- Email: [email protected]
A preliminary survey of anthropogenic gadolinium in water and sediment of a constructed wetland
Journal of Environmental Management 255 (2020) 109897
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: http://www.elsevier.com/locate/jenvman
Short communication
A preliminary survey of anthropogenic gadolinium in water and sediment of a constructed wetland Anthony J. Altomare, Nicholas A. Young, Melanie J. Beazley * Department of Chemistry, University of Central Florida, Orlando, 4111 Libra Drive, Room 255, Florida, 32816, USA
A R T I C L E I N F O
A B S T R A C T
Keywords: Gadolinium Emerging contaminants Constructed wetland Rare earth elements Environmental contaminants
Gadolinium (Gd) is a rare earth element used in magnetic resonance imaging (MRI) contrast agents that has recently been identified as an emerging contaminant of concern due to its possible toxic effects and accumulation in the environment. The objectives of this preliminary study were to determine the occurrence and fate of Gd in surface water and sediment of a constructed wetland that receives effluent from a wastewater treatment plant. The rate of anthropogenic Gd entering the wetland was determined to be approximately 25 g Gd day 1, with surface water concentrations in the parts per trillion. Anthropogenic Gd concentrations in surface waters decreased as a function of distance from the inlet site to near the outfall, and were three orders of magnitude higher in sediment than in surface water suggesting that the wetland was providing a sink for Gd possibly through plant uptake and incorporation in organic biomass. An anthropogenic Gd anomaly was observed with an average GdAnt/GdGeo ratio of 5.34. Sediment with higher total organic carbon (TOC) tended to be higher in anthropogenic Gd, suggesting that Gd sequestration may occur through uptake by plants and/or through floc culation and deposition of natural organic matter.
1. Introduction Gadolinium (Gd) is a rare earth element (REE) commonly used in magnetic resonance imaging (MRI) contrast agents. Gadolinium-based contrast agents (GBCA) consist of organic ligands such as Gd-DTPA, Gd-DOTA, and Gd-DO3A-butrol complexed with Gd that protect the body from the toxic effects of free Gd3þ ion (Pagano et al., 2015). GBCAs are administered to patients prior to MRI imaging but are not metabo lized in humans. Instead, the GBCA complex is removed by the renal and hepatic systems and excreted in urine and feces (Le Fur and Caravan, 2019). GBCAs have recently been identified as emerging environmental contaminants of concern due to their presence in medical waste and sewage and the potential toxicity of free dissolved Gd (Rogowska et al., 2018). Although GBCAs are generally regarded as safe for medical use, there are a number of health effects related to Gd exposure (Le Fur and Caravan, 2019). GBCAs are known to accumulate in bone and brain tissues of patients at concentrations in the ng kg 1 to μg kg 1 range (Murata et al., 2016), with greater accumulation possible in patients with renal conditions and renal failure. The Gd3þ ion is widely consid ered to be more toxic than GBCA complexes and can interfere with
biological processes that utilize Ca2þ channels (Caravan et al., 1999). Free Gd has been shown to cause inhibition of Vibrio fischeri (Kurvet et al., 2017) at as low as 6.4 mg L 1 as compared to GBCAs, which have shown no-observed-effect concentrations (NOEC) as high as 937 mg L 1 in aquatic organisms (Neubert et al., 2008). Both GBCAs and free Gd have been shown to accumulate in bivalves (Perrat et al., 2017) and other terrestrial and aquatic organisms (Lingott et al., 2016). Currently, the chemical breakdown of GBCAs, both in municipal wastewater treatment systems and the environment, is not well under stood. Several studies have shown that dissolved Gd free ions and GBCA complexes are not efficiently removed in wastewater treatment plant (WWTP) processes (Kunnemeyer et al., 2009; Rabiet et al., 2009). As a result, there is potential for considerable release of Gd into the envi ronment. Studies performed in South Korea (Song et al., 2017), Prague, €ller et al., 2002), and Brazil (Pedreira et al., 2018) Czech Republic (Mo have shown Gd concentrations in WWTP effluents ranging from 39 to 410 ng L 1. Though the toxicity of GBCA degradation byproducts is not known, nor do researchers fully understand the effects of the anthro pogenic release of Gd into the environment, occurrence of Gd has been used previously to track wastewater release in the environment (Goulle et al., 2011; Pedreira et al., 2018). Since the behavior of Gd-containing
* Corresponding author. 4111 Libra Dr, Room 255, Orlando, FL, 32816, USA. E-mail addresses: [email protected] (A.J. Altomare), [email protected] (N.A. Young), [email protected] (M.J. Beazley). https://doi.org/10.1016/j.jenvman.2019.109897 Received 27 July 2019; Received in revised form 13 November 2019; Accepted 18 November 2019 Available online 26 November 2019 0301-4797/© 2019 Elsevier Ltd. All rights reserved.
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
Fig. 1. Map of OEW showing cell numbers and the three major flow trains, North (blue), Central (yellow), and South (red). Sample sites for the current study are indicated with a star. Site locations include Cell 1 (28� 340 12.500 N 81� 000 45.200 W), Cell 14 (28� 340 39.000 N 80� 590 52.100 W), Cell 16B (28� 340 26.700 N 80� 590 24.800 W), and Cell 17 (28� 340 53.200 N 80� 590 24.100 W). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
materials in nature is largely unknown and varies depending on the system, it is important to understand the fate and transport mechanisms of anthropogenic Gd in the environment. Several studies have shown that hospital and WWTP effluents are major contributors to increases in anthropogenic Gd in the environment (Bau et al., 2006; Hatje et al., 2016; Luo et al., 2014). Anthropogenic Gd has primarily been found in fresh and coastal waters that receive WWTP effluents, where the levels of anthropogenic Gd in WWTP effluent were indirectly tied with higher hospital patient volume. These studies give further evidence that the major source of Gd released into the envi ronment originates from human wastewater containing GBCAs. These previous studies have primarily focused on Gd in surface waters with limited studies on Gd in sediments (da Silva et al., 2018; Rogowska et al., 2018). In the current study, the occurrence and fate of anthropogenic Gd in a constructed wetland that receives WWTP effluent is examined. As WWTP effluent is the primary source of surface water for this wetland, it allows for the analysis of anthropogenic Gd in a more controlled envi ronmental system as compared to large bodies of water such as rivers or coastal waters. The Orlando Easterly Wetlands (OEW) located east of Orlando, Florida is one of the largest constructed wetland treatment systems in Florida that receives wastewater effluent from the City of Orlando’s Iron Bridge Regional Water Reclamation Facility and has been in operation since 1987 (Wang et al., 2006). The purpose of this wetland is to polish nutrients from the WWTP effluent stream before it is discharged into the nearby St. John’s River (White et al., 2018). The average loss of total phosphorus load from the inflow to the outfall of the wetlands is 70.54% with flocculation and settling of organic material as the hypothesized sequestration mechanism. The rate of effluent delivery from the WWTP to the OEW is approximately 45,000 m3 per day at the inlet (White et al., 2018). From there, it flows through a series of cells, which are individual bodies of water separated by sand berms. The water follows three flow paths (Fig. 1) with residence times of 18.1, 37.6, and 54.6 days for the northern, central, and southern flow paths, respectively (Wang et al.,
2006). Since 2001, the OEW has conducted a wetlands rejuvenation project to increase the hydraulic efficiency and nutrient removal effectiveness of the wetland. Water within selected cells in the flow paths is drained and the top layers of sediment excavated and removed. New vegetation is planted, and the cells returned to service. Overall phosphorus removal and hydraulic efficiency has increased as a result of the rejuvenation project (Wang et al., 2006). The Iron Bridge WWTP services portions of the City of Orlando, as well as other portions of northern Orange and southern Seminole Counties. This area includes six hospitals and several smaller medical facilities that may serve as anthropogenic sources of Gd. As a result, Iron Bridge, and by extension OEW, may receive considerable amounts of anthropogenic Gd. As OEW receives water primarily from the Iron Bridge WWTP, it offers a unique natural system in which to study the behavior of anthropogenic Gd without the dilution effects of other sur face waters. Therefore, the objectives of this study were to determine the occurrence and fate of Gd in the water and sediment of the OEW con structed wetland. 2. Materials and methods 2.1. Sample site description The OEW is a constructed wetland located in Christmas, Florida 20 km east of the City of Orlando. The wetland was constructed in 1987 and receives wastewater effluent from the City of Orlando’s Iron Bridge Regional Water Reclamation Facility (Wang et al., 2006). Surface water and sediment cores were collected at the OEW in Cell 1 (28� 340 12.500 N 81� 000 45.200 W), Cell 14 (28� 340 39.000 N 80� 590 52.100 W), Cell 16B (28� 340 26.700 N 80� 590 24.800 W), and Cell 17 (28� 340 53.200 N 80� 590 24.100 W) (Fig. 1). These sample sites were selected for this pre liminary study as they represented a typical flow of water through the wetland and were also sites of previous studies conducted at the OEW. 2
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
Fig. 2. Log REE values obtained from PAAS-normalized concentrations in sediment. The anomaly of Gd suggests the presence of anthropogenic Gd relative to geologic Gd. Points where no value is shown indicate the REE was below the limit of detection. Sample legend indicates ‘Cell number-core replicate numbersediment depth’.
