-GR heat exchanger tube, designed to observe the formation of ice scale layers and bulk crystallization. The production of ice was measured every 20 minutes for 1 hour. The amount of ice scale layer was measured after the experiment. The study was conducted in two phases: the first phase used multicomponent industrial brine from the Tweeifontein Water Reclamation Plant with Total Dissolved Solids (TDS) concentrations of 103 g/L. This brine was pre-concentrated using the cascade method to obtain concentrations of 122 g/L, and 133 g/L TDS. The second phase investigated synthetic brines of Na2SO4H2O and MgSO4-H2O at a constant molality of 0.613 mol/kg and varying MgSO4 concentrations of 0.613, 1.130, and 1.47 mol/kg. In the industrial brines, increasing the total solute concentration, from TDS of 103 to 133 g/L, reduced ice scale layer amount by 72%, from 42 g to 11 g. At 103 g/L TDS, the ice scale layer contributed 62% of the total ice yield, while at higher concentrations, it contributed less (18% at 122 g/L and 9% at 133 g/L). The ice scale layer formed within 20 minutes for the brine with 103 g/L TDS, within 30-40 minutes for 122 g/L TDS, and within 40-60 minutes for 133 g/L TDS. This reduction in ice scaling was linked to decreased interfacial interactions between the brine and heat exchanger (HX) surface, likely due to increased surface tension (from 79 to 81 mN/m) and contact angle, which weakened the attraction between the brine and HX surface. However, the role of surface tension in ice scaling could not be fully isolated due to the influence of low ice seed loading (0.5 wt.%) and low mixing intensity (Re ~700), which led to uneven supersaturation and favored surface nucleation. Future studies should increase mixing intensity and ice seed loading to better understand the impact of surface tension on ice scaling. The differences in ice scale formation were not only due to delays in ice scale layer formation but also to variations in growth rate and delamination as solute concentration changed. Slower growth of the ice scale layer was observed in more concentrated brines, likely due to increased resistance to mass transfer of water molecules from the bulk solution to the ice scale layer. Ice scale layer from dilute brine adhered strongly to the PP-GR surface, making removal difficult, while scales from more concentrated brines (122 g/L and 133 g/L TDS) delaminated easily. This is likely due to reduced ice adhesion strength at higher concentrations, which disrupts water molecule orientation and weakens interfacial interactions. In lower concentrations, more ordered water layers enhance adhesion. Variations in contact angle may also influence this, with lower angles in dilute brine promoting stronger adhesion, while higher angles in concentrated brines reduce it. Other factors remain unclear. For the binary synthetic brines, it was found that increasing concentration (0.61 to 1.47 mol/kg) in synthetic binary MgSO4 brines decreased ice scale amount by 99% (from 82 g to 0.7 g). The ice scale layer formed within 20 minutes for the dilute brine (0.613 mol/kg), within 20-40 minutes for 1.13 mol/kg, and within 40-60 minutes for the 1.47 mol/kg. The reduction in ice scaling tendency with increased concentration for this brine was also attributed to the reduced interfacial interactions between the brine and HX surface. Increasing the concentration from 0.613 to 1.47 mol/kg increased the surface tension from 77 mN/m to 79.4 mN/m. This is thought to have reduced the strength of interfacial interactions between the brine and HX surface, possibly delaying the formation of ice scaling. Slower growth of ice scale layer was also observed at higher concentrations possibly due to increased resistance to mass transfer, with Mg²⁺ ions forming a diffusive boundary layer that delayed ice scale growth. It was also observed that the ice scale amount on the surface of PP-GR was 30% lower in MgSO4 brine compared to that of Na2SO4 brine at the constant molality of 0.613 mol/kg While the ice scale layer formed within 20 minutes for both solutions, Na2SO4 brine had a faster ice scaling formation, leading to a 65% decrease in ice production rate (from 2.60 to 0.90 g/min), whereas MgSO4 brine showed only a 22% decrease (from 2.04 to 1.58 g/min). Despite slower initial ice production, MgSO4 brine maintained a steadier rate due to slower growth and frequent delamination of ice scale layers. Mg2+ ions may have disrupted hydrogen-bonded networks within the liquid-like layer on the ice surface, weakening adhesion. Additionally, Mg2+ ions likely neutralized surface charges on the surface of ice, further reducing adhesion. Although the exact mechanism is unclear, both disruption of hydrogen bonds and electrostatic screening are likely to contribute to delamination. These findings suggest that ions with higher charge density, like Mg2+, may reduce ice scaling and allow for longer freeze crystallization periods. Overall, the current study has shown that selecting highly concentrated brine along with PP-GR would be beneficial in delaying the formation of ice scaling, reducing the growth rates of ice scale layers, and increasing the frequency of delamination of ice scale layers. Furthermore, the type of solute presents in a brine affects the formation of ice scaling as there are solutes that may have negligible or significant impact on ice scaling.
  • No Thumbnail Available
    Item
    Open Access
    Using simulation and lab validation to develop a MnCO3 Recovery Process using CO2
    (2025) Sibanda, Thabo; Lewis, Alison; Chivavava, Jemitias; Mgabhi, Senzo
    Manganese (Mn) is a critical metal in the production of lithium-ion battery (LiB) precursors due to its role in improving safety, stability and promoting higher efficiency and faster charging of LiBs. Battery-grade Mn or High Purity Manganese Sulphate Monohydrate (HPMSM, MnSO4·H2O), a key LiB precursor, requires low Mg and Ca content (< 0.01 wt.% each). Industrial MnSO4 pregnant leach solutions are a valuable source of HPMSM but conventional purification using electrowinning is energy-intensive, unsustainable, and environmentally harmful. Therefore, this study aimed to investigate the feasibility of chemical precipitation using a greenhouse gas (carbon dioxide gas, CO2) and ammonia (NH3) as a sustainable and cheaper alternative purification method. A high-concentration industrial MnSO4 pregnant leach solution containing at least 93.9 wt.% Mn2+, 2.23 wt.% Mg2+, and 0.14 wt.% Ca2+ was used. The results on the effect of pH from thermodynamic simulations were compared to experimental results. Experimental results investigated the effect of pH from 5.0 to 6.6 and CO2 bubbling times from 1 to 12 h using a 1.0 L semi-batch and continuously stirred glass reactor at ambient temperature and pressure. The CO2 was sparged at 0.4 L/min and the agitator speed was 500 rpm. The thermodynamic simulation predicted more than 94% Mn2+ recovery at pH > 5.0, with optimal Mn2+ selectivity at pH < 6.6. The experimental results showed optimal Mn2+ recovery of 61.3% at pH 6.6 and 8 h CO2 bubbling time, with the rejection of 57.6% Mg2+ and 46.3% Ca2+ from the MnCO3 precipitate, respectively. The discrepancy between simulation and experimental results was attributed to the slow dissolution rate of CO2. Finally, regardless of the CO2 bubbling time and pH, the washed MnCO3 precipitate contained at least 98.8% Mn, 0.15% Ca, and 0.05% Mg, meeting high-purity Mn specifications, but slightly lower than the requirements for battery grade Mn (> 99.9 wt.%, ultra-high-purity). The CO2 bubbling time and pH have a significant influence on both the recovery of Mn2+ and the rejection of Mg2+ and Ca2+. It is recommended that future work to explore the influence of pH 6.6-7.0, the effect of increasing partial pressure of CO2, and the use of nanobubbles to enhance the CO2 absorption. This study showed that carbonate precipitation using CO2 and NH3 can selectively recover Mn2+ from an industrial MnSO4 leachate containing high Mg2+ and Ca2+ impurities, offering a sustainable process with great potential for industrial application.