Optimization of electron donor for SRB: application in acid mine drainage treatment
Many industrial wastewaters, particularly from the mining, fermentation and food processing industry contain high sulphate (SO42-) concentrations. SO42- reduction (SR) is a serious environmental problem under non-controlled conditions, which can result in the release of toxic sulphide to the environment. Typical characteristics of SO42--rich wastewater are 0.4-20.8 g SO42-.L-1, low pH, high oxidative potential, low or negligible chemical oxygen demand (COD) and high heavy metals concentrations (acid mine drainage), that can dramatically damage the flora and fauna of water bodies. The aim of this PhD is to study the effect of electron donor supply on the biological SR process by sulphate reducing bacteria in bioreactors. The biological SR process was studied using carbohydrate based polymers (CBP) and lignocelulosic biowaste (L) as slow release electron donors (CBP-SRED and L-SRED, respectively) in batch bioreactors and continuously operated inverse fluidized bed bioreactors (IFBB). IFBB were vigorously tested for SR under high rate and transient (feast-famine) feeding conditions. In another bioreactor configuration, a sequencing batch bioreactor, the effect of the initial SO42- concentration on the reactor start-up was investigated. Besides, the effect of the initial concentration of electron donor, NH4+, and SO42- were evaluated in batch bioreactors as well.
The robustness and resilience of SR was demonstrated in IFBB using lactate as the electron donor wherein the SO42- removal efficiency (SRE) was comparable in the feast period (67 ± 15%) of IFBB2 to steady feeding conditions (71 ± 4%) in the same IFBB2 and to IFBB1, the control reactor (61 ± 15%). From artificial neural network modeling and sensitivity analysis of data of IFBB2 operation, it was envisaged that the influent SO42- concentrations affected the COD removal efficiency, the sulphide production and pH changes. In another IFBB3 at a COD:SO42- ratio of 2.3, SR under high rate conditions (HRT=0.125 d) was 4,866 mg SO42-. L-1 d-1 and a SRE of 79% was achieved. Besides, the Grau second order and the Stover-Kincannon substrate removal models fitted the high rate reactor performance with r2 > 0.96. The COD:SO42- ratio was the major factor affecting the SR.
In batch bioreactors, using filter paper as CBP-SRED, SR was carried out at > 98% SRE. Using scourer as L-SRED, a 95% SRE was achieved. However, when the scourer was used as the L-SRED carrier material in an IFBB4, the SR showed 38 (± 14) % SRE between 10-33 d of operation. The SR was limited by the hydrolysis-fermentation rate and, therefore, the complexity of the SRED. Concerning sequencing batch bioreactor operation, the SR process was affected by the initial SO42- concentration (2.5 g SO42-.L-1) during the start-up. The sequencing batch bioreactor performing at low SRE (< 70%) lead to propionate accumulation, however, acetate was the major end product when SRE was > 90%. In batch bioreactors, the NH4+ feast or famine conditions affected the SR rates with only 4% and the electron donor uptake was 16.6% greater under NH4+ feast conditions. The electron donor utilization via the SR process improved simultaneously to the decreasing initial electron donor concentrations.
This PhD research demonstrated that the SR process is robust to transient and high rate feeding conditions. Moreover, SR was mainly affected by the approach how electron donor is supplied, e.g. as SRED or as easy available electron donor, independently of the COD:SO42- ratio. Besides, the electron donor and SO42- utilization were affected by the lack or presence of nutrients like NH4+.