Water depths at Cells 1, 16B, and 17 were between 1 and 2 m at the sample site, while Cell 14 contained no surface water. The water column at these sites had an average pH of 6.90, with orthophosphate concen trations between 2.1 and 3.2 ppm, and reducing conditions and low dissolved oxygen levels except near vegetation. Sediment composition in Cells 14, 16B, and 17 was rich in organic matter near the surface and transitioned to sand near the bottom. Cell 1 had limited vegetative growth, due to high water flow rates, and thus was primarily sand. Sediment samples that were rich in organic matter on average were observed to have higher Fe and Al concentrations at 2.4 � 0.7 and 8.4 � 1.8 g kg 1, respectively, compared to sandy sediment samples that contained 0.34 � 0.29 and 0.97 � 1.2 g kg 1, respectively. Cell 1 re ceives inflow directly from the Iron Bridge WWTP. Cells 17 and 16b are located along the Northern and Central flow paths, respectively, in close proximity to the outfall site. Cell 14 is located along the Central flow path. The wetlands rejuvenation project affected two cells that were sampled in this study. Cell 1 and 17 were excavated and replanted in 2001 and 2007, respectively, and at the time of sampling Cell 14 had been drained of surface water and was drying prior to excavation. Cells 14 and 16b had not previously been excavated at the location where samples were taken and the sediment at these sites was a result of ver tical accretion since 1987.
(v/v) trace metal grade nitric acid (Fisher Scientific) prepared with ul trapure water (Elga Veolia). REE (La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Er, Tm, Yb, and Lu) concentrations were determined using a Thermo Fisher Scientific iCAP-Qc inductively coupled plasma mass spectrometer (ICPMS) with QCell technology and operated in kinetic energy discrimina tion (KED) mode of analysis with helium as the collision gas. Calibra tion, internal, and quality control standards (Inorganic Ventures) were prepared in 2% (v/v) HNO3 and calibration standards were prepared at concentrations of 10–100 ng L 1 (ppt). All REEs were measured with the exception of Pm, Tb, and Ho, which were either present as internal standards or not present in the analytical standard. Average concen trations were calculated as an average of three analytical runs for each sample (<5% RSD). The analytical limit of detection for this method was <2 ng L 1 (ppt) for Gd based on the standard deviation of five method blanks. 2.4. Sediment analysis Sediment core sections were analyzed for total acid extractable REE metals. Total acid extractable REE concentrations were determined by digesting ashed sediment in concentrated acid. Dry sediment (0.2–0.5 g) was combusted in a Thermo Fisher Lindberg Blue M furnace at 600 � C for 6 h, digested in 1:4 concentrated hydrochloric acid (trace metal grade; Fisher Scientific) to concentrated HNO3 at 80 � C in a water bath for 12 h, and the supernatant analyzed by ICP-MS. Method blank sam ples (without sediment) were processed the same as field samples to determine any contamination during sample preparation. REE concen trations were normalized to Post-Archean Australian Shale (PAAS) (McLennan, 1989) and geological Gd (GdGeo) was interpolated from the third order fit of the logarithm of the normalized REE elements as €ller et al. (2002). Anthropogenic Gd (GdAnt) was described by Mo determined as the difference between the measured Gd (GdMeasured) and GdGeo.
2.2. Sample collection Surface water and sediment cores were collected in October 2018. Samples were collected in acid-cleaned polypropylene collection bottles and core liners (7.5 cm dia.) and stored on ice after collection for transport to the lab. Water samples were collected near the water surface at each site (depth < 1 m). All sediment cores were hand-cored and capped at each site with sediment depth ranging from 10 to 14 cm. Duplicate sediment cores were extruded (on day of collection) in 1 cm sections for the first 6 cm and in 2 cm increments after. Each section was then centrifuged at 3750 RPM and the porewater collected as the supernatant.
2.5. Total organic carbon analysis
2.3. Water and porewater analysis
Total organic carbon (TOC) analysis was performed using a Teledyne Tekmar Lotix with Lotix Solid Sampler boat for solid samples. Samples were analyzed in triplicate using 2–50 mg dry sediment. Sediment was homogenized using a mortar and pestle and acidified with 22%
Surface waters and porewaters were filtered (0.2 μm pore size polypropylene membrane syringe filters; VWR) and diluted 1:1 with 4% 3
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
by interpolating from the third-order trend of the normalized Log(REE) values. The difference between GdGeo and Gdmeasured was used to calculate GdAnt (Fig. 2). 3.1. Dissolved Gd Gd concentrations in wetland water decreased from 555 ng L 1 at the inlet in Cell 1 to 422 ng L 1 near the outfall in Cell 17 (Fig. 3). Nearoutlet concentrations of Gd were comparable to a previous study where Gd concentrations of 410 ng L 1 were determined at a WWTP outfall along the Atlantic Ocean on the Brazilian coast (Pedreira et al., 2018), demonstrating that although WWTP systems are capable of precipitating Gd, anthropogenic Gd is still released into the environ ment. The speciation of Gd may play a role in its fate depending on whether it is in the chelated form or the ionic Gd3þ form. Previous studies have shown that chelated REEs tend to be less reactive and, therefore, more soluble than their ionic forms (Khan et al., 2017). Using the inlet value of ~0.56 μg L 1 Gd entering OEW and assuming that is a typical value throughout a given year, at approximately 45,000 m3 per day of effluent entering OEW that equates to a flux of approximately 25 g Gd per day. The presence of a distinct GdAnt anomaly may suggest that the wetland is a sink for Gd, most probably through adsorption, precipita tion, and/or complexation reactions. Previous studies on aqueous speciation have shown that the Gd3þ ion forms complexes with car bonate, as well as phosphate in systems where orthophosphate is present at high concentrations (Verplanck et al., 2005). In addition, precipita tion of REEs with phosphate has been identified as a possible mechanism for inorganic sedimentation of Gd (Byrne et al., 1996). The concentra tions of phosphorus entering the OEW may provide opportunity for Gd–P precipitation and/or complexation reactions. Total phosphorus concentrations in the surface water were measured (by ICP-MS) between 165 and 417 μg L 1; however, there was no correlation observed be tween dissolved P and dissolved Gd concentrations. While a previous study has attributed REE anomalies in estuarine sediment dredge spoils to phosphate fertilizers due to a positive correlation between REEs and phosphate (Xu et al., 2018), no such correlation was determined in OEW water and sediment. Typically, REEs in sediment are found in many forms, including the water-soluble, exchangeable, carbonate-bound, organic, and Fe–Mn bound fractions of extracted sediment (Wali et al.,
Fig. 3. Gd concentrations in water and sediment porewater. Surface water concentrations are shown at a depth of ( 1 cm) where 0 cm corresponds to the sediment-water interface. All values of depth �0 cm represent sediment pore waters. Error bars represent standard deviations between duplicate sediment cores. Missing data points represent sediment sections without enough pore water for collection.
