Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

Contents lists available at ScienceDirect

Journal of Trace Elements in Medicine and Biology

journal homepage: www.elsevier.com/locate/jtemb

Lanthanum and Gadolinium availability in aquatic mediums: New insights

to ecotoxicology and environmental studies
Cátia Figueiredo a, b, c, *, Tiago F. Grilo a, Clara Lopes b, Pedro Brito d, Miguel Caetano b, d,
Joana Raimundo b, d
a
MARE – Marine and Environmental Sciences Centre, Faculdade de Ciências da Universidade de Lisboa, Campo Grande 1749-016, Lisboa, Portugal
b
Division of Oceanography and Marine Environment, IPMA – Portuguese Institute for Sea and Atmosphere, Av. Alfredo Magalhães Ramalho, 6, 1495-165 Algés,
Portugal
c
UCIBIO, REQUIMTE, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa, 2829-516 Caparica, Portugal
d
CIIMAR – Interdisciplinary Centre of Marine and Environmental Research, Avenida General Norton de Matos S/N, 4450-208 Matosinhos, Portugal

A R T I C L E I N F O A B S T R A C T

Keywords: Studies dealing with Rare Earth Elements (REE) ecotoxicological behavior are scattered and with potential
Rare earth elements conflicting results. Climate change impacts on aquatic biota and is known to modify contaminants toxicokinetic.
Salinity Nevertheless, the current knowledge on the potential interactions between climate change and REE is virtually
Temperature
non-existent. Therefore, we focus our research on La and Gd as representatives of Light and Heavy REE that also
pH
Speciation
are of great environmental concern. Experiments on different mediums (fresh-, brackish- and seawater) were
designed to run at present-day and near-future conditions (T◦ =+4 ◦ C, pH=△-0.4). Sampling was taken at
different time scales from minutes to hours for one day. The main challenge was to evaluate the availability of La
and Gd under environmental conditions closely related to climate changes scenarios. Furthermore, this study will
contribute to the baseline knowledge by which future research towards understanding REE patterns and toxicity
will build upon. Lanthanum and Gd behave differently with salinity. Temperature also affects the availability of
dissolved La in freshwater. On the other hand, pH reduction causes the decrease of Gd in freshwater. In this
medium, concentrations reduce sharply, presumably due to sorption processes or precipitates. In the brackish
water experiment only the dissolved La levels in the Warming (T◦ =+4 ◦ C) and Warming & Acidification
(T◦ =+4 ◦ C, pH=△0.4) diminished significantly through time. Dissolved La and Gd levels in seawater were
relatively constant with time. The speciation of both elements is also of great relevance for ecotoxicological
experiments. The trivalent free ions (La3+ and Gd3+) were the most common species in the trials. However, as
ionic strength increases, the availability of other complexes rose, which should be subject of great attention for
upcoming ecotoxicological studies.

1. Introduction REE, La and Gd are used in numerous industrial processes and medical
treatments. Since the 1980’s Gd has been used as a contrasting agent in
Rare earth elements (REE) are a group of fifteen lanthanides plus magnetic resonance imaging (MRI). After being injected into the blood
scandium and yttrium, found in the Earth’s crust, that are irreplaceable stream, it is excreted through urine, reaching directly the aquatic
in the making of high-end modern technology. Increased usage of these environment practically unaltered by wastewater-treatment plants
chemical elements has raised concern about their role as environmental (WWTPs, Migaszewski and Gałuszka, 2016; [43]). Each MRI uses 1.2 g
emerging contaminants [34]. The ever-growing demand for modern of Gd, which accounts for 5% of the total Gd production [29]. Natural
electronic equipment, which production relies on REE, engenders environmental levels of Gd in river water range between 1 and 4 ng L-1
excessive amounts of e-waste and the REE recycling industry is [28,6], however, in sites with WWTPs effluents concentrations can
expanding, however, it is still at the development stage [22]. Among the range from 200 to 1100 mg L-1 [6]. Furthermore, with increasing

* Corresponding author at: MARE – Marine and Environmental Sciences Centre, Faculdade de Ciências da Universidade de Lisboa, Campo Grande 1749-016,
Lisboa, Portugal.
E-mail address: [email protected] (C. Figueiredo).

https://doi.org/10.1016/j.jtemb.2022.126957
Received 4 January 2022; Received in revised form 7 February 2022; Accepted 18 February 2022
Available online 23 February 2022
0946-672X/© 2022 The Authors. Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

