Page 1 of 6

Environmental impacts of aquifer thermal energy storage investigated by field and laboratory experiments

Matthijs BONTE1, Boris VAN BREUKELEN2, Pieter STUYFZAND1,2

1KWR Watercycle Research Institute Groningenhaven 7 Postbus 1072

3430 BB Nieuwegein e-mail: matthijs.bonte@kwrwater.nl, pieter.stuyfzand@kwrwater.nl

2Afdeling Hydrologie en Geo-Milieuwetenschappen, Vrije Universiteit Amsterdam De Boelelaan 1085

1081 HV, Amsterdam, Nederland e-mail: Boris.van.breukelen@falw.vu.n

Abstract Aquifer thermal energy storage (ATES) uses groundwater to store energy for heating or cooling purposes in the built environment. ATES systems are often located in the same aquifers used for public drinking water supply which has led to questions being raised by water authorities and companies on the environmental impacts of ATES. This contribution presents the results of on-going research aiming to elucidate the effects that ATES operation may have on chemical groundwater quality. Field data from an ATES site in the south of the Netherlands show that results in chemical quality perturbations due to homogenization of the present vertical quality gradient. Laboratory data derived from column experiments show that effects at temperatures below 25°C are limited, while high temperature ATES operation (60°C) can result in carbonate dissolution and precipitation, possibly clogging wells, desorption of alkali metals such as K and Li and trace metals like As. Keywords: aquifer thermal energy storage, ground source heat pumps, environmental impacts, groundwater quality. Introduction Aquifer thermal energy storage (ATES) uses groundwater to store energy for heating or cooling purposes in the urban environment. In the Netherlands, most ATES systems operate at a relatively small temperature ranging to several °C above and below the natural groundwater temperature. The maximum allowable injection temperature is set in provincial regulations and ranges between 25 and 30 °C (Bonte et al., 2008). Interest for high temperature ATES, between

40 to 80 °C is growing and pressure is increasing to allow this as well. Over the last decade, the number of ATES systems in the Netherlands has grown rapidly: from around 100 in 2000 to more than 1000 in 2010. This exponential growth has led to an increasing number of sites where ATES systems are planned or built near groundwater pumping stations used for public water supply. This often leads to questions by regulators regarding the risks of ATES in aquifers used for public water

Page 2 of 6

supply, in the Netherlands and abroad (Bonte and P. J. Stuyfzand, 2011). Most published research on this subject focuses on operational aspects, such as scaling due to mineral precipitation occurring at high temperature systems, and is based on either laboratory experiments (Griffioen and Appelo, 1993), chemical equilibrium modelling (Palmer and Cherry, 1984) or a combination of both (Arning et al., 2006). The results of these studies show for example that depending on aquifer geochemistry, precipitation of iron-carbonates can form a problem for ATES. A study that is more relevant for drinking water production is by Brons et al. (1991) who looked at mobilisation of organic carbon from sediments under increased temperature. Incubation experiments showed that at temperatures above 45°C organic carbon was mobilised from sediment resulting in an increased chemical oxygen demand potentially leading to values above the maximum allowable concentration (MAC) for drinking water. Microbiological research pertaining to ATES showed that although no evidence is observed for a growth of pathogens (Winters, 1992) or increasing cell counts (Schippers and Reichling, 2006), a considerable change in the microbiological community composition may occur (Sowers et al., 2006; Brielmann et al., 2009; Brielmann et al., 2011). This paper describes the preliminary outcomes of our on-going research on ATES systems in the Netherlands. Our research builds on the above mentioned studies and focuses on the risks that ATES may impose on groundwater quality, drinking water production and the sub-soil environment in general.

We use field data from an ATES site located in Eindhoven to investigate these effects on groundwater quality in a real system. This ATES system is located near public water supply well fields. In addition, laboratory experiments are carried out in which sediment cores collected in Helvoirt from the same aquifer as used at the ATES system in Eindhoven are subjected to different temperatures. Methods In order to assess the effects of the ATES system on groundwater quality, we sampled groundwater from ATES production wells and monitoring wells non-influenced by the ATES system. Water samples were analysed on a broad range of anorganic, organic and microbiological parameters. Figure 1 presents a hydrogeological cross section of the site in Eindhoven. More details on the site in Eindhoven are reported in Bonte et al. (2011).