phosphoric acid (Fisher Scientific) to remove inorganic carbon and combusted at a furnace and catalyst temperature of 800 � C. Organic carbon standards (Ricca Chemical) were prepared at concentrations of 100–2500 μg organic carbon and verified using a 1200 μg check standard. 3. Results and discussion Gd is a REE found naturally in the environment, and to distinguish naturally occurring REE from anthropogenic REE, the data was normalized to REE concentrations in PAAS. Gd was the only REE detected in OEW water and sediment porewater, so it was assumed that all measured dissolved Gd was anthropogenic. In the acid-extracted sediment samples, Gd had higher concentrations relative to the other REE, indicating an anthropogenic Gd anomaly. GdGeo was determined
Fig. 4. Anthropogenic Gd concentrations in duplicate sediment cores. Anthropogenic Gd was calculated as the difference between Geological Gd and Measured Gd. Error bars represent standard deviations between duplicate sediment cores. Sample legend indicates ‘Cell number-core replicate number’. 4
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
therefore, the buildup of sediment in the other cells was a result of vertical accretion since 1987 when the OEW was constructed. The depth of sediment accretion (below sediment-water interface) ranged from 4 cm at Cells 16B and 17 to 12 cm at Cell 14. Each of the sediment cores were sectioned down to about 14 cm and the porewaters removed. Sediment at Cell 1 (inlet) was primarily sandy with little organic matter and minimal porewater. Dissolved Gd tended to decrease below the sediment-water interface (Fig. 3) suggesting the loss of dissolved Gd into the sediments. Due to compaction, sediments were less saturated in portions of some of the cores and porewater was not available to collect as shown in Fig. 3 by missing data points. At the time of sample collection, Cell 14 was drained of surface water and was in the process of dry down in preparation for excavation. Dissolved Gd in porewaters below 6 cm at this cell was less than porewaters at the other sites sug gesting the dry down procedure may have affected sediment uptake. Concentrations of Gd as determined by ICP-MS of the acid digested sediment samples were normalized to PAAS values (Fig. 2). GdAnt was present in ppb (ug/kg) concentrations at each site (Fig. 4). It should also be noted that in some samples, GdGeo was observed to be less than 10 μg/ kg, resulting in GdAnt comprising a majority of the measured Gd. In samples where GdGeo was greater than 10 μg/kg, the average ratio of GdAnt to GdGeo was 5.34. In general, GdAnt decreased at greater sediment depths with the exception of samples taken from Cell 1. These trends coincided with changes in sediment composition, with samples con taining high amounts of sandy material containing lower amounts of GdAnt. In contrast, samples with a darker muck-like consistency, along with greater amounts of TOC, had higher concentrations of GdAnt. It was also observed that Cells 14 and 16b shared a similar trend with a decrease in GdAnt at 2 cm and a subsequent increase at 3–4 cm, which possibly indicates temporal changes in either Gd introduction or sedi mentation rates. Cell 14 displayed the highest overall average GdAnt concentrations. Cell 1 was primarily sand, while Cells 14, 16b, and 17 all
Fig. 5. Total organic carbon content in wetland sediment. Error bars represent standard deviations between duplicate sediment cores. Sample legend indicates ‘Cell number-core replicate number’.
2014; Wang et al., 2001). 3.2. Sediment Gd The OEW was constructed in 1987 and accretion, or biomass accu mulation, was measured by the depth of sediment in the collected cores above the compacted sand layer. Cells 1 and 17 were the only cells in the OEW to have a previous excavation (2001 and 2007, respectively);
Fig. 6. Correlations between Anthropogenic Gd and total organic carbon for each cell in the wetland. Correlation strength is displayed as adjusted (r2) values. The slope of the correlation is represented as (m). 5
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
contained several centimeters of organic-rich sediment before tran sitioning to the cell bed at greater depths. TOC data was collected for sediment samples from each of the four cells studied (Fig. 5). TOC trends were similar to those of GdAnt, with the lowest concentrations in Cell 1 that decreased at lower depths in the sediment in the other three cells. In addition, the trend of Gd decreasing at a depth of 2 cm was conserved for TOC concentrations in samples from Cells 14, 16b, and 17. This data also demonstrates the transition from organic-rich sediment to sand at the cell bed, which occurred at 12, 10, and 4–8 cm for Cells 14, 16b, and 17, respectively. These transitions correspond to decreases in GdAnt concentrations giving additional evi dence of the location of the original sand cell bed. The correlation between TOC and GdAnt is shown in Fig. 6 for each cell. TOC and GdAnt show conservative positive correlations in Cells 16b and 17, suggesting that Gd sedimentation may be linked to the deposi tion of natural organic matter (NOM) in this system. Cell 14, which was in the process of dry down, demonstrated only a moderate correlation of Gd with TOC. The evaporation of water in this cell would have removed Gd from the dissolved phase and onto sediment particles possibly dis rupting the TOC-Gd correlation. The slope of the relationship between GdAnt and TOC varied between cells that could be caused by a number of factors such as differences in flow and settling rates, vegetative species and density, dissolved Gd concentrations, and other reactions that may result in Gd precipitation. If Gd sedimentation occurs primarily through flocculation or complexation of Gd with NOM, this may explain the relatively low concentrations of Gd in Cell 1 in comparison to the other cells in this study. Cell 1 had very little vegetation, and thus a low flux of NOM compared to the other cells. Conversely, Cells 16b and 17 had large amounts of vegetation, which explains why the top layers of sediment at these sites contained up to 30–50% organic carbon by mass. An additional pathway for Gd into sediments and correlation with TOC is through plants. Plants have previously been shown to uptake chelated Gd through their root systems that eventually becomes distributed throughout the plant from the roots to the leaves (Lingott et al., 2016). The speciation of Gd in the environment whether as the free Gd3þ ion or in the chelated form such as a GBCA will have an important impact on its behavior in water and sediment. As a chelate, Gd is more soluble and less reactive with particle surfaces and, therefore, moves through a water system with less retention. However, as the charged ion, Gd is more reactive and available for precipitation, such as with phosphate, or adsorption onto sediment particles, and more prone to sequestration. More information is needed to determine the effects of WWTP and environmental processes on the chelated complexes of Gd.