production and demand, it is expected that both elements’ availabilities set to mimic climate change scenarios taking the IPCC end-century
in the environment will augment. This highlights the urgent need for a predictions [17]: temperature increase of up to 4 ◦ C and a pH
better understanding of their biogeochemical behavior. Although the decrease of up to 0.4 units. All the used material was previously
number of studies dealing with their ecotoxicological behavior has been decontaminated with HNO3 (20% v/v) and washed with ultra-pure
increasing, no hot spot species have been used, and the endpoints are water (18.2 MΩ.cm). Each beaker was filled with water (spring,
scattered and with potential conflicting results [18,34]. Therefore, it is brackish or seawater) and continuously aerated water and randomly
key to run more studies on different environmental levels and organ­ distributed in two temperature-controlled water baths. The environ­
isms. Additionally, the increased human footprint since the industrial mental conditions were altered to: i) Initial condition (Control T◦ C and
revolution has resulted in another great environmental concern: climate pH); ii) Acidification (pH=△0.4); iii) Warming (T◦ =+4 ◦ C) and iv)
change. Since then, human activities have continuously increased the Warming & Acidification (T◦ =+4 ◦ C, pH=△0.4). The experiment
emission of greenhouse gases that in turn trap solar energy contributing comprised twelve treatments as follow: i) Initial Cond; ii) Initial Cond +
to the well-known Greenhouse Effect [1]. Accordingly to the Intergov­ La; iii) Initial Cond + Gd; iv) Acidification; v) Acidification + La; vi)
ernmental Panel on Climate Change (IPCC), greenhouse gas emissions Acidification + Gd; vii) Warming; viii) Warming + La; ix) Warming +
are causing the warming of the planet and the oceanic uptake of CO2, Gd; x) Warming & Acidification; xi) Warming & Acidification + La and
which leads to pH dropping on a global scale [17]. These two phe­ xii) Warming & Acidification + Gd.
nomena are known as Ocean Warming and Acidification and, depending The temperature was kept stable in the two water baths (18 ◦ C ±
on the World’s region, can occur isolated or combined. Climate change 0.3 ◦ C for the control temperature treatments and 22 ◦ C ± 0.3 ◦ C for the
impacts aquatic biota, by altering their fitness, reproduction, recruit­ warming treatments) through submerged heaters (V2Therm 200 W,
ment, and distribution [5]. In the last decade, a growing body of evi­ TMC Iberia) and chillers (Hailea, HC-250 A). Water pH was automati­
dence became available on the broad potential negative impacts of cally regulated by the controlling system Profilux 3.1 T (GHL) connected
climate change on the toxicity of environmental contaminants (e.g., to pH probes (GHL) which monitored pH in each glass beaker every 2
[10]). It is also known that climate change can modify contaminants’ seconds. Upregulation was done with filtered air and down-regulated
toxicokinetics, enhancing their biological effects. Hence, the scientific via solenoid valves (Etopi) with CO2 injection. Hysteresis was kept at
community increasingly became aware of the need to investigate the ± 0.05 units of pH. Master parameters were measured with a portable
interaction between climate change and contaminants [36,44]. Never­ pH and temperature probe (VWR pHenomenal) and salinity with a
theless, the current knowledge on the potential interactions between refractometer (V2, TMC Iberia). Carbonate system speciation was
climate change and REE is virtually non-existent. The main challenge calculated from pH measurements and total alkalinity (Alkalinity
was to evaluate the availability of La and Gd under environmental checkers, Hanna® Instruments) using CO2SYS software [39]. Due to the
conditions closely related to climate changes scenarios. This study will natural pH fluctuations inherent in the types of water studied, the
also contribute to the baseline knowledge by which future research to­ freshwater pH was adjusted at 7.90 ± 0.10 for the Control pH treatments
wards understanding REE patterns and toxicity will build upon. Given and 7.52 ± 0.11 for the Acidification treatments. The brackish water pH
this context, we designed an experiment with Lanthanum (La) and was kept at 8.07 ± 0.05 for the Control pH treatments, and 7.70 ± 0.11
Gadolinium (Gd) levels (1.5 µg L-1 and 1 µg L-1, respectively) in fresh-, for the Acidification treatments and the seawater pH was maintained at
brackish- and saltwater. Concentrations were monitored during eight 8.29 ± 0.06 and 7.83 ± 0.11 for the Control pH and Acidification
sampling occasions over 24 hours in present-day and climate change treatments, respectively.
conditions (T◦ C=+4 ◦ C, pH=△0.4). The designed experiments also Water aliquots were sampled after 15 minutes (15’), 30 minutes
contribute to the discussion of methodology impairments in laboratory (30’), one hour (1 h), two hours (2 h), four hours (4 h), six hours (6 h),
experiments used to study the ecotoxicological effects of REE. twelve hours (12 h) and twenty-four hours (24 h) of the exposure.

2. Materials and methods 2.2. Lanthanum and Gadolinium quantification

2.1. Experimental setup Water aliquots (as triplicates) taken at each time were filtrated (0.45
µm, MF-Millipore ™, Merk) and acidified (2% Ultrapur® HNO3). Dis­
Although REE are known to behave coherently, their grouping is solved La and Gd levels were determined using a quadrupole ICP-MS
inconsistent among authors. Here, La and Gd were chosen as represen­ (Perkin-Elmer NexION 2000C), equipped with a cyclonic spray cham­
tatives of Light and Heavy rare earth elements [2], respectively. The ber, a concentric Meinhard nebulizer and a dual detector. The 139La and
158
dissolved concentration of both REEs vary in a broad range in the Gd were the quantified isotopes since they present minimum isobaric
aquatic environment from ng L-1 to hundreds of µg L-1 (e.g., [47]). and polyatomic interferences by setting the ratios 137Ba++/137Ba and
140 16
Therefore, it was optioned to select metal exposure concentrations in an Ce O/140Ce to 0.010 under routine operating conditions. Metals in
average value avoiding the extremes where negligible or excessive spring waters samples were measured by direct aspiration into the ICP-
changes may exist. The exposure experiment with La was performed MS, while brackish and seawater were diluted to salinity ~3 with Milli-
with 1.5 µg L-1 while the Gd exposure was 1 µg L-1 of Gd. Furthermore, Q water to minimize matrix interference.
the selected environmentally relevant exposure concentrations fall Quality Control (QC) solutions were run every 20 samples, and a 6-
within the exposure range of other experiments (e.g., [34]). Standard point calibration curve within a range of 1–1000 μg L− 1 was used for
solutions of LaCl3 and GdCl3 were used as a source of both metals. The quantification, using 115In as internal standard. Three reagent blanks
exposure experiments were conducted with freshwater (Sal=0), and random sample duplicates were included within each batch of 20
brackish water (Sal=15) and seawater (Sal=35) to reflect different samples to control the analytical quality of the method and accounted
aquatic environments, from rivers and lakes to estuaries and the open for less than 1% of the total La and Gd concentration in the samples. The
sea. Springwater collected at 41◦ 28’05.8"N and 8◦ 12’08.9"W (NE detection limits were 0.022, 0.024 and 0.034 µg L-1 for La and 0.006,
Portugal) was used for the freshwater experiment. For the seawater 0.016 and 0.010 µg L-1 for Gd in spring, brackish and seawater,
experiment, coastal water was collected in 38◦ 40’45.5"N and respectively.
9◦ 20’12.1"W (West Europe). Both freshwater and seawater were filtered
through 0.45 µm membranes before use. By mixing the seawater with 2.3. Statistical analyses
spring water, brackish water was obtained with salinity 15. The exper­
iments were performed in individual 1 L glass beakers. The different A one-way ANOVA approach was applied, for La and Gd levels, with
environmental conditions studied (temperature and acidification) were Salinity as the treatment factor to investigate differences between

