Recovery of oxalic acid from hydrochloric acid and cobalt containing hydrometallurgical side-stream by solvent extraction

[Display omitted] •Comparison of reagents for solvent extraction of oxalic acid.•1.83 M TBP proposed as extractant and pure water as stripping agent.•Verification of the process design in continuous bench-scale mixer-settlers.•More than 93% of oxalic acid recovered at 98.55% relative purity. Oxalic acid is a crucial precipitation and reducing agent in hydrometallurgical processes, and significant amounts of oxalic acid with a small content of metals are created as side streams. The extraction of oxalic acid from these side streams would have both economic and environmental benefits. Economically, it can provide a valuable source of revenue, and environmentally, it can reduce the amount of waste. In this research, the aim was to investigate recovery of 10.2 g l−1 of oxalic acid from 37.5 g l−1 hydrochloric acid solution containing 1.1 g l−1 Co with tributyl phosphate (TBP), trialkyl phosphine oxides (CYANEX 923), trioctylamine (TOA), and trioctylmethylammonium chloride (Aliquat 336) extractants. The effects of extractant concentration, equilibration time, and temperature were investigated by batch experiments. Oxalic acid distribution coefficient increased when temperature increased with CYANEX 923 and Aliquat 336 but decreased with TBP. Fast extraction kinetics was observed with both TBP and CYANEX 923 reagents, and the equilibrium was achieved within 1 min. The results of the equilibrium studies and theoretical McCabe–Thiele analysis showed that the effectiveness of extracting oxalic acid using 1.83 M TBP, 0.25 M of CYANEX 923, and 1.09 M Aliquat 336 in three counter-current stages would be 96.2 %, 93.4 %, and 88.2 % respectively. Furthermore, four counter-current stripping stages with pure water would result in 95.5 %, 88.4 %, and 89.1 % yields, respectively. Continuous counter-current experiments in a bench-scale mixer-settler cascade were performed with 1.83 M of TBP to investigate the effect of number of stages (N), residence time in mixer (τmix) and phase ratio (O/A) on the extraction and stripping efficiency of oxalic acid. Over 99 % of oxalic acid was extracted from the feed solution using a O/A = 1:1, τmix = 1 min, N = 4, and at T = 20 ± 1 °C. The co-extraction percentages for HCl and Co were 14.6 % and 5.1 %, respectively. 94.0 % of the oxalic acid could be stripped from the loaded reagent using pure water with O/A = 1:2, τmix = 1 min, N = 4, and at T = 20 ± 1 °C. Overall, the results of this work demonstrate the potential of the designed solvent