D. Sees of the Orlando Easterly Wetlands for providing access to the wetlands, assistance with sample collection, helpful discussions on the wetlands, and providing diagrams and maps. We also thank the two anonymous reviewers for their valuable comments to improve the manuscript. References Bau, M., Knappe, A., Dulski, P., 2006. Anthropogenic gadolinium as a micropollutant in river waters in Pennsylvania, and in Lake Erie, northeastern United States. Chem. Erde-Geochem. 66, 143–152. Byrne, R.H., Liu, X.W., Schijf, J., 1996. The influence of phosphate coprecipitation on rare earth distributions in natural waters. Geochim. Cosmochim. Acta 60, 3341–3346. Caravan, P., Ellison, J.J., McMurry, T.J., Lauffer, R.B., 1999. Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem. Rev. 99, 2293–2352. da Silva, Y., do Nascimento, C.W.A., da Silva, Y., Amorim, F.F., Cantalice, J.R.B., Singh, V.P., Collins, A.L., 2018. Bed and suspended sediment-associated rare earth element concentrations and fluxes in a polluted Brazilian river system. Environ. Sci. Pollut. Res. 25, 34426–34437. Goulle, J., Lacroix, C., Saussereau, E., Mahieu, L., Cellier, D., Spiroux, J., Guerbet, M., 2011. Gadolinium and platinum in wastewater and sludge as tracers of medical activities. Toxicol. Lett. 205. S125–S125. Hatje, V., Bruland, K.W., Flegal, A.R., 2016. Increases in anthropogenic gadolinium anomalies and rare earth element concentrations in San Francisco Bay over a 20 year record. Environ. Sci. Technol. 50, 4159–4168. Khan, A.M., Abu Bakar, N.K., Abu Bakar, A.F., Ashraf, M.A., 2017. Chemical speciation and bioavailability of rare earth elements (REEs) in the ecosystem: a review. Environ. Sci. Pollut. Res. 24, 22764–22789. Kunnemeyer, J., Terborg, L., Meermann, B., Brauckmann, C., Moller, I., Scheffer, A., Karst, U., 2009. Speciation analysis of gadolinium chelates in hospital effluents and wastewater treatment plant sewage by a novel HILIC/ICP-MS method. Environ. Sci. Technol. 43, 2884–2890. Kurvet, I., Juganson, K., Vija, H., Sihtm€ ae, M., Blinova, I., Syvertsen-Wiig, G., Kahru, A., 2017. Toxicity of nine (doped) rare earth metal oxides and respective individual metals to aquatic microorganisms Vibrio fischeri and Tetrahymena thermophila. Materials 10, 754. Le Fur, M., Caravan, P., 2019. The biological fate of gadolinium-based MRI contrast agents: a call to action for bioinorganic chemists. Metallomics 11, 240–254. Lingott, J., Lindner, U., Telgmann, L., Esteban-Fernandez, D., Jakubowski, N., Panne, U., 2016. Gadolinium-uptake by aquatic and terrestrial organisms-distribution determined by laser ablation inductively coupled plasma mass spectrometry. Environ. Sci.: Processes Impacts 18, 200–207. Luo, Y.L., Guo, W.S., Ngo, H.H., Nghiem, L.D., Hai, F.I., Zhang, J., Liang, S., Wang, X.C. C., 2014. A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment. Sci. Total Environ. 473, 619–641. McLennan, S.M., 1989. Rare-earth Elements in Sedimentary Rocks - Influence of Provenance and Sedimentary Processes. Mineralogical Society of America, Special Paper, Washington. M€ oller, P., Paces, T., Dulski, P., Morteani, G., 2002. Anthropogenic Gd in surface water, drainage system, and the water supply of the city of Prague, Czech Republic. Environ. Sci. Technol. 36, 2387–2394. Murata, N., Gonzalez-Cuyar, L.F., Murata, K., Fligner, C., Dills, R., Hippe, D., Maravilla, K.R., 2016. Macrocyclic and other non-Group 1 gadolinium contrast agents deposit low levels of gadolinium in brain and bone tissue: preliminary results from 9 patients with normal renal function. Invest. Radiol. 51 (7), 447–453. Neubert, C., Laenge, R., Steger-Hartmann, T., 2008. Gadolinium containing contrast agents for magnetic resonance imaging (MRI) - Investigations on the environmental fate and effects. In: Aga, D.S. (Ed.), Fate of Pharmaceuticals in the Environment and in Water Treatment Systems, pp. 101–120. Pagano, G., Guida, M., Tommasi, F., Oral, R., 2015. Health effects and toxicity mechanisms of rare earth elements-Knowledge gaps and research prospects. Ecotox. Environ. Safe. 115, 40–48. Pedreira, R.M.A., Pahnke, K., Boning, P., Hatje, V., 2018. Tracking hospital effluentderived gadolinium in Atlantic coastal waters off Brazil. Water Res. 145, 62–72. Perrat, E., Parant, M., Py, J.S., Rosin, C., Cossu-Leguille, C., 2017. Bioaccumulation of gadolinium in freshwater bivalves. Environ. Sci. Pollut. Res. 24, 12405–12415. Rabiet, M., Brissaud, F., Seidel, J.L., Pistre, S., Elbaz-Poulichet, F., 2009. Positive gadolinium anomalies in wastewater treatment plant effluents and aquatic environment in the Herault watershed (South France). Chemosphere 75, 1057–1064. Rogowska, J., Olkowska, E., Ratajczyk, W., Wolska, L., 2018. Gadolinium as a new emerging contaminant of aquatic environments. Environ. Toxicol. Chem. 37, 1523–1534. Song, H., Shin, W.J., Ryu, J.S., Shin, H.S., Chung, H., Lee, K.S., 2017. Anthropogenic rare earth elements and their spatial distributions in the Han River, South Korea. Chemosphere 172, 155–165. Verplanck, P.L., Taylor, H.E., Nordstrom, D.K., Barber, L.B., 2005. Aqueous stability of gadolinium in surface waters receiving sewage treatment plant effluent, boulder creek, Colorado. Environ. Sci. Technol. 39, 6923–6929. Wali, A., Colinet, G., Ksibi, M., 2014. Speciation of heavy metals by modified BCR sequential extraction in soils contaminated by phosphosypsum in Sfax, Tunisia. Environ. Res. Eng. Manag. 4, 14–26.
4. Conclusions The results of this preliminary study describe the occurrence and fate of anthropogenic Gd in a constructed wetland that receives effluent primarily from a WWTP. Concentrations of dissolved Gd in surface waters decreased between the inlet of the wetland and the outfall, suggesting the wetlands may act as a sink for Gd and is supported by an average GdAnt/GdGeo anomaly of 5.34 in the sediment. Concentrations of Gd were found to have a moderate correlation to TOC in some cells, suggesting the mechanism of Gd sequestration in the wetland may be through complexation with organic biomass or incorporation into biomass via plant uptake. However, these results are based on only one set of sampling timepoints. Further work is needed at this site with more extensive sampling throughout the OEW and over multiple timepoints in order to provide a better understanding of the role the wetlands play in Gd fate and transport. Acknowledgements This research was funded by Beazley Startup Funds from the Uni versity of Central Florida. The authors wish to thank Dr. Emily Heider for assistance with sample collection. The authors especially thank Mark 6
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
Wang, H., Jawitz, J.W., White, J.R., Martinez, C.J., Sees, M.D., 2006. Rejuvenating the largest municipal treatment wetland in Florida. Ecol. Eng. 26, 132–146. Wang, Z.W., Shan, X.Q., Zhang, S.Z., 2001. Comparison of speciation and bioavailability of rare earth elements between wet rhizosphere soil and air-dried bulk soil. Anal. Chim. Acta 441, 147–156. White, J.R., Sees, M., Jerauld, M., 2018. The long-term use of treatment wetlands for total phosphorus removal: can performance Be rejuvenated with adaptive
management? In: Nagabhatla, N., Metcalfe, C.D. (Eds.), Multifunctional Wetlands: Pollution Abatement and Other Ecological Services from Natural and Constructed Wetlands. Springer International Publishing Ag, Cham, pp. 121–140. Xu, N., Morgan, B., Rate, A.W., 2018. From source to sink: rare-earth elements trace the legacy of sulfuric dredge spoils on estuarine sediments. Sci. Total Environ. 637–638, 1537–1549.