2
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

salinities. This was followed by all pairwise comparisons between the brackish- and seawater (Tukey test, p = 0.53). On the other hand, Gd
predicted means of the salinity factor using Tukey’s procedure [9]. The concentrations presented significant differences between the three
analysis of variance two-way ANOVA was later used with temperature water types (Tukey test, p < 0.0001), with higher values in sea-, fol­
and pH as the treatment factors to identify differences in La and Gd lowed by brackish- and freshwater.
levels, between treatments, for each salinity, and all pairwise compari­
sons using Tukey post hoc tests were assessed. Lastly, to evaluate the 3.2. Temperature and pH
effect of time on La and Gd levels, a one-way ANOVA was applied for
each spiked treatment, followed by all Tukey’s pairwise comparisons. Temperature is a significant factor (p < 0.0001, Supplemental
Whenever necessary, data were Log transformed to conform with the Table 3) for the availability of dissolved La in freshwater, unlike pH
assumptions of normality (independence and homogeneity of vari­ alone (p = 0.12). All treatments showed differences amongst them,
ances). Statistical analyses were performed at a significance level of 0.05 except Control T◦ C and pH vs Acidification (p = 0.74). For dissolved Gd
using the open-source software InVivoStat, version 4.1. in freshwater, levels were significantly affected only by the interaction
Temperature * pH (p = 0.021). In fact, it was only when the Gd con­
3. Results centrations from the Control T◦ C and pH were compared with the
Acidification treatment that significant differences were found (p =
A summary of seawater physicochemical parameters and specifics of 0.011).
the carbonate system for all treatments and salinities (mean ± standard Temperature, pH, and their interaction did not significantly influ­
deviation) is presented in Table 1. Mean and standard deviation of ence dissolved La or Gd levels in brackish- and seawater (p > 0.05).
dissolved La and Gd levels (µg L-1) in freshwater (salinity 0), brackish
water (salinity 15), and seawater (salinity 35) sampled at 15’, 30’, 1 h, 2
3.3. Time course evolution of La and Gd
h, 4 h, 6 h, 12 h and 24 h are presented in Supplemental Table 1. The
ANOVA results are shown in Supplemental Table 2, 3 and 4 (A and B).
For salinity 0 treatments without La or Gd added, both elements’
concentrations were below detection limit (0.022 µg L-1 for La and
3.1. Salinity 0.006 µg L-1 for Gd). Conversely, the La and Gd levels in different
treatments varied significantly through time (p < 0.0001, Supplemental
Different salinities affected dissolved La and Gd levels (p < 0.0001, Table 4A). The uppermost concentration of La was registered in the
Supplemental Table 2). During the first 24 h, La concentrations found in Warming (1.3 ± 0.054 µg L-1), after 15 minutes of exposure, followed by
freshwater are significantly lower than in brackish- and seawater (Tukey Warming & Acidification (1.1 ± 0.039 µg L-1) treatment (Fig. 1 A,
test, p < 0.0001). However, the levels did not vary significantly between Supplemental Table 1).

Table 1
Summary of physicochemical parameters and specifics of carbonate system for all treatments and salinities (mean ± standard deviation). Measured temperature (ºC),
pH (full scale) and total alkalinity (TA, μmol kg-1 SW) were used to calculate carbonate system parameters: pCO2 (carbon dioxide partial pressure, μatm) and TC (total
inorganic carbon, μmol kg-1 SW).
Treatment Temperature (ºC) pH (total scale) TA (μmol kg-1 SW) TC (μmol kg-1 SW) pCO2 (μatm)

Sal 0 Control Present-day 18 ± 0.2 7.89 ± 0.10 520 323 230

Acidification 18 ± 0.2 7.54 ± 0.09 360 401 1016
Warming 22 ± 0.4 7.87 ± 0.11 580 477 276
Warming & Acidification 22 ± 0.2 7.52 ± 0.10 700 745 1346
La Present-day 18 ± 0.2 7.9 ± 0.08 700 305 303
Acidification 18 ± 0.2 7.53 ± 0.12 620 498 644
Warming 22 ± 0.3 7.91 ± 0.09 780 475 337
Warming & Acidification 22 ± 0.4 7.52 ± 0.12 600 650 656
Gd Present-day 18 ± 0.3 7.89 ± 0.13 380 384 243
Acidification 18 ± 0.4 7.49 ± 0.09 520 547 729
Warming 22 ± 0.4 7.91 ± 0.09 540 533 216
Warming & Acidification 22 ± 0.2 7.49 ± 0.12 680 846 992
Sal 15 Control Present-day 18 ± 0.3 8.12 ± 0.06 1760 1695 320
Acidification 18 ± 0.2 7.68 ± 0.12 1760 1936 1003
Warming 22 ± 0.3 7.98 ± 0.10 1760 1670 458
Warming & Acidification 22 ± 0.4 7.72 ± 0.13 1980 2038 1161
La spiked Present-day 18 ± 0.3 8.11 ± 0.04 1700 1632 317
Acidification 18 ± 0.3 7.67 ± 0.15 1840 1918 1024
Warming 22 ± 0.2 8.13 ± 0.05 1740 1655 310
Warming & Acidification 22 ± 0.4 7.73 ± 0.10 1920 2068 946
Gd spiked Present-day 18 ± 0.4 8.01 ± 0.04 1680 1635 405
Acidification 18 ± 0.3 7.69 ± 0.09 1860 1914 1091
Warming 22 ± 0.3 7.99 ± 0.03 1800 1718 461
Warming & Acidification 22 ± 0.3 7.70 ± 0.06 1980 1993 1046
Sal 35 Control Present-day 18 ± 0.4 8.22 ± 0.04 2700 2325 283
Acidification 18 ± 0.3 7.89 ± 0.10 2980 2711 773
Warming 22 ± 0.3 8.18 ± 0.05 2740 2314 318
Warming & Acidification 22 ± 0.3 7.88 ± 0.11 2960 2605 788
La spiked Present-day 18 ± 0.4 8.20 ± 0.07 2520 2248 278
Acidification 18 ± 0.3 7.86 ± 0.13 2620 2579 818
Warming 22 ± 0.2 8.19 ± 0.08 2600 2289 292
Warming & Acidification 22 ± 0.3 7.79 ± 0.09 2680 2648 904
Gd spiked Present-day 18 ± 0.3 8.17 ± 0.06 2580 2324 311
Acidification 18 ± 0.3 7.78 ± 0.10 2640 2740 907
Warming 22 ± 0.4 8.21 ± 0.08 2780 2454 296
Warming & Acidification 22 ± 0.3 7.77 ± 0.15 2700 2752 960

3
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

Fig. 1. Time course evolution of a) Lanthanum and b) Gadolinium levels (µg L-1) in spiked treatments with 1.5 µg L-1 and 1 µg L-1, respectively, for different sampling
times (15’, 30’, 1 h, 2 h, 4 h, 6 h, 12 h and 24 h) at salinity 0. Values represent means ± SD.

