Bacterial dynamics during aerobic and anoxic biodegradation of MTBE at bench and pilot-scale

Abstract

Unintentional contamination of groundwater with gasoline oxygenates is a problem that emerged in the 1990s. The fuel oxygenate methyl tert-butyl ether (MTBE) and aromatic hydrocarbons benzene, toluene, ethylbenzene and xylenes (BTEX) are major gasoline constituents that co-occur in groundwater when storage tanks leak. Several physical and chemical treatment methods have been implemented for remediating gasoline impacted sites, however they are subject to high operational and construction costs as well as unreliable performance. Bioremediation has emerged as a promising alternative and low cost option, however, an understanding of the microbiology of MTBE remediation is needed in order to optimize this treatment in the field. Three main challenges are addressed in this study: the lack of appropriate microorganisms, anoxic conditions, and the presence of other gasoline components, such as BTEX, that potentially inhibit MTBE bioremediation. Specifically, the objectives of this study are to: 1.) determine the effect of microbial community and reactor conditions on aerobic MTBE biodegradation in the presence of BTEX; 2.) explore the potential of wetlands for bioremediating MTBE in the presence of other contaminants under low-oxygen conditions; 3.) explore the potential of anaerobic permeable reactive barriers (PRBs) for bioremediating MTBE; 4.) isolate a pure culture capable of anoxic MTBE biodegradation; and 5.) characterize bacterial dynamics during biodegradation of MTBE under ferric and sulfate reduction conditions. The overall goal is to support practical treatments for long-term sustainability. The effect of microbial community and reactor conditions on aerobic MTBE biodegradation in the presence of BTEX was addressed with a batch experiment, where the MTBE removal rates of two aerobic cultures was compared in the presence and the absence of BTEX. Culture MO was previously enriched in MTBE and culture MB was enriched in MTBE and BTEX. A semi-batch reactor was set up using the MO culture. BTEX was added in this reactor. The composition of the microbial consortia was explored using a combination of 16S rRNA gene cloning and quantitative polymerase chain reaction targeting the known MTBE-degrading strain PM1T. It was observed that the composition and diversity of the microbial community in the reactors played an important role on its ability to remove MTBE and BTEX simultaneously. The removal of these contaminants was feasible in the reactors that had a smaller concentration of PM1T, but other MTBE degraders were present (Hydrogenophaga sp.), and in the reactors that had a more diverse microbial community. The potential of wetlands for bioremediating MTBE in the presence of other contaminants under low-oxygen conditions was demonstrated with a mesocosm study, using scaled wetlands. The effects of natural attenuation, bioaugmentation and biostimulation were studied and compared. Water coming from the influent of an existing air stripping system in a refinery near Denver, CO. was used as the source water for this experiment. Two mixed cultures were used as the inocula in this test. An anaerobic Fe-reducing MTBE-degrading culture and culture MO (previously enriched in MTBE). High removal rates were observed in the scale wetlands, however, due to a high concentration of PM1 in the refinery water it was not possible to separate the effect of bioaugmentation or biostimulation from natural attenuation. The potential of anaerobic permeable reactive barriers (PRBs) for bioremediating MTBE was studied first with a batch experiment in order to determine if the anaerobic Fe-reducing MTBE-degrading culture was able to degrade MTBE under Fe(III) and SO4 conditions; and then with a column study where two reactive materials were compared. The two reactive materials were selected based on two criteria cost and reactivity. The two slow release electron acceptor sources used in the experiment were Fe(OH)3 and CaSO4. The columns were inoculated with the anaerobic Fe-reducing culture. It was observed that the anaerobic consortia was able to degrade MTBE under both reducing conditions in the batch reactors, however MTBE was not removed in the columns, with the exception of the beginning of the experiment, when the columns were spiked with humic acids. It was assumed that the malfunction of the columns was due to the failure of providing the right conditions for the bacteria to remove MTBE, since the solubility of the reactive media was very low and probably did not release enough electron acceptor to be used for the bacteria. A pure culture isolate of anoxic MTBE biodegradation was isolated from an anaerobic Fe-reducing culture and an aerobic culture (MO). The purity of the isolates was demonstrated with scanning electron microscopy (SEM), denaturing gel gradient electrophoresis (DGGE) and direct sequencing of the 16S rRNA gene. A phylogenetic three was constructed. The cells were observed to have a rod shape with an approximate length of 1.5 μm and width of 0.5 μm. No flagella were observed. The cells were determined to be Gram-variable. Based on 16S rRNA sequencing, the closest match for the pure culture was Desulfosporosinus meridiei S5 (AF076248). The pure culture was observed to be able to degrade MTBE under Fe(III) and SO4 reducing conditions. The characterization of microbial community dynamics during biodegradation of MTBE under ferric and sulfate reduction conditions was studied using semi-batch reactors. A semi-batch reactor was set up using Fe2(SO4)3 as the electron acceptor. Once this reactor was actively removing MTBE, its culture was used to seed four sets of batch reactors (nine in total). In the first set, the effect of having one electron acceptor (Fe(III) or SO4) and sodium sulfide on the removal of MTBE was studied. In the second set, the same conditions were studied but in the absence of sodium sulfide. In the third set the presence of BTEX on the removal of MTBE in the presence and absence of sodium sulfide was determined. In the last set of reactors, the effect of adding two electron acceptors (Fe(III) and SO4) using a different salt for each electron acceptor (i.e. Na2SO4 and FeCl3) was studied. It was observed that the presence of one electron acceptor (vs. both), sodium sulfide and BTEX reduced the removal rate of MTBE. It was also observed that the oxidation-reduction potential also had an effect on the removal of MTBE. Total inhibition or a great reduction in the removal rate was observed in the reactors which source water was more reduced. Overall it was observed that the removal of MTBE is feasible under aerobic and anaerobic conditions and in the presence of BTEX, and that overall success may be enhanced by several factors. First, it is important to recognize that the characteristics of the microbial consortium do matter. Second, the reactor configuration is also important for controlling the composition and performance of the microbial consortium.