7
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: http://www.elsevier.com/locate/jenvman
Short communication
A preliminary survey of anthropogenic gadolinium in water and sediment of a constructed wetland Anthony J. Altomare, Nicholas A. Young, Melanie J. Beazley * Department of Chemistry, University of Central Florida, Orlando, 4111 Libra Drive, Room 255, Florida, 32816, USA
A R T I C L E I N F O
A B S T R A C T
Keywords: Gadolinium Emerging contaminants Constructed wetland Rare earth elements Environmental contaminants
Gadolinium (Gd) is a rare earth element used in magnetic resonance imaging (MRI) contrast agents that has recently been identified as an emerging contaminant of concern due to its possible toxic effects and accumulation in the environment. The objectives of this preliminary study were to determine the occurrence and fate of Gd in surface water and sediment of a constructed wetland that receives effluent from a wastewater treatment plant. The rate of anthropogenic Gd entering the wetland was determined to be approximately 25 g Gd day 1, with surface water concentrations in the parts per trillion. Anthropogenic Gd concentrations in surface waters decreased as a function of distance from the inlet site to near the outfall, and were three orders of magnitude higher in sediment than in surface water suggesting that the wetland was providing a sink for Gd possibly through plant uptake and incorporation in organic biomass. An anthropogenic Gd anomaly was observed with an average GdAnt/GdGeo ratio of 5.34. Sediment with higher total organic carbon (TOC) tended to be higher in anthropogenic Gd, suggesting that Gd sequestration may occur through uptake by plants and/or through floc culation and deposition of natural organic matter.
1. Introduction Gadolinium (Gd) is a rare earth element (REE) commonly used in magnetic resonance imaging (MRI) contrast agents. Gadolinium-based contrast agents (GBCA) consist of organic ligands such as Gd-DTPA, Gd-DOTA, and Gd-DO3A-butrol complexed with Gd that protect the body from the toxic effects of free Gd3þ ion (Pagano et al., 2015). GBCAs are administered to patients prior to MRI imaging but are not metabo lized in humans. Instead, the GBCA complex is removed by the renal and hepatic systems and excreted in urine and feces (Le Fur and Caravan, 2019). GBCAs have recently been identified as emerging environmental contaminants of concern due to their presence in medical waste and sewage and the potential toxicity of free dissolved Gd (Rogowska et al., 2018). Although GBCAs are generally regarded as safe for medical use, there are a number of health effects related to Gd exposure (Le Fur and Caravan, 2019). GBCAs are known to accumulate in bone and brain tissues of patients at concentrations in the ng kg 1 to μg kg 1 range (Murata et al., 2016), with greater accumulation possible in patients with renal conditions and renal failure. The Gd3þ ion is widely consid ered to be more toxic than GBCA complexes and can interfere with
biological processes that utilize Ca2þ channels (Caravan et al., 1999). Free Gd has been shown to cause inhibition of Vibrio fischeri (Kurvet et al., 2017) at as low as 6.4 mg L 1 as compared to GBCAs, which have shown no-observed-effect concentrations (NOEC) as high as 937 mg L 1 in aquatic organisms (Neubert et al., 2008). Both GBCAs and free Gd have been shown to accumulate in bivalves (Perrat et al., 2017) and other terrestrial and aquatic organisms (Lingott et al., 2016). Currently, the chemical breakdown of GBCAs, both in municipal wastewater treatment systems and the environment, is not well under stood. Several studies have shown that dissolved Gd free ions and GBCA complexes are not efficiently removed in wastewater treatment plant (WWTP) processes (Kunnemeyer et al., 2009; Rabiet et al., 2009). As a result, there is potential for considerable release of Gd into the envi ronment. Studies performed in South Korea (Song et al., 2017), Prague, €ller et al., 2002), and Brazil (Pedreira et al., 2018) Czech Republic (Mo have shown Gd concentrations in WWTP effluents ranging from 39 to 410 ng L 1. Though the toxicity of GBCA degradation byproducts is not known, nor do researchers fully understand the effects of the anthro pogenic release of Gd into the environment, occurrence of Gd has been used previously to track wastewater release in the environment (Goulle et al., 2011; Pedreira et al., 2018). Since the behavior of Gd-containing
* Corresponding author. 4111 Libra Dr, Room 255, Orlando, FL, 32816, USA. E-mail addresses: [email protected] (A.J. Altomare), [email protected] (N.A. Young), [email protected] (M.J. Beazley). https://doi.org/10.1016/j.jenvman.2019.109897 Received 27 July 2019; Received in revised form 13 November 2019; Accepted 18 November 2019 Available online 26 November 2019 0301-4797/© 2019 Elsevier Ltd. All rights reserved.
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
Fig. 1. Map of OEW showing cell numbers and the three major flow trains, North (blue), Central (yellow), and South (red). Sample sites for the current study are indicated with a star. Site locations include Cell 1 (28� 340 12.500 N 81� 000 45.200 W), Cell 14 (28� 340 39.000 N 80� 590 52.100 W), Cell 16B (28� 340 26.700 N 80� 590 24.800 W), and Cell 17 (28� 340 53.200 N 80� 590 24.100 W). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
materials in nature is largely unknown and varies depending on the system, it is important to understand the fate and transport mechanisms of anthropogenic Gd in the environment. Several studies have shown that hospital and WWTP effluents are major contributors to increases in anthropogenic Gd in the environment (Bau et al., 2006; Hatje et al., 2016; Luo et al., 2014). Anthropogenic Gd has primarily been found in fresh and coastal waters that receive WWTP effluents, where the levels of anthropogenic Gd in WWTP effluent were indirectly tied with higher hospital patient volume. These studies give further evidence that the major source of Gd released into the envi ronment originates from human wastewater containing GBCAs. These previous studies have primarily focused on Gd in surface waters with limited studies on Gd in sediments (da Silva et al., 2018; Rogowska et al., 2018). In the current study, the occurrence and fate of anthropogenic Gd in a constructed wetland that receives WWTP effluent is examined. As WWTP effluent is the primary source of surface water for this wetland, it allows for the analysis of anthropogenic Gd in a more controlled envi ronmental system as compared to large bodies of water such as rivers or coastal waters. The Orlando Easterly Wetlands (OEW) located east of Orlando, Florida is one of the largest constructed wetland treatment systems in Florida that receives wastewater effluent from the City of Orlando’s Iron Bridge Regional Water Reclamation Facility and has been in operation since 1987 (Wang et al., 2006). The purpose of this wetland is to polish nutrients from the WWTP effluent stream before it is discharged into the nearby St. John’s River (White et al., 2018). The average loss of total phosphorus load from the inflow to the outfall of the wetlands is 70.54% with flocculation and settling of organic material as the hypothesized sequestration mechanism. The rate of effluent delivery from the WWTP to the OEW is approximately 45,000 m3 per day at the inlet (White et al., 2018). From there, it flows through a series of cells, which are individual bodies of water separated by sand berms. The water follows three flow paths (Fig. 1) with residence times of 18.1, 37.6, and 54.6 days for the northern, central, and southern flow paths, respectively (Wang et al.,
2006). Since 2001, the OEW has conducted a wetlands rejuvenation project to increase the hydraulic efficiency and nutrient removal effectiveness of the wetland. Water within selected cells in the flow paths is drained and the top layers of sediment excavated and removed. New vegetation is planted, and the cells returned to service. Overall phosphorus removal and hydraulic efficiency has increased as a result of the rejuvenation project (Wang et al., 2006). The Iron Bridge WWTP services portions of the City of Orlando, as well as other portions of northern Orange and southern Seminole Counties. This area includes six hospitals and several smaller medical facilities that may serve as anthropogenic sources of Gd. As a result, Iron Bridge, and by extension OEW, may receive considerable amounts of anthropogenic Gd. As OEW receives water primarily from the Iron Bridge WWTP, it offers a unique natural system in which to study the behavior of anthropogenic Gd without the dilution effects of other sur face waters. Therefore, the objectives of this study were to determine the occurrence and fate of Gd in the water and sediment of the OEW con structed wetland. 2. Materials and methods 2.1. Sample site description The OEW is a constructed wetland located in Christmas, Florida 20 km east of the City of Orlando. The wetland was constructed in 1987 and receives wastewater effluent from the City of Orlando’s Iron Bridge Regional Water Reclamation Facility (Wang et al., 2006). Surface water and sediment cores were collected at the OEW in Cell 1 (28� 340 12.500 N 81� 000 45.200 W), Cell 14 (28� 340 39.000 N 80� 590 52.100 W), Cell 16B (28� 340 26.700 N 80� 590 24.800 W), and Cell 17 (28� 340 53.200 N 80� 590 24.100 W) (Fig. 1). These sample sites were selected for this pre liminary study as they represented a typical flow of water through the wetland and were also sites of previous studies conducted at the OEW. 2
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
Fig. 2. Log REE values obtained from PAAS-normalized concentrations in sediment. The anomaly of Gd suggests the presence of anthropogenic Gd relative to geologic Gd. Points where no value is shown indicate the REE was below the limit of detection. Sample legend indicates ‘Cell number-core replicate numbersediment depth’.
