Bioremediation of chlorinated ethenes in aquifer thermal energy storage

<strong>Subjects:</strong> bioremediation; biodegradation; environmental biotechnology, subsurface and groundwater contamination; biological processes; geochemistry; microbiology The combination of enhanced natural attenuation (ENA) of chlorinated volatile organic compounds (CVOCs) and aquifer thermal energy storage (ATES) appears attractive because such integration provides a promising solution for redevelopment of urban areas in terms of improving the local environmental quality as well as achieving sustainable energy supply. It will reduce the current negative interference between groundwater contaminants and ATES systems that arises from the rapid increase of ATES system numbers and generally long duration of contaminated groundwater treatments. However, currently the implementation of the combined system is at an initial stage, and still requires comprehensive study before advancing to mature application. Studies should specifically focus on understanding of the basic biogeochemical processes in aquifer systems under conditions of ATES and enhanced bioremediation and their mutual impacts when combined in ATES-ENA. To this end, the research as reported in this thesis employed laboratory experiments and modeling approaches focused on finding the essential process factors involved in the combined system, revealing possible drawbacks, and providing a better understanding to design alternative options on better operation of the combined system. Chapter 2 assessed the limiting factor for reductive dechlorination of PCE in an Fe(III) reducing aquifer, being the typical type of subsurface in the Netherlands. A step-wise batch study was performed which consisted of redox conditioning by lactate and ascorbic acid, followed by reductive dechlorination in different scenarios. For the sediment material sampled from the Fe(III) reducing aquifer, conditioning of the redox potential could stimulate PCE dechlorination. It was concluded that 75 µmol electron equivalents per gram dry mass of aquifer material was the threshold to obtain a redox potential of -450 mV, which is theoretically suitable for PCE reductive dechlorination. However, dechlorinating bacteria required for fully reductive dechlorination are generally lacking in Fe(III) reducing aquifers. Without bioaugmentation of dechlorinating bacteria, PCE could only be reduced to TCE or cis-DCE. The step-wise approach and findings obtained from different scenarios tested in this study are relevant for improving the cost-effectiveness of the design and operation of in situ bioremediation. The redox potential of an aquifer can be used as a general indicator to evaluate the potential for CVOCs reductive dechlorination. For achieving specific goals of in situ bioremediation projects at different CVOCs contaminated sites with various environmental conditions, the balance between cost, benefit, and potential risks (e.g. bio‑chemical well clogging due to bacteria growth and precipitation of metal-oxides) should be estimated before the design and operation of the ATES-ENA systems. This chapter provides insights into the essential factors that determine the feasibility of ATES-ENA. In Chapter 3, the two most important impacts of ATES on enhanced bioremediation of CVOCs were investigated using batch experiments. Besides, another type of underground thermal energy storage system, the borehole thermal energy storage (BTES) was also studied as a comparison to ATES. Here cis-DCE was targeted as it is commonly found to accumulate in the subsurface due to incomplete dechlorination. Compared to a natural situation (NS) with sufficient electron donor and bioaugmentation at a constant temperature of 10 ˚C, we assessed the effect of ATES by exchanging liquid between bottles kept at 25 and 5 ˚C, and the effect of BTES by alternating temperature between 25 and 5 ˚C periodically. Under ATES warm condition, cis-DCE was dechlorinated to ethene and at an increasing rate with each liquid exchange, despite no biodegradation being observed under ATES cold condition. The overall removal rate under alternating ATES conditions reached 1.83 μmol cis‑DCE/day, which was over 1.5 and 13 times faster than those in BTES and NS conditions. Most probably growth of biomass occurred under ATES warm condition, leading to an autocatalytic increase in conversion rates due to higher biomass concentration. Comparison between batches with or without <em>Dehalococcoides</em> inoculum revealed that their initial presence is a determining factor for the dechlorination process. Temperature then became the dominant factor when <em>Dehalococcoides</em> concentration was sufficient. The results also indicated that <em>Dehalococcoides </em>was preferentially attached to the soil matrix. This chapter highlights the importance of the dynamic temperature regimes in ATES on the bioremediation of CVOCs and recommends to implement biostimulation actions in the ATES warm well. Further impacts of ATES related to change in redox condition on bioremediation of CVOCs, with focus on microbial responses of <em>Dehalococcoides</em>, were explored in Chapter 4. In this chapter, we adopted a recirculating column experiment with a flow rate of 10 mL/min (representing the flow velocity at a distance of 1.3 m from the center of a typical ATES well) to simulate the ATES system. To mimic potential periodic redox fluctuations that accompany ATES, serial additions of lactate and nitrate were performed. Firstly, also at the relatively high liquid velocity (compared to normal bioremediation conditions) complete reductive dechlorination from cis-DCE to ethene was achieved in the column system. However, dechlorination was immediately terminated by subsequent nitrate addition due to direct interruption of <em>Dehalococcoides</em> retention to the soil matrix. In our column system, which was much more homogeneous than subsurface in reality, repeated interruption of dechlorination via <em>Dehalococcoides</em> was extremely severe. Such