The present-day treatment registered 0.64 ± 0.031 µg L-1, and the percentage is 45% in the Control TºC and pH treatment (0.83
Acidification treatment 0.70 ± 0.011 µg L-1, after the same exposure ± 0.036 µg L-1 at 24 h), 43% in the Warming treatment (0.85
time. In the course of 24 hours, the levels of dissolved La at present-day ± 0.01 µg L-1 at 24 h), 27% in the Acidification treatment (1.1
conditions dropped 67% to 0.20 ± 0.010 µg L-1, being the treatment ± 0.025 µg L-1 at 24 h) and 59% in the Warming & Acidification treat­
that registered the largest decrease. The La concentrations in the Acid­ ment (0.62 ± 0.075 µg L-1 at 24 h). The time course evolution of La in
ification treatment fell overtime 57% to 0.30 ± 0.007 µg L-1, and in the Warming and Warming & Acidification conditions varied significantly
Warming & Acidification treatment dropped 56% to 0.44 ± 0.011 µg L- throughout the 24 h (p = 0.002 and p = 0.001, respectively).
1
. After 24 h, the La values in the Warming treatment listed 0.61 The baseline Gd levels in the salinity 15 experiments ranged between
± 0.011 µg L-1, having dropped 53% then the values registered at 15’. 0.02 ± 0.001 µg L-1 and 0.09 ± 0.007 µg L-1 (Supplemental Table 1). At
Similarly, Gd levels also registered a great percentage decrease overtime 15 minutes Gd levels registered 1.0 ± 0.043 µg L-1 in the Acidification
(Fig. 1B). The Warming & Acidification treatment registered the most treatment, 0.98 ± 0.052 µg L-1 in the present-day treatment, 0.95
significant Gd decrease (77%) from 0.56 ± 0.028 µg L-1 at 15’ to 0.13 ± 0.029 µg L-1 in the Warming treatment and 0.85 ± 0.017 µg L-1 in the
± 0.004 µg L-1 at 24 h, followed by the Acidification treatment (66%) Warming & Acidification treatment (Fig. 2B). During the 24 hours
from 0.38 ± 0.027 µg L-1 to 0.13 ± 0.006 µg L-1, over the same period. experiment the Gd in these treatments decreased 44%, 47%, 55% and
The highest Gd level in the Warming treatment registered 0.47 49%, to 0.56 ± 0.014 µg L-1, 0.52 ± 0.022 µg L-1, 0.43 ± 0.007 µg L-1
± 0.007 µg L-1 at 15’, and the lowest at 24 h (0.17 ± 0.007 µg L-1, 64% and 0.43 ± 0.021 µg L-1, respectively. However, these time variations
decrease). The Control TºC and pH treatment, at salinity 0, presented the have no significance at the statistical level (p > 0.05).
minor Gd decline over time (42%) ranging from 0.43 ± 0.033 µg L-1 at The baseline La levels in salinity 35 fluctuated between 0.04
15’ to 0.25 ± 0.005 µg L-1 at 24 h (Supplemental Table 1). ± 0.001 µg L-1 and 0.09 ± 0.006 µg L-1 (Supplemental Table 1).
For salinity 15, the baseline La levels in the control treatments varied The percentage of La also decrease over time as found in the other
between 0.03 ± 0.001 µg L-1 and 0.09 ± 0.007 µg L-1 (Supplemental salinities, but not progressively (Fig. 3A).
Table 1). After 15 minutes of exposure, the La concentrations were The Acidification treatment presented 1.5 ± 0.030 µg L-1 of dis­
higher than those observed in the salinity 0 trial (Fig. 2A) and similar to solved La at 15’ and 1.0 ± 0.030 at 24 h, representing a 29% decrease.
the four treatments. The Warming & Acidification treatment showed a 23% La reduction
Despite presenting globally higher La levels, the percentage of La from 1.4 ± 0.020 µg L-1 at 15’ to 1.1 ± 0.034 µg L-1 at 24 h. Lanthanum
decreased over time is comparable to that shown in salinity 0. The exposed to the Warming conditions peaked at 15’ (1.3 ± 0.064 µg L-1)

Fig. 2. Time course evolution of a) Lanthanum and b) Gadolinium levels (µg L-1) in spiked treatments with 1.5 µg L-1 and 1 µg L-1, respectively, for different sampling
times (15’, 30’, 1 h, 2 h, 4 h, 6 h, 12 h and 24 h) at salinity 15. Values represent means ± SD.

4
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

Fig. 3. a) Time course evolution of a) Lanthanum and b) gadolinium (µg L-1) levels in spiked with 1.5 µg L-1 and 1 µg L-1, treatments, respectively, in different
sampling times (15’, 30’, 1 h, 2 h, 4 h, 6 h, 12 h and 24 h) at salinity 35. Values represent means ± SD.

and decreased 15% on the course of 24 h (1.1 ± 0.040 µg L-1). The The second lowest Gd decrease occurred in the Warming treatment
lowest La percentage drop (14%) was exhibited by the Control TºC and (14%), with 1.1 ± 0.031 µg L-1 after 15 minutes and 0.95 ± 0.080 µg L-1
pH treatment (1.2 ± 0.050 µg L-1 at 15’ to 1.0 ± 0.025 µg L-1 at 24 h). after 24 hours. The Acidification and Warming & Acidification treat­
This up and down variability of La concentration in all treatments had ments registered a Gd decrease of 21%. The first sampled aliquots
no significance (p > 0.05). registered 1.1 ± 0.041 µg L-1 and 1.0 ± 0.048 µg L-1, and 24 h upon the
The Gd levels in the control treatments swung from 0.02 spike both treatments registered 0.87 ± 0.033 µg L-1 and 0.79
± 0.001 µg L-1 to 0.09 ± 0.010 µg L-1 (Supplemental Table 1). The Gd ± 0.053 µg L-1, respectively. The 24 h experiment did not showcase
level at the Control TºC and pH was the most stable of the experiment, significant differences in Gd levels in the studied treatments (p > 0.05).
having ranged from 1.1 ± 0.034 µg L-1 to 1.1 ± 0.057 µg L-1 (Fig. 3B).

Table 2
Percentage distribution among La and Gd species for La (1.5 µg L-1) and Gd (1 µg L-1) treatments for salinity 0, 15 and 35.
Treatment La3+ LaCl2+ LaOH2+