Water depths at Cells 1, 16B, and 17 were between 1 and 2 m at the sample site, while Cell 14 contained no surface water. The water column at these sites had an average pH of 6.90, with orthophosphate concen trations between 2.1 and 3.2 ppm, and reducing conditions and low dissolved oxygen levels except near vegetation. Sediment composition in Cells 14, 16B, and 17 was rich in organic matter near the surface and transitioned to sand near the bottom. Cell 1 had limited vegetative growth, due to high water flow rates, and thus was primarily sand. Sediment samples that were rich in organic matter on average were observed to have higher Fe and Al concentrations at 2.4 � 0.7 and 8.4 � 1.8 g kg 1, respectively, compared to sandy sediment samples that contained 0.34 � 0.29 and 0.97 � 1.2 g kg 1, respectively. Cell 1 re ceives inflow directly from the Iron Bridge WWTP. Cells 17 and 16b are located along the Northern and Central flow paths, respectively, in close proximity to the outfall site. Cell 14 is located along the Central flow path. The wetlands rejuvenation project affected two cells that were sampled in this study. Cell 1 and 17 were excavated and replanted in 2001 and 2007, respectively, and at the time of sampling Cell 14 had been drained of surface water and was drying prior to excavation. Cells 14 and 16b had not previously been excavated at the location where samples were taken and the sediment at these sites was a result of ver tical accretion since 1987.
(v/v) trace metal grade nitric acid (Fisher Scientific) prepared with ul trapure water (Elga Veolia). REE (La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Er, Tm, Yb, and Lu) concentrations were determined using a Thermo Fisher Scientific iCAP-Qc inductively coupled plasma mass spectrometer (ICPMS) with QCell technology and operated in kinetic energy discrimina tion (KED) mode of analysis with helium as the collision gas. Calibra tion, internal, and quality control standards (Inorganic Ventures) were prepared in 2% (v/v) HNO3 and calibration standards were prepared at concentrations of 10–100 ng L 1 (ppt). All REEs were measured with the exception of Pm, Tb, and Ho, which were either present as internal standards or not present in the analytical standard. Average concen trations were calculated as an average of three analytical runs for each sample (<5% RSD). The analytical limit of detection for this method was <2 ng L 1 (ppt) for Gd based on the standard deviation of five method blanks. 2.4. Sediment analysis Sediment core sections were analyzed for total acid extractable REE metals. Total acid extractable REE concentrations were determined by digesting ashed sediment in concentrated acid. Dry sediment (0.2–0.5 g) was combusted in a Thermo Fisher Lindberg Blue M furnace at 600 � C for 6 h, digested in 1:4 concentrated hydrochloric acid (trace metal grade; Fisher Scientific) to concentrated HNO3 at 80 � C in a water bath for 12 h, and the supernatant analyzed by ICP-MS. Method blank sam ples (without sediment) were processed the same as field samples to determine any contamination during sample preparation. REE concen trations were normalized to Post-Archean Australian Shale (PAAS) (McLennan, 1989) and geological Gd (GdGeo) was interpolated from the third order fit of the logarithm of the normalized REE elements as €ller et al. (2002). Anthropogenic Gd (GdAnt) was described by Mo determined as the difference between the measured Gd (GdMeasured) and GdGeo.
2.2. Sample collection Surface water and sediment cores were collected in October 2018. Samples were collected in acid-cleaned polypropylene collection bottles and core liners (7.5 cm dia.) and stored on ice after collection for transport to the lab. Water samples were collected near the water surface at each site (depth < 1 m). All sediment cores were hand-cored and capped at each site with sediment depth ranging from 10 to 14 cm. Duplicate sediment cores were extruded (on day of collection) in 1 cm sections for the first 6 cm and in 2 cm increments after. Each section was then centrifuged at 3750 RPM and the porewater collected as the supernatant.
2.5. Total organic carbon analysis
2.3. Water and porewater analysis
Total organic carbon (TOC) analysis was performed using a Teledyne Tekmar Lotix with Lotix Solid Sampler boat for solid samples. Samples were analyzed in triplicate using 2–50 mg dry sediment. Sediment was homogenized using a mortar and pestle and acidified with 22%
Surface waters and porewaters were filtered (0.2 μm pore size polypropylene membrane syringe filters; VWR) and diluted 1:1 with 4% 3
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
by interpolating from the third-order trend of the normalized Log(REE) values. The difference between GdGeo and Gdmeasured was used to calculate GdAnt (Fig. 2). 3.1. Dissolved Gd Gd concentrations in wetland water decreased from 555 ng L 1 at the inlet in Cell 1 to 422 ng L 1 near the outfall in Cell 17 (Fig. 3). Nearoutlet concentrations of Gd were comparable to a previous study where Gd concentrations of 410 ng L 1 were determined at a WWTP outfall along the Atlantic Ocean on the Brazilian coast (Pedreira et al., 2018), demonstrating that although WWTP systems are capable of precipitating Gd, anthropogenic Gd is still released into the environ ment. The speciation of Gd may play a role in its fate depending on whether it is in the chelated form or the ionic Gd3þ form. Previous studies have shown that chelated REEs tend to be less reactive and, therefore, more soluble than their ionic forms (Khan et al., 2017). Using the inlet value of ~0.56 μg L 1 Gd entering OEW and assuming that is a typical value throughout a given year, at approximately 45,000 m3 per day of effluent entering OEW that equates to a flux of approximately 25 g Gd per day. The presence of a distinct GdAnt anomaly may suggest that the wetland is a sink for Gd, most probably through adsorption, precipita tion, and/or complexation reactions. Previous studies on aqueous speciation have shown that the Gd3þ ion forms complexes with car bonate, as well as phosphate in systems where orthophosphate is present at high concentrations (Verplanck et al., 2005). In addition, precipita tion of REEs with phosphate has been identified as a possible mechanism for inorganic sedimentation of Gd (Byrne et al., 1996). The concentra tions of phosphorus entering the OEW may provide opportunity for Gd–P precipitation and/or complexation reactions. Total phosphorus concentrations in the surface water were measured (by ICP-MS) between 165 and 417 μg L 1; however, there was no correlation observed be tween dissolved P and dissolved Gd concentrations. While a previous study has attributed REE anomalies in estuarine sediment dredge spoils to phosphate fertilizers due to a positive correlation between REEs and phosphate (Xu et al., 2018), no such correlation was determined in OEW water and sediment. Typically, REEs in sediment are found in many forms, including the water-soluble, exchangeable, carbonate-bound, organic, and Fe–Mn bound fractions of extracted sediment (Wali et al.,
Fig. 3. Gd concentrations in water and sediment porewater. Surface water concentrations are shown at a depth of ( 1 cm) where 0 cm corresponds to the sediment-water interface. All values of depth �0 cm represent sediment pore waters. Error bars represent standard deviations between duplicate sediment cores. Missing data points represent sediment sections without enough pore water for collection.