Figure 1: Hydrogeological cross section of the ATES study site in Eindhoven. The column experiments were carried out with sediments collected using Ackerman core sampling. In order to keep sediments anaerobic, N2 gas was bubbled through the drilling work water and the Ackerman cores were transported to the laboratory in a chilled cooler filled with N2 gas. The sediment cores were taken from two depths in the fluvial Sterksel formation, in a coarse

Page 3 of 6

and fine part of the formation. Sediment of cores collected at the same depth were unwrapped in a glovebox under a N2 atmosphere, mixed thoroughly to obtain a homogeneous sample and repacked in four Ackerman cores with 1 cm increments. After addition of each layer, the sediment was compacted to avoid the formation of preferential flow paths (Rinck-Pfeiffer et al., 2000). The columns were stored at 4 ° C to avoid microbiological changes until they are used (Castro et al., 2002). Five subsamples were taken from the bulk sample and analysed for moisture content, organic matter and carbonate content using thermo gravimetric analyses (TGA), grain size distribution and clay content using laser particle sizing, total element content using X-ray fluorescence (XRF) and leachable element content by ICP-MS analyses of an aqua regia digest. We note at the time of writing of this abstract the results of XRF and aqua regia+ICP are not available yet.

Figure 2: Column testing set-up.

The cores were placed in the experimental set-up shown in figure 2. The four cores were maintained at temperatures of 5, 12, 25 and 60°C and flushed with groundwater sampled from a filter constructed at the depth from which the sediment cores were collected. A mixture of N2/CO2 gas (99%/1%) was bubbled through the holding tank to keep the water anaerobic and at constant pH during the experiment. Dissolved oxygen (D.O.) is continually measured in both the holding tank and in-line in the effluent of core 2. D.O. remained below detection limits during the experiments (holding tank < 0,05 mg/l, effluent core 2 < 0,01 mg/l). The sediments were flushed with a flux of 0.26 ml/min resulting in a residence time of 38.5 hours. In- and effluent samples were taken in containers pre-filled with N2 gas and analysed for major, trace elements and dissolved organic carbon. Results from ATES field site Data for a selection of elements collected from the ATES production at the field site in Eindhoven are shown in figure 3. The blue shadings in the figure show the observed concentration range in ambient groundwater (measured in reference monitoring wells). The minimum and maximum temperatures observed at the site are 7 and 22°C, respectively. The data show a cyclic character for many elements. This cyclic character is explained by to the relatively high groundwater flow velocity in the aquifer at the site: The ATES system extracts groundwater from the different depths within aquifer having different qualities, mixes it and injects it back into the aquifer. The injected groundwater mixture has a different quality than the ambient

Page 4 of 6

groundwater: leveling out the natural vertical concentration gradient.

Figure 3: Observed groundwater quality in two monitoring filters in two ATES production wells with ambient concentrations shown with blue shading. The ATES system introduces a fraction of shallow groundwater with a different chemical composition at greater depth, and a fraction of deep groundwater at shallow depth. At the depth of monitoring wells shown in figure 3, this leads to increased concentrations of SO4, Cl, Na, K and other elements, while concentrations of CH4 and Fe decreased. After some time, during the storage phase of ATES operation, the injected groundwater drifts away with the natural groundwater flow and groundwater quality returns to ambient conditions. The data show that ATES operation changes the natural redox zoning in the aquifer, which may trigger secondary reactions, for example mobilization of trace metals and organic carbon. At this site this is however hard to investigate in detail: none monitoring wells turned out to be close enough to capture the injected ATES water.

Overall, the observed concentration changes at this site are sufficiently small to keep groundwater suitable for drinking water production from a physical-chemical point of view. Preliminary results from laboratory experiments Table 1 shows some characteristics of the two sediment types investigated. The deeper sediment has higher carbonate, organic carbon and clay content and is expected to more reactive than the coarser shallower sediment. Table 1: Sediment characteristics Sediment Organic

Carbon (LOI550)

CaCO3% Clay% D50 (µm)

A (42 m) 0.76% 6.01% 3.7% 187

B (33 m) 0.48% 0.53% 2.2% 244

The data analyses of the column experiments is at the time of writing of this abstract still in progress. In figure 4 we present a preliminary selection of some of the effluent curves from the column experiments for sediment B to illustrate a number of processes thought to occur. The first two processes which may occur relate to silicate and carbonate mineral phases: Silicate minerals become more soluble at higher temperatures while carbonate minerals become less soluble. Si and K in effluent of 60°C core during the first 5 pore flushes (PFs) are almost twice the concentration of effluent obtained from cores 1 to 3. The differences may however also be explained by de- and adsorption of dissolved Si and K on exchangers and other reactive surfaces. We are currently pursuing the construction of a kinetic chemical model to simulate various competing processes. Ca and HCO3 concentrations in effluent of 60°C core are relatively low, possible the

Page 5 of 6

result of precipitation of carbonate minerals such as calcite and dolomite (Mg is not shown but shows a similar pattern to Ca). DOC of the two lowest temperature cores is on average equal to the influent, DOC of core 3 is slightly higher (0.1 mg/l) while DOC of core 4 is increased by 46% from 2.4 to 3.5 mg/l. The increase in DOC observed in effluents of sediment A (higher organic carbon) show an even greater increase in DOC with increasing temperature. These results are consistent with the findings of Brons et al. (1991).

