Science of the Total Environment 675 (2019) 165–175

Contents lists available at ScienceDirect

Science of the Total Environment

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

Nanoremediation and long-term monitoring of brownfield soil highly

polluted with As and Hg
M. Gil-Díaz a,⁎, E. Rodríguez-Valdés b, J. Alonso a, D. Baragaño b, J.R. Gallego b, M.C. Lobo a
a
IMIDRA, Finca “El Encín”, 28800 Alcalá de Henares, Madrid, Spain
b
INDUROT and Environmental Biotechnology & Geochemistry Group, University of Oviedo, C/Gonzalo Gutiérrez Quirós s/n, 33600 Mieres, Asturias, Spain

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Two plots from a highly polluted

brownfield were treated with commer-
cial nZVI
• nZVI significantly reduced As and Hg
mobility at field conditions
• The nZVI effectiveness depended on the
degree of pollution
• It was evidenced the stability of the
nanoremediation for 32 months

a r t i c l e i n f o a b s t r a c t

Article history: In the last decade, several laboratory-scale experiments have shown the use of nanoscale zero-valent iron (nZVI)
Received 13 March 2019 to be effective in reducing metal(loid) availability in polluted soils. The present study evaluates the capacity of
Received in revised form 10 April 2019 nZVI for reducing the availability of As and Hg in brownfield soils at a pilot scale, and monitors the stability of
Accepted 11 April 2019
the immobilization of these contaminants over a 32 month period. To the best of our knowledge, this is the
Available online 13 April 2019
first study to apply nZVI to metal(loid)-polluted soils under field conditions. Two sub-areas (A and B) that dif-
Editor: Patricia Holden fered in pollution load were selected, and a 5 m2 plot was treated with 2.5% nZVI (by weight) in each case
(Nanofer 25S, NanoIron). In sub-area A, which had a greater degree of pollution, a second application was per-
Keywords: formed eight months after the first application. Overall, the treatment significantly reduced the availability of
Arsenic both As and Hg, after only 72 h, although the effectiveness of the treatment was highly dependent on the degree
Mercury of initial contamination. Sub-area B (with a lower level of pollution) showed the best and most stable immobili-
Availability zation results, with As and Hg in toxicity characteristics leaching procedure (TCLP) extracts decreasing by 70%
Field scale and 80%, respectively. In comparison, the concentrations of As and Hg in sub-area A decreased by 65% and 50%,
nZVI
respectively. Based on our findings, the use of nZVI at a dose of 2.5% appears to be an effective approach for
the remediation of soils at this brownfield site, especially in sub-area B. For sub-area A, a higher dose of nZVI—
or its use in combination with other remediation strategies—should be tested.
© 2019 Elsevier B.V. All rights reserved.

1. Introduction

⁎ Corresponding author. Since the onset of industrialization two centuries ago, the release of
E-mail address: mar.gil.diaz@madrid.org (M. Gil-Díaz). metal(loid)s and various organic compounds from industrial processes