Sal 0 La Present day conditions 93.5 0.13 6.4

Acidification 97.2 0.13 2.7
Warming 91.9 0.13 8.0
Warming & 96.5 0.13 3.4
Acidification
Gd3þ GdCl2þ GdOH2þ
Gd Present day conditions 62.6 0.052 37.4
Acidification 80.7 0.067 19.2
Warming 55.5 0.046 44.5
Warming & 75.8 0.063 24.2
Acidification
La3þ LaOH2þ LaF2þ LaCl2þ La LaSO4þ LaHCO32þ La LaCO3þ
(SO4)2- (CO3)2-
Sal 15 La Present day conditions 48.0 1.4 0.63 0.69 3.8 41.5 0.063 0.021 3.8
Acidification 49.6 0.59 0.65 0.71 4.0 42.9 0.066 – 1.6
Warming 45.9 1.7 0.62 0.67 4.3 42.8 0.065 0.024 3.9
Warming & 47.5 0.72 0.64 0.69 4.5 44.3 0.069 – 1.7
Acidification
Gd3þ GdHCO32þ GdOH2þ GdF2þ GdF2þ GdCl2þ Gd(SO4)2- GdSO4þ Gd GdCO3þ
(CO3)2-
Gd Present day conditions 37.5 0.052 9.8 0.014 2.6 0.33 2.5 33.3 0.25 13.7
Acidification 43.7 0.061 4.6 0.017 3.0 0.39 2.9 38.8 0.048 6.4
Warming 34.8 0.052 12.1 0.012 2.4 0.31 2.6 33.7 0.28 13.7
Warming & 41.3 0.063 5.7 0.015 2.9 0.36 3.1 39.9 0.054 6.6
Acidification
La3þ LaOH2þ LaF2þ LaCl2þ La LaSO4þ LaHCO32þ La LaCO3þ
(SO4)2- (CO3)2-
Sal 35 La Present day conditions 43.0 1.2 0.73 8.6 6.4 34.3 0.077 0.072 5.64
Acidification 44.8 0.50 0.76 9.0 6.7 35.8 0.083 0.013 2.43
Warming 43.9 1.4 0.71 8.3 7.1 32.6 0.079 0.081 5.76
Warming & 45.9 0.60 0.75 8.7 7.4 34.0 0.086 0.014 2.50
Acidification
Gd3þ GdHCO32þ GdOH2þ GdF2þ GdF2þ GdCl2þ Gd(SO4)2- GdSO4þ Gd GdCO3þ
(CO3)2-
Gd Present day conditions 33.8 0.062 8.0 0.028 3.0 4.1 4.1 26.3 0.84 19.8
Acidification 40.8 0.077 3.8 0.033 3.6 4.9 4.9 31.8 0.17 9.9
Warming 34.2 0.062 9.9 0.024 2.8 3.7 4.3 24.4 0.91 19.7
Warming & 41.8 0.079 4.8 0.029 3.4 4.6 5.3 29.8 0.19 10.0
Acidification

5
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

3.4. Lanthanum and Gd speciation during the 24 h experiment. On other hand, the IPCC projection of 4 ºC
increase may be less extreme as the temperature variation exhibited in
Lanthanum and Gd speciation analysis were carried out in Visual different seasons, or extreme events, which suggest an increased
MINTEQ program (version 3.1), using the measured average pH, tem­ removal of La and Gd from the dissolved fraction. Sholkovitz [45]
perature, and La and Gd values for each treatment. The concentrations determined how pH, salinity and colloids affect REE fractionation in the
of major elements (i.e., Ca2+, Na+, Cl-, CO32-, Mg2+, S042-) from the river and estuarine waters and described that both fractionation and
spring water and seawater [41] was used as components for the speci­ concentration of dissolved REE in freshwaters are pH-dependent.
ation calculus [21]. The modelled percentage distribution among La and Moermond et al. [35] and Åström and Corin [4] stated that REE con­
Gd species for all salinities is presented in Table 2. Despite La3+ and centration has an inverse correlation to river water pH and Goldstein
Gd3+ being dominant in the three studied salinities, their percentage and Jacobsen [20] measured REE in several rivers in North America and
varied greatly between them. At salinity 0, La3+ was most present in the found that low pH rivers present high REE concentrations while rivers
Acidification (97.2%) and Warming & Acidification treatment (96.5%), with a pH of 8 exhibited low light REE concentrations and were HREE
while presenting the lowest percentage in the Warming treatment enriched. These findings are based on broad changes of pH, which was
(91.9%) followed by 93.5% in the Control TºC and pH treatment. The ion not mimicked in our study since we searched for REE differences asso­
Gd3+ followed this trend, constituting 80.7% of species in the Acidifi­ ciated with only increased acidification of 0.4 units. Therefore, it was
cation treatment and 75.8% in the Warming & Acidification treatment. expected no significant differences of La (p > 0.05) at salinity 0 in the
In general, the fraction of the Gd free ion in solution was lower than the Control TºC and pH and Acidification treatments. Otherwise, this was
trivalent La. The lowest percentage of Gd3+ was 55.5% in the Warming not followed by Gd as their levels in the Acidification treatment were
treatment. At present-day conditions (Control T◦ C and pH), 62.6% of significantly lower (p = 0.011) than the ones at the Control TºC and pH.
species were Gd3+. The second most present form for both elements was Goldstein and Jacobsen [20] showed that in freshwater medium the
the complexed LaOH2+ and GdOH2+, but with great percentage differ­ temperature is a major driver affecting dissolved La availability pre­
ences between them. sumably because minor pH variation was found in their work. The
At salinity 15, both La3+ and Gd3+ were present at lower percentages Warming and Warming & Acidification treatments presented signifi­
than at salinity 0. The ions La3+ and Gd3+ were present at the lowest cantly higher La levels than their Control TºC counterparts. The toxicity
percentage in the Warming treatment (45.9% and 34.8%, correspond­ of several metals is pH-modulated [33] however very few studies re­
ingly) and the highest percentage in the Acidification treatment (49.6% ported this for REE, and are majorly focused on extreme events such as
and 43.7%, respectively). The second most common species in salinity acid rains [31]. Rivers are the major source of elements to the oceans
