Inoculation of granular activated carbon in a fixed bed with S-triazine-degrading bacteria as a water treatment process

Two bacterial strains (SL1: Rhodococcus rhodochrous, WT1: Acinetobacter junii) capable of biodegrading atrazine and simazine in surface water (1–10 μg l−1) were inoculated into fixed beds of granular activated carbon (GAC, 20 min empty bed contact time, EBCT). Eluted SL1, WT1 and indigenous bacteria were enumerated by spread plate on nutrient agar. Bacteria attached to GAC were desorbed in 50 mM Tris buffer (pH 7.5, recovery 75 ± 5%) before enumeration. Atrazine and simazine in treated water and adsorbed to GAC were extracted and determined by GC (detection limit 0.003 μg l−1). 48 h after inoculation with 9.8 × 108 SL1 or 1.3 × 109 WT1, 10% of the SL1 (1 ± 0.04 × 107 g−1 dry wt GAC) and 0.77% of the WT1 (1 ± 0.04 × 106 g−1 dry wt of GAC) remained in the fixed beds. No WT1 were detected in the treated water over a 28 day operating period although they were present on the GAC (2.8 ± 0.22 × 104 g−1 dry wt). An influent concentration of 2 mg l−1 atrazine and 2 mg l−1 simazine (in dH2O) was used to induce breakthrough exceeding 0.003 μg l−1 of s-triazines in treated water, and the effluent concentration of inoculated and non-inoculated columns was compared over 40 d (27 min EBCT). Up to day 18, WT1 reduced s-triazine concentration (maximum 0.15 ± 0.061 μg l−1 atrazine, 1.1 ± 0.08 μg l−1 simazine) in treated water compared to a non-inoculated column (maximum 0.49 ± 0.061 μg l−1 atrazine, 2.1 ± 0.08 μg l−1 simazine). After day 18 indigenous bacteria had acclimated to biodegrade the s-triazines and effluent concentrations were the same for both treatments. Total biodegradation of adsorbed s-triazine (87 mg atrazine and 87 mg simazine per fixed bed) ranged from 19.5 to 32% of each herbicide for both inoculated and non-inoculated GAC. Influent containing 10 μg l−1 atrazine and 10 μg l−1 simazine did not cause s-triazine breakthrough in treated water (< 0.003 μg l−1) over 20 days (40 min EBCT). Biodegradation of adsorbed atrazine and simazine (13.9 μg g−1 GAC for each s-triazine) by isolate SL1 (2 × 107 cells g−1 dry wt) was 53 ± 12.2% atrazine and 58 ± 8.2% simazine and was significantly greater than that of non-sterile, non-inoculated GAC (6 × 106 indigenous bacteria g−1 dry wt; 0% atrazine, 32 ± 8.2% simazine). Inoculation of GAC packed beds with s-triazine degraders reduced the amount of s-triazine adsorbed and treated water concentration within 40 days of inoculation and may extend bed life of GAC for surface water treatment.