Treatment Of Petroleum Refinery Wastewater By Anoxic-Aerobic Sequential Moving Bed Reactors

No Thumbnail Available
Journal Title
Journal ISSN
Volume Title
Simulated petroleum refinery effluent with varied sulfide (0-1200 mg/L), phenol (0-1200 mg/L), diesel (0-700 mg/L), ammonia-N (0-450 mg/L) and nitrate-N (0-2000 mg/L), was assimilated in anoxic sequencing batch reactors to maximize elemental sulfur (S0) generation. The nitrate-N to sulfide molar ratio of 3.04 was found to be optimum, which correspond to influent concentrations of sulfide 750 mg/L, phenol 750 mg/L, diesel 300 mg/L, ammonia-N 350 mg/L and nitrate-N 1000 mg/L with maximum S0 generation of 300 ± 12 mg/L. Feed with the optimized concentrations of pollutants was fed to an anoxic disc bed reactor (A1) and complete removal of sulfide was achieved at 20 rpm speed and 2.5d HRT with partial removal of organics and ammonia. Residual organics and ammonia were removed to the dischargeable limit in a downstream aerobic moving bed reactor (A2) at 20h HRT with overall system HRT of 80h. Final effluent COD from the system was less than 100 mg O2/L and the conical bottom in both the reactors was useful for effective solids (biomass + elemental S0) separation. Effect of various hydrocarbons [kerosene (K), heavy engine oil (HO), mixtures of kerosene, heavy engine oil and diesel (K+HO and K+HO+D) and crude oil, 300 mg/L] with different density and viscosity were studied in the anoxic-aerobic system. Density and viscosity of the hydrocarbons decreased in the order of crude oil > HO > K+HO+D > K+HO > K. Organics degradation in A1 was hampered, sloughing of the biomass was triggered and decrease in solids retention time (SRT) was observed with increase in density and viscosity of the hydrocarbons. Effect of hydrocarbons was not observed in A2 and more than 99% removals of sulfide, organics and ammonia-N were achieved by the anoxic-aerobic sequential moving bed reactors at 80h HRT. Anoxic biomass was dominated by Pseudomonas aeruginosa SKM2013 and Pseudomonas aeruginosa SC2013, where the later showed chemolithotrophic characteristics. Aerobic biomass was dominated by Lysinibacillus sp. H200-510, Stenotrophomonas sp. LW-34, Pseudomonas aeruginosa strain ISB4 and Pseudomonas aeruginosa strain LZS8436. The first three aerobic strains were able to degrade ammonia-N and the fourth one was able to utilize nitrite-N as electron acceptor in the presence of organics. During the treatment of real automobile service station wastewater, complete oxidations of phenol, hydrocarbons and nitrate-N were achieved at 22.5h HRT in A1 and residual ammonia-N was oxidized at 7.5h HRT in A2, at an overall HRT of 30h by the anoxic-aerobic sequential moving bed reactors. Sludge released from A1 and A2 were suggested for their suitability for utilization in agricultural lands after metals analysis. Removal of organics deteriorated in A1 with increase in NaCl. More than 30 g/L of NaCl hampered the performance of A2 and there was decrease in the removals of organics and ammonia-N by the anoxic-aerobic system. Organics removal and denitrification in A1 and ammonia-N nitrification in A2 were hampered with crude oil shock loads of 600 mg/L and 900 mg/L applied in two phases. Recovery of A1 and A2 were achieved after the withdrawal of NaCl dose and each shock load, suggesting complete recovery of the anoxic-aerobic system. Heterotrophic activity of both anoxic and aerobic biomass were higher whenever degradation of organics were higher. Similarly, chemolithotrophic activity of anoxic biomass was higher when sulfide oxidation and conversion of sulfide to sulfate were higher. Direct dependence of biomass activity with the removal of COD was observed. Hence, biomass activity test can be used as an indirect tool to predict bioreactor performance.
Supervisor: S. Chakraborty