15 were La and Gd sulfate complexes: LaSO4+ and GdSO4+, respectively. and the behavior of REE in freshwater should be further investigated.
The percentage of the free ions La3+ and Gd3+ species decreased The concentration of La and Gd decreased greatly over 24 h, at salinity
further at salinity 35. The highest percentage of La3+ occurred in the 0, in all studied treatments (Supplemental Table 4A and B), which bring
acidification treatment (43.7%) and lowest in the Warming treatment additional challenges in ecotoxicology studies. Furthermore, this
(34.8%). The trivalent Gd3+ peaked in the Warming & Acidification decrease has not reached a plateau pointing to continuous removal of
treatment (41.8%) and was lowest in the Control TºC and pH treatment both elements from the dissolved fraction. Therefore, in 24 h studies, or
(33.8%). longer, freshwater biota will be subject to different contamination levels
and biological effects may change. The desired REE exposure concen­
4. Discussion tration should be monitored in a shorter time scale or additional actions
of changing the medium and re-spiking should be done to maintain
The REE show very similar geochemical behavior, although the conditions.
lanthanide contraction (ionic radius decrease from La to Lu) leads to Regarding our results for salinity 15 (brackish water), the percent­
gradual changes in their physicochemical properties. The trivalent REE ages of La and Gd removal was lower than those found in freshwater
ions present different affinity towards surface adsorption and pointing to higher stability of both elements in the dissolved fraction.
complexation with ligands (e.g., carbonates, phosphates and hydrox­ Most studies showed a higher removal of REE in the mixing area of fresh
ides) in solution due to this lanthanide contraction, which ultimately to seawater [15,30,40]. This removal is mainly related to the increase of
leads to HREE greater stability in solution and preferential adsorption of particulate material, generation of colloids and precipitation of Al, Fe
LREE on colloids and suspended particulate material [16]. Furthermore, and Mn compounds [27,46]. These new particles promote sorption and
REE bioavailability varies with ionic and ligand loads and decreases precipitation reactions that scavenge REE into the solids. In our systems
with free ions and organic or inorganic complexes [8]. This organic this perturbation of REE equilibria do not exit and therefore, dissolved
complexation are different across La to Lu series [11,14]. Therefore, it is species tend to be more stable in solution presumably due to increased
challenging to fully understand REE behavior in natural waters due to complexation with chloride and sulphate ions. In the present study,
the formation of complexes, colloids, ion exchange and adsorption, temperature significantly affect the La levels in freshwater (Supple­
which in turn can lead to fractionation [7]. mental Table 3). Furthermore, seasonality and tides, which in turn may
We showed that La and Gd behave differently from salinity 0–35. Our impose broader variability of temperature, may further impacting REE
results showed that a temperature increase of 4ºC in freshwater rises the availability. Additionally, at salinity 15 the La levels in the Warming and
availability of dissolved La. However, the decrease of 0.4 pH units has Warming & Acidification treatment varied significantly through time,
not changed the availability of both elements. On the other hand, Gd which highlights the effect of temperature in their availability. Dissolved
levels at present-day settings were higher than the ones at the Acidifi­ La and Gd levels in seawater showed lower variability than at salinity
cation treatment, both at the same T◦ C (18 ◦ C). Zhang et al. [50] studied 0 and 15 since the percentage loss was the lowest and concentrations did
the REE dissolved distributions in small streams in China and found that not vary between treatments. In these seawater experiment, the speci­
anthropogenic Gd presented seasonal variations, being, for example, ation of both elements is much different from the observed in the other
diluted by fresh rainwater. Additionally, the studied water masses trials. Lower percentages of trivalent La and Gd free ions in solution and
exhibited HREE enrichment due to adsorption reactions that decreased increased proportion of sulphate and chloride complexes were found in
the LREE in solution. At present-day conditions, for salinity 0, La saltwater. The quantification of the total concentrations of REE in water
decreased 67% over the 24 h, while the Gd decreased 42%. This is not directly linked to their potential interactions with the environ­
discrepancy may be related to increased La adsorption to all systems in ment, bioaccumulation, and toxicological effects. Oral et al. [37]
the aquaria (glass, plastic air tubes, pH probe, etc.). Other hypothesis is described the toxicity of REE on sea urchin embryos and sperm and
the removal of both elements through the precipitation as hydroxides reported that levels ranged between 0.02 and 2 µg g-1 dry weight (dw),

