Investigation into the use of cement kiln dust in high density sludge (HDS) treatment of acid mine water

- Cement kiln dust (CKD) was tested at bench-scale in the high density sludge process.
- Lower arsenic concentrations and elevated turbidity were found versus lime addition.
- Lower sludge volumes were also found with CKD addition compared to lime.
- High solids loading with CKD addition caused reduced water content of flocs/sludge.
- High solids loading increased particle interactions and compression settling rate.

The purpose of this study was to investigate the potential to replace lime with cement kiln dust (CKD) in high density sludge (HDS) treatment of acid mine drainage (AMD). The bench-scale study used two water samples: AMD sampled from a lead-zinc mine with high concentrations of iron (Fe), zinc (Zn), and arsenic (As) (Fe/Zn-AMD) and a synthetic AMD solution (Syn-AMD) spiked with ferric sulfate (Fe2(SO4)3). Arsenic was found to be significantly reduced with CKD-HDS treatment of Fe/Zn-AMD compared to lime-HDS treatment, to concentrations below the stringent mine effluent discharge regulation of 0.10 mg As/L (i.e., 0.04 ± 0.02 mg/L). Both CKD- and lime-HDS treatment of the two AMD samples resulted in settled water Fe concentrations above the stringent discharge guideline of 0.3 mg Fe/L. CKD addition in the HDS process also resulted in high settled water turbidity, above typical discharge guidelines of 15 mg TSS/L. CKD-HDS treatment was found to result in significantly improved settled solids (i.e., sludge) quality compared to that generated in the lime-HDS process. HDS treatment with CKD resulted in 25-88% lower sludge volume indices, 2 to 9 times higher % wet solids, and 10 to 20 times higher % dry solids compared to lime addition. XRD and XPS testing indicated that CKD-HDS sludge consisted of mainly CaCO3 and SiO2 with Fe3+ precipitates attached at particle surfaces. XRD and XPS testing of the lime-HDS generated sludge showed that it consisted of non-crystalline Fe oxides typical of sludge formed from precipitates with a high water concentration. Increased sedimentation rates were also found for CKD (1.3 cm/s) compared to lime (0.3 cm/s). The increased solids loading with CKD addition compared to lime addition in the HDS process was suggested to both promote surface complexation of metal precipitates with insoluble CKD particles and increase compression effects during Type IV sedimentation. These mechanisms collectively contributed to the reduced water content of CKD-HDS sludge. The results of this study suggest that solids loading is a significant factor in increased sludge density found with the HDS process compared to conventional lime precipitation-sedimentation.

325