Figure 4: Effluent data of column experiments on sediment B. HCO3 is in meq/l. Effluent Fe concentrations show a pattern which is not easily explained: all effluents are depleted in iron relative to the influent with the 12°C core showing the greatest depletion and the 25°C core the least. After 7 PFs all cores show a rapid further depletion. As is not present in influent but is considerably increased above the drinking water limit of 10 µg/l for both 25 and 60°C

cores. In the 12°C As is not significantly above the detection limit, but interestingly however, in the 5°C core it is.

Conclusions

Field data collected at an ATES system in Eindhoven show that the groundwater circulation by the ATES system can impact chemical groundwater quality by introducing shallow groundwater with a different chemical composition at greater depth and vice versa. However, the observed concentration changes are sufficiently small to keep groundwater suitable for drinking water production. Laboratory experiments up to now suggest that high temperature thermal energy storage at 60°C significantly impacts groundwater quality, increasing DOC and arsenic. Also at 25°C storage effects on arsenic and DOC are visible, albeit much less pronounced. Concluding, both field and laboratory data suggest that ATES operation can impact upon groundwater chemical quality. Effects are obviously highest at high temperature storage and are potentially irreversible, especially for DOC mobilisation. DOC in turn can act as a carrier for trace elements which means that mobilised trace elements will remain mobile and are not re-sorbed or re-precipitated. These aspects should be considered when planning or designing an ATES system, especially in aquifers used for water supply.

Acknowledgements

This project was funded through the joint water research program of the Dutch water supply companies.

Page 6 of 6

References Arning, E., Kölling, M., Schulz, H.D.,

Panteleit, B. and Reichling, J., 2006. Einfluss oberflächennaher Wärmegewinnung auf geochemische Prozesse im Grundwasserleiter. Grundwasser, 11(1): 27-39

Bonte, M., 2011. Effects of aquifer thermal energy storage on groundwater quality and the consequences for drinking water production: a case study from the Netherlands. Water Science & Technology 63(9): 1922-1931.doi:10.2166/wst.2011.189

Bonte, M. and P. J. Stuyfzand, A.H., and P. Van Beelen., 2011. Underground thermal energy storage: environmental risks and policy developments in the Netherlands and European Union. . Ecology and Society 16(1): 22

Bonte, M., Van Den Berg, G. and A.M., W., 2008. Underground thermal energy storage in relation to groundwater protection (in Dutch). Bodem, 5/2008: 22-26

Brielmann, H., Griebler, C., Schmidt, S.I., Michel, R. and Lueders, T., 2009. Effects of thermal energy discharge on shallow groundwater ecosystems. Fems Microbiology Ecology, 68(3): 273-286.10.1111/j.1574-6941.2009.00674.x

Brielmann, H. et al., 2011. Oberflächennahe Geothermie und ihre potenziellen Auswirkungen auf Grundwasserökosysteme. Grundwasser: 1-15.10.1007/s00767-011-0166-9

Brons, H.J., Griffioen, J., Appelo, C.A.J. and Zehnder, A.J.B., 1991. (Bio)geochemical reactions in aquifer material from a thermal

energy storage site. Water Research, 25(6): 729-736

Castro, H., Queirolo, M., Quevedo, M. and Muxí, L., 2002. Preservation methods for the storage of anaerobic sludges. Biotechnology Letters, 24(4): 329-333

Griffioen, J. and Appelo, C.A.J., 1993. Nature and extent of carbonate precipitation during aquifer thermal energy storage. Applied Geochemistry, 8(2): 161-176

Palmer, C.D. and Cherry, J.A., 1984. Geochemical reactions associated with low-temperature thermal energy storage in aquifers. Can. Geotech. J., 21(3): 13.10.1139/t84-051

Rinck-Pfeiffer, S., Ragusa, S., Sztajnbok, P. and Vandevelde, T., 2000. Interrelationships between biological, chemical, and physical processes as an analog to clogging in aquifer storage and recovery (ASR) wells. Water Research, 34(7): 2110-2118

Schippers, A. and Reichling, J., 2006. Laboruntersuchungen zum Einfluss von Temperaturveränderungen auf die Mikrobiologie des Untergrundes. Grundwasser, 11(1): 40-45

Sowers, L., York, K.P. and Stiles, L., 2006. Impact of thermal buildup on groundwater chemistry and aquifer microbes, Ecostock 2006, New Jersey.

Winters, A.L., 1992. Summary of Research on microbiological processes, Dept. Biological Sciences. Univ. of Alabama., PNL-7980