6
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

which were between 4 and 7 orders of magnitude above the concen­ With virtually non-existent knowledge on the potential interactions
tration on the water. Studies evaluating the REE bioaccumulation and between climate change and REE, we introduced new insights regarding
toxicity should investigate and report their availability in the water to REE availability and speciation in three salinities at present-day and
provide a more comprehensive overview of these processes. near-future conditions. Nevertheless, the range of REE levels varies
Yang et al. [49] established a static aquatic microcosm to evaluate greatly depending on the aquatic system, and hence a broader assort­
the distribution of five REEs, including La and Gd. They found that the ment of concentrations should be tested in upcoming studies.
sediment was the major reservoir for REE with 80–90% of both elements Climate change is forecasted to rise the occurrence and harshness of a
sediment, and a significant proportion in water reached 8% for La and wide array of pathogens, such as virus, parasites, toxins and to exacer­
17% for Gd. This partitioning of REE in aquatic environments is key to bate the effects of pollutants. As verified, REE interactions with climate
understanding the intricacies of REE availability, accumulation, and change variables are likely to occur. The lack of data and studies on both
toxicity. Trifuoggi et al. [48] studied REE (Y, La, Ce, Nd, Sm, Eu and Gd) problematics highlights the urgent need to further assess this chal­
toxicology in the early life stages of two sea urchin species. They lenging topic.
observed different toxicity patterns for individual REEs, and these pat­
terns vary for each studied sea urchin species and treatment protocol. CRediT authorship contribution statement
This highpoint the distinct toxicity patterns among REE and the urgent
need to further study their behavior. For example, Gd-based contrast Cátia Figueiredo: Conceptualization, Methodology, Validation,
agents are chelated and used as contrast media for magnetic resonance Formal analysis, Investigation, Writing – original draft, Funding acqui­
imaging (MRIs) that are globally considered safe to humans at recom­ sition. Tiago F. Grilo: Conceptualization, Methodology, Validation,
mended dosing levels [42]. This stability in the aquatic environment Resources, Writing – review & editing, Funding acquisition. Clara
(sediment, water, suspended matter) is nevertheless understudied. Still, Lopes: Validation, Writing – review & editing. Pedro Brito: Validation,
it is known that the decomposition of these compounds may lead to the Writing – review & editing. Miguel Caetano: Conceptualization, Vali­
release of free Gd ions to the solution, which is a more toxic form. Perrat dation, Investigation, Resources, Writing – review & editing, Funding
et al. [38] studied the Gd bioaccumulation in the digestive gland and acquisition. Joana Raimundo: Conceptualization, Validation, Re­
gills of two edible bivalve species (i.e., Dreissena rostriformis bugensis and sources, Writing – review & editing, Funding acquisition.
Corbicula fluminea) and showed that it could be accumulated in these
tissues even as only present as the stable Gd-contrast agents, raising
Declaration of Competing Interest
concerns regarding their toxicity. Our results showed that at salinity 35,
several Gd complexes are theoretically stable, which raises the relevance
The authors declare that they have no known competing financial
of studying each element speciation for ecotoxicological studies.
interests or personal relationships that could have appeared to influence
REE speciation depends on salinity, pH and the presence of anions
the work reported in this paper.
[32,3]. The speciation, in turn, regulates solubility and bioavailability
[26]. Solubility is also dependent on pH and temperature. Speciation
Acknowledgements
analysis conducted by Johannesson et al. [25] in the organic-rich waters
evidenced that dissolved REE bound with organic matter. Therefore, a
This work was supported by Fundação para a Ciência e Tecnologia
clear understanding of metal speciation during the exposure phase in
(FCT), through the project Climatoxeel (PTDC/AAG-GLO/3795/2014)
ecotoxicology trials is key to broadening our knowledge of toxicity to
awarded to Tiago F. Grilo and the strategic project UID/MAR/04292/
aquatic organisms. Carbonates, phosphates and hydroxides may
2019 granted to MARE. Cátia Figueiredo acknowledges the FCT-PhD
decrease dissolved REE concentration in water [13,19]. However, the
grant SFRH/BD/130023/2017 and the Early Career Research Grant
dissolved complexes LaSO4+ and GdSO4+ were the second most common
awarded by National Geographic Society.
species at salinity 15 and 35. The availability of chloride and carbonate
anions in brackish and seawater leads to a change of speciation of La and
Gd with a significant decrease of free ions, which are considered the Appendix A. Supporting information
most toxic species [12,24]. The trivalent REE ions, such as La3+,
chemically resemble Ca2+, with comparable ionic radii, and are poten­ Supplementary data associated with this article can be found in the
tial competitors for Ca uptake [23]. This could inhibit Ca-channels, with online version at doi:10.1016/j.jtemb.2022.126957.
deleterious effects on biota. Furthermore, free Gd3+ is extremely toxic
[42]. However, even with greater REE removal at lower salinities over References
24-hours, the high proportion of trivalent ions at salinity 0 may influ­
[1] L. Al-Ghussain, Global warming: review on driving forces and mitigation, Environ.
ence the toxicity responses and should be the subject of great attention Prog. Sustain. Energy 38 (2019) 13–21.
for upcoming ecotoxicology studies. [2] I. Anastopoulos, A. Bhatnagar, E.C. Lima, Adsorption of rare earth metals: a review
of recent literature, J. Mol. Liq. 221 (2016) 954–962.
[3] M. Aqeel Ashraf, M. Jamil Maah, I. Yusoff, M. Ghararibreza, Speciation of heavy
5. Conclusions metals in the surface waters of a former tin mining catchment, Chem. Speciat.
Bioavailab. 24 (2012) 1–12.
The findings of this work showed that temperature affects the [4] M. Åström, N. Corin, Distribution of rare earth elements in anionic, cationic and
particulate fractions in boreal humus-rich streams affected by acid sulphate soils,
availability of La in freshwater systems and the pH decrease caused a Water Res. 37 (2003) 273–280.
reduction in Gd concentrations. La and Gd levels decreased over 24 h, at [5] M. Barange, T. Bahri, M.C. Beveridge, K.L. Cochrane, S. Funge-Smith, F. Poulain,
salinity 0, which should lead to short-time scale monitoring of REE Impacts of Climate Change on Fisheries and Aquaculture: Synthesis of Currrent
Knowledge, Adaptation and Mitigation Options, FAO, 2018.
exposure concentrations. At salinity 15, La dissolved levels in the [6] M. Bau, P. Dulski, Distribution of yttrium and rare-earth elements in the Penge and
Warming (T◦ =+4 ◦ C) and Warming & Acidification (T◦ =+4 ◦ C, Kuruman iron-formations Transvaal Supergroup, South Africa, Precambrian Res.
pH=△0.4) treatments decreased significantly through time, high­ 79 (1996) 37–55.
[7] R. Biddau, R. Cidu, F. Frau, Rare earth elements in waters from the albitite-bearing
lighting the interaction between climate change and the availability of
granodiorites of Central Sardinia, Italy, Chem. Geol. 182 (2002) 1–14.
this element in this medium. Dissolved La and Gd levels in seawater [8] J. Borrego, B. Carro, N. López-González, J. De La Rosa, J. Grande, T. Gómez, M. De
were kept at high levels and more stable with time. Nonetheless, with La Torre, Effect of acid mine drainage on dissolved rare earth elements
demonstrated REE speciation changes from salinity 0–35, coming geochemistry along a fluvial–estuarine system: the Tinto-Odiel Estuary (SW Spain),
Hydrol. Res. 43 (2012) 262–274.
studies on the effects on biota should consider the baseline information [9] H. Braun, The collected works of John W. Tukey: vol. VIII, multiple comparisons
presented in this study. 1948-1983, J. R. Stat. Soc. Ser. A Stat. Soc. (158) (1995) 629.