https://doi.org/10.1016/j.scitotenv.2019.04.183
0048-9697/© 2019 Elsevier B.V. All rights reserved.
166 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175
such as manufacturing, mining, road transport, military activities, and
agriculture have led to large increases in contaminated land and
water. In fact, there are about 2.5 million potentially contaminated
sites in the European Union, of which approximately 14% (340,000
sites) are expected to be contaminated and likely to require remediation
(van Liedekerke et al., 2014). According to the European Environment
Agency (EEA), the management of contaminated land in Europe costs
an estimated € 6.5 billion per year (EEA, 2014).
Mining and metallurgy are important sources of soil pollution. In this
context, the El Terronal mine (Asturias, Northern Spain) was the second
largest Hg producer in Spain in the 1960s and early 1970s, and it also
produced As compounds (Fernandez-Martinez et al., 2014; Gallego
et al., 2015). The crash in Hg prices due to international health concerns
over their use led to the cessation and abandonment of mining-
metallurgical activities in this area in 1974. Mining waste, as well as
the mining facilities, remained on site until 2002, when spoil heap
waste was partially isolated in an on-site security landfill, however, pol-
luted soil and waste persist in the area, thus promoting the diffusion of
pollutants that have a severe environmental impact (Fernandez-
Martinez et al., 2014; Gallego et al., 2015; Gonzalez-Fernandez et al.,
2018). In addition to As and Hg, the brownfield site contains organic
contaminants including benzo(a)pyrene and other Polycyclic aromatic
hydrocarbons (PAHs), which are sub-products of the pyrometallurgical
process (Gallego et al., 2015). The selection of remediation strategies is
often more complicated in the case of multiple pollutants since their ef-
ficiency is influenced by the synergistic and antagonistic interactions
between pollutants (Ye et al., 2017a,b). However, given the heavy pol-
lutant load of the El Terronal brownfield site, remediation is required.
Arsenic is a metalloid that is highly toxic to biota and humans. In
humans, As can induce skin lesions, cancer, and has deleterious effects
on the respiratory, circulatory, nervous, and reproductive systems
(Zhang and Selim, 2008). The oxidation states of As are −3, 0, +3,
and +5, of which As0 and As3+ are characteristic of reducing environ-
ments. In fact, under most environmental conditions, As is found as
As5+, such as in the H2AsO− 4 species, while As
3+
, as found in the
0
H2AsO3 species, is dominant only in low-pH and low-Eh environments.
As3+ shows greater toxicity and mobility in soils than As5+ (Hopkins
et al., 2009; Kabata-Pendias, 2011; Zhang and Selim, 2008). The most
mobile forms of As are the complex anions AsO− 3− 2−
2 , AsO4 , HAsO4 , and
H2AsO− 4 , which are sorbed preferentially between pH 7 and 9. The be-
havior of arsenate resembles that of phosphates and vanadates
(Kabata-Pendias, 2011; Zhang and Selim, 2008). Although As is also
found combined with some metals (i.e., Fe, Pb, and Cu), arsenopyrite
(FeAsS) is the most common As mineral. As-rich pyrite is common in
sedimentary formations rich in organic matter, especially shale, coal,
and peat deposits (Kabata-Pendias, 2011).
Mercury is also highly toxic to humans and ecosystems. Mercury
compounds in nature exist both as inorganic salts and organic com-
pounds (Adriano, 2001; Kabata-Pendias, 2011). Although all forms of
Hg are highly toxic, organic Hg compounds, such as methyl-Hg, are
the most dangerous due to their lipophilic nature, which enhances bio-
accumulation (Adriano, 2001; Kabata-Pendias, 2011; Xu et al., 2015).
Once methyl-Hg is formed, it cycles through the environment for thou-
sands of years, exposing humans and other species to potentially toxic
levels for multiple generations (Science Communication Unit, 2013).
The behavior of Hg in soils is highly dependent on its form, which is a
function of soil properties, Hg concentrations, and the presence of
other ions, such as Cl−. In general, simple cations of Hg dominate in aer-
obic soil conditions and complex anions dominate in anoxic/anaerobic
conditions, while methylated Hg compounds predominate in soils of in-
termediate redox states (Kabata-Pendias, 2011; Yin et al., 1997).
Unlike organic pollutants, metal(loid)s do not show chemical or bi-
ological degradation and thus can remain in soils for extended periods
of time (Adriano et al., 2004; Bolan et al., 2014). Currently, most legisla-
tion related to metal(loid)s in soils is based on their total concentration,
which is a poor indicator of their deleterious effects. In this regard, metal
(loid) bioavailability is increasingly being used as a key indicator of the
potential risk to both environmental and human health (Adriano et al.,
2004), and some countries are introducing this concept into their na-
tional legislation (Kumpiene et al., 2017). According to Vangronsveld
and Cunningham (1998), metal(loid) bioavailability refers to the frac-
tion of the total chemical load that can interact with a biological target.
In other words, bioavailability refers to uptake by living organisms
(Adriano et al., 2004).
A wide variety of methodologies using selective chemical
extractants have been used to estimate metal(loid) availability in soil
samples (Abedin et al., 2012; Bolan et al., 2014; Fang et al., 2007; Gil-
Diaz et al., 2017a; Hammer and Keller, 2002; Jing et al., 2008). Sequen-
tial extraction methods provide information on metal(loid) concentra-
tions in the different soil fractions and, thus, on the concentrations
that could be mobilized under a range of conditions (i.e., via reduction,
oxidation, and acidification) (Kumpiene et al., 2017). Tessier's extrac-
tion procedure is the most widely used sequential extraction method,
and includes five fractions, namely (in decreasing order of availability)
exchangeable, carbonate bound, Fe and Mn oxide bound (reducible), or-
ganic matter bound (oxidizable), and residual (Bacon and Davidson,
2008; Tessier et al., 1979). The two first fractions of the method (ex-
changeable and carbonate) are considered the most bioavailable metal
(loid) fractions (Abedin et al., 2012; Gallego et al., 2015).
Immobilization techniques involve the addition of an agent that re-
duces the bioavailability of the metal(loid), thereby decreasing poten-
tial transfer to plants as well as leachability to groundwater through
precipitation, adsorption, and complexation reactions. A variety of tradi-
tional and new materials, such as phosphate compounds, liming mate-
rials, organic matter, biochar, metal oxides, and highly porous metal-
organic frameworks, can immobilize metal(loid)s in soils (Bolan et al.,
