MODELLING AND OPTIMIZATION OF HEXAVALENT CHROMIUM [CR (VI)] REMOVAL FROM AQUEOUS SOLUTION USING MODIFIED AND NON-MODIFIED RICE HUSK: INTEGRATED RESPONSE SURFACE METHODOLOGY AND ARTIFICIAL NEURAL APPROACHES

Summary

Hexavalent chromium [Cr(VI)] in industrial wastewater requires treatment methods that are efficient, affordable and adaptable to resource-constrained settings. This study compared nitric-acid-modified and non-modified rice husk in a 29-run response surface design comprising initial chromium concentration (20-100 mg/L), pH (2-10), adsorbent dose (0.10-1.00 g) and contact time (10-70 min). Residual dissolved chromium and percentage removal were modelled with quadratic response surface methodology (RSM) and a 4-10-1 artificial neural network (ANN). Chemical modification lowered the mean residual concentration from 34.26 to 20.59 mg/L and increased mean removal from 45.66% to 67.00%, a gain of 21.33 percentage points. The maximum experimental removal was 98.2% for modified rice husk. At the replicated centre point, modified and non-modified materials produced 68.77 ± 0.31% and 44.10 ± 0.17% removal, respectively (n = 3). For modified rice husk, the RSM calibration yielded R² = 0.9940 and RMSE = 1.3923 mg/L, while leave-one-out cross-validation yielded R²cv = 0.9701 and RMSEcv = 3.1111 mg/L. The corresponding ANN calibration/split metrics were R² = 0.9292 and RMSE = 4.7909 mg/L. Solution pH was the dominant factor, followed by initial concentration, and the concentration-pH interaction was highly significant. Constrained optimization located the best model-based operating region at pH 2.0, 100 mg/L, 1.00 g and 70 min. The results demonstrate that nitric-acid modification improves rice-husk performance and that RSM offers reliable interpolation within the investigated domain, although validation with real wastewater, chromium-speciation analysis, regeneration testing and continuous-flow trials remains necessary.

Index Terms

Browse additional articles authored by one or more contributors to this paper.