7
C. Figueiredo et al. Journal of Trace Elements in Medicine and Biology 71 (2022) 126957

[10] A.H. Buckman, S.B. Brown, J. Small, D.C. Muir, J. Parrott, K.R. Solomon, A.T. Fisk, [31] C. Liang, W. Wang, Antioxidant response of soybean seedlings to joint stress of
Role of temperature and enzyme induction in the biotransformation of lanthanum and acid rain, Environ. Sci. Pollut. Res. 20 (2013) 8182–8191.
polychlorinated biphenyls and bioformation of hydroxylated polychlorinated [32] T. Liang, S. Zhang, L. Wang, H.-T. Kung, Y. Wang, A. Hu, S. Ding, Environmental
biphenyls by rainbow trout (Oncorhynchus mykiss), Environ. Sci. Technol. 41 biogeochemical behaviors of rare earth elements in soil–plant systems, Environ.
(2007) 3856–3863. Geochem. Health 27 (2005) 301–311.
[11] R.H. Byrne, B. Li, Comparative complexation behavior of the rare earths, Geochim. [33] A.T. Luís, B. Bonet, N. Corcoll, S.F. Almeida, E.F. Da Silva, E. Figueira, H. Guasch,
Cosmochim. Acta 59 (1995) 4575–4589. Experimental evaluation of the contribution of acidic pH and Fe concentration to
[12] T. Das, A. Sharma, G. Talukder, Effects of lanthanum in cellular systems, Biol. the structure, function and tolerance to metals (Cu and Zn) exposure in fluvial
Trace Elem. Res. 18 (1988) 201–228. biofilms, Ecotoxicology 23 (2014) 1270–1282.
[13] J. Delgado, R. Pérez-López, L. Galván, J.M. Nieto, T. Boski, Enrichment of rare [34] N. Malhotra, H.-S. Hsu, S.-T. Liang, M.J.M. Roldan, J.-S. Lee, T.-R. Ger, C.-D. Hsiao,
earth elements as environmental tracers of contamination by acid mine drainage in An updated review of toxicity effect of the rare earth elements (REEs) on aquatic
salt marshes: a new perspective, Mar. Pollut. Bull. 64 (2012) 1799–1808. organisms, Animals 10 (2020) 1663.
[14] P. Ebrahimi, M. Barbieri, Gadolinium as an emerging microcontaminant in water [35] C.T. Moermond, J. Tijink, A.P. van Wezel, A.A. Koelmans, Distribution, speciation,
resources: threats and opportunities, Geosciences 9 (2019) 93. and bioavailability of lanthanides in the Rhine-Meuse estuary, The Netherlands,
[15] H. Elderfield, R. Upstill-Goddard, E. Sholkovitz, The rare earth elements in rivers, Environ. Toxicol. Chem. Int. J. 20 (2001) 1916–1926.
estuaries, and coastal seas and their significance to the composition of ocean [36] P.D. Noyes, M.K. McElwee, H.D. Miller, B.W. Clark, L.A. Van Tiem, K.C. Walcott, K.
waters, Geochim. Cosmochim. Acta 54 (1990) 971–991. N. Erwin, E.D. Levin, The toxicology of climate change: environmental
[16] Y. Erel, E.M. Stolper, Modeling of rare-earth element partitioning between particles contaminants in a warming world, Environ. Int. 35 (2009) 971–986.
and solution in aquatic environments, Geochim. Cosmochim. Acta 57 (1993) [37] R. Oral, P. Bustamante, M. Warnau, A. d’Ambra, M. Guida, G. Pagano, Cytogenetic
513–518. and developmental toxicity of cerium and lanthanum to sea urchin embryos,
[17] C.B. Field, Climate Change 2014–Impacts, Adaptation and Vulnerability: Regional Chemosphere 81 (2010) 194–198.
Aspects, Cambridge University Press, 2014. [38] E. Perrat, M. Parant, J.-S. Py, C. Rosin, C. Cossu-Leguille, Bioaccumulation of
[18] C. Figueiredo, J. Raimundo, A.R. Lopes, C. Lopes, N. Rosa, P. Brito, M. Diniz, gadolinium in freshwater bivalves, Environ. Sci. Pollut. Res. 24 (2017)
M. Caetano, T.F. Grilo, Warming enhances lanthanum accumulation and toxicity 12405–12415.
promoting cellular damage in glass eels (Anguilla anguilla), Environ. Res. 191 [39] Pierrot, D., Lewis, E., Wallace, D., 2006. MS Excel program developed for CO2
(2020), 110051. system calculations. ORNL/CDIAC-105a. Carbon Dioxide Information Analysis
[19] A. Foucault-Collet, K.A. Gogick, K.A. White, S. Villette, A. Pallier, G. Collet, Center, Oak Ridge National Laboratory, US Department of Energy, Oak Ridge,
C. Kieda, T. Li, S.J. Geib, N.L. Rosi, Lanthanide near infrared imaging in living cells Tennessee 10.
with Yb3+ nano metal organic frameworks, Proc. Natl. Acad. Sci. U.S.A. 110 [40] O.S. Pokrovsky, L.S. Shirokova, J. Viers, V. Gordeev, V.P. Shevchenko,
(2013) 17199–17204. A. Chupakov, T. Vorobieva, F. Candaudap, C. Causserand, A. Lanzanova, Fate of
[20] S.J. Goldstein, S.B. Jacobsen, Rare earth elements in river waters, Earth Planet. Sci. colloids during estuarine mixing in the Arctic, Ocean Sci. 10 (2014) 107–125.
Lett. 89 (1988) 35–47. [41] M. Pontié, J.S. Derauw, S. Plantier, L. Edouard, L. Bailly, Seawater desalination:
[21] J.P. Gustafsson, Visual MINTEQ 3.0 User Guide, KTH, Department of Land and nanofiltration—a substitute for reverse osmosis? Desalin. Water Treat. 51 (2013)
Water Recources, Stockholm, Sweden, 2011. 485–494.
[22] N. Haque, A. Hughes, S. Lim, C. Vernon, Rare earth elements: overview of mining, [42] M. Rogosnitzky, S. Branch, Gadolinium-based contrast agent toxicity: a review of
mineralogy, uses, sustainability and environmental impact, Resources 3 (2014) known and proposed mechanisms, Biometals 29 (2016) 365–376.
614–635. [43] J. Rogowska, E. Olkowska, W. Ratajczyk, L. Wolska, Gadolinium as a new
[23] H. Herrmann, J. Nolde, S. Berger, S. Heise, Aquatic ecotoxicity of lanthanum–a emerging contaminant of aquatic environments, Environ. Toxicol. Chem. 37
review and an attempt to derive water and sediment quality criteria, Ecotoxicol. (2018) 1523–1534.
Environ. Saf. 124 (2016) 213–238. [44] D. Schiedek, B. Sundelin, J.W. Readman, R.W. Macdonald, Interactions between
[24] S. Hirano, K.T. Suzuki, Exposure, metabolism, and toxicity of rare earths and climate change and contaminants, Mar. Pollut. Bull. 54 (2007) 1845–1856.
related compounds, Environ. Health Perspect. 104 (1996) 85–95. [45] E.R. Sholkovitz, The aquatic chemistry of rare earth elements in rivers and
[25] K.H. Johannesson, J. Tang, J.M. Daniels, W.J. Bounds, D.J. Burdige, Rare earth estuaries, Aquat. Geochem. 1 (1995) 1–34.
element concentrations and speciation in organic-rich blackwaters of the Great [46] N. Tepe, M. Bau, Behavior of rare earth elements and yttrium during simulation of
Dismal Swamp, Virginia, USA, Chem. Geol. 209 (2004) 271–294. arctic estuarine mixing between glacial-fed river waters and seawater and the
[26] A.M. Khan, N.K.A. Bakar, A.F.A. Bakar, M.A. Ashraf, Chemical speciation and impact of inorganic (nano-) particles, Chem. Geol. 438 (2016) 134–145.
bioavailability of rare earth elements (REEs) in the ecosystem: a review, Environ. [47] G. Trapasso, S. Chiesa, R. Freitas, E. Pereira, What do we know about the
Sci. Pollut. Res. 24 (2017) 22764–22789. ecotoxicological implications of the rare earth element gadolinium in aquatic
[27] S. Kulaksız, M. Bau, Contrasting behaviour of anthropogenic gadolinium and ecosystems? Sci. Total Environ. (2021), 146273.
natural rare earth elements in estuaries and the gadolinium input into the North [48] M. Trifuoggi, G. Pagano, M. Guida, A. Palumbo, A. Siciliano, M. Gravina, D.
Sea, Earth Planet. Sci. Lett. 260 (2007) 361–371. M. Lyons, P. Burić, M. Levak, P.J. Thomas, Comparative toxicity of seven rare earth
[28] K. Kummerer, E. Helmers, Hospital effluents as a source of gadolinium in the elements in sea urchin early life stages, Environ. Sci. Pollut. Res. 24 (2017)
aquatic environment, Environ. Sci. Technol. 34 (2000) 573–577. 20803–20810.
[29] J. Kunnemeyer, L. Terborg, B. Meermann, C. Brauckmann, I. Moller, A. Scheffer, [49] X. Yang, D. Yin, H. Sun, X. Wang, L. Dai, Y. Chen, M. Cao, Distribution and
U. Karst, Speciation analysis of gadolinium chelates in hospital effluents and bioavailability of rare earth elements in aquatic microcosm, Chemosphere 39
wastewater treatment plant sewage by a novel HILIC/ICPMS method, Environ. Sci. (1999) 2443–2450.
Technol. 43 (2009) 2884–2890. [50] J. Zhang, Z. Wang, Q. Wu, Y. An, H. Jia, Y. Shen, Anthropogenic rare earth
[30] M.G. Lawrence, B.S. Kamber, The behaviour of the rare earth elements during elements: gadolinium in a small catchment in Guizhou Province, Southwest China,
estuarine mixing—revisited, Mar. Chem. 100 (2006) 147–161. Int. J. Environ. Res. Public Health 16 (2019) 4052.