2014; Kumpiene et al., 2008; Lee et al., 2011; Li et al., 2018; Querol
et al., 2006). Recently, the use of nanoscale zero-valent iron (nZVI) to re-
duce metal(loid) mobility in polluted sites has been extensively studied.
Although most nanoremediation research studies—and many field ap-
plications—have been performed using water samples (Bardos et al.,
2015; Karn et al., 2009; Klimkova et al., 2011; Li et al., 2014, 2017;
Mueller et al., 2012; O'Carroll et al., 2013; Tosco et al., 2014; Wu et al.,
2019; Zou et al., 2016), more attention has recently been paid to the
nanoremediation of metal(loid)-polluted soils and sediments (Boente
et al., 2018; Cao et al., 2018; Gil-Diaz et al., 2014a,b, 2016, 2017a,b;
Gong et al., 2017; Huang et al., 2016; Madhavi et al., 2014; Singh et al.,
2012; Vitkova et al., 2017; Wang et al., 2014; Xue et al., 2018a,b,c;
Zhang et al., 2010). Experiments conducted at the laboratory scale
have found that nZVI is effective at immobilizing Ag, Al, As, Be, Cd, Cr,
Cu, Hg, Ni, Pb, U, V, and Zn in water samples. Studies of metal(loid) im-
mobilization in soils have concluded that the effectiveness of nanoparti-
cles is highly dependent on the soil properties, metal(loid)
characteristics, dose of nZVI applied, and the presence of other metal
(loid)s in the soil (Gil-Diaz et al., 2017a,b; Vitkova et al., 2017). One of
the most important limitations of immobilization strategies is that the
temporal stability of the immobilization is unknown because most stud-
ies are performed over short periods of time. Thus, monitoring to eval-
uate metal availability over extended time periods is required.
To date, most applications of nZVI on metal(loid)-polluted soils have
been performed at the laboratory scale and few data are available re-
garding the effectiveness of this approach at the pilot and field scale.
Similarly, information about the long-term stability of metal(loid) im-
mobilization is lacking. In these regards, previous laboratory experi-
ments performed using two soils from the El Terronal brownfield site
have shown that nZVI is effective in As and Hg immobilization using
Tessier's sequential extraction procedure (Gil-Diaz et al., 2017a). The
same study showed that treatment with nZVI did not have a significant
effect on soil pH nor electrical conductivity, nor on As and Hg speciation.
Based on these preliminary findings, the goals of the present study
were: i) to study the effectiveness of nZVI for As and Hg immobilization
in two areas of this brownfield site at a pilot scale and ii) to examine the
 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175
stability of As and Hg immobilization by monitoring the availability of
these pollutants over a period of 32 months. To the best of our knowl-
edge, this is the first study to perform nZVI remediation of metal
(loid)-polluted soils under field conditions.
2. Materials and methods
2.1. Nanoscale zero-valent iron
A stabilized aqueous dispersion of nZVI NANOFER 25S commercial-
ized by NANO IRON s.r.o. (Rajhrad, Czech Republic) was used in the
pilot experiment. The Fe(0) content was 14–18% (by weight) and iron
oxides accounted for 2–6% (by weight). More information about the
physical and chemical properties of nZVI is available at http://
nanoiron.cz/en/products/zero-valent-iron-nanoparticles/nanofer-25s.
2.2. Pilot experiment
The pilot study area was located in the municipality of Mieres, situ-
ated in the Carboniferous Central Basin of the Asturias Region in North-
ern Spain (Fig. 1A and B). The climate in this area is oceanic with mild
temperatures (yearly average = 12.5 °C). Precipitation typically varies
between 1000 and 1200 mm per year (Gonzalez-Fernandez et al.,
2018).
Industrial and mining activities associated with Hg ores were carried
out at the El Terronal site until its closure in the 1970s. This brownfield
land contains metallurgical dust, slag, and other mining wastes gener-
ated by former Hg mining and metallurgical factory activities, which
were abundant throughout the area (Gallego et al., 2015). Years of
treated gas emissions from a pyrometallurgical plant channeled by an
old duct (chimney), dust dispersion from mineral transport works,
Fig. 1. (A) Location of “El Terronal” site in Spain (Asturias region, Mieres municipality). (B) Site location in the San Tirso River valley (see González-Fernández et al., 2018 for details).
(C) Detailed aerial view where the brownfield site is contoured (shown in black). (D) Photographs of the parcels preparation and application of nanoparticles (the two sub-areas selected
for the treatments are shown).
167
and smelting waste accumulation have all had a dramatic effect on
soil quality, ground and surficial waters, and sediments of the nearby
river (Fig. 1C, see Gonzalez-Fernandez et al., 2018, for details).
The brownfield site presents a heterogeneous distribution of con-
taminants, especially in the case of As. Thus, two sub-areas, hereafter
called A and B, were selected for the experiment (Fig. 1C). Sub-area A
was nearer to the main pollution sources and showed a higher concen-
tration of As and Hg than sub-area B (Table 1). Bulk soil samples from
each area were collected from the surface layer (0–20 cm depth), air-
dried, and sieved (b2 mm) prior to analysis. Three composite samples
from each area were collected and their physico-chemical soil proper-
ties were analyzed following the official methodology in Spain
(Table 1) (MAPA, 1994). Soil texture was measured using the Robinson
pipette method; the concentrations of total metal(loid)s in the samples
were analyzed by acid digestion in a microwave reaction system
(Multiwave Go, AntonPaar GmbH), as previously explained in Gil-Diaz
et al. (2017a), and the concentrations of Cd, Cr, Cu, Ni, Pb, and Zn
were quantified by flame atomic absorption spectrometry (AAS),
while As and Hg concentrations were measured by graphite furnace
atomic absorption spectrometry (GFAAS) and inductively coupled
plasma-mass spectrometry (ICP-MS), respectively.
A 5 m2 plot in each sub-area was used for the nanoremediation
treatment; nZVI was applied using a 2.5% (by weight) dosage based
on effectiveness and economic viability criteria (Gil-Diaz et al., 2017a).
The commercial suspension was diluted with water (1:1) and evenly
distributed over the surface of the plot. It was then incorporated into
the surface layer (0–20 cm) of the soil using a rotavator (Fig. 1D). In
sub-area A (the most polluted), nZVI was reapplied (using the same
dose) eight months after the first application and following the same
methodology. To evaluate the effectiveness of the nanoremediation
treatment at reducing As and Hg availability, and to determine the
168 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175