phosphoric acid (Fisher Scientific) to remove inorganic carbon and combusted at a furnace and catalyst temperature of 800 � C. Organic carbon standards (Ricca Chemical) were prepared at concentrations of 100–2500 μg organic carbon and verified using a 1200 μg check standard. 3. Results and discussion Gd is a REE found naturally in the environment, and to distinguish naturally occurring REE from anthropogenic REE, the data was normalized to REE concentrations in PAAS. Gd was the only REE detected in OEW water and sediment porewater, so it was assumed that all measured dissolved Gd was anthropogenic. In the acid-extracted sediment samples, Gd had higher concentrations relative to the other REE, indicating an anthropogenic Gd anomaly. GdGeo was determined
Fig. 4. Anthropogenic Gd concentrations in duplicate sediment cores. Anthropogenic Gd was calculated as the difference between Geological Gd and Measured Gd. Error bars represent standard deviations between duplicate sediment cores. Sample legend indicates ‘Cell number-core replicate number’. 4
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
therefore, the buildup of sediment in the other cells was a result of vertical accretion since 1987 when the OEW was constructed. The depth of sediment accretion (below sediment-water interface) ranged from 4 cm at Cells 16B and 17 to 12 cm at Cell 14. Each of the sediment cores were sectioned down to about 14 cm and the porewaters removed. Sediment at Cell 1 (inlet) was primarily sandy with little organic matter and minimal porewater. Dissolved Gd tended to decrease below the sediment-water interface (Fig. 3) suggesting the loss of dissolved Gd into the sediments. Due to compaction, sediments were less saturated in portions of some of the cores and porewater was not available to collect as shown in Fig. 3 by missing data points. At the time of sample collection, Cell 14 was drained of surface water and was in the process of dry down in preparation for excavation. Dissolved Gd in porewaters below 6 cm at this cell was less than porewaters at the other sites sug gesting the dry down procedure may have affected sediment uptake. Concentrations of Gd as determined by ICP-MS of the acid digested sediment samples were normalized to PAAS values (Fig. 2). GdAnt was present in ppb (ug/kg) concentrations at each site (Fig. 4). It should also be noted that in some samples, GdGeo was observed to be less than 10 μg/ kg, resulting in GdAnt comprising a majority of the measured Gd. In samples where GdGeo was greater than 10 μg/kg, the average ratio of GdAnt to GdGeo was 5.34. In general, GdAnt decreased at greater sediment depths with the exception of samples taken from Cell 1. These trends coincided with changes in sediment composition, with samples con taining high amounts of sandy material containing lower amounts of GdAnt. In contrast, samples with a darker muck-like consistency, along with greater amounts of TOC, had higher concentrations of GdAnt. It was also observed that Cells 14 and 16b shared a similar trend with a decrease in GdAnt at 2 cm and a subsequent increase at 3–4 cm, which possibly indicates temporal changes in either Gd introduction or sedi mentation rates. Cell 14 displayed the highest overall average GdAnt concentrations. Cell 1 was primarily sand, while Cells 14, 16b, and 17 all
Fig. 5. Total organic carbon content in wetland sediment. Error bars represent standard deviations between duplicate sediment cores. Sample legend indicates ‘Cell number-core replicate number’.
2014; Wang et al., 2001). 3.2. Sediment Gd The OEW was constructed in 1987 and accretion, or biomass accu mulation, was measured by the depth of sediment in the collected cores above the compacted sand layer. Cells 1 and 17 were the only cells in the OEW to have a previous excavation (2001 and 2007, respectively);
Fig. 6. Correlations between Anthropogenic Gd and total organic carbon for each cell in the wetland. Correlation strength is displayed as adjusted (r2) values. The slope of the correlation is represented as (m). 5
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
contained several centimeters of organic-rich sediment before tran sitioning to the cell bed at greater depths. TOC data was collected for sediment samples from each of the four cells studied (Fig. 5). TOC trends were similar to those of GdAnt, with the lowest concentrations in Cell 1 that decreased at lower depths in the sediment in the other three cells. In addition, the trend of Gd decreasing at a depth of 2 cm was conserved for TOC concentrations in samples from Cells 14, 16b, and 17. This data also demonstrates the transition from organic-rich sediment to sand at the cell bed, which occurred at 12, 10, and 4–8 cm for Cells 14, 16b, and 17, respectively. These transitions correspond to decreases in GdAnt concentrations giving additional evi dence of the location of the original sand cell bed. The correlation between TOC and GdAnt is shown in Fig. 6 for each cell. TOC and GdAnt show conservative positive correlations in Cells 16b and 17, suggesting that Gd sedimentation may be linked to the deposi tion of natural organic matter (NOM) in this system. Cell 14, which was in the process of dry down, demonstrated only a moderate correlation of Gd with TOC. The evaporation of water in this cell would have removed Gd from the dissolved phase and onto sediment particles possibly dis rupting the TOC-Gd correlation. The slope of the relationship between GdAnt and TOC varied between cells that could be caused by a number of factors such as differences in flow and settling rates, vegetative species and density, dissolved Gd concentrations, and other reactions that may result in Gd precipitation. If Gd sedimentation occurs primarily through flocculation or complexation of Gd with NOM, this may explain the relatively low concentrations of Gd in Cell 1 in comparison to the other cells in this study. Cell 1 had very little vegetation, and thus a low flux of NOM compared to the other cells. Conversely, Cells 16b and 17 had large amounts of vegetation, which explains why the top layers of sediment at these sites contained up to 30–50% organic carbon by mass. An additional pathway for Gd into sediments and correlation with TOC is through plants. Plants have previously been shown to uptake chelated Gd through their root systems that eventually becomes distributed throughout the plant from the roots to the leaves (Lingott et al., 2016). The speciation of Gd in the environment whether as the free Gd3þ ion or in the chelated form such as a GBCA will have an important impact on its behavior in water and sediment. As a chelate, Gd is more soluble and less reactive with particle surfaces and, therefore, moves through a water system with less retention. However, as the charged ion, Gd is more reactive and available for precipitation, such as with phosphate, or adsorption onto sediment particles, and more prone to sequestration. More information is needed to determine the effects of WWTP and environmental processes on the chelated complexes of Gd.