repeated interruption by nitrate dosing eventually led to less easily reversible while requiring more efforts for recovering dechlorination. In addition, the hypothesis of the immobility of <em>Dehalococcoides</em> was further confirmed by the microbial analysis of microorganism in the liquid phase where only less than 0.1% of the <em>Dehalococcoides</em> inoculum could be found back. Although some field studies demonstrated easier regeneration of <em>Dehalococcoides</em> in the subsurface after suffering oxidant, results from this chapter emphasized the sensitive resilience of <em>Dehalococcoides</em> which needs careful consideration in biostimulated ATES condition, and a functional combined system requires dedicated ATES operation and monitoring on the aquifer geochemical conditions. The major concern on possible negative impact of enhanced bioremediation on ATES is biological clogging attributed to biomass growth. As chemical clogging due to Fe(III) precipitates is a common problem in the functioning of ATES, the clogging issues (both biological and chemical) should be addressed before practical application. The potential clogging issues in the combined system were then researched in Chapter 5 using the same recirculating column system as in the previous chapter. For this purpose, two flow rates, 10 and 50 mL/min, were implemented. In the two columns, enhanced biological activity and chemically promoted Fe-oxide precipitation were studied by addition of lactate and nitrate respectively. Pressure drop (∆P) between the influent and effluent of the columns was monitored to indicate clogging of the system. The results showed no increase in ∆P during the period of enhanced biological activity, with large amount of lactate and active inoculum being added, even when the concentration of total bacteria in the liquid phase increased by four orders of magnitude. Nitrate addition, however, caused significant increase of ∆P. Remarkably, in the column with higher flow rate (50 mL/min), an unforeseen blow-up occurred at the end of experiment, as the buildup of pressure in the system was higher than the strength of the glass column. However, in the column with flow rate of 10 mL/min, high pressure buildup caused by nitrate addition could be alleviated by lactate addition. Such finding indicates that the risk of biological clogging related to biostimulation is relatively small, because by maintaining a low redox condition biostimulation itself may counter chemical clogging in ATES. Nevertheless, acknowledging that a column system cannot fully mimic real ATES conditions, additional tests are necessary to further investigate the clogging issues in the combined system. In Chapter 6, we performed a simulation of ATES-ENA with a reactive transport model, using ATES as the engineering tool for lactate injection in a hypothetical TCE contaminated aquifer which is assumed to be homogeneous. Many relevant processes in the combined system were simulated, such as TCE, cis-DCE and VC dechlorination, sulphate and Fe(III) reduction, organic acid fermentation and oxidation and growth of different biomass. In total 15 scenarios are considered in the model, including variations in lactate dosage (three concentration levels: 3.8, 1.9 and 0.38 mmol/L), temperature (three pairs for the ATES cold/warm well: 5/15 ˚C, 10/10 ˚C, 5/25 ˚C), biomass mobility (purely mobile or immobile), and pH limitation on Fe(III) reduction (absence and presence of such an effect). In the five years’ simulation by the model, complete dechlorination to ethene was achieved within 1 year, in the influence zone of the ATES wells, for the reference scenario with 3.8 mmol/L lactate, 5/15 ˚C ATES well temperatures and mobile biomass. Scenarios with lower dosage of lactate gave results with less dechlorination progress. Growth of biomass, especially iron reducer and lactate fermenter, was significant also in the first year (for both mobile and immobile biomass scenarios). Biomass also spread throughout the influence volume of ATES for both warm and cold wells. However, scenarios with different well-temperature pairs did not noteworthy differ in dechlorination progress. This could probably be due to biomass concentration being the limiting factor in this model setup, while temperature was not. Such situation was quite different than that in Chapter 3, of which experiment with bioaugmentation in the beginning. Besides, the model here could not include the important autocatalytic process (Chapter 3) which generated much faster dechlorination than just could be realized by only temperature increase in this chapter. In general, the modeling in this chapter suggests that applying ATES as engineering tool for biostimulation (substrate injection and bioaugmentation) can be a cost-effective approach to support the combined system. Eventually in Chapter 7, overall discussions upon results gained from previous chapters were integrated and the research questions as presented in the introduction are reiterated. In addition, recommendation upon future study, and wider implications with future perspective for practical application are also discussed. We concluded that redox condition is the most essential factor in the ATES-ENA system. The mutual impacts of ATES and ENA were revealed to be quite positive. Elevated temperature in the ATES warm well synergizing with groundwater transport can provide “1 + 1 > 2” effect. Besides, ENA can probably reduce risk of chemical clogging in ATES, instead of causing biological clogging. The further investigation was recommended to perform with larger scale pilot tests. Finally, a brief review of possible applications was given for two countries, the Netherlands and China, which both have dense groundwater and subsurface contaminations around urban areas. The ATES technology is much more mature in the Netherlands, whereas in China, the advantage is the more flexible usage of subsurface. For both countries, ATES-ENA can provide cost‑effective outcomes on both energy production and groundwater management.