Table 1 with a pH close to neutrality and an organic matter content higher

Physical and chemical properties of the soil samples collected from the brownfield (n = 3). than 3%. Soil from sub-area A showed a greater content of N, Ca, Na,
Properties A B GRLa (mg/kg) Mg, and K, as well as heavy metals including Cr, Cu, Ni, Pb, and Zn. The
pH 7.01 7.17 –
mean concentrations of As and Hg showed high pollution levels in
EC (dS m−1) 0.14 0.13 – both areas, but particularly in soil A, which exceeded the maximum
N (%) 0.20 0.16 – levels permitted by regional (BOPA, 2014) and international regulations
OM (%) 3.32 3.14 – (Crommentuijn et al., 2000; MEF, 2007). In addition, data for Hg and As
Ca (mg/kg) 2798 1240 –
chemical speciation, and their notable availability in this area, have been
Mg (mg/kg) 265 98 –
Na (mg/kg) 40 19 – previously reported (Fernandez-Martinez et al., 2014; Gallego et al.,
K (mg/kg) 397 137 – 2015; Rumayor et al., 2017). Thus, the high level of contamination mea-
As (mg/kg) 43,300 7280 200 sured for this brownfield site poses a high ecological risk and requires
Cd (mg/kg) 3.8 3.8 200 remediation (Gonzalez-Fernandez et al., 2018).
Cr (mg/kg) 62 37 10,000
Cu (mg/kg) 41 27 4000
Hg (mg/kg) 2200 1300 100 3.1. Evolution of As and Hg availability
Ni (mg/kg) 55 27 6500
Pb (mg/kg) 137 56 800 Figs. 2 and 3 show monitoring data for As and Hg concentrations in
Zn (mg/kg) 1227 139 10,000
the TCLP extracts and the EX+CB fractions from the soil samples from
a
GRL: Generic reference levels for metal(loid)s according to regional legislation (BOPA, the two areas, respectively. In general, a significant decrease in available
2014). As and Hg was observed over time, and the most significant reduction
was observed in soil B. The greatest quantitative reduction was ob-
stability of the immobilizing effect, three composite soil samples were served 72 h after initial nZVI application. Soil A, having a higher As con-
collected from the surface layer (0–20 cm) of each area at different centration, was subjected to two applications of nZVI at a dose rate of
times. In sub-area A, samples were collected before treatment (0 h) 2.5% (a total of 5%). Subsequently, the effect of applying two 2.5% nZVI
and 72 h, 1 month, 2 months, and 8 months after the first treatment doses or a single 5% dose on As and Hg immobilization in soil from
and 72 h, 1 month, 2 months, 8 months, 12 months, 17 months, and sub-area A was studied in the laboratory, showing no significant differ-
24 months after the second treatment. In sub-area B, samples were col- ence (p N 0.05). Seventy-two hours after the first treatment, the As con-
lected before treatment (0 h) and after 72 h, 1 month, 2 months, centration in the TCLP extract decreased from 1033 mg/kg to 616 mg/kg
8 months, 14 months, 24 months, and 32 months. Soil samples were (a decrease of 40%), and the TCLP extract showed a mean As concentra-
taken to the laboratory, air dried, and sieved (b2 mm) prior to the anal- tion of 441 mg/kg after eight months. In comparison, the two first frac-
ysis of As and Hg availability and soil physico-chemical properties. tions obtained by using Tessier's extraction procedure (EX+CB)
showed a lower decrease of available As (30%) 72 h after the first nZVI
2.3. As and Hg availability application. In the case of Hg, a decrease of 39% and 54% was observed
based on the TCLP extracts and EX+CB fractions, respectively, at the
The availability of As and Hg was evaluated using the toxicity charac- same sampling time. Tessier's fractions showed greater Hg availability
teristics leaching procedure (TCLP) described by the USEPA (Method compared with the TCLP extracts due to the higher concentration of ex-
1311, 1992). In addition, the first two fractions of Tessier's extraction changeable Hg. Despite the application of nZVI, the concentrations of
procedure were also analyzed (Tessier et al., 1979), namely exchange- available As and Hg in soil A were still very high and a second applica-
able (EX) and carbonate-bound (CB) fractions, which are related to tion was, therefore performed eight months after the first. Arsenic de-
the most available fractions and are potentially available for plants creased by around 40% and 35% in the TCLP extracts and EX+CB
and other soil organisms (Abedin et al., 2012; Gallego et al., 2015). fractions, respectively, 72 h after the second application. Monitoring
Both methods are described in detail by Gil-Diaz et al. (2014a). Arsenic over 24 months showed that As immobilization remained stable, falling
and Hg were measured in all the extracts following the previously ex- by 212–355 mg/kg in the TCLP extracts and 331–414 mg/kg in the EX
plained methodologies. +CB fractions. These results show that a high content of available As
was still present in sub-area A after the nanoremediation treatment.
2.4. Iron availability To achieve a more effective reduction of available As, a higher dose of
nZVI or nZVI application in combined with other remediation strategies
The impact of the nZVI application on Fe availability in soil was ex- would be required in this subzone. In this regard, higher As immobiliza-
amined using the TCLP. Iron was measured in the TCLP extracts by tion in soil samples from the same sub-area was found in a previous
flame ASS. laboratory-scale experiment using 5% and 10% doses of nZVI (Gil-Diaz
et al., 2017a). In that study, treatment with 5% nZVI was more effective
2.5. Scanning Electron Microscopy (SEM) analysis of the soil samples than in the present experiment under field conditions, since As in the
TCLP extracts and EX+CB fractions were reduced to mean values of
A JEOL JSM-5600 SEM coupled to an Energy Dispersive X-ray (EDX) 140 mg/kg and 229 mg/kg, respectively. These results highlight the
analyzer (INCA Energy 200) was used to study the interaction of the need for long-term pilot-scale studies before the selection of a remedi-
nanoparticles with the polluted soils. Samples were embedded in a ation strategy. In previous research, a 10% nZVI dose proved to be the
resin matrix and were Au-coated before analysis. most effective approach for reducing As availability at a laboratory
scale; however, a 10% dose is economically unviable at the field scale.
2.6. Statistical analysis In addition, this dosage implies a high Fe input, which can decrease
soil pore size. With respect to Hg, the second application of nZVI used
Statistical analysis, including analysis of variance (ANOVA), was con- here did not greatly affect its availability in the test soils.
ducted using SPSS 23.0.0.0 (SPSS Inc., IBM Company). The high organic matter content found in the soil samples from both
sub-areas may have limited the capacity of nZVI to immobilize As, as
3. Results and discussion humic acids can compete with As for adsorption sites on the surface of
nZVI (Grafe et al., 2001, 2002; Sadiq, 1997). This observation can be ex-
The physico-chemical properties of the studied soils are shown in plained by the anionic nature of humic acids and the strong affinity of
Table 1. The soils of both sub-areas A and B had a silt loam texture their carboxylic and phenolic functional groups with iron oxide surfaces
 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175
Fig. 2. Monitoring of As and Hg in TCLP extracts in soil samples from sub-area A and B throughout the experiment.
(Grafe et al., 2001; Gustafsson and Bhattacharya, 2007). Gustafsson and
Bhattacharya (2007) described the adsorption of humic acid functional
groups onto the surfaces of oxides based on a monodentate complex;
the complex formed between the iron oxide surface and humic acid is
stable, and, thus, difficult to desorb unless pH is increased considerably,
which leads to an increase in the polarity of the humic acids and an in-
creased negative charge on the oxide surface (Avena and Koopal, 1998).
In addition, soil organic matter can form an aqueous complex with As
(Redman et al., 2002).
In sub-area B, the application of nZVI was more effective at reducing
As and Hg availability than in sub-area A (Figs. 2 and 3). Arsenic de-
creased by 74% and 51% according to the TCLP and EX+CB fractions, re-
spectively, whereas Hg showed a greater decrease with reductions of up
to 86% and 93%, respectively. The stability of the immobilization was
greater in sub-area B than in sub-area A (Figs. 2 and 3). According to
the TCLP and EX+CB fractions, As concentrations were in the ranges
of 11–27 mg/kg and 35–61 mg/kg, respectively. In the case of Hg, con-
centrations ranged between 0.37 and 0.79 mg/kg in the TCLP extracts
169
Sub-area A
Sub-area B
and between 6.8 and 26 mg/kg in the EX+CB fractions. The greater ef-
fectiveness of the treatment in soil B was due mainly to its lower level
of initial contamination; sub-area A, which was nearer to the source of
contamination, showed a higher concentration of As and Hg. In particu-
lar, the total As concentration in soil A was almost six times higher than
in soil B. In addition to As and Hg, the brownfield site included organic
contaminants such as PAHs, with sub-area A showing higher concentra-
tions of these compounds (Gallego et al., 2015). Interactions between
pollutants have a significant influence on the efficiency of soil remedia-
tion (Ye et al., 2017a,b). Therefore, the PAHs identified at this site could
interact with iron nanoparticles, thereby decreasing their reactivity.
3.2. Impact on soil aggregates
The results obtained in the SEM analysis are shown in Fig. 4. Accord-
ing to SEM images and EDX analysis, the soil samples showed nanopar-
ticle aggregates that contained As and Hg (Fig. 4A, B, C, and E). The
microanalysis of three zones of the aggregates showed that As mainly
170 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175
Fig. 3. Monitoring of As and Hg in more available fractions (EX+CB) using Tessier's fractionation in soil samples from sub-area A and B throughout the experiment.
occurred at the surface in association with iron oxides (Fig. 4B and C). In
the center of the nanoparticle aggregates, Fe occurred in the form of me-
tallic iron (Fe0), whereas on the surface, Fe was oxidized, forming iron
oxides and hydroxides that complexed with As. This distribution was
observed in the EDX mapping (Fig. 4D, F, and G) in which Fe is shown
in red and As in green. In this respect, the mechanism of interaction be-
tween As and nZVI was adsorption onto iron oxides, forming the shell of
nZVI. Previous laboratory experiments have shown that As mainly oc-
curs as arsenate (N98%) at this site since the nanoremediation treatment
did not reduce As5+ to As3+ or As0 (Gil-Diaz et al., 2017a).
Mercury, in the form of sulfur compounds and elemental Hg, was
also found in association with Fe (Fig. 4E, F, and G). The elemental
map in Fig. 4F (Hg is shown in red and Fe in green) shows the interac-
tion between Hg and Fe within the nanoparticles (Fig. 4H). The reduc-
tion of Hg2+ to the metallic form was likely the main removal
mechanism of ionic Hg since redox reactions are favorable based on
standard redox potentials (E0Hg = 0.86 V; E0Fe = −0.44 V; ΔE0 =
1.30 V) (O'Carroll et al., 2013; Xue et al., 2018a).
Sub-area A
Sub-area B
3.3. Iron availability
The addition of nZVI to the soil did not lead to a large increase in Fe
availability, as shown by the TCLP results. After the nanoremediation
treatment, a slight increase of available Fe was detected; however, this
effect was not maintained over time. This might be due to the neutral
pH conditions, which favor the precipitation of Fe. Some variability
was observed in the Fe-TCLP content, which was in the range of
1.0–11 mg/kg in both sub-areas throughout the experiment.
3.4. Impact on soil properties
The effects of the nanoremediation treatment on soil properties are
shown in Tables 2 and 3 for soil A and B, respectively. The application
of nZVI under the experimental conditions did not greatly affect soil
properties with the exception of Na, the availability of which signifi-
cantly increase in both sub-areas (p b 0.001). This is attributed to the
 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175 171