D. Sees of the Orlando Easterly Wetlands for providing access to the wetlands, assistance with sample collection, helpful discussions on the wetlands, and providing diagrams and maps. We also thank the two anonymous reviewers for their valuable comments to improve the manuscript. References Bau, M., Knappe, A., Dulski, P., 2006. Anthropogenic gadolinium as a micropollutant in river waters in Pennsylvania, and in Lake Erie, northeastern United States. Chem. Erde-Geochem. 66, 143–152. Byrne, R.H., Liu, X.W., Schijf, J., 1996. The influence of phosphate coprecipitation on rare earth distributions in natural waters. Geochim. Cosmochim. Acta 60, 3341–3346. Caravan, P., Ellison, J.J., McMurry, T.J., Lauffer, R.B., 1999. Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem. Rev. 99, 2293–2352. da Silva, Y., do Nascimento, C.W.A., da Silva, Y., Amorim, F.F., Cantalice, J.R.B., Singh, V.P., Collins, A.L., 2018. Bed and suspended sediment-associated rare earth element concentrations and fluxes in a polluted Brazilian river system. Environ. Sci. Pollut. Res. 25, 34426–34437. Goulle, J., Lacroix, C., Saussereau, E., Mahieu, L., Cellier, D., Spiroux, J., Guerbet, M., 2011. Gadolinium and platinum in wastewater and sludge as tracers of medical activities. Toxicol. Lett. 205. S125–S125. Hatje, V., Bruland, K.W., Flegal, A.R., 2016. Increases in anthropogenic gadolinium anomalies and rare earth element concentrations in San Francisco Bay over a 20 year record. Environ. Sci. Technol. 50, 4159–4168. Khan, A.M., Abu Bakar, N.K., Abu Bakar, A.F., Ashraf, M.A., 2017. Chemical speciation and bioavailability of rare earth elements (REEs) in the ecosystem: a review. Environ. Sci. Pollut. Res. 24, 22764–22789. Kunnemeyer, J., Terborg, L., Meermann, B., Brauckmann, C., Moller, I., Scheffer, A., Karst, U., 2009. Speciation analysis of gadolinium chelates in hospital effluents and wastewater treatment plant sewage by a novel HILIC/ICP-MS method. Environ. Sci. Technol. 43, 2884–2890. Kurvet, I., Juganson, K., Vija, H., Sihtm€ ae, M., Blinova, I., Syvertsen-Wiig, G., Kahru, A., 2017. Toxicity of nine (doped) rare earth metal oxides and respective individual metals to aquatic microorganisms Vibrio fischeri and Tetrahymena thermophila. Materials 10, 754. Le Fur, M., Caravan, P., 2019. The biological fate of gadolinium-based MRI contrast agents: a call to action for bioinorganic chemists. Metallomics 11, 240–254. Lingott, J., Lindner, U., Telgmann, L., Esteban-Fernandez, D., Jakubowski, N., Panne, U., 2016. Gadolinium-uptake by aquatic and terrestrial organisms-distribution determined by laser ablation inductively coupled plasma mass spectrometry. Environ. Sci.: Processes Impacts 18, 200–207. Luo, Y.L., Guo, W.S., Ngo, H.H., Nghiem, L.D., Hai, F.I., Zhang, J., Liang, S., Wang, X.C. C., 2014. A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment. Sci. Total Environ. 473, 619–641. McLennan, S.M., 1989. Rare-earth Elements in Sedimentary Rocks - Influence of Provenance and Sedimentary Processes. Mineralogical Society of America, Special Paper, Washington. M€ oller, P., Paces, T., Dulski, P., Morteani, G., 2002. Anthropogenic Gd in surface water, drainage system, and the water supply of the city of Prague, Czech Republic. Environ. Sci. Technol. 36, 2387–2394. Murata, N., Gonzalez-Cuyar, L.F., Murata, K., Fligner, C., Dills, R., Hippe, D., Maravilla, K.R., 2016. Macrocyclic and other non-Group 1 gadolinium contrast agents deposit low levels of gadolinium in brain and bone tissue: preliminary results from 9 patients with normal renal function. Invest. Radiol. 51 (7), 447–453. Neubert, C., Laenge, R., Steger-Hartmann, T., 2008. Gadolinium containing contrast agents for magnetic resonance imaging (MRI) - Investigations on the environmental fate and effects. In: Aga, D.S. (Ed.), Fate of Pharmaceuticals in the Environment and in Water Treatment Systems, pp. 101–120. Pagano, G., Guida, M., Tommasi, F., Oral, R., 2015. Health effects and toxicity mechanisms of rare earth elements-Knowledge gaps and research prospects. Ecotox. Environ. Safe. 115, 40–48. Pedreira, R.M.A., Pahnke, K., Boning, P., Hatje, V., 2018. Tracking hospital effluentderived gadolinium in Atlantic coastal waters off Brazil. Water Res. 145, 62–72. Perrat, E., Parant, M., Py, J.S., Rosin, C., Cossu-Leguille, C., 2017. Bioaccumulation of gadolinium in freshwater bivalves. Environ. Sci. Pollut. Res. 24, 12405–12415. Rabiet, M., Brissaud, F., Seidel, J.L., Pistre, S., Elbaz-Poulichet, F., 2009. Positive gadolinium anomalies in wastewater treatment plant effluents and aquatic environment in the Herault watershed (South France). Chemosphere 75, 1057–1064. Rogowska, J., Olkowska, E., Ratajczyk, W., Wolska, L., 2018. Gadolinium as a new emerging contaminant of aquatic environments. Environ. Toxicol. Chem. 37, 1523–1534. Song, H., Shin, W.J., Ryu, J.S., Shin, H.S., Chung, H., Lee, K.S., 2017. Anthropogenic rare earth elements and their spatial distributions in the Han River, South Korea. Chemosphere 172, 155–165. Verplanck, P.L., Taylor, H.E., Nordstrom, D.K., Barber, L.B., 2005. Aqueous stability of gadolinium in surface waters receiving sewage treatment plant effluent, boulder creek, Colorado. Environ. Sci. Technol. 39, 6923–6929. Wali, A., Colinet, G., Ksibi, M., 2014. Speciation of heavy metals by modified BCR sequential extraction in soils contaminated by phosphosypsum in Sfax, Tunisia. Environ. Res. Eng. Manag. 4, 14–26.
4. Conclusions The results of this preliminary study describe the occurrence and fate of anthropogenic Gd in a constructed wetland that receives effluent primarily from a WWTP. Concentrations of dissolved Gd in surface waters decreased between the inlet of the wetland and the outfall, suggesting the wetlands may act as a sink for Gd and is supported by an average GdAnt/GdGeo anomaly of 5.34 in the sediment. Concentrations of Gd were found to have a moderate correlation to TOC in some cells, suggesting the mechanism of Gd sequestration in the wetland may be through complexation with organic biomass or incorporation into biomass via plant uptake. However, these results are based on only one set of sampling timepoints. Further work is needed at this site with more extensive sampling throughout the OEW and over multiple timepoints in order to provide a better understanding of the role the wetlands play in Gd fate and transport. Acknowledgements This research was funded by Beazley Startup Funds from the Uni versity of Central Florida. The authors wish to thank Dr. Emily Heider for assistance with sample collection. The authors especially thank Mark 6
A.J. Altomare et al.
Journal of Environmental Management 255 (2020) 109897
Wang, H., Jawitz, J.W., White, J.R., Martinez, C.J., Sees, M.D., 2006. Rejuvenating the largest municipal treatment wetland in Florida. Ecol. Eng. 26, 132–146. Wang, Z.W., Shan, X.Q., Zhang, S.Z., 2001. Comparison of speciation and bioavailability of rare earth elements between wet rhizosphere soil and air-dried bulk soil. Anal. Chim. Acta 441, 147–156. White, J.R., Sees, M., Jerauld, M., 2018. The long-term use of treatment wetlands for total phosphorus removal: can performance Be rejuvenated with adaptive
management? In: Nagabhatla, N., Metcalfe, C.D. (Eds.), Multifunctional Wetlands: Pollution Abatement and Other Ecological Services from Natural and Constructed Wetlands. Springer International Publishing Ag, Cham, pp. 121–140. Xu, N., Morgan, B., Rate, A.W., 2018. From source to sink: rare-earth elements trace the legacy of sulfuric dredge spoils on estuarine sediments. Sci. Total Environ. 637–638, 1537–1549.
7