Abstract
The production of sodium chromate (Na2CrO4) and its chemistry have been extensively studied by previous researchers. These studies have contributed to a better understanding of the reaction mechanisms and the optimisation of the Na2CrO4 formation process under different operating conditions. One of the main challenges is to maximize Na2CrO4 yields while minimizing the environmental impact of Cr6+.
In this study, chromite samples roasted with sodium chloride (NaCl) at temperature ranges 800 – 1200 ℃ for a period of 1 – 4 hours. These samples were pulverized to -75 μm particle, and a pellet was formed after mixing with the salt. Leaching of sodium chromate was performed using water as a lixiviant, with a 1:10 S/L ratio. Conversion to dichromate was achieved using varying concentrations of sulphuric acid. Leaching tests were conducted in batches, assessing the effects of acid concentrations (0 – 18 M), temperature (25 – 50 ℃), and time (5 – 45 minutes). Filtrate analysis was done using AAS, while residue analysis utilized XRF, XRD, and SEM-EDS. Iron, aluminium, magnesium, and chromium were the target metal ions, and their dissolved percentages were calculated based on initial mass percentages before leaching.
The thermodynamic effect of temperature has a significant impact on the formation of sodium chromate during the roasting stage. Intermediate temperature ranges (1000 - 1100 °C) promote the formation of sodium chromate, while higher temperatures (>1100 °C) result in the formation of unwanted metal chlorides, inhibiting the desired complex formation. The optimal temperature determined through thermodynamic and complexation studies is 1050 °C. Sodium chromate was successfully formed and identified as the most stable and feasible chromite complex. The roasting time and Cr:NaCl ratio are important parameters interacting with temperature for the formation of sodium chromate. An optimal roasting time of 2 hours and a Cr:NaCl ratio of 1:1.5 at the optimum temperature setting result in rapid conversions of insoluble chromite to water-soluble sodium chromate, achieving a purity of 96%.
After performing kinetics studies, it was discovered that the conversion of sodium chromate to sodium dichromate is influenced by temperature and sulfuric acid concentration. The dissolution of chromium follows a chemical reaction model, and the presence of ferric chloride and ferric sulphate affects the conversion of sodium dichromate. Optimum conditions for dissolution are achieved at a temperature of 45 °C and an acid concentration of 18 M.
Optimisation of the production of sodium dichromate salts using quantum machine learning algorithms
6 | P a g e
In terms of comparison to the conventional methods, simulations using Artificial Neural Networks (ANN) representing conventional approaches and Quantum Approximate Optimisation Algorithm (QAOA) representing quantum approaches were conducted. The QAOA model outperformed ANN in terms of computation time, error distribution, comparison metrics, and convergence trajectories, indicating its superior optimisation capabilities for the roasting process.
For comparison, the obtained results from ANN were assessed against those from QAOA, where they were observed to converge around optimum settings of roasting temperature of 1048.00°C, leaching time of 45.00 minutes, leaching temperature of 40.00°C, roasting time: 2.50 hours, acid concentration of 12.00M and Cr: NaCl Ratio of 1.50. The developed sensitivity analysis revealed that the above-mentioned optimum parameters may be ascendingly ranked as follows: roasting temperature, concentration of H2SO4, Cr:NaCl ratio, roasting time, leaching temperature, and leaching time. The predicted yields of metals in the sodium dichromate were simulated to be 7.89%, 6.39%, 96.95%, and 4.24% for Fe, Al, Cr, and Mg, respectively.
It was recommended that further research should explore and develop advanced quantum machine learning (QML) algorithms specifically for optimizing the production of sodium chromate and dichromate salts. Experimental validation at an industrial scale is necessary to verify the effectiveness of proposed optimisation strategies. Optimisation of other parameters such as pressure, reaction time, and additives should be investigated using QML algorithms. Future research should prioritize sustainable production methods and consider integrating QML algorithms with process control and monitoring systems. Collaboration and knowledge sharing among researchers and industry experts are essential for accelerating progress in this field.