Fig. 4. SEM images of soil after the treatment with nZVI (A), where the red square indicates a magnification of the area shown in (B); EDS results of image B (C); color mapping showing the
distribution of As (green) and Fe (red) (D); SEM image of soil treated with nZVI (E) where the red square indicates a magnification of the area shown in (F); EDS results of image F (G);
color mapping showing the distribution of Hg (red) and Fe (green).
172 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175

Fig. 4 (continued).

fact that the organic stabilizer in commercial nanoparticles contains Na. months. After the second application of nZVI, available Na again in-
In sub-area A, available Na increased from 40 to 105 mg/kg 72 h after creased (to 129–156 mg/kg) and a decrease was not observed until
the treatment, while it decreased over time to 57 mg/kg after eight eight months later. After two years of monitoring, the concentrations
 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175 173

Fig. 4 (continued).

of available Na were at similar levels to those measured before the
nanoremediation treatment, probably due to losses by leaching. The be-
havior of available Na in sub-area B was similar, with concentrations in-
creasing from 19 to 167 mg/kg 72 h after the nZVI treatment and
decreasing over time. Thirty-two months after the nZVI application,
the concentrations of available Na were similar to those measured be-
fore the treatment. Previous studies have also reported increases in
available Na after soil treatment with these commercial nanoparticles
Table 2
Monitoring of the physico-chemical soil properties of soil samples from sub-area A
through the experiment.
(Gil-Diaz et al., 2014b, 2016). The mean values of available Na were
within the normal range for these soils.
4. Conclusions
To the best of our knowledge, this is the first nanoremediation ex-
periment to evaluate the effectiveness of nZVI in the immobilization of
heavy metal(loid)s under field conditions and the long-term
(32 month) stability of these effects. The effectiveness of nZVI applica-
tion was primarily dependent on the degree of contamination. A signif-
icant decrease in As and Hg availability was observed just 72 h after the

Time pH EC N OM K Na Ca Mg Table 3
(dS/m) (%) (%) (mg/kg) (mg/kg) (mg/kg) (mg/kg) Monitoring of the physico-chemical soil properties of soil samples from sub-area B
0 6.94 0.190 0.20 3.06 472 40.0 3871 341 through the experiment.
72 h 7.08 0.193 0.18 2.61 461 105 3726 329
Time pH EC N OM K Na Ca Mg
1m 6.94 0.127 0.20 3.04 400 84.0 3359 321
(dS/m) (%) (%) (mg/kg) (mg/kg) (mg/kg) (mg/kg)
2m 7.25 0.119 0.20 2.93 394 82.0 3195 368
8m 7.13 0.124 0.21 2.15 400 57.7 3278 307 0 7.08 0.157 0.16 3.47 125 19.3 1488 93.7
72 h−2 7.13 0.167 0.19 2.28 506 129 3858 330 72 h 7.12 0.258 0.14 3.05 166 167.0 1417 102
1 m−2 7.11 0.203 0.23 2.59 409 155 2747 266 1m 6.98 0.185 0.14 2.96 152 103.3 1242 99.7
2 m−2 7.20 0.206 0.23 2.27 438 156 3088 291 2m 7.43 0.135 0.13 2.89 150 94.0 1320 118
8 m−2 7.29 0.120 0.21 2.48 393 94.3 2109 248 8m 6.98 0.091 0.15 2.94 130 26.3 1390 101
12 m−2 7.32 0.150 0.21 2.58 417 86.0 2556 257 14 m 6.93 0.110 0.14 3.05 120 30.0 1052 96.0
17 m−2 7.41 0.110 0.19 2.25 398 87.5 2657 257 24 m 7.18 0.103 0.14 2.85 145 16.7 1085 106
24 m−2 7.59 0.150 0.22 2.32 465 36.0 2939 370 32 m 6.62 0.130 0.15 3.29 111 16.0 1073 98
174 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175
nZVI application. In sub-area B, which had a lower level of As and Hg
contamination, the application of nZVI (2.5%) achieved the best immo-
bilization results and a greater degree of stability over time compared
to sub-area A. The reduced effectiveness of remediation in sub-area A,
despite two applications of nZVI with a combined dose of 5%, is likely at-
tributable to the higher initial pollutant load in this area. These results
show that the use of 2.5% nZVI can provide an effective remediation ap-
proach for the soils at this brownfield site (sub-area B) and, similarly,
such an approach could be considered as a potential starting point for
other polluted sites. Given the greater pollutant load in the soil from
sub-area A, we propose that higher doses of nZVI—or its application in
combination with other remediation strategies—should be tested to re-
duce the risks associated with the mobility of metal(loid) pollutants.
Acknowledgements
This work was supported by Projects REHABILITA CTM2016-78222-
C2-1-R (AEI/FEDER, UE), FP-16-NANOREMED (IMIDRA, Comunidad de
Madrid, Spain), BIONANOWASH CTM2016-75894-P (AEI/FEDER, UE)
and I+DARTS (LIFE11 ENV/ES/000547, EC). Diego Baragaño obtained
a grant (FPU-16/05450) from the FPU program financed by the
“Ministerio de Educación, Cultura y Deporte”, Spain. We would also like
to thank the Electronic Microscope and the Environmental Assay Units
of the Scientific and Technical Services of the University of Oviedo for
technical support.
References
Abedin, J., Beckett, P., Spiers, G., 2012. An evaluation of extractants for assessment of
metal phytoavailability to guide reclamation practices in acidic soilscapes in northern
regions. Can. J. Soil Sci. 92, 253–268.
Adriano, D.C., 2001. Trace Elements in Terrestrial Environments: Biogeochemistry, Bio-
availability, Risk of Metals. second ed. Springer-Verlag, New York.
Adriano, D.C., Wenzel, W.W., Vangronsveld, J., Bolan, N.S., 2004. Role of assisted natural
remediation in environmental cleanup. Geoderma 122, 121–142.
Avena, M.J., Koopal, L.K., 1998. Desorption of humic acids from an iron oxide surface. En-
viron. Sci. Technol. 32, 2572–2577.
Bacon, J.R., Davidson, C.M., 2008. Is there a future for sequential chemical extraction? An-
alyst 133, 25–46.
Bardos, P., Bone, B., Cernik, M., Elliott, D.W., Jones, S., Merly, C., 2015. Nanoremediation
and international environmental restoration markets. Remediat. J. 25, 83–94.
Boente, C., Sierra, C., Martinez-Blanco, D., Menendez-Aguado, J.M., Gallego, J.R., 2018.
Nanoscale zero-valent iron-assisted soil washing for the removal of potentially
toxic elements. J. Hazard. Mater. 350, 55–65.
Bolan, N., Kunhikrishnan, A., Thangarajan, R., Kumpiene, J., Park, J., Makino, T., et al., 2014.
Remediation of heavy metal(loid)s contaminated soils - to mobilize or to immobi-
lize? J. Hazard. Mater. 266, 141–166.
BOPA, Boletín Oficial del Principado de Asturias, April 21, 2014. Generic reference levels
for heavy metals in soils from principality of Asturias, Spain, vol. 91. https://sede.as-
turias.es/bopa/2014/04/21/2014-06617.pdf (accessed March 2019).
Cao, Y., Zhang, S., Zhong, Q., Wang, G., Xu, X., Li, T., et al., 2018. Feasibility of nanoscale
zero-valent iron to enhance the removal efficiencies of heavy metals from polluted
soils by organic acids. Ecotoxicol. Environ. Saf. 162, 464–473.
Crommentuijn, T., Sijm, D., de Bruijn, J., van den Hoop, M., van Leeuwen, K., van de
Plassche, E., 2000. Maximum permissible and negligible concentrations for metals
and metalloids in the Netherlands, taking into account background concentrations.
J. Environ. Manag. 60, 121–143.
EEA, 2014. Soil contamination widespread in Europe. https://www.eea.europa.eu/high-
lights/soil-contamination-widespread-in-europe, Accessed date: March 2019.
Fang, J., Wen, B., Shan, X.-Q., Lin, J.-M., Owens, G., 2007. Is an adjusted rhizosphere-based
method valid for field assessment of metal phytoavailability? Application to non-
contaminated soils. Environ. Pollut. 150, 209–217.
Fernandez-Martinez, R., Loredo, J., Ordonezb, A., Rucandio, I., 2014. Mercury availability
by operationally defined fractionation in granulometric distributions of soils and
mine wastes from an abandoned cinnabar mine. Environ. Sci.: Processes Impacts
16, 1069–1075.
Gallego, J.R., Esquinas, N., Rodriguez-Valdes, E., Menendez-Aguado, J.M., Sierra, C.,
2015. Comprehensive waste characterization and organic pollution co-
occurrence in a Hg and As mining and metallurgy brownfield. J. Hazard. Mater.
300, 561–571.
Gil-Diaz, M., Alonso, J., Rodriguez-Valdes, E., Pinilla, P., Lobo, M.C., 2014a. Reducing the
mobility of arsenic in brownfield soil using stabilised zero-valent iron nanoparticles.
J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 49, 1361–1369.
Gil-Diaz, M., Perez-Sanz, A., Vicente, M.A., Lobo, M.C., 2014b. Immobilisation of Pb and Zn
in soils using stabilised zero-valent iron nanoparticles: effects on soil properties.
Clean: Soil, Air, Water 42, 1776–1784.
Gil-Diaz, M., Diez-Pascual, S., Gonzalez, A., Alonso, J., Rodriguez-Valdes, E., Gallego, J.R., et
al., 2016. A nanoremediation strategy for the recovery of an As-polluted soil.
Chemosphere 149, 137–145.
Gil-Diaz, M., Alonso, J., Rodriguez-Valdes, E., Gallego, J.R., Lobo, M.C., 2017a. Comparing
different commercial zero valent iron nanoparticles to immobilize As and Hg in
brownfield soil. Sci. Total Environ. 584, 1324–1332.
Gil-Diaz, M., Pinilla, P., Alonso, J., Lobo, M.C., 2017b. Viability of a nanoremediation process
in single or multi-metal(loid) contaminated soils. J. Hazard. Mater. 321, 812–819.
Gong, X.M., Huang, D.L., Liu, Y.G., Zeng, G.M., Wang, R.Z., Wan, J., et al., 2017. Stabilized
nanoscale zerovalent iron mediated cadmium accumulation and oxidative damage
of Boehmeria nivea (L.) Gaudich cultivated in cadmium contaminated sediments. En-
viron. Sci. Technol. 51, 11308–11316.
Gonzalez-Fernandez, B., Rodriguez-Valdes, E., Boente, C., Menendez-Casares, E.,
Fernandez-Brana, A., Gallego, J.R., 2018. Long-term ongoing impact of arsenic con-
tamination on the environmental compartments of a former mining-metallurgy
area. Sci. Total Environ. 610, 820–830.
Grafe, M., Eick, M.J., Grossl, P.R., 2001. Adsorption of arsenate (V) and arsenite (III) on goe-
thite in the presence and absence of dissolved organic carbon. Soil Sci. Soc. Am. J. 65,
1680–1687.
Grafe, M., Eick, M.J., Grossl, P.R., Saunders, A.M., 2002. Adsorption of arsenate and arsenite
on ferrihydrite in the presence and absence of dissolved organic carbon. J. Environ.
Qual. 31, 1115–1123.
Gustafsson, J.P., Bhattacharya, P., 2007. Geochemical modelling of arsenic adsorption to
oxide surfaces. In: Bhattacharya, P., Mukherjee, A.B., Bundschuh, J., Zevenhoven, R.,
Loeppert, R.H. (Eds.), Arsenic in Soil and Groundwater Environment: Biogeochemical
Interactions, Health Effects and Remediation. 9. Elsevier Science Bv, Amsterdam,
pp. 159–206.
Hammer, D., Keller, C., 2002. Changes in the rhizosphere of metal-accumulating plants ev-
idenced by chemical extractants. J. Environ. Qual. 31, 1561–1569.
Hopkins, J., Watts, P., Hosford, M., 2009. Contaminants in soil: updated collation of toxico-
logical data and intake values for humans. Inorganic Arsenic. Science Report:
SC050021/TOX 1. Environment Agency http://webarchive.nationalarchives.gov.uk/
20140328084622/http:/www.environment-agency.gov.uk/research/planning/
64002.aspx, Accessed date: March 2019.
Huang, D.L., Xue, W.J., Zeng, G.M., Wan, J., Chen, G.M., Huang, C., et al., 2016. Immobiliza-
tion of Cd in river sediments by sodium alginate modified nanoscale zero-valent iron:
impact on enzyme activities and microbial community diversity. Water Res. 106,
15–25.
Jing, Y.D., He, Z.L., Yang, X.E., Sun, C.Y., 2008. Evaluation of soil tests for plant-available
mercury in a soil-crop rotation system. Commun. Soil Sci. Plant Anal. 39, 3032–3046.
Kabata-Pendias, A., 2011. Trace Elements in Soils and Plants. 4th edn. CRC Taylor & Francis
Group, Boca Raton, FL, p. 2011.
Karn, B., Kuiken, T., Otto, M., 2009. Nanotechnology and in situ remediation: a review of
the benefits and potential risks. Environ. Health Perspect. 117, 1823–1831.
Klimkova, S., Cernik, M., Lacinova, L., Filip, J., Jancik, D., Zboril, R., 2011. Zero-valent iron
nanoparticles in treatment of acid mine water from in situ uranium leaching.
Chemosphere 82, 1178–1184.
Kumpiene, J., Lagerkvist, A., Maurice, C., 2008. Stabilization of As, Cr, Cu, Pb and Zn in soil
using amendments - a review. Waste Manag. 28, 215–225.
Kumpiene, J., Giagnoni, L., Marschner, B., Denys, S., Mench, M., Adriaensen, K., et al., 2017.
Assessment of methods for determining bioavailability of trace elements in soils: a
review. Pedosphere 27, 389–406.
Lee, S.-H., Park, H., Koo, N., Hyun, S., Hwang, A., 2011. Evaluation of the effectiveness of
various amendments on trace metals stabilization by chemical and biological
methods. J. Hazard. Mater. 188, 44–51.
Li, S.L., Wang, W., Yan, W.L., Zhang, W.X., 2014. Nanoscale zero-valent iron (nZVI) for the
treatment of concentrated Cu(II) wastewater: a field demonstration. Environ. Sci.:
Processes Impacts 16, 524–533.
Li, S.L., Wang, W., Liang, F., Zhang, W.X., 2017. Heavy metal removal using nanoscale zero-
valent iron (nZVI): theory and application. J. Hazard. Mater. 322, 163–171.
Li, J., Wang, X.X., Zhao, G.X., Chen, C.L., Chai, Z.F., Alsaedi, A., et al., 2018. Metal-organic
framework-based materials: superior adsorbents for the capture of toxic and radioac-
tive metal ions. Chem. Soc. Rev. 47, 2322–2356.
Madhavi, V., Prasad, T., Reddy, B.R., Reddy, A.V.B., Gajulapalle, M., 2014. Conjunctive effect
of CMC-zero-valent iron nanoparticles and FYM in the remediation of chromium-
contaminated soils. Appl. Nanosci. 4, 477–484.
MAPA, 1994. Métodos Oficiales de Análisis. vol. III. Secretaría General Técnica Ministerio
de Agricultura, Pesca y Alimentación, Madrid, pp. 219–324.
MEF, Ministry of the Environment Finland, 2007. Government Decree on the Assessment
of Soil Contamination and Remediation Needs (214/2007, March 1, 2007).
Mueller, N.C., Braun, J., Bruns, J., Cernik, M., Rissing, P., Rickerby, D., et al., 2012. Applica-
tion of nanoscale zero valent iron (NZVI) for groundwater remediation in Europe. En-
viron. Sci. Pollut. Res. 19, 550–558.
O'Carroll, D., Sleep, B., Krol, M., Boparai, H., Kocur, C., 2013. Nanoscale zero valent iron and
bimetallic particles for contaminated site remediation. Adv. Water Resour. 51,
104–122.
Querol, X., Alastuey, A., Moreno, N., Alvarez-Ayuso, E., Garcia-Sanchez, A., Cama, J., et al.,
2006. Immobilization of heavy metals in polluted soils by the addition of zeolitic ma-
terial synthesized from coal fly ash. Chemosphere 62, 171–180.
Redman, A.D., Macalady, D.L., Ahmann, D., 2002. Natural organic matter affects arsenic
speciation and sorption onto hematite. Environ. Sci. Technol. 36, 2889–2896.
Rumayor, M., Gallego, J.R., Rodriguez-Valdes, E., Diaz-Somoano, M., 2017. An assessment
of the environmental fate of mercury species in highly polluted brownfields by
means of thermal desorption. J. Hazard. Mater. 325, 1–7.
Sadiq, M., 1997. Arsenic chemistry in soils: an overview of thermodynamic predictions
and field observations. Water Air Soil Pollut. 93, 117–136.
 M. Gil-Díaz et al. / Science of the Total Environment 675 (2019) 165–175
Science Communication Unit, University of the West of England, Bristol, September 2013.
Science for Environment Policy In-depth Report: Soil Contamination: Impacts on
Human Health. Report Produced for the European Commission DG Environment.
http://ec.europa.eu/science-environment-policy, Accessed date: March 2019.
Singh, R., Misra, V., Singh, R.P., 2012. Removal of Cr(VI) by nanoscale zero-valent iron
(nZVI) from soil contaminated with tannery wastes. Bull. Environ. Contam. Toxicol.
88, 210–214.
Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential extraction procedure for the spe-
ciation of particulate trace-metals. Anal. Chem. 51, 844–851.
Tosco, T., Papini, M.P., Viggi, C.C., Sethi, R., 2014. Nanoscale zerovalent iron particles for
groundwater remediation: a review. J. Clean. Prod. 77, 10–21.
USEPA, US Environmental Protection Agency, 1992. Method 1311: toxicity characteristic
leaching procedure. https://www.epa.gov/hw-sw846/sw-846-test-method-1311-
toxicity-characteristic-leaching-procedure, Accessed date: March 2019.
van Liedekerke, M., Prokop, G., Rabl-Berger, S., Kibblewhite, M., Louwagie, G., 2014. Prog-
ress in the Management of Contaminated Sites in Europe. Joint Research Centre, Re-
port EUR 26376 EN. https://doi.org/10.2788/4658 http://bookshop.europa.eu/en/
progress-in-the-management-of-contaminated-sites-in-europe-pbLBNA26376/,
Accessed date: March 2019.
Vangronsveld, J., Cunningham, S.D., 1998. Metal-contaminated soils. In Situ Inactivation
and Phytorestoration. Springer, New York, pp. 1–15.
Vitkova, M., Rakosova, S., Michalkova, Z., Komarek, M., 2017. Metal(loid)s behaviour in
soils amended with nano zero-valent iron as a function of pH and time. J. Environ.
Manag. 186, 268–276.
Wang, Y., Fang, Z., Kang, Y., Tsang, E.P., 2014. Immobilization and phytotoxicity of chro-
mium in contaminated soil remediated by CMC-stabilized nZVI. J. Hazard. Mater.
275, 230–237.
Wu, Y.H., Pang, H.W., Liu, Y., Wang, X.X., Yu, S.J., Fu, D., et al., 2019. Environmental reme-
diation of heavy metal ions by novel-nanomaterials: a review. Environ. Pollut. 246,
608–620.
175
Xu, J.Y., Bravo, A.G., Lagerkvist, A., Bertilsson, S., Sjoblom, R., Kumpiene, J., 2015. Sources
and remediation techniques for mercury contaminated soil. Environ. Int. 74, 42–53.
Xue, W.J., Huang, D.L., Zeng, G.M., Wan, J., Cheng, M., Zhang, C., et al., 2018a. Performance
and toxicity assessment of nanoscale zero valent iron particles in the remediation of
contaminated soil: a review. Chemosphere 210, 1145–1156.
Xue, W.J., Huang, D.L., Zeng, G.M., Wan, J., Zhang, C., Xu, R., et al., 2018b. Nanoscale zero-
valent iron coated with rhamnolipid as an effective stabilizer for immobilization of Cd
and Pb in river sediments. J. Hazard. Mater. 341, 381–389.
Xue, W.J., Peng, Z.W., Huang, D.L., Zeng, G.M., Wan, J., Xu, R., et al., 2018c.
Nanoremediation of cadmium contaminated river sediments: microbial response
and organic carbon changes. J. Hazard. Mater. 359, 290–299.
Ye, S.J., Zeng, G.M., Wu, H.P., Zhang, C., Dai, J., Liang, J., et al., 2017a. Biological technologies
for the remediation of co-contaminated soil. Crit. Rev. Biotechnol. 37, 1062–1076.
Ye, S.J., Zeng, G.M., Wu, H.P., Zhang, C., Liang, J., Dai, J., et al., 2017b. Co-occurrence and in-
teractions of pollutants, and their impacts on soil remediation-a review. Crit. Rev. En-
viron. Sci. Technol. 47, 1528–1553.
Yin, Y.J., Allen, H.E., Huang, C.P., Sanders, P.F., 1997. Adsorption/desorption isotherms of
Hg(II) by soil. Soil Sci. 162, 35–45.
Zhang, H., Selim, H.M., 2008. Reaction and transport of arsenic in soils: equilibrium and
kinetic modeling. In: Sparks, D.L. (Ed.), Advances in Agronomy. vol. 98. Elsevier Ac-
ademic Press Inc., San Diego, pp. 45–115 98.
Zhang, M.Y., Wang, Y., Zhao, D.Y., Pan, G., 2010. Immobilization of arsenic in soils by sta-
bilized nanoscale zero-valent iron, iron sulfide (FeS), and magnetite (Fe3O4) parti-
cles. Chin. Sci. Bull. 55, 365–372.
Zou, Y., Wang, X., Khan, A., Wang, P., Liu, Y., Alsaedi, A., et al., 2016. Environmental reme-
diation and application of nanoscale zero-valent iron and its composites for the re-
moval of heavy metal ions: a review. Environ. Sci. Technol. 50, 7290–7304.